key: cord-0025146-vdbc62bw
authors: nan
title: Proceedings of the World Molecular Imaging Congress 2020, October 7-9, 2020: General Abstracts
date: 2022-01-04
journal: Mol Imaging Biol
DOI: 10.1007/s11307-021-01691-0
sha: 3422ac0599c181d1ece68e27e74cdac5780ae3ca
doc_id: 25146
cord_uid: vdbc62bw

nan

also result in cancer cells escaping immune surveillance through increased expression of PD-L1. Metabolic characterization showed that cells with downregulated Chk-α levels shifted towards a more immunosuppressive profile through metabolic reprogramming, increasing the production of metabolites linked to decreased effectiveness of T-cells11. Our data also suggest that low levels of PD-L1 could skew cancer cells towards a more immunosuppressive profile by significant changes in the lipid profile12. These results provide new insights in the role of PD-L1 in the cancer metabolome that may be exploited to improve the outcome of treatment with immune checkpoint inhibitors. Acknowledgement: This work was supported by NIH R35CA209960 and R01CA82337. JPT was founded by Fundación Alonso Martín Escudero and MSCA. Figure 1 . Metabolic heat map generated from quantitative analysis of high-resolution 1H MR spectral data of the (a) aqueous and (b) lipid phase, displaying differences in the metabolic profile of MDA-MB-231 cells. The heat map displays metabolites from untreated cells, cells transfected with 100 nM luciferase siRNA for 48h, cells transfected with 100 nM Chk-a siRNA for 48h, cells transfected with 100 nM PD-L1 siRNA for 48h, and cells transfected with a mixture of 50 nM PD-L1 #1 and 50 nM Chk-a siRNA for 48h. Heat maps were created using MATLAB software (MATLAB R2012b, MathWorks) to visualize the metabolic patterns. Due to the high dynamic range of metabolites, we normalized the highest intensity of a metabolite in each of the four groups to 100%. This normalization provides a dynamic range between 0 -100%, allowing a better presentation of heat maps. The heat map represents the average of 3-6 replicates per group. The integral area under the peak was normalized to the number of cells for each sample. TSP dissolved in D2O was used as a quantitative reference in the spectral analysis. *p ≤ 0.05 **p ≤ 0.01 ***p ≤ 0.001 compared to untreated + cells transfected with luciferase siRNA cells. GPC: glycerophosphocholine, PC: phosphocholine, Cho: choline, GSH: glutathione, GSSG: oxidized glutathione, MTA: S-methyl-5′-thioadenosine. Lipids (-CH3): methyl groups of fatty acids, Lipids (-CH2-): methylene groups of fatty acids (truncated), OOC-CH2: methylene groups at the a position of the carboxylic function, OOC-CH2-CH2: methylene groups at the b position of the carboxylic function, ARA: arachidonic acid, EPA: eicosapentaenoic acid, PtdEA: phosphatidylethanolamine, PtdCholine: phosphatidylcholine, (CH=CH-CH2-CH=CH)n: diallylic methylene protons, CH=CH-CH2: methylene groups at the a position of a double bond, CH = CH: fatty acid double bonds. mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science, 2004 . 304(5676): p. 1497 -1500 , N. C.C.N. NCCN Clinical Practice Guidelines in Oncology. Non-small Cell Lung Cancer Version 2.2020 . [cited 2019 December 23, 2019 ; Available from: https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. 4. Taniguchi, K., et al., Intratumor heterogeneity of epidermal growth factor receptor mutations in lung cancer and its correlation to the response to gefitinib. Cancer science, 2008. 99 (5) Abstract Body : Introduction Stroke is responsible for over 5.5 million annual deaths worldwide [1] with 15-20% being attributable to the rupture of atherosclerotic plaques in the carotid artery [2] . Rupture is strongly correlated to the presence of intraplaque hemorrhage (IPH) sites inside these plaques [3] . As such, researchers have developed a technique that uses a T1-weighted gradient echo MRI sequence (Siemens MPRAGE) to detect IPH and identify those at risk of stroke [4] . While promising, this method cannot be widely implemented in clinical settings unless there exists a method of standardizing images acquired across a variety of MR scanner models. With currently available atherosclerosis phantoms failing to include IPH [5, 6] , the objective of this research was to synthesize two types of novel phantoms that replicate IPH and thus, by allowing for better image standardization, facilitate clinical IPH imaging. The first type of phantom was an IPH site size-assessment tool that modelled IPH sites of varying diameters, and the second consisted of anatomical replicas of IPH-positive plaques. Methods IPH sites appear as hyperintense regions on T1-weighted images as they possess short T1 relaxation times. As such, a material that mimics the imaging properties of IPH was created by adding GdCl3 to a 0.15%-alginate, 1.5%-agarose gel. Alginate (an anionic polymer) was added to the gel to prevent the diffusion of Gd3+ ions and thus preserve the signal intensity of IPH sites over time. To demonstrate the efficacy of this diffusion-prevention strategy, gels doped with different GdCl3 concentrations were created, embedded in an agar block, and their T1 relaxation times tracked over time. Once the IPH material was developed, the IPH site size-assessment phantom was fabricated. This was done by casting the IPH material into 1-8 mm diameter cylinders and embedding them in 2% agar. Lastly, the anatomical atherosclerosis phantoms were created by casting multi-component plaques, layer-by-layer, using 3D printed moulds. Each replica plaque possessed an atheroma cap made of 2% agar, a lipid core consisting of a 50% canola oil-water emulsion solidified with 1.5% agarose, and an IPH site. Once fabricated, plaques were slid inside a mock, 8 mm diameter, blood vessel (made of 2% agar). All phantoms were imaged using the MPRAGE pulse sequence. Results Results showed that gels made from the IPH-mimicking material (with varying concentrations of GdCl3) maintained their T1 relaxation times over time. Over a 14-day period, the average T1 relaxation time of these sites was found to decrease by 2.32%, a statistically insignificant margin. Interestingly, when alginate was not added to the IPH gel, a hypointense region appeared at the centre of the site, indicating diffusion of Gd3+ ions out of the site. The IPH-mimicking gel was then used to create a phantom that possessed IPH sites of varying diameters ( Figure A) . Each site mimicked the hyperintense signal of clinical IPH while also demonstrating statistically insignificant changes in signal intensity when stored for 7 days. Lastly, two anatomical atherosclerosis phantoms, each mimicking plaques with either a large or small IPH site (Figures B and C) , were made. Similar to clinical plaques, these phantoms possessed hyperintense IPH sites, medium-intensity vessel walls, and a hypointense lumen. delivered by tumor-feeding blood vessels and intra-tumoral accumulation. Primary organ distribution was found in the liver, spleen and lungs with low uptake in kidneys following tailvein injections, manifesting a hepatic-biliary clearance. FeS2-PEG8 displayed superior, stable and repetitive photothermal effects and photothermal therapy was successfully implemented in breast cancer xenograft models accordingly, resulting in statistically significant decreases of tumor burden. CONCLUSION Since breast tumors are most frequently located relatively superficially (esp. after breast compression during the imaging, which is performed routinely during mammography), our approach holds a realistic chance for clinical use. Its abundance, low toxicity and NIR light-harvesting capability have endowed FeS2 with translational opportunities after PEG modification, also because MSOT has been a readily-translatable imaging modality. Eradication of breast tumors without remnant cancer cells that may cause recurrence and metastatic disease may be achieved by post-surgical photothermal Reactive thiol group is then activated by 1,4-dithiothreitol (DTT) through S-S bond cleavage, followed by precipitation with diethyl ether, and then reacts with FeS2 in chloroform by forming new S-S bonds to achieve covalent functionalization. (B) Schematic illustration of the photovoltaic module consisting of FeS2 coating on a fluorine-doped tin oxide (FTO) glass substrate, a Ag/AgCl reference, and a Pt wire counter electrode in an alkaline electrolyte (pH = 13.5) with SO32−/S2O32− as the sacrificial agent. (C) Current-time characteristics of FeS2 nanocrystals on FTO under on/off cycles of a 808 nm NIR laser illumination at a constant bias of 0.1 V. (D) The optoacoustic spectrum of FeS2-PEG8 shares similar features with its absorption spectrum. Inset shows single-wavelength optoacoustic images of FeS2-PEG8. (E) Multispectral optoacoustic stability. FeS2-PEG8 is subject to continuous spectral sweeping from 700-900 nm by a pulse laser of multispectral optoacoustic tomography (MSOT). 78.9% of the original MSOT signal is retained post exposure. Inset shows corresponding images at various time points. (F) Xray diffraction patterns. 2θ peaks of FeS2 nanocrystals closely match the indexed planes of a FeS2 pyrite reference standard (ICSD 53935), while covalent conjugation of 8-arm-PEGs well preserves the single-phase pyrite crystal structure. Additionally, two distinct peaks around 27º corresponding to branched PEGs can be clearly observed only after PEGylation1. (G) FeS2-PEG8 shows strong optical absorption in the NIR region. Characteristics of the absorption spectrum for FeS2-PEG8 are almost identical in chloroform and aqueous solution, also resembling to the pristine FeS2 nanocrystals. (H) Detection of different concentrations of FeS2-PEG8 in an agarose phantom by MSOT. (I) Uptake of FeS2-PEG8 by MDA-MB-231 human breast cancer cells was confirmed by MOST imaging of cell phantoms. (J) Cellular uptake was further confirmed by Turnbull's Blue staining for intracellular Fe2+ species. FeS2-PEG8 is mainly localized both near the cell membrane and within the cytoplasm. Such endocytosis can be effectively blocked by pharmacological inhibitors of 5-(N-ethyl-N-isopropyl)-amiloride (EIPA) and wortmannin, indicating that PI3K-dependent macropinocytosis is one possible endocytic mechanism. (K) Specific MSOT signals of FeS2-PEG8 can be observed in the tumor and lungs post intravenous injeciton. Tumor feeding blood vessels also displayed increased signals after injection. (L) Ex vivo confirmation by histology. Turnbull's Blue staining with H&E as background contrast reveals distribution of FeS2-PEG8 in the tumor and the tumor-feeding blood vessels. H&E staining is included as reference. (M) Single-wavelength optoacoustic transverse images of a NU/J mouse 1 h following intravenous injection. Enhanced optoacoustic contrast is observed primarily in the liver and spleen with minor distribution in the lungs. (N) In vivo optoacoustic spectra of FeS2-PEG8 remain identical in distributed organs, also bearing resemblance to the in vitro spectrum. (O) MSOT imaging of a tissue mimicking phantom containing FeS2-PEG8 at different depths. The MSOT signals clearly decrease with increasing depths as expected, but positive signals can still be detected at a rather deep inclusion around 20 mm. (P) Optical spectra of major intrinsic photoabsorbers in biological tissues, oxygenated (HbO2) and deoxygenated (Hb) hemoglobins. The distinctly different exogenous spectrum of FeS2-PEG8 in the NIR imaging window can therefore be spectrally unmixed. (Q) In vivo MSOT imaging discloses FeS2-PEG8 is primarily accumulated in the liver, spleen and lungs, but not the kidneys, following intravenous injeciton. Therefore, the predominant pathway is through the hepatic-biliary clearance. time, which was unexpected as iron content dilutes with cell proliferation6, however, death of iron-labeled cells and subsequent cell uptake by tumour-associated macrophages or circulating tumour cells returning to the primary tumour is a possible explanation for this finding. Using MPI in conjunction with BLI provided valuable information on both the iron content and the viability of the labeled MFP tumours and revealed that our cell populations were viable and proliferating. The ability to employ dual modality imaging of PDX cells provides a powerful system to study models that better predict metastatic behaviour for identification of biomarkers and personalized therapies targeting breast cancer brain metastasis.

References: 1. DeRose, Y. S., Wang, G., Lin, Y. C., Bernard, P. S., Buys, S. S., Ebbert, M. T., ... & Neumayer, L. (2011) . Tumor grafts derived from women with breast cancer authentically reflect tumor pathology, growth, metastasis and disease outcomes. Nature medicine, 17(11), 1514. 2. Cassidy, J. W., Caldas, C., & Bruna, A. (2015) . Maintaining tumor heterogeneity in patient-derived tumor xenografts. Cancer research, 75(15), 2963-2968 3. Zhang, X., Claerhout, S., Pratt, A., Dobrolecki, L. E., Petrovic, I., Lai, Q., ... & Wong, H. (2013) . A renewable tissue resource of phenotypically stable, biologically and ethnically diverse, patient-derived human breast cancer xenograft models. Cancer research 4. Landis, M. D., Lehmann, B. D., Pietenpol, J. A., & Chang, J. C. (2013) . Patient-derived breast tumor xenografts facilitating personalized cancer therapy. Breast Cancer Research, 15(1), 1. 5. Reyal, F., Guyader, C., Decraene, C., Lucchesi, C., Auger, N., Assayag, F., ... & De Cremoux, P. (2012) . Molecular profiling of patient-derived breast cancer xenografts. Breast cancer research, 14(1), R11 6. Economopoulos, V., Chen, Y., McFadden, C., & Foster, P. J. (2013) . MRI detection of nonproliferative tumor cells in lymph node metastases using iron oxide particles in a mouse model of breast cancer. Translational oncology, 6(3), 347. Abstract Body : Introduction Liquid biopsies is a technique that involves the use of molecular biomarkers to characterize tumours.1 Focused ultrasound in conjunction with phospholipid microbubbles circulating in the bloodstream, can promote the release of molecular biomarkers from the brain into the bloodstream by locally disrupting the blood brain barrier in rodent models.2 However, as the concentration of biomarkers that are released at safe acoustic pressures is low, it can be challenging to detect these biomarkers upon dilution in circulation. 3 The aim of this study was to develop a system for encapsulating biomarkers at their source, prior to dilution, for their subsequent isolation and identification using echogenic microbubbles. Our hypothesis is that after exposure to focused ultrasound, phospholipid microbubble debris can spontaneously form liposomes that encapsulate local biomarkers in situ. By co-injecting fluorescent molecules with the microbubbles, the newly formed liposomes will also contain fluorescent molecules enabling identification and isolation with optical techniques after blood sampling. We evaluate the capture of a fluorescent identifier and a surrogate biomarker (i.e., DsRed2) after phospholipids microbubble exposure to focused ultrasound as a function of acoustic pressure in a flow channel. Methods Perfluorobutane (C4F10, Synquest Laboratories) microbubbles were synthesized in-house using the phospholipids DPPA, DPPC, and DPPE-mPEG5k (Avanti Polar Lipids). U87-RFP cells (Anticancer) were grown in DMEM with 10% fetal bovine serum (Wisent). To verify echogenicity, microbubbles (gas volume concentration = ~1.4x106 µm3/mL) were flowed through a tissue-mimicking flow phantom (8mm channel, Model 523A, ATS Laboratories-Phantoms) and imaged using an iU22 clinical ultrasound system (Philips) with an L9-3 transducer (MI = 0.1). Microbubbles under B-mode imaging were assessed in 20s segments every two minutes over 14-minute intervals and repeated at 0, 30, 60, and 90 minutes after the initial microbubble activation. Analysis was conducted using MATLAB (MathWorks). For cavitation experiments, microbubbles with DsRed2 as a surrogate protein biomarker and fluorescein as a fluorescent identifier were flowed through a 1.5% agarose ultrasound phantom with a 1.4mm channel. Focused ultrasound (1.15 MHz, duty cycle of 1%, 0-3.04 MPa) was applied with a transducer diameter of 75mm and a focal length of 60mm (FUS Instruments, Toronto, CAN) . Microbubbles and cavitation debris were separated and washed with phosphate buffered saline (pH 7.4) using centrifugation. Size distribution and concentration of isolated microbubbles and collected material post-cavitation were measured using Multisizer 4e (Beckman-Coulter) with 10 and 20μm apertures. Microscopy was conducted using an Axiovert 200M (Zeiss). Results Acoustically responsive perfluorobutane microbubbles were synthesized (concentration 8.2±0.7x109 microbubbles/mL, peak diameter of 1.1±0.1μm). Synthesized microbubbles retained 31±4% of their starting population over 55min in vitro and were verified to be echogenic through contrast ultrasound. Microbubble cavitation by focused ultrasound resulted in the formation of new particles, ranging in diameter from ~1000nm to below to limit of measurement (i.e., ±3nm. The concentration of the particles formed increased with focused ultrasound (2.50±0.42x107, 4.26±0.02x107, 4.78±0.06x107, and 4.6±0 .22x107 particles/mL, from pressures of 0, 1.06, 2.07, and 3.07 MPa, respectively) . Fluorescence microscopy showed that the particles formed after cavitation in the presence of both DsRed2 and fluorescein were able to successfully capture both molecules in comparison to control microbubbles (not exposed to focused ultrasound). Conclusions Synthesized phospholipid-stabilized microbubbles were echogenic and stable, similar to commercial ultrasound contrast agents. After microbubble exposure to focused ultrasound, new particles were formed however more acoustic pressure did not result in significantly more particles. These particles encapsulated local molecules from the microenvironment at pressures upwards of 1.06 MPa. In the future, the potential of this system to encapsulate biomarkers in vivo with fluorescent identifiers as a tool for cancer diagnosis treatment planning will be evaluated. (measured up to 55 min) . Microscope images of particles formed after cavitation of microbubbles at 3.07 MPa in the presence of fluorescent molecules. (C) Particles were presumed to be liposomes as they were characteristically difficult to observe under DIC conditions and were able to co-localize water-soluble fluorescent molecules as shown by (D) fluorescein inside the liposome aggregate was observed with fluorescence microscopy (E), along with the DsRed2 protein. Abstract Body : Colon cancer is one of the leading cancers with high morbidity and mortality, and early diagnosis of colon cancer still faces serious challenges to improve accuracy and specificity. At present, the most-common clinical diagnostic methods for colon cancer, including colonoscopy and biopsy, have non-negligible drawbacks and limitations, such as false positive or false negative results. In this work, we choose hydrogen sulfide (H2S) as a specific mark of colon tumor and develop a H2S-activated NIR-II-Emitting fluorescent probe, achieving a noninvasive diagnostic method with high sensitivity and specificity for colon cancer. Compared with current NIR fluorescent probes (700-900 nm), the second near-infrared wavelength window (NIR-II, 1,000-1,700 nm) fluorescence has low scattering and absorption in tissues, making it has higher resolution and sensitivity in vivo and more accurate for reflection of lesions. Among them, the lanthanide downconversion nanoparticals (DCNPs) have advantages of low toxicity, long stokes shift, the adjustable size and so on. As we know, Prussian blue (PB), a clinical drug approved by FDA, have great biosafety, compatibility and stable structure. More importantly, PB possesses strong NIR-II absorbance and can be decomposed by the redox reaction of Fe3+ and H2S in aqueous solution with a high sensitivity. Consequently, luminescence of PB-coated DCNPs can be quenched by Förster resonance energy transfer (FRET) and be recovered once H2S activation. Herein, we synthesized core-shell-structure lanthanide nanoparticles as luminescent materials, which can be excited by 808 nm laser to emit 1060 nm fluorescence. Then we developed a new method to grow PB on the surface of DCNP (Fig. 1a) . DCNP @ PB has the size with about 120 nm, and has a stable crystal structure. Further, the probe showed increased luminescence by 30 fold after H2S response (Figs. 1b and c) , and excellent imaging with high signal-to-noise (>10) in mice model of colon tumor ( Fig. d and e) . Thus, PB-coated DCNPs may be an ideal NIR-II imaging probe with high specificity and sensitivity for diagnosis and imaging-guided surgery in colon cancer.

References: [1] R. Labianca, et al. Colon Cancer. Crit. Rev. Oncol./Hematol. 2010, 74, 106-133 . Abstract Body : Zinc is one of the most essential and abundant microelements in the human body, required for hundreds of cellular processes and a cofactor for more than 300 enzymes (1) . The prostate gland contains the highest levels of zinc in the body and it has been shown that these levels are markedly reduced in prostate cancer while maintaining high levels in other pathologically benign prostatic conditions. Recently, we demonstrated that the prostate secretes high levels of zinc when stimulated by a bolus of glucose in fasted animals (2) . This mechanism allowed for extracellular molecular probes to detect secreted zinc in the prostate and thus detect malignant lesions by MRI. Here we designed a Mn-based zinc responsive MRI contrast agent to detect zinc secretion in the prostate by MRI. The mechanism of detection relies on the idea that a chelated paramagnetic ion, for example Mn(II), can undergo transmetallation by high concentrations of zinc. This zinc displacement would then result in an increase in transverse (r2) and longitudinal (r1) relaxivity rendering it more efficient at shortening the neighboring water protons longitudinal (R1) and transverse (R2) relaxation rates. And this in turn translates into an increase in signal in a T1-weighted MRI image ( Figure 1a ). Here, we took advantage of this displacement by choosing the appropriate chelate in order to obtain an optimal displacement rate to image zinc secretion in the prostate. MnEDTA is commercially available, low cost, and easy to manufacture. Because EDTA has a higher affinity for zinc than manganese, Mn(II) can be displaced when exposed to high concentrations of zinc (3) . Figure 1b shows the ∆R2 of MnEDTA and MnDOTA, which is a stable chelate unable to undergo transmetallation after zinc exposure and thus can be used as a negative control. As expected, ∆R2 of MnEDTA increased slowly after adding zinc while MnDOTA exhibited almost no change in R2 indicating that MnEDTA is better suited for imaging zinc secretion. Figure 1c shows the protocol for the in vivo MRI studies on healthy male mice. Briefly, we infused MnEDTA, and MnDOTA i.v for 20 minutes until the concentration of probe in the prostate reached a steady state, then we administered a bolus of glucose or saline i.p while continuously infusing Mn-probe for another 20 minutes at which time we stopped the infusion. Images were collected serially throughout the duration of the infusion (40 mins) and during an additional 20 minutes without infusion to monitor probe wash out. Figure 1d shows the change in prostate signal normalized to back muscle for both MnEDTA and MnDOTA after glucose or saline injections. It is evident that the prostate gland (more specifically the ventral lobe) is significantly more enhanced for the animals who received glucose and MnEDTA compared to MnDOTA. This indicates that the zinc secreted as a result of glucose stimulation displaces the Mn from EDTA but not from DOTA. Figure 1e shows the quantitative results where it is evident that in the ventral lobe only the MnEDTA plus glucose group shows significant increase in signal after ~30 minutes post glucose stimulation. In conclusion, these results suggest that by careful selection of a chelate, it is References: 1. Iwano, S., et al., Abstract Body : Introduction In preclinical research, non-invasive imaging is often used to observe physiological processes. For clinical applications most imaging devices are considered to be safe nowadays. However, in preclinical research necessary anesthesia, contrast agent application or the handling procedure can induce stress1 and, therefore, influence animal welfare and study results. Low-dose radiation (2 and affect animal behavior3. So far, no systematic study has been published, examining the effect of repeated CT measurements on animal welfare and study results in mice. Thus, we investigated the influence of standard contrast-enhanced FMT/CT and CT protocols on welfare and physiological parameters. Methods For this purpose, healthy BALB/C mice were scanned with FMT/CT (4 mGy) or CT (50 mGy) three times per week over a four weeks period. Imaging was performed under isoflurane anesthesia for 30 min. Since mice hair scatters light, animals that underwent optical imaging (FMT/CT) were repeatedly shaved before the measurements and ICG (120 nmol/kg BW) was used as contrast agent. Regular CT scans were performed with Imeron 400 (2 mg/kg BW). A Rotarod test and blood pressure measurements were performed twice per week to investigate behavioral alterations. The general condition of the animals was evaluated daily using a score sheet. For the investigation of physiological parameters, blood was taken for hemograms and organs like liver, spleen, lung, heart and kidneys were removed for histological examinations. Results/ Discussion The investigations have shown that even after repeated contrast-enhanced FMT/CT and CT measurements neither the Rotarod tests nor the blood pressure measurements showed any alterations. However, according to the score sheets, the burden of the animals measured with FMT/CT (4 mGy) exceeded a medium range after first time shaving, which can be attributed to a reversible weight loss after the procedure. Furthermore, alterations of physiological parameters were observed. Mice imaged by CT (50 mGy) showed lower liver, spleen and kidney weights than control animals. Moreover, a reduction of the white blood cell count and an increase of leukocyte infiltration in liver and spleen were observed in FMT/CT and CT imaged mice (figure 1) which are in line with the literature4. Conclusion Based on the results the burden of the animals during CT imaging can be considered mild, whereas FMT/CT imaging led to a medium burden. The observed alterations in physiological parameters need to be further investigated. Therefore, DNA strand breakage and fecal corticosterone metabolites examinations are currently ongoing. Furthermore, healthy mice will be imaged without contrast agents to unravel the mechanisms behind the observations and distinguish between possible influences of the contrast agent and the radiation dosage. Moreover, the influence of CT measurements on study results will be investigated in 4T1-breast cancer bearing mice. Acknowledgement This research was funded by the Deutsche Forschungsgesellschaft (DFG) within the framework of the research group FOR2591 (project number 321137804) and the Research Training Group 2375 "Tumortargeted Drug Delivery". Figure 1 : Influence of contrast-enhanced FMT/CT and CT imaging on physiological parameters and animal welfare. A: Repeated CT measurements did not increase the overall burden of the animals, but first time shaving for FMT/CT measurements led to a reversible decrease of body weight. B: Liver weight significantly decreased in the Imeron 400 CT group in comparison to the other investigated groups. C: Mice showed a lower spleen weight after repeated CT measurements with Imeron 400 contrast enhancement. D: The white blood cell count was slightly decreased in CT investigated mice in comparison to the control and isoflurane group. E: Leukocyte infiltration (CD45) in mice livers was significantly increased in CT and CT/FMT investigated mice in comparison to the control group. Abstract Body : Introduction: Synchronous bilateral breast cancer (SBBC) is the detection of malignant lesions in both breasts at the time of diagnosis or within 3-12 months. It occurs in approximately 2% of breast cancer patients1, and compared to women with unilateral breast cancer, SSBC patients have higher rates of metastasis and lower overall survival2,3. Studies show that the worse prognosis is not due to increased tumor aggressiveness, but due to the combined effect of two tumors resulting in a higher chance of metastasis3,4. However, little is known about the metastatic properties of cells from each tumor and whether metastatic cells from one tumor contribute to the growth of metastases from the other tumor. In this study the first mouse model of SBBC was developed by orthotopically implanting human breast cancer cells into bilateral mammary fat pads. The objective was to characterize the distribution of metastatic cells from each tumour using non-invasive dual-bioluminescence imaging (dual-BLI) with the highly sensitive reporters Akaluc and Antares2, as well as endpoint fluorescence microscopy. Methods: MDA-MB-231 human breast cancer cells were engineered with lentivirus to co-express either Antares2 and the fluorescence reporter zsGreen (zsG), or Akaluc and the fluorescence reporter tdTomato (tdT). Transduced cells were FACS sorted to obtain stable engineered cell populations. Female nod-scid-gamma (NSG) mice received orthotopic injections of 3x105 zsG/Antares2 cells and 3x105 tdT/Akaluc cells into contralateral fourth mammary fat pads (day 0; n=10). Antares2 and Akaluc dual-BLI was performed weekly for up to 6 weeks. Three mice were sacrificed on day 29 upon detection of both Antares2-and Akaluc-expressing lung metastasis. Dual-BLI was performed on the remaining seven mice until endpoint as determined by primary tumor necrosis (up to day 42). Mammary fat pad tumors and lungs were fixed, cryopreserved, and sectioned, and nuclei were stained with Hoechst. The cell composition of micrometastases in the lungs, identified as clusters of cells with a diameter greater than 200 µm, was analyzed using fluorescence microscopy. Results: Metastases were first detected in the lungs in six of the 10 mice on day 15 using Akaluc-BLI, with all mice showing Akaluc signal in the lungs by day 29. Antares2-BLI detected lung metastases in four of the 10 mice by day 21, and in six of seven mice by day 35 (Fig. 1A) . Fluorescence microscopy of the lungs of mice sacrificed on day 42 showed both zsG-and tdT-expressing micrometastases, with the majority (60%) of micrometastases composed of cells from both primary tumors (n=2; Fig. 1B /C) . Discussion: This study was the first to characterize metastatic SBBC in an animal model. The high proportion of metastases composed of cells from both primary tumors in preliminary results may be explained by metastatic cross-seeding, a phenomenon where circulating tumor cells that have disseminated from a primary tumor preferentially seed onto an established metastatic lesion. This process may contribute to tumour heterogeneity in SBBC patients with inter-tumoral molecular subtype discordance and lead to treatment resistance. These unexpected findings may explain the worse outcome of SBBC patients and should provide future insight on patient treatment and management. 

Biyue Zhu, Massachusetts General Hospital, 413897689@qq.com Category: Neuroscience Abstract Body : Bioluminescence imaging has changed daily practice in preclinical research of cancers and other diseases in the last decades; however, it has been rarely applied in preclinical research of Alzheimer's disease (AD) . In this report, we demonstrated that bioluminescence imaging could be used to report the levels of amyloid beta (Ab) species in vivo. We hypothesized that AkaLumine, a newly discovered substrate for luciferase, could bind to Ab aggregates and plaques. We further speculated that the Ab species have the reservoir capacity to sequester and release AkaLumine to control the bioluminescence intensity, which could be used to report the levels of Abs. Our hypotheses have been validated via in vitro tests, mimic phantom imaging, and in vivo imaging using transgenic AD mice that were virally transduced with aka Luciferase (AkaLuc), a new luciferase that generates bioluminescence in the near infrared window. As expected, compared to the control group, we observed that the Ab group showed lower bioluminescence intensity due to AkaLumine sequestering at early time points, while higher intensity due to AkaLumine releasing at later time points. Our approach can be easily implanted into daily imaging experiments. We believe that our method has tremendous potential to change daily practice of preclinical AD research. Genes Dev 17, 545-580 (2003) . 12. Hochgrafe, K. & Mandelkow, E.M. Making the brain glow: in vivo bioluminescence imaging to study neurodegeneration. Mol Neurobiol 47, 868-882 (2013) . 13. Dawson, T.M., Golde, T.E. & Lagier-Tourenne, C. Animal models of neurodegenerative diseases. Nature neuroscience 21, 1370-1379 (2018) . 14. King, A. The search for better animal models of Alzheimer's disease. Nature 559, S13-S15 (2018). Abstract Body : Objectives: PSMA has emerged as a relevant target for clinical prostate cancer (PC) imaging, although variability in its overexpression as it relates to response to androgen deprivation therapy (ADT), disease progression and genotypic signatures remain to be established. Therefore, assessing PC with additional metabolic imaging agents in the same patient would address the heterogeneity found in PC and may be informative in prognosis and treatment management. With the development of PC LuCaP patient derived xenograft (PDX) mouse models [1] and organoid cultures designed to preserve the genomic integrity of the PDX [2] , the in vitro screening of multiple imaging agents as it relates to disease diversity is possible. Herein, in vitro protocols were established using the PSMA PET agent [18F] DCFPyL (FDC) and metabolic PET agents [18F] Fluorodeoxyglucose (FDG), [18F] Fluoroglutamine (Glu), and [18F] Fluoroarachidonic Acid (ArA) in LuCaP organoids to characterize PSMA and its relationship with metabolic markers. Methods: FDC, Glu and ArA were synthesized following published methods [3] [4] [5] with radiochemical yields of 15-43% and purities >98%. PC3/PSMA+ [PSMA-transfected;+control) and LuCaP PDX models (4 adenocarcinomas [AD]; 1 neuroendocrine [NE]) were evaluated for PSMA with FDC in vivo and in vitro (binding assays [Bmax] ), in association with phenotype, metastatic location, androgen receptor (AR, immunohistochemistry [IHC]) [1] , and ADT response (Fig.1) . ADT response was determined by measuring tumor volumes (caliper) in PDX-bearing mice (control and ADT treated [Degarelix, 25mg/kg]) over a 3-week ADT paradigm (Fig.1) . Organoids of the same PDX were cultured (7 days) and evaluated for PSMA (FDC; Bmax) and metabolic agents (FDG, Glu, and ArA uptake; 2h; Fig.1) . Results: PSMA levels detected by FDC in PDXs and PDX-derived organoids were comparable and corresponded with in vivo PET images (Fig.1 ). Three ADs, LuCaP77, 141 and 136, exhibited high ADT sensitivity and intense AR IHC but varied in PSMA: LuCaP136 was 50-100 fold lower than LuCaP141 and 77. The ADT non-responsive AD LuCaP167 had high AR with the lowest PSMA, whereas the other ADT non-responder, NE LuCaP145.1, exhibited negligible AR but moderate PSMA levels. FDG uptakes of LuCaP77, 145.1, and 167 were 3-fold higher than LuCaP141 and 136, indicating non-correspondence to PSMA or ADT sensitivity. ArA uptakes were higher than FDG and Glu for all LuCaP PDXs, suggesting that PC organoids import and possibly metabolize more arachidonic acid compared to glucose and glutamine. Furthermore, the NE LuCaP145.1 exhibited higher FDG, Glu, and ArA uptake compared to AD LuCaP models, indicating that it may be more metabolically active. These results highlight the potential clinical value of combining FDC imaging with metabolic imaging as a means to characterize PC heterogeneity. Conclusions: The favorable comparison of PSMA levels between in vitro organoids and in vivo PDXs indicate that organoid models may be useful for the direct comparison of multiple PET agents towards the generation of a database to study PC biology.

The use of patient-derived genomically-heterogeneous models coupled with PET imaging is anticipated to provide insights into clinical PC. [168] [169] [170] [171] [172] [173] [174] [175] F., et al., fluoroarachidonic acid: tissue biodistribution and incorporation into phospholipids. Biol Pharm Bull, 1996. 19 (10) [PSMA-transfected;+control) and LuCaP PDX mouse models in vivo (A) reflect PSMA levels (Bmax) determined from in vitro FDC saturation binding assays using corresponding LuCaP PDX and organoid membrane preparations (fmoles PSMA/µg protein membrane prep) from either fresh/frozen excised whole mouse xenografts or PDX-derived organoids, respectively (B). An imaging agent grid was generated (B) detailing the PC model, original patient histology and metastatic location [1] , in vivo response to ADT (Degarelix) and AR protein levels determined by immunohistochemistry (AR IHC) in vivo [1] ; PSMA levels in PDXs and PDX-derived organoids were paired with metabolic PET agent uptake (% uptake per 106 live cells = [(cpmuptake) / (cpmtotal added) x (live cellstotal added)] x (106live cells) x 100) after 120 min of incubation in organoids. PC3/PSMA+ cells were maintained as a single cell suspension from 2D culturing for metabolic agent uptake assays. 

Lucero Aceves-Serrano, University of British Columbia, aceves.lucero@alumni.ubc.ca Category: Neuroscience Abstract Body : Introduction: Positron emission tomography tracers that target the translocator protein (TSPO) have become an essential tool in the assessment of neuroinflammation. However, tracer quantification can be difficult as it traditionally requires arterial cannulation. Thus, four alternative quantitative and semi-quantitative blood-free measurements of the TSPO tracer [11C]-PBR28 binding were evaluated and compared to the standard plasma input measurements. Methods: Thirteen [11C]-PBR28 scans from rhesus monkeys were acquired along with metabolite and plasma activity used to create arterial input functions (AIF). Total distribution volume (VT) and non-displaceable binding potential (BPND) were quantified with AIF (VTAIF, BPNDAIF) (1) . For BPNDAIF, activity in white matter (WM) was used to estimate VND. Four additional metrics were computed: VT and BPND using a population-derived metabolitecorrected plasma input function (VTPBIF, BPNDPBIF), standard uptake value (SUV), SUV ratio to WM (SUVr), and BPND using WM time activity curve as tissue input function (BPNDWM) (2) using (i)the 20-50 min and the (ii) 30-70 minutes post-injection time frames. 4 animals were scanned twice, providing test-retest reproducibility data. Results: 20-50 min postinjection time frames; Parent and metabolite fractions were similar in arterial (N=9) and venous samples (N=3). There was good reproducibility between the test and retest scans for BPNDAIF (ICC=0.72) BPNDPBIF (ICC=0.69) SUVr (ICC=0.85) and BPNDWM (ICC=0.86) but not for SUV (ICC=0.22). There was a good correlation between the VTAIF and VTPBIF within individual animals (r=0.98±0.01) but not across animals (r=0.52) (Fig. 1) . Correlation between the BPNDAIF and BPNDPBIF was good both within (r=0.98±0.01) and across animals (r=0.97). Furthermore, the correlation was moderate for SUV, SUVr, and BPNDWM within animals (r=0.8±0.09), moderate for BPNDWM (r = 0.71) across animals and poor for SUV and SUVr across animals ( r=0.21). 30-70 min time frames; BPNDAIF, BPNDPBIF, and SUV showed increased variability compared to using the 20-50 min time frame suggesting that some [11C]-PBR28 metabolite(s) can cross the blood-brain barrier, as their effect would be most impactful on data acquired later in the scan . Conclusion: In monkeys, the stability of peripheral metabolism permits the use of a PBIF to measure BPND. The high correlation between BPNDAIF and BPNDPBIF and the good test-retest value suggests that a population-derived curve may be a good alternative to individual sampling, as previously shown in humans (3) . Furthermore, these results suggest that using earlier (20-50 min) time frames might yield more reliable measurements than using later (30-70 min) time frames. Acknowledgments: The authors would like to acknowledge TRIUMF for providing the radiotracer, and Brain Canada for funding this project.

References: 1. J. Logan et al., J. Cereb. Blood Flow Metab. 10, 740-747 (1990) . 2. J. Logan et al., J. Cereb. Blood Flow Metab. 16, 834-840 (1996) Abstract Body : Differentiating amyloid beta (Ab) subspecies Ab40 and Ab42 has long been considered as an impossible mission with small-molecule probes. In this report, based on recently published structures of Ab fibrils, we designed iminocoumarin-thiazole (ICT) fluorescence probes to differentiate Ab40 and Ab42, among which Ab42 has much higher neurotoxicity. We demonstrated that ICTAD-1 robustly responds to Ab fibrils, evidenced by turn-on fluorescence intensity and red-shifting of emission peaks. Remarkably, ICTAD-1 showed different spectra towards Ab40 and Ab42 fibrils. In vitro results demonstrated that ICTAD-1 could be used to differentiate Ab40/42 in solutions. Moreover, our data revealed that ICTAD-1 could be used to separate Ab40/42 components in plaques of AD mouse brain slides. In addition, two-photon imaging suggested that ICTAD-1 was able to cross the BBB and label plaques in vivo. Interestingly, we observed that ICTAD-1 was specific toward plaques, but not cerebral amyloid angiopathy (CAA) on brain blood vessels. Given Ab40 and Ab42 species have significant differences of neurotoxicity, we believe that ICTAD-1 can be used as an important tool for basic studies and has the potential to provide a better diagnosis in the future. Ann Neurol. 1999, 46 (6) , 860-6. 4 . Lue, L. F., Kuo, Y.M., Roher, A.E., Brachova, L., Shen, Y., Sue, L., Beach, T., Kurth, J.H., Rydel, R.E., Rogers, J., Soluble amyloid beta peptide concentration as a predictor of synaptic change in Alzheimer's disease. Am J Pathol. 1999, 155 (3) , 853-62. 5. Terry, R., Masliah E, Salmon DP, Butters N, DeTeresa R, Hill R, Hansen LA, Katzman R., Physical basis of cognitive alterations in Alzheimer's disease: synapse loss is the major correlate of cognitive impairment. Ann Neurol. 1991, 30 (4) , 572-80. 6 Abstract Body : Purpose/Objectives: Magnetic Resonance Image guided stereotactic body radiotherapy (MRgRT) is an emerging technology increasingly utilized in localized pancreatic adenocarcinoma (PDAC) [1] . Given the variable response rates and short progression times in PDAC, there is an unmet clinical need for a method to assess early RT response and ultimately predict survival that may allow us to better personalize therapy. We hypothesize that quantitative image feature analysis (radiomics) of the daily MR scans before and during MRgRT may be used to extract information related to early treatment response. Materials/Methods: 9 standard histogram [2] radiomic features were extracted from MRgRT images [3] (MRidian System, ViewRay, 0.35T, TRUFI sequence, 1.5mm/3mm voxel size/slice thickness) from n=19 locally advanced and borderline resectable patients, scanned at simulation (SIM) ~14 days before SBRT start, and prior to each of 5 daily fractions (F1-F5) of 10Gy to Gross Tumor Volume (GTV). 9 patients had CA19-9 serum levels recorded pre/post induction chemotherapy (IC) pre RT (median days IC end to RT start:17). They were stratified for IC response by the CA19-9 [4] ratio post/pre IC (good if median). Normal tissue region (ROI_tis) was defined as whole abdomen minus GTV in 10 slices around the tumor center (Fig.1A) . Spatial stability was quantified as Kendall's Concordance Coefficient (CC) between feature values when GTV Region of Interest (ROI) was moved +/-1.5mm in x and y. Temporal image stability was quantified as coefficient of variation (CoV) of mean ROI_tis signal values SIM-F5. Significance of changes SIM-F1 and F1-F5 were assessed using two-sided paired t-test. Results: All considered features apart from min, max and kurtosis showed very good spatial stability with CC>0.95. Mean signal intensity ratio F1/SIM in GTV correlates strongly with corresponding changes in ROI_tis (Pearson r=0.87, p Suppl.Tab.1, Fig.1B) . No trends were seen in ROI_tis (IQR p=0.50, SD p=0.48) . Patients with good biochemical response to IC [5] (post/pre CA19-9 ratio Fig.1C) , consistently with the observed heterogeneity drop after XRT. Conclusion: With MRgRT traditionally only used for spatial dose planning, these results present a novel framework for MRgRT image quantification using histogram analysis. Efficient normalization method was found to remove technical value drift and improve sensitivity. Significant changes in heterogeneity metrics were observed prior and after MRgRT, found to correlate with tumor response to induction chemotherapy, suggesting that our methods can be used for surrogate treatment response quantification. More work is required to understand the biological basis of the observed trends.

References: 1. Rudra S, Jiang N, Rosenberg SA, Olsen JR, Roach MC, Wan L, et al. Using adaptive magnetic resonance image-guided radiation therapy for treatment of inoperable pancreatic cancer. Cancer Med 2019;8:2123-32 2. Just N. Improving tumour heterogeneity MRI assessment with histograms. Br J Cancer. 2014;111(12) : 2205-13. 3 . Mutic S, Dempsey JF. The ViewRay system: magnetic resonance-guided and controlled radiotherapy. Semin Radiat Oncol 2014; 24:196-9 4 . Hess V, Glimelius B, Grawe P, Dietrich D, Bodoky G, Ruhstaller T, tumour-marker response to chemotherapy in patients with advanced pancreatic cancer enrolled in a randomised controlled trial. Lancet Oncol. 2008; 9(2) :132-8. Abstract Body : BACKGROUND: External beam radiotherapy (XRT) is a widely used cancer treatment, yet responses vary dramatically between patients. In this work, a novel Magnetic Resonance Imaging (MRI) approach is presented, utilizing the histogram analysis of a clinicallyused T2 relaxation measurement to provide a quantitative measure of tumor heterogeneity [1] , indicative of early response to radiation, which can be used to adapt the treatment and maximize its effect. Relationship to treatment response, robustness, and biological underpinnings of the proposed biomarker candidate are shown. METHODS: MRI was used to measure the response to single dose 10Gy X-ray irradiation (XRAD 320, Precision X-Ray, CT) in two murine subcutaneous models of pancreatic ductal adenocarcinoma (Panc02, n=11 and BXPC3, n=8). T2 relaxation mapping (MSME, TR=3427ms, 32xTE=7-224ms, 128x128 points) was performed on a 7T preclinical MRI (Bruker) every 72h after treatment, and quantitative histogram analysis was used to identify most sensitive metrics of response. H&E stained sections were processed in Tissue Studio (Definiens) and MATLAB to derive local cellular density maps at resolution matching MRI, approximated as the nuclear fraction, a pixel area fraction occupied by nuclei. RESULTS: 72h after XRT clear changes in tumor T2 maps were observed, apparent in histogram analysis (Fig.1 ). Interquartile range (IQR) of T2 showed excellent sensitivity for detection of changes due to early XRT response in both tumor types, compared to other histogram metrics (Tab.1). Non-irradiated tumors (n=6 Panc02) showed no change in the metrics. Histogram analysis of local cell density maps and T2 MRI maps showed matching IQR differences between the tumor types with elevated IQR in the heterogeneous BXPC3 tumors (Suppl. Fig.1 ). Additionally, histological co-registration showed higher IQR in areas of necrosis (7.4±1.1ms vs. 4.0±0.4ms, p=0.025) suggesting that the changes are driven by radiation induced cell death. Importantly, a significant correlation was shown between early IQR change and tumor volume change ratio at the point of regrowth (Pearson r=-0.67, p=0.035 Panc02, r=-0.70, p=0.036 BXPC3, Suppl. Fig.2 ), illustrating the utility of the metric for prediction of further tumor response. Furthermore, the IQR showed excellent test-retest agreement, superior to other metrics (Tab.1, last column). CONCLUSION: The study presents a novel, robust method and an associated imaging biomarker for measurement of early radiotherapy response in murine models of pancreatic cancer, based on readily translatable MRI T2 mapping. Histological comparison reveals the biological underpinnings of the metric, showing that T2 IQR is sensitive to cellular heterogeneity, strongly affected by necrosis as caused by irradiation [2] . Importantly, the predictive power of the IQR for tumor response was shown. This approach can be directly validated clinically thanks to the advent of MRI-guided radiotherapy [3, 4] . Further, the method can be used for patient stratification, with escalation of radiation dose when a poor response to standard therapy protocol is predicted by the model. Abstract Body : Background. One of the most exciting uses of 18F-FDG PET in cancer is as an early functional biomarker of drug efficacy1. However, identifying for which therapies and in which cancers 18F-FDG PET can perform this function remains a significant challenge. We have an incomplete understanding of how cancers regulate glucose consumption and only lowthroughput methods for measuring this key pathway. High-throughput screens are powerful tools to identify new regulators of a pathway. Objective. To develop and validate a highthroughput assay for measuring glucose consumption, and to conduct a high-throughput screen to identify new compounds that regulate glucose consumption and for which 18F-FDG PET could function as a biomarker of efficacy in lung cancer. Methods. We adapted and optimized a luminescence assay for measuring glucose consumption to achieve the necessary properties to use in a high-throughput screen. We screened 3555 clinically relevant compounds against three genetically distinct lung cancer cell lines (H460, HCC827, and A549) and measured changes in glucose consumption using our high-throughput assay. We validated 15 of the compounds in cell culture by measuring IC50 values and further validated one compound, Milciclib, in vivo in H460 cells using 18F-FDG PET. We overexpressed GLUT1 in H460 cells to evaluate whether blocking glucose consumption was necessary for Milciclib to block cell growth. Results. We identified conditions that yielded a high-throughput assay for measuring glucose consumption with a Z-factor of 0.59 ± 0.05, indicating a large dynamic range and low variability. We discovered over 100 clinically relevant compounds across three genetically distinct lung cancer cell lines that blocked glucose consumption by >50%. Surprisingly, very few of the compounds overlapped between the different cell lines, suggesting that the genetic background of a cancer strongly determines the mechanisms used to regulate glucose consumption. The most potent inhibitors of glucose consumption include topoisomerase inhibitors such as camptothecin (IC50: 464 and 98 nM in HCC827 and H460 cells, respectively) and the tyrosine kinase inhibitor dovitinib (IC50: 554 and 1511 nM in HCC827 and H460 cells, respectively). Milciclib is a CDK and TRK inhibitor in clinical trials. Milciclib inhibits glucose consumption in H460 cells in culture with an IC50 of 1.5 mM by decreasing GLUT1 levels and decreases 18F-FDG consumption in H460 xenografts by 30.4 ± 2.5%. GLUT1 overexpression blocked Milciclib from inhibiting glucose consumption and inhibiting cell growth, suggesting that blocking glucose consumption is a major mechanism through which Milciclib limits cell growth2. Innovation. We developed a new method to rapidly identify compounds that block cancer cell glucose consumption and for which 18F-FDG PET could be used an early functional biomarker. Impact. We anticipate that this method will enable studies that expand the use of 18F-FDG PET and suggest new therapies for which 18F-FDG PET could function as a biomarker of drug efficacy in cancer.

Evaluation of calcification in the heart of partially nephrectomized mice by computed tomography and 18F-NaF positron emission tomography Rira Watanabe, Daiichi Sankyo, watanabe.rira.vi@daiichisankyo.co.jp

Abstract Body : Coronary artery calcification is routinely evaluated by an unenhanced low-dose CT scan, while sodium fluoride-18 (18F-NaF) positron emission tomography (PET), which can highlight active mineralization, are getting more popular and applied in the clinic1). In the course of drug discovery and development, preclinical CT/18F-NaF PET study is demanded before clinical implementation to ensure the usefulness of conducting the imaging studies in the clinic especially for evaluation of drug efficacy with consideration for the mechanism of action of the drug. In this study, we demonstrated usefulness of micro CT/18F-NaF PET for calcification in the heart in a nephrectomized animal model. Five/six nephrectomized DBA/2J mice (Nx mice) on high-phosphorus and calcium diet (acceleration diet) was used since they develop severe coronary artery calcification. Each 4 Nx mice were on the acceleration diet for 1, 2, 3 or 14 weeks and each 2 sham-operation mice were fed standard-chow for 3 or 14 weeks. Micro-CT images of the chest were taken with cardiac synchronization and calcification volume of the heart was calculated by imaging analysis. The day after CT scanning, 18F-NaF was injected to all mice. 18F-NaF uptake was assessed by 10-minute 2D-PET scanning followed by dissection of the heart for counting radioactivity accumulated. The dissected hearts were then dried completely and analyzed for calcium content by colorimetric analysis. Results showed calcium content increased over the course of the study. Micro-CT and 18F-NaF 2D-PET imaging were able to visualize heart calcification. Micro-CT demonstrated mild to severe calcification in the heart of Nx mice after 2, 3 or 14 weeks of the acceleration diet-loading, while calcification was not apparent in the heart of sham-operation mice and Nx mice after 1 week of the diet-loading. 18F-NaF 2D-PET imaging also showed higher signals around the heart of Nx mice than shamoperation mice, however, delineation of the heart was difficult because of the limitation of 2D-PET: overlay of the physiological accumulation of 18F-NaF on the bone and low resolution. Calcification volume measured by micro-CT and 18F-NaF accumulation in the heart correlated well with calcium content respectively (R = 0.90 and 0.95). However, limitation of detection by CT were suggested for small calcification at early phase of this animal model whereas high sensitivity and good linearity of 18F-NaF accumulation was indicated. In this model mice, calcium content in the heart was considerably variable among individuals from early to late phase, so development of heart calcification should be ideally assessed by individual basis. These in vivo imaging methods are considered useful for serial monitoring of progression of lesions because micro-CT is minimally invasive and can be applied consecutively. This imaging modality could be useful for allocation of individual model mice into groups to evaluate anticalcification efficacy of drugs with small number of animals. In assessment of mineralization, combination of quantitative micro-CT and 18F-NaF accumulation would be useful to assess early and late phase calcification and allow us to evaluate anti-calcification effect of compounds under development in preclinical, which eventually encourages clinical application of these imaging analyses. Abstract Body : Background: Cerenkov luminescence tomography (CLT) provides an effective way to harness the advantages of both nuclear imaging and optical imaging. The previous studies have shown that CLT is highly sensitive and timesaving for in vivo tumor detection. However, the accuracy of CLT is still limited, because the conventional mathematical functions are difficult to precisely describe the photon propagation in vivo. Therefore, the calculation error is inevitable. Herein, a machine learning method was developed to perform in vivo CLT. Using the ANN reconstruction method, the calculation error could be avoided and the reconstruction accuracy was able to be improved. Evaluation with Monte-Carlo simulation and glioma mouse models have shown that the ANN CLT approach owned significantly higher reconstruction accuracy compared with the conventional methods. Methods: As shown in Figure 1A , the developed ANN for CLT contained five different layers. From left to right, the first one was the input layer, and then there were three hidden layers followed by the output layer. For a welltrained ANN CLT model, the reconstructed Cerenkov luminescence sources could be generated when a surficial optical flux was given. Monte-Carlo simulation data were generated for the ANN to learn the relationship between detected signals and the actual photon sources. Approximately 6,000 Monte-Carlo CLI data were simulated, which covered nearly 500 different positions in the mouse brain. To take the simulation randomness into account, three different source shapes (ellipsoid, cylinder and cube) were used, and five repetitive simulations were conducted in each source position. Molecular optical simulation environment (MOSE) version 2.3 and a standard digital mouse model was applied to perform the simulation. The ANN CLT reconstruction performance was evaluated through photon sources in different numbers, depth and shapes. Conventional propagation model based reconstruction method was also applied for contrast. Moreover, orthotropic glioma mouse models (n=4) were constructed for in vivo validation. 4×106 U87-MG glioma cells mixed with PBS (10 μL) were injected in the mouse brains. 11C-methionine (11C-MET, 600 μCi, 100 μL) was injected through the tail vein and performed in vivo CLI after 20 min, followed by CT imaging. H&E staining results cryo-slices of the mouse head were obtained as the "golden standard" to validate the reconstruction accuracy. Results: Figure 1B illustrates the CLT results using the ANN CLT method. Explicit consistency can be observed between the simulated tumor tissues and the reconstructed results. Compared with the conventional CLT method, the ANN CLT results provided more accurate tumor reconstruction results, both in tumor positions and tumor shapes. These observations demonstrated the proposed reconstruction strategy was effective and promising. The improvements of the ANN CLT strategy were also be seen from the in vivo imaging on mice. The reconstruction accuracy was verified by the cryo-slice H&E results. Higher reconstruction accuracy was shown in the ANN CLT, compared with the conventional method. This outcome indicated that the proposed ANN CLT method was feasible for in vivo tumor imaging and was able to enhance the tumor diagnostic precision. Conclusion: In this study, a novel CLT reconstruction method based on ANN was developed, which could avoid the reconstruction errors generated from the calculation of photon propagation. The improvements of the ANN CLT were presented by both Monte-Carlo simulation and in vivo imaging on orthotropic glioma bearing mice. Compared with the conventional CLT method, the proposed ANN CLT reconstruct the inner photon sources with higher accuracy. The in vivo evaluation on mouse models further indicated that the ANN CLT effectively detected the inner tumor tissues. We believe this CLT method could promote the application of CLT, and boost the combination of artificial intelligence and molecular imaging.

References: [1] [1] . It is difficult to accurately characterize the MCI heterogeneity by using the current subtyping method [2] . Magnetic resonance imaging (MRI) combined with various machine learning algorithms provides tools to investigate alterations in brain structure and function of MCI patients [3] . Hence, this work aims to propose a new method for subtyping the MCI patients based on MRI image features. Figure 1 shows the overall design of the study. The data used in this study were approved by the ADNI database (http://adni.loni.usc.edu/). Standard T1-weighted imaging was obtained by 3D-MPRAGE with slightly different resolutions across patients. FreeSurfer was used to calculate seven kinds of features that reflect brain structure. Then, quantitative imaging features were clustered by using consensus clustering algorithm to identify and validated MCI subtypes in a 3T discovery cohort (n=295) and a heterogeneous 1.5T validation cohort (n=202). Four clusters were selected (cluster number k=4), and each cluster represented a MCI subtype. As shown in Figure p1A (supplemental data), the four-cluster solution had the smallest incremental changes in the area (delta area) under the cumulative distribution function (CDF) curves. The consensus matrices showed that the four-cluster solution had the maximum consensus within clusters and the minimum rate of ambiguity across 10,000 bootstraps. This resulted in 30 patients in subtype 1, 128 patients in subtype 2, 89 patients in subtype 3 and 48 patients in subtype 4 in the discovery cohort. In order to evaluate the stability and repeatability of the clustering model (Figure p1B ), we independently validated the same clustering process in the 1.5T validation cohort. As shown in Figure 2B , the optimal cluster size in this cohort was also four. To compare differences across the subtypes, we performed ANOVA and Kaplan-Meier survival analysis on the neuropsychological measures by pooling both cohorts together. As shown in Figure p2A in the ADAS11, ADAS13, FAQ and GDS, subtype 1 had the highest value, while subtype 3 had the lowest value. On the other hand, in the MMSE, ADNI_MEN, ADNI_EF, RAVLT and Logical Memory II tests, subtype 3 had the highest value, and subtype 1 had the lowest value. These results indicate that patients in subtype 1 have the most severe impairments in cognition, memory, and executive function, while the patients in subtype 3 had the least impairment at baseline diagnosis. The value of subtypes 2 and 4 on cognition, memory and executive function was between subtypes 1 and 3, showing moderate impairment. ADNI_EF can significantly distinguish the differences between subtypes in post hoc tests, suggesting that the differences between the subtypes may be more concentrated in executive functions. Furthermore, in line with previous results, conversion differences also existed across the subtypes (Figure p2B ). Patients in subtype 1 had a higher probability of converting to AD, and the rate of conversion was faster than in other subtypes. Most patients in subtype 3 were able to maintain stable MCI status. The other two subtypes had moderate conversion rates that did not significantly differ from each other. We speculate that the difference between subtype 2 and subtype 4 may be reflected in more detailed aspects, such as a specific behavioral domain, and it may take longer to observe the differences. For example, the survival curve showed that the survival difference between subtype 2 and subtype 4 in MCI status became larger after 100 months. This result requires more future exploration in neuropsychological and imaging association studies. Abstract Body : Introduction The amino acid transporter system xc-is over-expressed in multiple cancer types, providing intracellular cysteine for glutathione biosynthesis. System xcactivity in tumours is altered in response to chemotherapy and in drug-resistant cells, which can be imaged by PET using the L-glutamate analogue [18F] FSPG (1) (2) . Little is known regarding transporter-recognition of non-natural amino acids or their utility for cancer imaging. Motivated by this, we compared the in vitro and preclinical characteristics of [18F] FSPG to its stereoisomer [18F] FRPG in models of cancer and inflammation. Methods [18F] FRPG and [18F] FSPG uptake was assessed in H460 lung cancer cells (3 × 104 cells/well, 0.25 MBq; 1h) , with efflux measured 30 min after removal of exogenous activity. Specificity of [18F] FRPG for system xc-was further examined following transporter inhibition and blocking studies with system xc-substrates and inhibitor (1 mM) . Tissue uptake of [18F] FRPG and [18F] FSPG was quantified in mice bearing subcutaneous A549, H460, VCAP and PC3 tumours by gamma counting 60 min post-injection. Additionally, mice bearing A549 tumours were dynamically imaged by PET/CT over 60 min following injection of radiotracer (3 MBq) . Radiometabolite analysis was performed on tissue and blood samples following imaging. Lung inflammation was induced in BALB/c mice through intratracheal administration of lipopolysaccharide (LPS; 1.5mg/kg). [18F] FRPG and [18F] FSPG uptake in LPS-treated lungs were compared to PBS controls by PET, 24 h post induction of inflammation (3 MBq; . Finally, [18F] FRPG uptake was evaluated in an orthotopic H460 lung cancer model (3 MBq; . Results and discussion [18F] FRPG uptake was specific for the glutamate/cystine antiporter system xc- (Fig 1A) , matching that of [¹⁸F]FSPG (3) . In H460 cells, [18F] FRPG uptake was lower than [18F] FSPG over 60 min (7.9 ± 0.2% radioactivity/100,000 cells vs. 15.1 ± 0.7% radioactivity/100,000 cells; n = 3; p = 0.004; Fig 1B) . However, the percentage of radiotracer retained in cells 30 min after removal of exogenous activity was higher for [18F] FRPG than [18F] FSPG (91% vs. 69%; n = 3; p = 0.002; Fig 1C) . In vivo biodistribution studies showed rapid clearance of [18F] FRPG from the blood, liver and pancreas. Tumour uptake of [18F] FRPG was lower compared to [18F] FSPG in prostate VCAP tumours and lung A549 tumours ( Fig 2D&F) . However, [18F] FRPG tumour-to-blood was either elevated or equal to that of [18F] FSPG ( Fig 2E) . [18F] FRPG and [18F] FSPG showed in vivo metabolism to a single metabolite, with [18F] FRPG showing a higher percentage of parent tracer in tumours compared to [18F] FSPG (Sup data 1). Radiotracer uptake in inflammatory cells may complicate the interpretation of tumour uptake. However, [18F] FRPG uptake in orthotopic tumours was 2.5fold higher than inflamed tissue, allowing for excellent tumour visualisation (Sup data 2). Conclusion These results confirm the specificity of [18F] FRPG for system xc-and desirable properties for cancer imaging. High [18F] FRPG uptake was measured in a range of human tumours, with background tissue uptake reduced in comparison to [18F] FSPG. The uptake of [18F] FRPG in LPS-induced inflammation increased compared to healthy tissue, however, tumour uptake exceeded these values significantly. Future work will determine the redox sensitivity of [18F] FRPG and its ability to image response and resistance to therapy. [18F] FRPG in in vitro and in vivo cancer models. A. [18F] FRPG uptake in H460 cells was blocked in the presence of FSPG, L-glutamate and the system xc-inhibitor p-carboxyphenylglycine (CPG) (1 mM; 30 min uptake; n = 3). B. Time course of [18F] FRPG and [18F] FSPG uptake over 90 min in H460 cells (n = 3). C. [18F] FRPG and [18F] FSPG retention following 30 min efflux in media alone or with the addition of L-glutamate. D. Radiotracer uptake in a range of subcutaneous tumour models, 60 min post-injection. E. Tumour to blood ratios for [18F] FRPG and [18F] FSPG. F. PET/CT maximum intensity projection images representing the same mouse bearing an A549 s/c tumour imaged with [18F] FRPG and [18F] FSPG (40-60 min p.i.). Images were taken from the same animal 24 h apart. **, p < 0.01; ***, p < 0.001. L-Glu, L-glutamate; SG, salivary glands; P, pancreas; B, bladder. Abstract Body : Ever since PARP inhibitors emerged as reliable drugs for cancer treatment, huge effort was taken to broaden their use and develop PET imaging agents for detection and characterization of malignant lesions. Up until now, two gold-standard PARP radiotracers are applied in clinical settings, [18F] PARPi [1] and [18F] FTT [2] , while ongoing research focuses on optimized compounds and derivatives. Still, both tracers suffer from strict hepatobiliary clearance in rodents, limiting their use for abdominal lesions due to high background uptake. Herewith, we present an logP-optimized PARPi variant, [18F] FPyPARP (logP 2.49), with expected shift towards renal excretion, and the first side-by-side comparison of [18F] PARPi (logP 3.36) and [18F] FTT (logP 3.09) together with the novel [18F] FPyPARP in a standardized mouse study. [18F] FPyPARP was synthesized in a one-pot reaction using the synthon [18F] FPyTFP, in a synthesis time of 71±4 min with a decay-corrected radiochemical yield of 12±7 % and a molar radioactivity of 31±15 GBq/µmol (n=6). Additionally, the rather complicated [18F] PARPi synthesis was automated using [18F] SFB, while the original one-step [18F] FTT synthesis was simply adopted. HCC1937 cells, chosen because of their decent PARP expression, were subcutaneously injected into female NOD.CB17-Prkdcscid/J mice (n=5). After xenograft growth, the mice were injected with 13.0±0.5 MBq of the respective radiotracer and subjected to dynamic (n=2) or static (n=3) PET scans with anatomical MR imaging. Subsequent ex vivo biodistribution studies and analysis of the time-activity-curves indeed revealed clearly improved renal clearance as indicated by visible bladder uptake and comparable but heterogeneous radiotracer uptake in the xenograft. Nevertheless, all radiotracers showed high abdominal uptake especially in liver, spleen, intestine and kidney, indicating that the shift to renal excretion was only partial. The tumor-to-muscle ratios of the [18F] FPyPARP cohort (2.10±0.41) were significantly (p=0.0085) lower compared to [18F] PARPi (5.30±1.57) due to high muscle uptake. In contrast, the tumor-to-blood ratios were increased (2.33±1.47, 3.39±0.47 and 2.97±0.92, respectively, for [18F] PARPi, [18F] FPyPARP and [18F] FTT), suggesting good blood clearance and possibly lower unspecific background. In conclusion, we developed an alternative PARP radiotracer with a facilitated synthesis and validated it in a side-by-side comparison to the current benchmark tracers. Our tracer exhibits similar results in detecting PARP expression and met the main goal of this work by demonstrating a shift to partial renal clearance. Taking into account that first clinical data for [18F] PARPi show ~30% renal clearance in humans [3] , an even higher rate for [18F] FPyPARP is likely. This provides great potential for clinical PARP imaging of abdominal lesions and thus might aid developing new strategies in personalized cancer therapy.

Surachet Imlimthan, Inselspital, Bern University Hospital, surachet.imlimthan@extern.insel.ch Category: New Chemistry, Biology & Bioengineering Abstract Body : Theranostic nanoparticle-based drug delivery systems (DDS) allows the diagnosis and treatment of cancer using a single pharmaceutical platform.1 Non-invasive nuclear molecular imaging techniques, such as PET and SPECT have great utility to visualize and quantitatively determine the biological events in many diseases in vivo.2 Nanomaterials tagged with a suitable radioactive isotope can be traced after systemic administration to determine their biodistribution, specific uptake, elimination, and metabolism. Moreover, they can be used to detect and monitor pathological and molecular changes related to chemo-and radiotherapeutic delivery, providing the information as theranostic nanosystems in vivo. Nanocrystalline cellulose (CNC) is the most abundant organic polymer that has excellent physicochemical properties, including facile and cost-effective preparation, biodegradability, biocompatibility, and versatile chemical accessibility. CNC has a rod-like shape with the typical dimensions of 5-10 nm in width and 100-300 nm in length, which is suitable for nanoscale DDS applications.3 Previously, we reported the in vivo behavior of 111In-labeled CNC in both healthy and tumor-bearing animal models. The results showed that CNC had a transient uptake and retention in the lung, suggesting that they could be used to target the highly vascularized metastases through the EPR effect.4 If undetected early, melanoma can easily metastasize from the primary site to other organs within the body, such as lung, spleen, bone marrow, and lymph nodes. In melanoma, the activating BRAF V600E mutation is the most common oncogene expressed by the majority of the cells (> 90%).5 Recently, vemurafenib, a kinase inhibitor targeting BRAF V600E mutation, got FDA approval for metastatic melanoma treatment in humans. However, the therapeutic efficacy of vemurafenib chemotherapy is limited by drug pharmacokinetics in vivo and the development of resistance after repeated cycles of treatment.6 In this work, we have further developed theranostic CNC-based DDS targeting lung metastatic melanoma for potential synergy of concomitant chemo-and radiotherapy. First, we radiolabeled CNC with 177Lu, (t1/2 = 6.73 d) that emits both beta and gamma radiation, allowing both diagnosis (with SPECT) and radiotherapy. Then, vemurafenib was entrapped to 177Lu-labeled CNC surface in the presence of cationic polymer poly-L-lysine in a one-pot reaction (177Lu-CNC-V). We then evaluated the in vitro drug release profiles in physiological media, cytotoxicity, cellular uptake, and cell survival as well as the biodistribution of 177Lu-CNC-V in BRAF V600E mutation-harboring YUMM1.G1 lung metastatic melanoma murine model. Vemurafenib exhibited sustained release behavior under physiological conditions, suggesting the prolongation of drug half-life in vivo. The in vitro cell viability was decreased in murine YUMM1.G1 and human A375 melanoma (BRAF V600E positive) cell lines while there was no effect observed in murine B16-F10 skin melanoma (wild-type BRAF), demonstrating the specificity of the construct. Moreover, theranostic 177Lu-CNC-V demonstrated better cellular internalization compared to free 177Lu and 177Lu-CNC. The cell survival assay was carried out to confirm the synergistic therapeutic effects of vemurafenib chemotherapy and 177Lu radiotherapy in both YUMM1.G1 and A375 cell lines. The results revealed that cells treated with 177Lu-CNC-V had the lowest survival fraction (SF) after 24 hours of treatment, suggesting a superior therapeutic efficacy of 177Lu-CNC-V. Further, the ex vivo biodistribution results were in agreement with our previous studies where the radiolabeled CNC showed a transient accumulation in the lung before clearance through MPS organs. Radiation dosimetry estimation showed a high absorbed dose in the spleen after 72 h post-injection. Overall, the in vitro and in vivo results suggest the potential use of the developed theranostic CNC as single chemo-and radiotherapeutic DDS platform for enhanced targeting metastatic melanoma in the lung. A long-term survival study with the theranostic DDS in the YUMM1.G1 model is underway.

References: 1. Ahmed N., et al., Drug Discov. Today 2012 , 17(17-18), 928-934. 2. Kircher M. and Lapa C., Curr. Cardiovasc. Imaging Rep. 2017 , 10(2):6. 3. Sacui I. A., et al., ACS Appl. Mater. Int. 2014 , 6(9), 6127-6138. 4. Imlimthan S., et al., Biomacromolecules 2019 , 674-683. 5. Erdei E. and Torres S. M., Expert Rev. Anticancer Ther. 2010 , 10(11), 1811 -1823 . 6. Johannessen C. M., et al., Nature 2010 Abstract Body : Introduction Cardiac imaging technologies, including echocardiography and cardiac magnetic resonance imaging (cMRI), enable the non-invasive detection of changes in heart function that may indicate disease progression. These modalities detect changes in structure and anatomy, and thus there is a critical need for the detection of biochemical and molecular changes that precede gross structural changes. The development of imaging agents that target molecular changes, combined with hybrid imaging technology, will be a powerful means to address this need. There is evidence that the cardiac growth hormone secretagogue receptor (GHSR) could be a potential molecular imaging target2-4. An array of imaging agents are being characterized for detection of myocardial GHSR by positron emission tomography (PET) in vivo in preclinical studies. With one of these novel agents, termed 18F-LCE470, I will be using the emerging technology of hybrid PET and MRI (PET/MRI) to detect and quantify changes in regional distribution of myocardial GHSR in a canine model of heart failure. Methods Female hounds (11-12 months of age, n=4) were used in this study. A myocardial infarct (MI) was induced by 2-hour occlusion of the left descending coronary artery, followed by reperfusion. Dogs were imaged at specific timepoints: 18F-LCE470 for GHSR (baseline, days 3, 21, months 4 and 11 post-MI) and 13N-NH3 for perfusion (day 21, 4 and 11 months post-MI). For PET/MRI, dogs were injected with 100-150 MBq of the tracer, immediately followed by a 1hour or 30 min (13N-NH3) dynamic PET scan. Cine MRI maps were obtained to calculate heart function. The contractile function of the left and right ventricles was determined by CircleCVI software. The regional distribution of GHSR in the left ventricle, represented by the volume of distribution (DV) of 18F-LCE470, was determined by two-tissue compartmental modeling based on reversible binding of the tracer. Polar maps representing a 3D version of the entire left ventricle myocardium [Fig1] were segmented initially into 17 segments, which allowed for 3 regions of delineation: Infarct (apex and part of the left coronary artery [LAD] near apex); Remote (remaining LAD and right coronary artery); and territory served by the left circumflex artery (LCX). Remote and LCX regions were kept the same across all time points, but infarct regions were recalculated for each dog as infarct size was highly variable. Results PET/MR images and regional compartmental analysis showed differences in distribution patterns of 18F-LCE470 (DV, ml/ml) and 13N-NH3 (flow, ml/(ml*min)) [Fig1] . There was relatively low uptake of both tracers in the infarct region, and 13N-NH3 was taken up uniformly in noninfarcted tissue. In contrast, . Therefore, we can infer that 18F-LCE470 uptake represents GHSR density in each region of the left ventricle. In one dog with a large infarct ("Hope"), absolute GHSR density was significantly elevated in the LCX compared to the infarct and remote regions at all time points post-MI. In another dog with a small infarct ("Merry"), GHSR density was significantly elevated in LCX from infarct at all time points post-MI, and from remote regions at day 3 ( Fig 2B) . These changes in tracer uptake were seen in the absence of changes in the left ventricular and right ventricular ejection fraction [Fig 3] . Conclusions Quantitative compartmental analysis of a novel small molecule PET tracer, 18F-LCE470, revealed changes in regional distribution of GHSR in the canine heart after a heart attack. These results illustrate regional dynamics of GHSR post-MI and may indicate changes in GHSR signaling in the LCX region. This PET molecular imaging tool may potentially detect subclinical biochemical changes in myocardium prior to structural changes detected by conventional imaging methods used for diagnosis of heart disease. show uptake in the left ventricle. White arrow -infarct, blue arrows -remote tissues. Figure 2 . Representative graphs of volume of distribution (receptor density) for a large infarct ("Hope") and small infarct ("Sunny") showing significant differences between LCX to both remote and infarct regions at all time points post-MI with bigger differences seen in the larger infarct area. Colour Timepoints: Black -Baseline; Red -Day 3; Green -Day 21; Blue -Month 11 (*p Figure 3 . Heart function data using cine MRI images where there is no significant difference in the left ventricular or right ventricular ejection fractions at any timepoint.

A Novel Method for Optical Imaging of PET Isotopes in Biological Systems. PLoS One 5, e13300. 5. Arroyo, A.D., Guzmán, A.E., Kachur, A.V., et al. (2019) Abstract Body : In principle, Magnetic Particle Imaging (MPI), using superparamagnetic nanoparticles as an imaging tracer, is touted as quantitative and MPI signal can be equated to tracer quantity, given proper signal calibration. Yet, MPI signal properties have never been characterized for magnetic nanoparticles undergoing biodegradation, and deviations from the original calibration would impact quantification. We performed serial in vitro and in vivo MPI on various magnetic nanoparticles and measured MPI signal and nanoparticle resolution as a function of nanoparticle degradation and iron oxide dissolution. The findings of this work are critical to interpreting MPI data, particularly where long term biological processes are being studied. First, in Experiment 1, we measured change in total signal intensity, peak signal intensity and nanoparticle resolution of four magnetic nanoparticles with varying degradation rates (Table 1 ), in solutions that mimicked extracellular conditions (PBS, pH 7.4), harsh endosomal/lysosomal conditions (50 mM citrate buffer, pH 5.5) or mild endosomal/lysosomal conditions (10 mM citrate buffer, pH 6.5), during incubation at 37℃ for 4200 hours. The principle findings were: in 50 mM citrate, pH 5.5, particles exhibited varied reduction in total and peak MPI signal intensity and changes in nanoparticle resolution over 4200 hours. In 10 mM citrate, pH 6.5, significantly less decrease in signal intensities and changes in nanoparticle resolution were measured. In PBS, MPI signal and nanoparticle resolution remained constant for the first 1500 hours for all particles. Unpredictably, Vivotrax exhibited 30% increase in total signal intensity at 1500 hours, coincident with a drop in peak signal intensity of ~30%, and a doubling in nanoparticle resolution that was maintained through 4200 hours. Experiments where the dextran coating was enzymatically removed showed that this signal increase was resultant from nanocrystal aggregation. Enzymatically responsive MPI experiments are thus suggested. In general, for all four particle types, TEM confirms nanoparticle decomposition during the experiment, albeit to different degrees depending on the nanoparticle type and the incubation solution and the disappearance in signal intensity is well explained by nanoparticles decomposition and dissolution of the Fe3O4 nanocrystals in citrate at pH 5.5. To further probe MPI signal change during degradation, in Experiment 2, we measured MPI signal for PLGA encapsulated Fe3O4 nanoparticles (CLIPs) as a function of PLGA molecular weight and size of Fe3O4 nanocrystal during degradation. We show that signal can either increase, decrease or remain constant, depending on the MW of the PLGA polymer, the degradation rate encouraged by the buffer, and the size of the Fe3O4 nanocrystal. In general: 1) 8 nm Fe3O4 nanocrystals exhibited decrease in MPI signal when degrading in PBS and phosphate buffer, pH 6.5, and 25 nm Fe3O4 nanocrystals exhibited increase in MPI signal; 2) high MW PLGA slowed degradation rate versus 5 kDa PLGA. These signal changes are in line with our current understanding of magnetic Relaxometry and the transition from Neel relaxation to Brownian relaxation. Lastly, we performed in vivo MPI on mice injected with the fast degrading Vivotrax (n=2) and the slow degrading Magnefy (n=2) to determine how well the in vitro experiments predict in vivo biodegradation. Immediately after injection, mice had high total signal intensity at the location of the liver, which decreased over 3082 hours. Vivotrax exhibited slower MPI signal loss in vivo (31.7 d half-life) as compared to Vivotrax incubated in vitro in 50 mM citrate, pH 5.5 (13.2 d half-life Abstract Body : Background, Motivation and Objective Stage IV breast cancer, which has a high risk of invasion, often develops into metastases in distant organs, especially in the lung, and this could threaten the lives of women. [1] Traditional chemotherapeutics is the standard clinic treatment of stage IV breast cancer, while suffers from low delivery efficiency to the tumor site with significant variation among different patients. [2] Microtubules are key components of the cytoskeleton and play a crucial role in mitotic cell division, inhibited by vinca alkaloids, such as hydrophobic vinorelbine (VIN). [3] Moreover, biosynthesis of DNA can be inhibited by doxorubicin (DOX), a routinely used chemotherapeutic agent. [4] However, currently, owing to low penetration and limited distribution of them in tumor, the chemotherapy efficacy in metastatic lesion is low. [5] Therefore, a kind of site-specific new platforms that can shrink the primary tumor, and target metastases while minimizing unwanted distribution in normal tissues is urgently needed. Fibronectin, a class of adhesive glycoproteins, plays a major role in extracellular matrix functions of cancer cells such as cell adhesion, proliferation, and migration, highly expressed in the invasive or metastatic tumor sites. Herein, a cocktail targeted dual-acting micelle encapsulating doxorubicin and vinorelbine functionalized with CREKA peptides served to target the fibronectin-expressing tumor area was rationally fabricated, termed CTM. Statement of Contribution/Methods CREKA peptide and DSPE-PEG-MAL were mixed and stirred for 4h, formed DSPE-PEG-CREKA (DPC). Then the equal molar ratio of DOX and VIN were incorporated into DPC aqueous solution with acoustics, prepared CTMs respectively. Results/Discussion The as-prepared CTMs are ultra-small with diameters around 11.0 nm in uniformity shown in TEM image (Fig. 1a) , demonstrated that the efficient assembled CTMs have potential for targeted drug delivery to tumors. The fluorescence microscopy showed significant DOX uptake in cells treated with CTMs at 6 h, compared with low uptake in free chemotherapy agents group (FC, including free DOX and VIN in 1:1 molar ratio). (Fig. 1b) Due to high efficiency of CTMs uptake, the microtubule network of 4T1 cells was inhibited by CTMs treatment, while with moderate microtubule disruption in FC treatment. (Fig. 1c ) It could be illustrated that CTMs promoted the hydrophobic VIN delivery efficiency and made the significant effect. The 4T1 metastatic tumor mice model was built and evaluated the CTMs biodistribution and cocktail targeted treatment efficacy in vivo. Twenty-four hours after intravenous drug administration, CTMs were enriched in lung metastatic tumor lesion, compared to the ineffectiveness enrichment in FC group. (Fig. 1d) suggested that the simple small CTMs opened a new avenue for drug delivery in metastasis. Importantly, CTMs showed a significant high efficacy in suppressing metastatic tumors in vivo, compared to FC group at the end of the treatment. (Fig. 1e ) Meanwhile, the reduction in invasive metastatic foci was apparent in CTM-treated mice, which the CTM-treated residual lesion area was approximately one-third of that in FC group. (Fig. 1f ) Moreover, the CTM-treated micrographs showed neither noticeable damage nor inflammation in major organs, (Fig. 1g) suggesting CTMs had few pathological effects on normal tissues and biosafety in vivo. Conclusion In summary, the dual-acting CTM for cocktail targeted therapy was successfully prepared, which could overcome the shortcomings of traditional chemotherapy in metastasis. The aforementioned persuasive data shows that the CTM is a great promising further modality for high efficacy breast carcinoma metastasis therapeutics.

References: [1] Weigelt, B., Peterse, J. L. & van 't Veer, L. J. Breast cancer metastasis: markers and models. Nat Rev Cancer 5, 591-602 (2005) . [2] Szakacs, G., Paterson, J. K., Ludwig, J. A., Booth-Genthe, C. & Gottesman, M. M. Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5, 219-234 (2006) . [3] Rossi, A. et al. Single agent vinorelbine as first-line chemotherapy in elderly patients with advanced breast cancer. Anticancer Res 23, 1657 Res 23, -1664 Res 23, (2003 . [4] Hameed, S. et al. Self-assembly of porphyrin-grafted lipid into nanoparticles encapsulating doxorubicin for synergistic chemo-photodynamic therapy and fluorescence imaging. Theranostics 8, 5501-5518 (2018 In this stage PC progresses rapidly and metastasizes. Common chemotherapeutic drugs, such as docetaxel and cabazitaxel, are not efficient to control the disease because of low therapeutic activity and high systemic toxicities of these drugs. Therefore, novel rationally designed targeted therapeutics and therapeutic strategies are urgently needed for management of aggressive PC. 5D3 is a novel monoclonal antibody (mAb) rose against PSMA antigen, which is a marker of aggressive PC.3 5D3 shows higher binding affinity and fast internalization characteristics in PSMA(+) PC cells. This mAb is robust and stable in chemical conjugation reactions, and doesn't change the binding affinities after conjugations.4,5 Hence, 5D3 is an ideal candidate for the development of antibody-drug conjugates (ADC) for PC therapy. In this study, we conjugated 5D3 mAb with anti-tubulin agent mertansine (DM1) and developed a novel ADC ( Figure 1A ). Novel 5D3-DM1 ADC was evaluated in PSMA(+) human PC subcutaneous mouse models. Methods Anti-PSMA mAb, 5D3, 5B1 and F(ab')2 of 5D3 were labeled with AlexaFluor-488. A comparative in vitro cellular study in PSMA(+) PC3-PIP cells was performed for understanding the internalization characteristics. For the internalization study, PSMA(+) PC3-PIP cells in chamber slides were treated with 5D3(AF-488)2, 5B1(AF-488)2, and F(ab')2(AF-488)2 and imaged using a confocal fluorescence microscope to identify the specifics for ADC development. The best candidate, 5D3 mAb was first functionalized with MCC groups by reacting with sulfo-SMCC-NHS and conjugated with DM1 drug molecules. 5D3-DM1 ADC was purified and characterized by MALDI-TOF and DLS. The degree of drug conjugation per mAb was 2.8 in the synthesized ADC. PSMA(+) PC3-PIP PC mouse models were used for the evaluation of 5D3-DM1 ADC with PSMA(-) PC3-Flu mouse models as the control. Mice in both groups were administered with two doses of 5D3-DM1 ADC two week apart. Therapeutic response was examined based on the change of tumor volumes. Results and Discussion 5D3 mAb shows fastest internalization characteristics and higher cytoplasmic accumulation in PSAM(+) PC3-PIP cells compared to anti-PSMA 5B1 mAb and F(ab')2 fragments of 5D3 ( Figure 1B ). After drug conjugation, molecular weight of ADC was measured by MALDI-TOF ( Figure 1C ) and the degree of drug conjugation per mAb was calculated based on the molecular weight change. In the treatment study, PSMA(+) PC3-PIP tumors showed significant growth reduction compared to PSMA(-) PC3-Flu tumors ( Figure 1D ). After intravenous administration, 5D3-DM1 internalizes fast in PSMA(+) tumor cells, is processed, and releases DM1 drugs arresting the cancer cell mitosis. A Kaplan-Meier graph was plotted considering 4-flod tumor growth with respect to the initial tumor volume as the terminal point for euthanization unless animals die naturally ( Figure 1E ). The plot reveals significantly higher survival rate in PSMA(+) tumor bearing mice after the 5D3-DM1 treatment compared to the PSMA(-) tumor bearing control mice and untreated control mice. No adverse systemic effects of the treatment were observed in either animal group. Conclusions 5D3 has high binding affinities on PSMA(+) PC-PIP cells and shows fast internalization properties. 5D3 mAb is stable and robust in conjugation reactions and can be conjugated with DM1 without changing its binding affinities with PSMA. 5D3-DM1 ADC has shown high therapeutic efficacy on PSMA(+) PC3-PIP prostate cancer subcutaneous mouse models exhibiting a reduction of tumor growth and increasing the survival.

References: References 1. Cho, S et al. Mol Cancer Ther 2018 , 17, 2176 -2186 2. Semenas, J et al. Curr Drug Targets 2012 , 13, 1308 -1323 3. Novakova, Z et al. Prostate 2017 , 77, 749-764 4. Banerjee, S R et al. J Nucl Med 2019 , 60, 400-406 5. Hapuarachchige, S et al. Mol Pharm 2020 Image/ Figure Abstract Body : Evaluation of lymph nodes plays a key role in staging patients' cancer, and helps determine adjuvant therapy, which has been shown to affect overall survival. In breast cancer for example, sentinel lymph node biopsy is the standard of care where the sentinel or few of the closest draining lymph nodes are assessed.1 In head and neck2, colorectal3, and pancreatic cancer4, however, it is common for dozens of lymph nodes to be removed and pathologically evaluated. The conventional method of interrogating a single lymph node is both time and labor intensive, and so tissues are often under-sampled and high rates of false negatives occur.

Optical imaging offers a potential solution for whole node evaluation, but application is limited because of scatter in biological tissue. Various methods of scatter rejection exist to improve spatial resolution, often with an interplay between performance and technical complexity. Previously, we have demonstrated through Monte Carlo simulation5 and preliminary studies6 that a low-cost angular domain optical projection tomography imaging system (ADEPT) and simple filtered backprojection (FBP) is sufficient to detect and localize clinically relevant metastases in lymph nodes. Here, we evaluate performance of the ADEPT system to determine if and how the proof of concept study results can be improved, and to push the limits of sensitivity and specificity.

Solid cylindrical resin phantoms with varying diameters ( 6, 8, 10, 12 mm) were created with scattering and absorbing agents added to match the optical properties and size range of lymph nodes (previously determined experimentally5). A single ~1.5 mm diameter well was drilled at half the radius of each; these were filled with capillary tubes containing 10 µM of fluorescent IRDye-800CW to represent metastases. Each of the phantoms underwent tomographic transmission and fluorescence imaging on the ADEPT system for open aperture (conventional imaging) and closed aperture (angular domain imaging) configurations. Images were collected in 5˚ intervals over 360˚ and FBP reconstruction was employed.

Results demonstrated that the fluorescent inclusions could easily be observed for all phantom sizes and both aperture setups [ Fig. 1(a) ]. Quantitatively, as shown in the bar plots below, angular domain imaging with a closed aperture provided higher contrast (with the exception of the 12 mm sample) and more accurate spatial resolution, where full width half maximum values of the detected inclusions were closer to the expected range of 1.1 -1.2 mm based on the inner diameter of the capillary tubes (closed: 0.93 ± 0.06 mm, open: 2.61 ± 0.09 mm). While results of contrast-to-noise ratio (CNR) showed better performance for the open setup, it should be noted that exposure time was not scaled; as such, if noise characteristics were normalized, closed CNR values would be increased by a factor of about three. This however, demonstrates an important tradeoff of angular restriction -balancing the number of rejected and detected photons (improved spatial resolution vs increased noise) -that should be considered in system design. Overall, these findings along with demonstrations of accurate geometric and fluorescence concentration reconstructions (see supplementary) support the potential of the ADEPT system for improving lymph node pathology. (7) Abstract Body : Objective: Noninvasive detection of vascular abnormalities that manifest in a variety of conditions such as atherosclerosis, tumor initiation and progression are of critical importance for disease prognosis and patient care. [1, 2] To this end, magnetic resonance angiography (MRA) is a dynamic technique that can be used to acquire high resolution angiographic images as well as to evaluate the hemodynamic effects of various vascular lesions. [3, 4] Despite the fact that a number of contrast agents based on low molecular weight (LMW) Gd(III) polyamino polycarboxylic acid chelates (e.g. Gd-DTPA and Gd-DOTA) have received clinical approval for Magnetic Resonance Imaging (MRI), [5] these are not ideal for MRA due to their non-specific biodistribution and minimal vascular retention. [6] The design of contrast agents for MRA can be approached either by the attachment of multiple LMW Gd3+chelate to a macromolecular core [7] or by utilizing a receptor induced magnetization enhancement strategy where a LMW Gd3+-chelate can bind to plasma proteins (e.g. Human Serum Albumin or HSA) in a non-covalent fashion. [8] In this work, we propose to combine both these strategies in the following three steps; firstly, a novel DTPA based ligand with a lipophilic moiety is designed for non-covalent binding with HSA. Secondly, this ligand is appended onto the twelve vertices of a closo-B12(OH)12 core in a radial fashion resulting in a multimeric medium sized configuration, termed by us as closomers. [9] Finally, 12-fold Gd-complexation yields the MRA closomer contrast agent CCA-I. In mice injected with CCA-I, significant contrast enhancement is observed in the vena cava, heart, liver, kidneys and the bladder with minimal Gd retention in tissues 24 h post-injection. Methods, Results and Discussion: A novel amphiphilic ligand, MRA Ligand-I is synthesized by incorporating a hydrophobic substituent comprised of a 3, 3-diphenylpropan-1-amine moiety off the DTPA backbone. Subsequent reaction with GdCl3.6H2O affords the corresponding Gd-MRA Ligand-I chelate, although the aqueous solubility of this final complex is not optimal. To impart better aqueous solubility, a short oligoethylene glycol (OEG) linker with a -NH2 group is introduced as a side chain resulting in the synthesis of MRA Ligand-II. To this end, Closomer-I, a B12-analogue with pendant meta-chloro phenylcarbonate arms, is reacted with MRA Ligand-II to obtain the 12-fold functionalized carbamate entity, Closomer-II in 80% yield. Next, removal of all tert-butyl ester groups from Closomer-II, followed by subsequent reaction with GdCl3·6H2O leads to the formation of CCA-I in 79% yield. Next, the percentage binding of the CCA-I with HSA is tested and found to be significantly high (97%) as compared to its parent ligand Gd-MRA Ligand-I (44%) and is also slightly higher compared to that of MS-325 (88%), a blood pool MRA contrast agent approved for clinical use.

[10] T1-weighted MRI phantom images of CCA-I at 7 T at various Gd3+ ion conc. in PBS and HSA/PBS (4.5% w/v = 0.67 mM) shows noticeable contrast enhancement for the solutions of CCA-I in presence of HSA. Finally, representative T1-weighted MRA scans of mice injected with CCA-I at a Gd dose of 0.1 mmol/kg display significant contrast enhancement of blood vessels that persisted for about 10 min post-injection (p.i.) and diminish over a 1 h time period. Some contrast enhancement in the liver tissues at 3 h p.i. is indicative of the hepatic uptake of CCA-I that is common for compounds with a considerable amphiphilic component. Further, contrast observed in the bladder at 3 h p.i. indicates the renal system is the primary clearance route for CCA-I. Conclusion: In summary, we report the synthesis, relaxivity measurements and in vivo MRA imaging results of an amphiphilic variation of the DTPA ligand, appended on a multimeric closomer core. Abstract Body : Purpose Radioluminescence imaging has been utilized for imaging 18F-FDG uptake in tumor by detecting the scintillation light converted from ionizing radiation sources with a scintillator. The modality suffers from low spatial resolution due to the air gap between the scintillator and the imaging animal. Here we present a painted radioluminescence imaging (PRLI) to enhance spatial resolution and optical conversion efficiency by directly painting a gadolinium oxysulfide: terbium (Gd2O2S: Tb) scintillator on the body surface of small animals. Procedures The spatial resolution, imaging depth and signal linearity of PRLI were first characterized with in vitro phantoms. In vivo experiments utilizing a nude mouse injected with the 18F-FDG in matrigel ball was then conducted to verify the proposed PRLI. Alternative modalities including flexible radioluminescence imaging (Flex-RLI) and Cerenkov luminescence imaging (CLI) were also performed to characterize PRLI. Results With respect to spatial resolution, full-width half maximum values across a 1 mm capillary tube were 2.1, 5.2 and 1.2 mm for PRLI, Flex-RLI and CLI respectively. The imaging depth of PRLI can reach to 9mm in biological tissue medium. In vivo, the signal intensity of PRLI was 1555.6 and 21.6 times stronger than CLI and flex-RLI, respectively. The spatial resolution of PRLI was similar with CLI and higher than flex-RLI for the mouse imaging. Conclusions PRLI provides higher spatial resolution than flex-RLI, and has highest signal intensity comparing with CLI and flex-RLI for imaging 18F-FDG. Abstract Body : Introduction Sentinel node procedures (SNP) are performed with the use of tracer-agents, mainly radio-colloid and/or blue dye, to identify and resect the sentinel lymph node (SLN) in oncology patients. Fluorescent agents have emerged as a new tracer-agent to identify the SLN intra-operatively with near-infrared imaging. Our aim is to compare the SLN detection rate and adverse events of fluorescent agents to current "golden standards" (blue dye and/or radio-colloid) for the SNP by means of a systematic review and meta-analysis without any restrictions based on tumor type. Methods A systematic search in PubMed, Embase and The Cochrane Library was performed. Articles that compared the detection rates of fluorescent agents with radio-colloid and/or blue dye were included. Risk of bias was assessed using the QUADAS-2 tool. Meta-analyses were performed for breast, gynecological and dermatological cancer using a random effects model. Results In total 6195 articles were screened which resulted in a final inclusion of 55 articles. These studies concerned patients with breast (n=26), gynecological (n=13), dermatological (n=12), urological (n=2) and miscellaneous (n = 2) cancer types. All studies used indocyanine green (ICG) as fluorescent agent. Meta-analyses comparing ICG with blue dye showed a significant and clinically relevant difference in detection rate ranging from 11% up to 36% in favor of ICG, for breast, dermatological, and gynecological cancer (see figure for breast cancer). Meta-analyses comparing ICG with radio-colloid did not show any significant differences, with the exception of ICG versus radio-colloid + blue dye for the bilateral SLN detection in gynecological cancer. The only reported side effect reported of ICG was temporary skin staining. For blue dye, a few cases of permanent tattooing were reported. Conclusion Near-infrared fluorescence imaging using ICG provides a higher SLN detection rate compared to blue dye for the SNP in a range of different tumor types. SLN detection rates of ICG are comparable to radio-colloid. Due to their complementary characteristics in terms of spatial resolution and transdermal sensitivity, we suggest to use a combination of both ICG and a radio-colloid. In facilities with limited or no access to radiocolloids, ICG appears to be a good alternative for blue dye, especially in patients with gynecological cancer and in patients with other cancers with fairly predictable locations of the SLN. Abstract Body : Immune checkpoint inhibitors have shown remarkable success in the treatment of some cancer types such as melanoma. However, for some patients, such as those with triple negative breast cancer (TNBC), the benefits of this therapy remain underwhelming1. This poor response has been largely attributed to tumor heterogeneity and limited immune cell infiltration2. An understanding of mechanisms that drive the heterogeneity of response to immune checkpoint inhibitors can lead to improved therapeutic outcomes. Hypoxia and acidic extracellular pH (pHe) are frequently encountered in tumors3. Here, for the first time, we have used combined PET-MR and optical imaging to understand the role of hypoxia and pHe in PD-L1 expression using a syngeneic TNBC model in immune competent mice. Methods: Murine 4T1 breast cancer cells were engineered to express firefly luciferase (luc) under the control of hypoxia response elements (HRE). Tumors were generated by inoculating 2 x 106 cells into the upper mammary fat pad of female Balb/c mice (n=5). Tumor growth was monitored over time and tumor hypoxia visualized and measured with in vivo bioluminescence imaging (BLI) using an IVIS Spectrum scanner. Once the presence of hypoxia was detected in tumors with BLI, mice were intravenously injected with 14 ± 2 MBq of [64Cu]Avelumab, a PD-L1 antibody, and imaged at 24h and 48h post injection using a simultaneous 7T Bruker PET-MR scanner. Tumor pHe was assessed by acquiring chemical exchange saturation transfer (CEST) images following an 80µL intravenous injection of 100% iopamidol. A RARE pulse sequence with rectangular saturation transfer module of 2 sec duration and 3 mT power was used to acquire CEST images for ~30 min repeatedly at 2 saturation offsets 4.2, 5.5 ppm and last 5 M0 images. Other imaging parameters were TR/TE = 10000/6.23 ms; matrix size = 48x32; rare factor = 16; slice thickness = 2 mm and FOV = 12.6x8.6 mm2 respectively. Mice were subsequently sacrificed, tumors were sliced in the corresponding orientation of the PET-MR images, and ex vivo BLI of tumor slices was performed. BLI and PET/MR images were co-registered using MATLAB software. Masks were then generated from BLI images using ImageJ to define areas based on the level of tumor hypoxia. These optical masks were then overlaid onto the registered PET/MR slice and used as the region of interest (ROI) to calculate PET PD-L1 signal and CEST pH levels within those respective areas. Results: Tumor hypoxia was visualized with in vivo BLI as shown in Figure  1A . Representative in vivo PET-MR images of [64Cu]Avelumab distribution in mice are presented in Figure 1B . Representative ex vivo (BLI) and corresponding in vivo images of the tumor are displayed in Figure 1C . We observed significantly higher [64Cu]Avelumab uptake in tumor regions with the highest BLI intensity (75-100% mask) compared to areas with the lowest intensity (0-25%) (p < 0 .05) ( Figure 1D ). Furthermore, we found a significant positive correlation between the mean BLI signal and the mean PET PD-L1 signal in the tumors (p < 0 .05). Our preliminary data from corresponding CEST images suggests that hypoxic regions with high PD-L1 expression are also more acidic compared to normoxic regions ( Figure 1D ).

Conclusions: These results identify, for the first time, the close association between areas of hypoxia and increased PD-L1 expression, suggesting that hypoxia increases PD-L1 expression. These results have direct relevance to treatment with PD-L1 or PD-1 immune checkpoint antibodies, since poorly perfused hypoxic tumor regions will have higher PD-L1 expression but lower delivery of the antibody, allowing cells in hypoxic tumor areas to escape immune surveillance. Future studies should investigate the role of hypoxia in regulating immune checkpoints, and the use of hypoxic cell targeting in improving the outcome of immune checkpoint inhibitor treatment. (C) . These images show the close association between hypoxia and PD-L1. BLI and PET/MR images were co-registered using MATLAB software. Masks generated from BLI images were used to define areas based on the level of tumor hypoxia. These optical masks were overlaid on the co-registered PET/MR slice and used to calculate the PET PD-L1 signal and CEST pHe (D) . Abstract Body : Introduction: Magnetic particle imaging (MPI ) is a new imaging modality that sensitively and specifically detects superparamagnetic iron oxide nanoparticles (SPIONs) within the body. Our lab has been developing cellular magnetic resonance imaging (MRI) tools for well over a decade using both SPIONs and fluorine-19 (19 F)-based contrast agents for numerous important applications, including tracking of immune and stem cells used for cellular therapies. We have shown that SPION-based MRI cell tracking has very high sensitivity, but low specificity. Contrastingly, our work has demonstrated that 19 F-MRI cell tracking has high specificity, but low sensitivity. SPION-MPI cell tracking could overcome the challenges of MRIbased cell tracking allowing for both high sensitivity and high specificity cell detection. This study is the first to demonstrate that micron-sized iron oxide particles (MPIO) can be used for quantitative MPI of iron-labeled cancer cells in the mouse brain. Methods: Human breast cancer cells (231BR) were labeled with micron-sized iron oxide nanoparticles (MPIO), Immune compromised mice (NSG) were injected with either 2.5 x 10 5 or 5.0 x 10 5 cells. MPIO-labeled cells were administered intracardially using ultrasound guidance. MRI was performed in vivo the same day at 3T using a balanced steady-state free precession (bSSFP) sequence. After imaging, the mice were euthanized, and the brains were fixed and removed for MPI scans. MPI was performed on a Momentum TM scanner. MRI images were used to quantify the percentage of black pixels which represent the black voids created by iron labeled cells. MPI images were used to quantify the amount of MPI signal which is related linearly to the amount of iron. In a second experiment, NSG mice were also injected with 5 x 10 4 4T1BR cells, a murine breast cancer cell line, labeled with either MPIO or the SPION ferucarbotran (Vivotrax) which is currently the gold standard SPION for MPI. MRI and MPI was performed in vivo. Results: Figure 1 shows representative MRI and MPI images. Signal voids due to MPIO-labeled cells can be detected in the in vivo brain MRI of mice injected with either 2.5 x 105 or 5.0 x 105 cells ( Figure 1A ,B). Ex vivo MPI also showed signal in the brain from iron, which is visible as a hot spot ( Figure 1C ,D). More black pixels were measured in the brain MRI of mice receiving an injection of 5.0 x 105 cells. This agreed with MPI data which showed that there is less iron in the brains of mice injected with 2.5 x 10 5 cells compared to those injected with 5.0 x 10 5 cells. In the second experiment, in vivo MRI was able to detect signal voids in the brains of mice injected with 5.0 x 10 4 cells which were labeled with either MPIO or Vivotrax (Figure 2A , C) , although voids were fainter in Vivotrax labeled cells. In vivo MPI signal was only detectable in mice injected with MPIO-labeled cells ( Figure 2B ). Conclusions: This is the first example of the use of MPIO for cell tracking with MPI. The major advantage of MPI is the ability to quantify iron content and estimate the number of iron labeled cells. With an intracardiac cell injection, approximately 15% of the injected cells are expected to arrest in the brain vasculature. For our lowest cell injection of 5.0 x 10 4 cells, this is ~10 000 cells. The next steps include increasing sensitivity for MPI by testing different MPI tailored iron nanoparticles. Figure [1-13C] pyruvate, the rate of conversion of pyruvate to lactate (kPL) via lactate dehydrogenase (LDH) can be computed per voxel in patients. However, a high noise floor can obscure the peaks of downstream metabolites including lactate and alanine, limiting the achievable spatial resolution and rendering kPL estimations inaccurate in these cases. While surface transmit/receive (T/R) coils can be used to improve the sensitivity and spatial resolution in HP experiments, these coils produce an inhomogeneous B1+ profile, resulting in varying flip angles across the regions of interest. Hence, in computing the kPL for each voxel, it is critical to correct the nominal flip angle based off the coil's B1+ profile. The goal of this project was to develop a novel post-processing pipeline to improve spectral SNR using tensor rank truncation and to correct HP kPL images for B1 variations in human studies. Methods & Results: In this case study, a patient with pancreatic neuroendocrine tumor liver metastases was imaged with HP [1-13C] pyruvate. An axial T1-weighted spoiled gradient-echo anatomical reference was acquired with the target lesions, approximately 5 cm deep, highlighted ( Figure 1a) . A 13C T/R surface coil was placed on the patient's abdomen on the right side closest to the target lesion to ensure coverage by the coil's T/R field. [1-13C] pyruvate and [1-13C] lactate signals were acquired with a single-slice 2D echoplanar spectroscopic imaging sequence over 1 minute with a temporal resolution of 3 s and a spatial resolution of 1.2 cm isotropic. Constant flip angles of 10° for pyruvate and 20° for lactate were applied. The B1+ profile of the 13C T/R coil was measured from a separate phantom scan using the double angle method.1 Vitamin E capsules within the coil, serving as fiducial markers, were visible in the T1-weighted anatomical images acquired during the patient scan (yellow dots in Figure 1b ). Using these markers, the B1+ profile was rotated and translated in MATLAB to match the coil's position during the patient scan as shown in Figure 1b . The adjacent color bar indicates the scaling factor relative to the nominal flip angle. The EPSI post-processing pipeline consisted of a B0-correction, zero-and first-order phase corrections, and a tensor rank truncation method.2 In an HP tensor, the spectral, spatial, and temporal dynamics are inherently low-rank allowing for most ranks to be truncated, preserving selected principal components. The number of ranks preserved in each dimension was optimized using a bias-variance tradeoff algorithm. The before-and after-processing spectra are shown in Figure 1c -d. Pyruvate and lactate images summed over time after processing are shown in Figure 1e -f. The brightest voxels correspond with the tumors of interest. After correcting the flip angles based on the oriented B1+ profile, kPL values for each voxel were then computed using an inputless two-site model to generate kPL maps.3 The kPL maps before and after this correction are shown in Figure 1g -h. For a healthy voxel over-flipped by 3%, the kPL before and after correction were 0.025 and 0.027 s-1, a 4.91% under-estimation without B1-correction. For a tumor voxel under-flipped by 3%, the kPL before and after correction were 0.044 and 0.042 s-1, a 5.22% over-estimation without B1-correction. Discussion: The HP 13C EPSI post-processing pipeline developed here recovers metabolite signals allowing for improved kPL estimations. Over-flipping led to an underestimation of kPL (a 4.91% under-estimation without the coil-correction) whereas under-flipping led to overestimation (a 5.22% overestimation without the coil-correction), which agree with simulations.3 Acknowledgements: This work was supported by NIH grants U01 EB026412, R01 CA183071, and P41 EB013598. [1-13C] lactate images summed over time after processing, respectively, with low SNR voxels dropped (pyruvate SNR threshold: 10.5; lactate SNR threshold: 3.74). (g-h) kPL maps before and after the flip angle correction, respectively. For a healthy voxel under-flipped by 3%, the kPL before and after correction were 0.025 and 0.027 s-1, a 4.91% under-estimation without B1-correction. For a tumor voxel overflipped by 3%, the kPL before and after correction were 0.044 and 0.042 s-1, a 5.22% overestimation without B1-correction. , characterized by excellent kinetic parameters for molecular imaging due to their size and robustness. They offer considerable potential in diagnostic and therapeutic applications [1] . Lymphadenectomy, the surgical removal of lymph nodes (LNs), is performed in clinically node-negative patients who have a high chance of expressing nodal metastases. For quality assurance, a minimum number of LNs in the area of interest needs to be harvested. Samples of resected tissue are controlled for signs of cancer to assess its progress and guide further treatment. Intraoperative identification of LNs can be a complex and demanding procedure. We propose to use fluorescence molecular imaging to highlight all LNs within the surgical field to facilitate node picking during surgery. To realize LN staining we fluorescently labeled a previously developed [2] [3] anti-mannose receptor (MR, CD206) Nb that yields significant uptake in LNs. MATERIALS AND METHODS: The Nb MMR3.49 with high affinity for MR as well as a non-targeting control Nb were fluorescentlylabeled with Cy5 or IRDye800CW. One hour after intravenous injection of 2 nmol of the labeled compound in healthy mice (n=3 per group), axillary, inguinal and popliteal LNs were imaged in situ and ex vivo with the KIS (Kaer Labs, excitation 640nm, collection 820nm) (for detection of IRDye800CW signals). The translational nature of the approach was studied by upscaling the experiment to a healthy porcine model (n=1, 35 kg). Here, LN mapping in the pelvic area, following intravenous injection of 73 nmol of MMR3.49-Cy5, was evaluated 90 minutes after injection. LNs, muscle and fat tissue were procured for further ex vivo assessment. Fluorescent signals and target-to-background ratios (TBRs) were quantified using ImageJ software by superimposing them on white light images to anatomically localize the highlighted structures. RESULTS: Both Cy5 and IRDye800CW-labeled anti-MR Nbs enabled clear and specific in situ visualization of the LNs in mice, with ex vivo TBRs of 2.64±0.63 and 4.62±0.50 respectively. No uptake in LNs was observed for the control Nb (TBR 1.40±0.18 and 1.14±0.32) ( Figure 1A ). Comparably, in the pig, MMR3.49-Cy5 helped to realize distinct in vivo fluorescent staining of pelvic LNs ( Figure 1B) . TBRs of the excised LNs were respectively 3.49 with muscle, and 2.98 with fat used as background. CONCLUSIONS: Intravenous injection of fluorescently-labeled anti-MR Nbs enables intraoperative localization of LNs within the whole surgical view. This could potentially increase the accuracy of extended nodal dissections. Early data suggests the imaging findings in mice are scalable to porcine models.

References: [1] S. Muyldermans, "Nanobodies: Natural Single-Domain Antibodies," Annu. Rev. Biochem., vol. 82, no. 1, pp. 775-797, 2013. [2] A. Blykers et al., "PET imaging of macrophage mannose receptor-expressing macrophages in tumor stroma using 18F-radiolabeled camelid single-domain antibody fragments," J. Nucl. Med., vol. 56, no. 8, pp. 1265 -1271 . Abstract Body : Purpose: An organ's microenvironment plays a critical role in becoming a preferred site for metastasis and contributes towards tumor progression and response to therapy1,2. Bone and liver are common sites for prostate cancer (PCa) metastasis3. Patients with liver metastases have poor prognosis4 relative to patients with bone metastases alone and have shown variable response to treatments5. There is a current gap in knowledge of the mechanistic underpinnings, such as the tumor-microenvironment and metabolism, which will be critical in evaluating diagnosis and treatment response of metastatic PCa. Potent androgen pathway inhibitors have shown to induce neuroendocrine prostate cancer (NEPC), a lethal subtype of castration-resistant prostate cancer with poor survival and limited treatment options6,7. Hyperpolarized 13C (HP) MRI is a valuable technique for dynamic, real-time and non-invasive evaluation of metabolism in-vivo8 for metastases which are often not amenable to biopsy. In this study, we analyzed the differences in metabolism of NEPC tumor models grown in three different sites and their response to chemotherapy. Methods: NEPC tumor grafts were established in the renal capsule (tumor sections9), bone and liver (isolated tumor cells10, 2×105 in 20μL) using LuCaP9311, LTLs 352 and 61012 PDXs (patient-derived xenografts). Baseline and post-one-week (bone and liver mice treated with 60mgs/kg carboplatin) imaging were conducted on a Bruker 3T scanner with dual-tuned 1H/13C volume coil. T2-weighted images and diffusion-weighted images were acquired for tumor delineation and generation of apparent diffusion coefficient (ADC) maps. Hyperpolarization was performed using a 3.35T dynamic nuclear polarizer. The copolarized 80mmol/L [1-13C] pyruvate and 13C urea were rapidly dissolved with dissolution buffer and injected into mouse over 12s. Dynamic 13C spectra were acquired using a 2D-CSI sequence. Metabolite maps were obtained via SIVIC13 and analyzed using MATLAB. Statistical analysis was conducted via t-tests, pyruvate-to-lactate conversion rate, kPL was calculated as described previously14 and normalized to ADCs to adjust for the differential cellularity of the tumors9. Tumors were harvested for immunohistochemical staining. Results: The T2-weighted images were used to visualize the tumor (Fig1) and calculate volumes (TableP1). All the PDXs implanted in kidney had similar kPL and ADCs (Fig1). The L93 kidney tumors had significantly lower kPL as compared to the bone and liver tumors. Bone tumors had the lowest ADCs when compared to kidney (p < 0 .05) and liver tumors (not significant). LTL610 bone tumors also had higher kPL and lower ADC than kidney, but not significant. LTL610 bone tumors had significantly lower ADC as compared to L93 bones (FigP2). Tumor volumes post-carboplatin increased for all the PDXs in bone and liver (FigP3C). LTL610 bone tumor kPL decreased by 41% (p < 0 .05) from baseline (FigP3A). The ADCs for LTL610 and L93 bone post-carboplatin increased significantly by 15% and 24% respectively (FigP3B). Discussion: ADC maps show that bone tumors had higher cellularity as compared to kidney (L93 and LTL610) and liver tumors (L93). The L93 kPL indicates that bone and liver tumors were significantly more metabolic than kidney tumors. The kPL of bone tumors could partly be attributed to increased cellularity or contamination of the muscle cells (due to the infiltrative tumor) which are naturally higher in LDHA. Significant decrease in kPL of LTL610 bone tumors post-carboplatin indicates response, while the liver tumors show a slight increase alluding to differential impact due to the tumor-microenvironment. These changes are currently being evaluated by protracted follow up on tumor volume. Ongoing biochemical and immunohistochemical assays will further inform on the mechanism behind the higher kPL of bone tumors. We will investigate the molecular and metabolic mechanisms responsible for the differential response to chemotherapy of the tumors as a result of the tumor site. HP-MRI has the potential to provide early metabolic readout of metastases and aid in monitoring therapy. Abstract Body : Minimally invasive stimulation of select neurons with great spatiotemporal precision remains a challenging feat in neuroscience. The field is a high priority research area since neural circuit manipulation is heavily utilized in neuroscience research. Further, neurostimulation shows promising results in treatment of a number of conditions, including Parkinson's disease, epilepsy, or depression. The technique has also been increasingly used in brain-machine interfacing applications and retinal or cochlear implants. However, established neurostimulation methods suffer from numerous limitations constraining the spatial resolution, penetration depth, or chronic use. Ultrasound (US) appears to be a modality with great potential for neuroscience applications as it offers important advantages over electrical or light-based methods. First, US has low attenuation in soft tissue, allowing up to 5-7 cm penetration and, thus, enabling non-invasive stimulus delivery into regions deep in the brain. Second, its spatial and temporal resolution are exquisite: the beam can be focused into a sub-mm spot at frequencies in the MHz range (which highly exceeds the maximum firing rates of neurons). A growing body of research suggests that ultrasound is capable of modulating neuronal activity by altering the neuron excitability. However, this method does not enable reliable activation of a small brain region due to the lack of US-sensitive ion channels. To enhance the sensitivity of neurons to ultrasound waves, we utilize piezoelectric nanoparticles molecularly targeted to the neuronal membranes. Through the direct piezoelectric effect, these nanoparticles transduce mechanical deformations into electric charges, activating voltage-gating ion channels. Here, we demonstrate the effects in cultured rat hippocampal neurons. We used 300-nm tetragonal barium titanate nanoparticles (BTNPs) and verified their crystal structure with X-ray powder diffraction as tetragonality is a necessity for the piezoelectric properties. To enhance the piezoelectricity of the particles, we electrically poled them in a 900 V/mm electric field at 110 °C. This procedure led to aligning of the ferroelectric domains, resulting in a 50-fold increase in the piezoelectric conversion. To prepare a stable nanoparticle dispersion, we coated the BTNPs with polyethylene glycol (PEG). Through the use of bifunctionalized PEG polymers, we were able to conjugate IgG antibodies to the BTNP surface to enable cell-specific binding. We utilized an anti-Neurofascin antibody which targets a cell adhesion molecule expressed in the axon initial segment. We incubated our BTNP-antibody conjugates with cultured hippocampal neurons and observed successful attachment to the membranes. In our studies of the ultrasound excitation of neuron bound BTNPs, we were interested in the effects on calcium signaling and glutamate release. For calcium studies, we utilized GCaMP6f, a genetic calcium indicator, expressed in the presynaptic bouton. To monitor the release of glutamate, we transfected the cells with a glutamate fluorescent reporter GluSnFr. We applied ultrasound stimulation with a focused ultrasound (FUS) transducer (Sonic Concepts H-151) while imaging the fluorescence of the genetically encoded reporters. We observed that FUS stimulation (5 x 1 ms pulses, 1.1 MHz frequency, 4.63 MPa focal pressure) induced a calcium influx into the presynaptic boutons. The ultrasound excitation of BTNPs was also found to cause neurotransmitter vesicle fusion and release of glutamate (single FUS pulse of 45.5 µs duration, 4.2 -4.6 MPa focal pressure, 1.1 MHz frequency). Ultrasound in the absence of the BTNPs was found not to be a reliable way of inducing calcium or glutamate response. These results indicate that FUS excitation of membrane bound BTNPs generates sufficient electric charges to activate primary hippocampal neurons. Combining the advantages of molecular targeting and the favorable properties of ultrasound, our technique has a strong potential to become widely utilized in minimally invasive neurostimulation applications. Abstract Body : Cytokines, which are rich in the tumor microenvironment (TME), are a broad category of small proteins (~5-20 kD) secreted by cells to influence the cellular behavior of surrounding cells. Because cytokines play an important role in tumor growth, immune evasion, and promotion of invasion and metastasis, they are attractive targets in cancer treatment. Here, we have developed an entirely new class of nanoparticles (NPs) to act as scavengers (Sca) for cytokines to disrupt multiple functions of cancer cells and their interactions with cytokines. Many reports indicate that the glycosaminoglycan (GAG) molecules such as heparin, demonstrate specific binding with cytokines and chemokines. Hence, we incorporated biocompatible heparin to construct the Sca-NPs to scavenge cytokines in the tumor. The heparin molecules were conjugated to nano-sized cores that were magnetic iron oxide nanoparticles (MIONPs) or dextran in our investigations, to improve the efficiency of capture. To shield the interaction between the Sca-NPs and proteins during circulation and transport, inert polyethylene glycol (PEG) polymers were conjugated to Sca-NPs through acid sensitive acetal bonds. This PEG shell is stable under normal conditions, but is cleaved in weak acid condition, such as in cancerous or inflamed tissues, to expose the Sca-NP to capture the cytokines ( Figure  1 ). Following degradation in the endosomal compartment, or following clearance by circulation, the Sca-NP bound cytokines and chemokines would be removed from the TME. These Sca-NPs were found to reduce the concentration of IL-6 in MB-MDA-231 cell medium. Following a 1 hour incubation, the Sca-NPs were removed from the medium, and the concentration of IL-6 decreased to 53% of control. In medium that was incubated with the PEG-Sca-NP, concentration of IL-6 changed slightly (reduced to 92% of control). When we treated the medium with cleaved PEG-Sca-NP (in which the PEG shield was removed), the IL-6 concentration was reduced to 56% of control. These Sca-NPs also were applied to scavenge IL-6 in mouse serum. Following treatment, the concentration of IL-6 was reduced to 70% of control. These Sca-NPs also can carry an imaging reporter to allow noninvasive detection of their localization in tumors. These scavenging NPs (Sca-NPs) may also have potential to reduce the cytokine storm to prevent organ failure. Abstract Body : Objective: Transdifferentiation of prostate adenocarcinoma to Neuroendocrine Prostate Cancer (NEPC) has emerged as one of the leading causes of resistance to androgen deprivation therapy (ADT) [1] . The patients have aggressive disease with visceral metastasis and have short survival time (7-10 months) [2] . Genomic and proteomic characterization of biopsy samples of NEPC lesions indicates loss of androgen receptor (AR) signaling [3] . Therefore, current PET imaging agents such as 18F-FDHT and 68Ga-PSMA11 that rely on functional AR cannot be used. Our goal is to develop a PET based molecular imaging agent that can uniquely identify NEPC lesions. Taking advantage of highly specific expression of Delta-like ligand 3 (DLL3) in NEPC lesions, we have developed zirconium-89 labeled immunoPET agent that can specifically identify NEPC lesions. Methods: We have used well characterized NCI-H660 cell line as representative NEPC model and compared it with AR dependent LNCaP and AR independent PC3 and DU145 cell lines. qPCR was used to measure relative levels of ARregulated gene transcripts (AR, PSMA, PSA) and NEPC marker DLL3 and normalized to b-actin in the cell lines. Relative protein levels of AR regulated genes and DLL3 were measured by western blot analysis of cellular extracts using b-actin as our control. ImmunoPET agent -89Zr-SC16 -was developed through the conjugation of the DFO chelator to SC16 (DLL3 specific) mAb and radiolabeled with zirconium-89. Saturation binding assay was performed on the cell line NCI-H660 to determine Bmax and Kd values. For in vivo PET imaging and biodistribution studies, NCI-H660 (DLL3/+) or DU145 (DLL3/-) xenografts were established in 6-8 week old male athymic nude mice. The mice were administered with 89Zr-SC16 and imaged at 24, 48, 72, 96, and 120 h post injection on a PET scanner and at chosen time points mice were euthanized and organs collected for biodistribution studies. Results: Saturation binding assay reveals that Kd =0.35 nM and Bmax = 863 fm/106 cells for 89Zr-SC16. In vitro studies indicated that NCI-H660 cell line was positive for DLL3 and negative for AR, PSA, and PSMA both at transcriptional and translational level. As expected all other cells except PC3 were negative for DLL3. In vivo PET imaging with 89Zr-SC16 showed clear delineation of NCI-H660 (DLL3/+) tumor xenografts. Biodistribution studies showed tumor uptake of 18.4 ± 3.8 %ID/g in the NCI-H660 tumors compared to 5.5 ± 0.5 %ID/g in the DU145 tumors, demonstrating the selective accumulation of the radiotracer in the DLL3-expressing tumors. Conclusion: Our findings demonstrate that only NEPC cells selectively express DLL3 and using DLL3 targeting PET agent 89Zr-SC16 we can non-invasively and uniquely identify NEPC lesions in vivo. Acknowledgement: Partial funding for these studies was provided by 2019 Geoffrey Beene Cancer Research Center Grant. Abstract Body : OBJECTIVES: 18F-FDG PET is accepted for the assessment of active cardiac sarcoidosis; however, up to 1/3 of patients fail to follow the required complex dietary restrictions, resulting in non-diagnostic scans. In addition, there are equivocal results. 18Flabeled FSPG, an L-glutamate derivative, is a promising radiotracer for PET imaging of the aminoacid antiporter system xC-that is involved in detoxification processes and balancing oxidative stress, including in sarcoidosis. No dietary restrictions are required prior to 18F-FSPG PET. This prospective pilot study evaluates the utility of 18F-FSPG PET for assessment of cardiac sarcoidosis, in comparison to negative or non-diagnostic 18F-FDG PET. MATERIALS AND METHODS: Patients referred to Nuclear Medicine to rule out active cardiac sarcoidosis were prospectively enrolled after the 18F-FDG PET/CT results were negative, equivocal or nondiagnostic. 18F-FSPG whole-body imaging started 46-67 minutes (mean±SD: 53.8±6.0), while dedicated cardiac imaging started 55-77 minutes (mean±SD: 65.9±5.9) after injection of 7.1-8.9 mCi (mean±SD: 8.1±0.5). The mean delay time between 18F-FDG PET and 18F-FSPG PET was 23.3 days. A state-of-the art SiPM-based PET/CT scanner (GE Discovery MI) was used for all scans. RESULTS: Data from 4 women and 11 men, 32-79 year-old (mean±SD: 58.2±12.9) were collected. Eight of the participants had no 18F-FDG uptake compatible with cardiac sarcoidosis; 2 of the 8 had 18F-FDG uptake in lung nodules and mediastinal lymph nodes, while 3 of the 8 had 18F-FDG uptake in mediastinal lymph nodes compatible with sarcoidosis. 18F-FSPG uptake was seen in the lung nodules and mediastinal lymph nodes in the same 5 participants with 18F-FDG uptake, while no 18F-FSPG uptake was identified in the myocardium. The remaining 7 patients had non-diagnostic cardiac 18F-FDG PET due to non-compliance with dietary instructions; 18F-FDG uptake in lung nodules and mediastinal lymph nodes was seen in 1 of these 7 patients. These lung nodules and mediastinal lymph nodes also had 18F-FSPG uptake. No 18F-FSPG myocardial uptake was identified in the 7 patients with non-diagnostic cardiac 18F-FDG PET. CONCUSIONS: 18F-FSPG may play a role in ruling out cardiac sarcoid involvement in a cohort with negative, equivocal or non-diagnostic 18F-FDG PET. Further evaluation in larger cohorts is needed to confirm our data. In addition, future studies should evaluate 18F-FSPG usage in patients with confirmed active cardiac sarcoidosis. Abstract Body : Cell-based immunotherapy refers to the administration of immune cells for the treatment of an ever-increasing list of cancerous, autoimmune, degenerative, and infectious diseases; some of which are rare and/or difficult to treat. For example, administration of peripheral blood mononuclear cells (PBMC) has been FDA approved for prostate cancer (1, 2) and regulatory T cells (Treg) are under development to replace druginduced immunosuppression for patients with autoimmune disease, organ transplantation, or blood cancers (3) . For immunotherapy, migration of administered cells to secondary lymphoid tissues is essential to elicit adaptive immune responses. It is crucial to know whether the cells have migrated to lymph nodes, and in what quantity, to determine the magnitude of this response (1, 2) . Researchers are now being challenged to verify the presence and quantity of cells in vivo to gain regulatory approval of cell therapies (4) . Fluorine-19 (19F) magnetic resonance imaging (MRI) has the potential to answer these fundamental questions. 19F images display distribution of perfluorocarbon-(PFC)-labeled cells and allow for measurements of cell number, providing information about cell delivery and engraftment (4, 5) . Our objective is to conduct longitudinal cell tracking of human PBMC and Treg in immunocompromised mice using 19F MRI at 3 Tesla (T), to study cell migration, persistence, and quantity in vivo. Methods: Following preparation of human PBMC (2) and Treg (3), cells were cocultured overnight with PFC (Celsense Inc.). PFC+ PBMC were injected into Nu/Nu mice in the right (3x106 cells) and left (6x106 cells) footpads (n=4) or by intraperitoneal injection (5x106 cells, n=2). In another cohort, a mixture of 10x106 PFC+ Treg and 10x106 unlabeled PBMC were administered intravenously (n=4) or subcutaneously (n=4). Unlabeled PBMC was included to repopulate lymphoid organs and promote Treg migration and retention at lymph nodes. 24 and 48 hours later, 1H and 19F images were acquired on a 3T clinical MRI using a 4.31 cm dual-tuned surface coil and a 3D balanced steady state free precession (bSSFP) sequence (1, 6) . 19F signal in images was quantified by comparison to reference tubes and 19F atoms/cell was measured by NMR. Results: Labeling was successful with 1.5x1012 19F spins/PBMC and 1.73x1012 19F spins/Treg, as determined by NMR. 19F images of mice receiving PFC+ PBMC show the accumulation of 19F signal in the popliteal and inguinal lymph nodes post-adoptive transfer. 19F images of mice receiving PFC+ Treg show the accumulation of 19F signal in inguinal and axillary lymph nodes. The included figure shows representative 19F images from each cohort and tables summarizing the quantification of PFC+ cells detected at lymph nodes. Discussion: We demonstrated detection and quantification of PBMC and Treg in vivo which accumulate at lymph nodes. This is the first study to conduct in vivo 19F imaging of Treg. This is also the first study to image 19F-positive PBMC in vivo under clinical conditions; and overall the third 19F cell tracking mouse study at 3T. By developing this technology at 3T, we move towards clinical translation. This work is central to understand the fate of cells after administration and to improve therapeutic benefit. For example, our images suggest that the route of cellular administration may impact the treatment efficacy (i.e. the number of cells reaching lymph nodes). 19F imaging of cellular immunotherapy may also play a clinical role, to inform physicians whether a patient may need repeat dosing or other interventions. Abstract Body : Certain types of chemotherapy are shown to induce anticancer immune response and are actively being investigated for their synergy with immunotherapy [1, 2] . Of the several hundred clinical trials exploring various combinations of chemo-and immunotherapy [3] very few utilize molecular imaging to evaluate immune response. Here, we assess the value of using [18F] F-AraG, an agent relatively specific for activated T cells [4] [5] [6] , as a non-invasive tool that could profile tumors based on the key players in adaptive antitumor response, CD8+ cells, and evaluate immunomodulatory effects of chemotherapy. To mimic the variations in cancerimmune phenotypes observed in the clinic, we imaged six syngeneic tumor models with different immune contextures: MC38 (colon), CT26 (colon), LLC (lung), A9F1 (immunogenic Lewis lung clone), 4T1 (breast), and B16F10 (melanoma). [18F] F-AraG revealed strikingly different uptake patterns in these tumors: signal in the tumor core (MC38 and A9F1), signal encircling the tumor (CT26), signal at the tumor margin (LLC and B16F10) and absence of signal (4T1). To evaluate the ability of [18F] F-AraG to report on the presence of CD8+ cells within the tumor microenvironment, the signal intensity was correlated with the number of lymphocytes isolated from the tumors one day post imaging. While [18F] F-AraG intensity did not correlate with the number of total lymphocytes (CD45+ cells), statistically significant correlation was found with the number of PD-1 positive CD8+ cells (r2 = 0.528, p To assess the ability of [18F] F-AraG to detect immunomodulatory effects of chemotherapy, we performed longitudinal imaging of tumor bearing mice (MC38, A9F1 and 4T1) undergoing two types of chemotherapy: oxaliplatin/cyclophosphamide, shown to induce immunogenic cell death and paclitaxel/carboplatin, reported to cause immunogenically silent tumor cell death. In MC38 model, paclitaxel/carboplatin did not result in appreciable change in signal post therapy, but oxaliplatin/cyclophosphamide treatment led to close to 2.4 fold higher [18F] F-AraG signal. The increase in signal post oxaliplatin/cyclophosphamide was also observed in A9F1, but not in 4T1 model. These imaging results corresponded to the differences in the tumor microenvironment that were detected between chemotherapies and tumor models. Compared to paclitaxel/carboplatin, oxaliplatin/cyclophosphamide therapy led to an increase in total lymphocytes in both MC38 and A9F1 tumors, but not in 4T1. In MC38 tumors that showed the most striking difference between the two chemotherapies, the oxaliplatin/cyclophosphamide group showed 27-fold higher ratio of effector CD8+ to regulatory CD4+FOXP3+ cells than the paclitaxel/carboplatin treated mice. Our data suggest [18F] F-AraG PET is a promising tool for CD8 profiling of the tumors and evaluation of immune induction strategies. The ability of [18F] F-AraG PET to assess the location and function of CD8+ cells, as well immune activity within tumors post immune priming therapy warrants further investigation into its utility for patient selection, evaluation of optimal time to deliver immunotherapies, and assessment of combinatorial therapy approaches.

George Firth, King's College London, george.firth@kcl.ac.uk Category: New Chemistry, Biology & Bioengineering Abstract Body : Introduction Manganese is a versatile essential metal that has found an important role at the heart of medical imaging, in the form of paramagnetic complexes as contrast agents for MRI and, most recently, as radionuclides for PET. Manganese-52 has favourable imaging properties (t1/2 = 5.6 days, β+ = 30%) that make it a desirable option for imaging long-lived biological processes, notably cell and antibody tracking.1,2 In order to permit successful immunoPET studies, high radiolabelling efficiency and good stability of the radioimmunoconjugate are paramount. In this study several different bifunctional chelators were labelled with 52Mn. The radiolabelling efficiency was determined and successful lead candidates where selected for in vivo stability and PET biodistribution studies. Methods 52Mn was produced at the Hevesy lab following the irradiation of pressed chromium powder (GE PETtrace; 16 MeV protons; 20 μA for 4 h). Several trastuzumab-like (Herzuma®, trastuzumab biosimilar) immunoconjugates of bifunctional chelators (BFCs) (DOTA, oxo-DOTA, NOTA, PCTA, DTPA, CDTA) were synthesised and radiolabelled following incubation at room temperature with [52Mn]MnCl2 using range of antibody concentrations (0.002-20 μM) and pH (5 and 7.2) . Reactions were quenched after 45 minutes with EDTA (50 mM). Labelling efficiencies from each dilution assay were determined by iTLC (iTLC-SG; 0.1 M citrate buffer). Chelator to antibody ratios were calculated by radiolabelling the immunoconjugates with 111In and co-incubating with varying concentrations of InCl3. The labelling efficiency at different indium to immunoconjugate ratios was calculated, and from this, the number of moles of indium specifically bound at each ratio was determined for each immunoconjugate. A known concentration of immunoconjugate thus allowed for the number of chelators to be determined. The biodistribution of [52Mn]DOTA-Her and [52Mn]PCTA-Her (1.5 MBq; 200 µL) selected as lead compounds on the basis of in vitro labelling efficiency were investigated in healthy BALB/c mice over 13 days by PET/CT. Ex vivo biodistribution at 13 days post injection was obtained following tissue dissection. [52Mn]MnCl2 PET images over 3 days and ex vivo biodistribution at 4 days are shown for comparison. Results/discussion The class of cyclic chelators tested demonstrated successful radiolabelling yields with quantitative labelling achieved at concentrations above 2 μM, compared to acyclic chelators which had considerably lower yields ( 90 % at 200 nm immunoconjugate concentration; room temperature; pH 5). Conjugation to each of the isothiocyanate-derived BFCs yielded approximately 5 chelators per antibody for DOTA-Her and PCTA-Her, and 7 for oxo-DOTA-Her and NOTA-Her using identical conjugation conditions for each BFC (Supplementary data 1A). In vivo biodistribution of the lead candidate, PCTA-Her, was investigated using PET imaging (Fig. 1B) . This was then compared to the gold standard chelator, DOTA, as a DOTA-Her BFC, which has previously demonstrated good in vivo stability.1 Both immunoconjugates demonstrated prolonged in vivo stability, with radioactivity present in the blood pool up to 166 h post injection. The subsequent time point at 318 h exhibits metabolism of the immunoconjugates, and is likely due to the biological half-life of the antibody and not the stability of the BFC. Conclusion The results presented here suggest that PCTA could offer advantages for long-term PET imaging, not from an in vivo stability perspective, as DOTA immunoconjugates already permit good in vivo stability, but from a radiolabelling efficiency perspective. The ability to quantitatively radiolabel PCTA-Her with 52Mn using mild conditions (room temperature, neutral pH, 2 μM BFC concentration) makes it a desirable choice for future applications in immunoPET. NOTA and oxo-DOTA are additional options with superior radiolabelling over DOTA, however their in vitro and in vivo stability are currently under investigation. Abstract Body : Introduction: Tumour-associated macrophages (TAMs) are the most abundant and influential cells infiltrating breast tumours. Accumulating evidence suggests that TAMs actively promote all aspects of tumour initiation, development, immunosuppression, metastasis, and resistance to therapies (1) (2) (3) (4) (5) (6) (7) . A meta-analysis of clinical data has indicated a strong link between TAM burden and the prognosis of breast cancer patients (8) . Considering the identification of TAMs as a biomarker and the emergence of TAM-targeted immunotherapies (6, 9) , a critical step involves the development of diagnostic tools to detect and quantify TAMs in vivo. MRI is a promising technique to monitor TAM burden because it has the ability to visualize and quantify the number of TAMs (10). The aim of this study is (1) to determine the density and distribution of TAMs in murine mammary tumours over time using fluorine-19 (19F) MRI and (2) to deplete macrophages by systemic administration of clodronate liposomes, then determine how this effects tumour growth and TAM presence using 1H/19F MRI. Methods: 3x105 4T1 breast cancer cells were implanted to the mammary fat pad of BALB/c mice (n=17). Phagocytic macrophages, including TAMs, were labeled through intravenous delivery of perfluorocarbon (V-Sense, CelSense Inc.). 24 hours later, 1H/19F images were acquired on a 3T clinical MRI using a dual-tuned surface coil and 3D balanced steady state free precession (bSSFP) sequence. Study 1: 1H/19F images were acquired 11 (n=4), 17 (n=5) and 20 (n=5) days after tumour implantation. Study 2: Clodronate liposomes (1mg) was delivered intravenously 5, 10, and 15 days after tumour implantation to induce apoptosis in phagocytic macrophages. 1H/19F images were acquired on day 17 (n =3). Regions of 19F signal were assessed for changes in spatial distribution, then delineated manually and quantified relative to reference tubes of known 19F content. Results: Study 1: Average tumour size increased from day 11 (162.13 mm3) to day 17 (468.50 mm3) and day 20 (788.62 mm3) (p Study 2: The average tumour volume in 1H images was 468.5 mm3 in control mice and 441.2 mm3 in mice administered clodronate (Fig E, p=0 .7133). 19F signal was found in the liver, spleen, and tumours of control mice and mice that receive clodronate due to the accumulation of perfluorocarbon ( Fig D) .19F signal measured in control tumours (1.01 x1019 19F spins) was not significant from mice administered clodronate (1.01 x1019 19F spins) (Fig F, p=0 .9941). 19F signal in the liver and spleen was 13.5 x1019 19F spins in control mice and 4.62 x1019 19F spins in mice administered clodronate (p=0.2511). Discussion: 19F MRI cell tracking can provide insights about the role of TAMs in the tumour microenvironment. We have shown that 19Flabeled TAMs are found predominately at the periphery of tumours and that TAM burden increases as tumours grow (between day 17 and 20). Future study will involve more timepoints to further understand the timing of TAM infiltration. 19F MRI can be used to provide a measure of TAM burden when testing novel therapies. In this study, we did not observe an effect of systemic administration of clodronate on TAMs, nor on macrophages in the liver and spleen.

More rigorous treatments, including combination treatments, may be required to deplete TAMs. With 19F MRI cell tracking, we can identify when TAMs persist following treatments to assess how TAMs contribute to tumour recurrence and resistance. . Antibody mediated MDSC depletion using anti-mouse granulocyte receptor-1 antigen (Gr1) Clone RB6-8C5 (Ly6GHi/Ly6CLo) has been used to effectively eliminate MDSCs in some tumor models, although further investigation must be carried out to validate its efficacy on both MDSC subtypes2. To achieve targeted near-infrared photoimmunotherapy (NIR-PIT)3, we synthesized an IR700-labeled Gr1 antibody-photosensitizer conjugate. This proof of principle study evaluates the specificity of the Gr1-IR700 conjugate to detect and eliminate splenic MDSCs in culture. Methods: Preparation of Gr1-IR700 and IgG-IR700 was carried out through the attachment of an NHS-activated NIR phthalocyanine dye, IR700, to the free amine residues on the RB6-8C5 monoclonal antibody and rat IgG2b,κ isotype control. Orthotopic 4T1 tumor models were established through the inoculation of 2 ×106 4T1 cells in the mammary fat pad of 4-6-week-old female, BALB/c mice. Mice bearing tumors with volumes ranging from 1.5 -2.0 cm3 were euthanized, and the spleens were harvested for MDSC isolation. After spleen dissociation and red blood cell lysis, immunomagnetic isolation of murine MDSCs was performed following the EasySep™ Mouse MDSC (CD11b+Gr1+) Isolation Kit protocol (77% purity shown in Figure 1 ). To validate in vitro targeting of MDSCs, viability assays were conducted. MSDCs were incubated at a density of 1 ×106 cells/mL with 1-10 ug/mL of Gr1-IR700. Reagent specificity was also determined using PBS and IgG-IR700 as controls. The cells were exposed to 16 J/cm2 of NIR irradiation. CCK-8 assays were used to measure cell viability. In ongoing in vivo PIT studies to demonstrate MDSC targeting, tumor-bearing mice will receive i.v. injection of 100 μg of Gr1-IR700 followed by splenic and tumor irradiation at 200 J/cm2. MDSC cell death will be characterized by fluorescence-activated cell sorting (FACS) analysis of spleens and tumors in PBS, IgG-IR700 and Gr1-IR700-treated groups. Results and Discussion: We confirmed that light-activated Gr1-IR700 caused Gr1-specific cell death in splenic MDSCs. Representative cell viability data for all groups (n = 2) are shown in Figure 2 . At the NIR light dose of 16 J/cm2, Gr1-mediated cell killing was not dependent on the conjugate concentration, and phototoxicity was partially inhibited by 5-fold excess RB6-8C5. These findings confirm the ability of Gr1-IR700 PIT to eliminate MDSCs, illustrating its potential application for targeting MDSCs in the TME. Abstract Body : Overexpression of cytosolic phospholipase A2 (cPLA2) is implicated in a variety of different cancer types such as triple-negative breast (TNBC) and non-small cell lung (NSCLC) cancers.1-3 NSCLC is the leading cause of cancer related death in the United States and is primarily treated by surgical resection of the primary tumor.4 Currently, surgeons rely on tissue palpation and visual inspection to guide surgical resection and identify malignant tissues intraoperatively. Previously, a novel cPLA2-activable fluorophore, DDAO-arachidonate, was developed for tumor margin detection in TNBC models. DDAO-arachidonate fluorescence is caged by esterification of DDAO to an unsaturated fatty acid. Enzymatic cleavage by cPLA2 releases DDAO resulting in a measurable fluorescence at 660 nm.5 The emission of DDAO in the visible spectrum of light is not ideal for in vivo imaging due to tissue absorption and scattering at these wavelengths. The purpose of the present study was twofold. First, we validated DDAO-arachidonate activation in human lung cancers in preparation for translation to imaging NSCLC. Second, new caged long wavelength fluorophores were synthesized and tested for NIR-II fluorescence to improve upon the in vivo optical properties of our probe. Liposomal formulations of DDAO-arachidonate were tested for size and zeta potential to demonstrate feasibility for in vivo applications. A solution of egg-phosphatidylcholine and DDAOarachidonate (95:5 mole%) in chloroform was dried under nitrogen before reconstituting in 0.1% PBS. Membrane extrusion was performed using a 100 nm filter. The size and zeta potential for the formulation were determined to be 105.7 nm and -15.13 ± 2.4 mV using a Malvern ZetaSizer Nano. To test feasibility of translation to NSCLC imaging, 20 nmol of liposomal DDAOarachidonate were painted on normal and cancerous human lung tissue in triplicate immediately following tissue resection. The tissue was incubated at 37°C for 3 hours to before imaging using the PerkinElmer IVIS Spectrum with 640 nm excitation and 680 nm emission. Fluorescence was seen only in the tumor tissue with a tumor to background ratio (TBR) of approximately 5.2. Second, a new class of caged cyanine dyes has been synthesized with emission in the NIR-II region. An absorbance spectrum for one of these compounds, JAM317, was obtained using a PerkinElmer Lambda35 with peak absorbance at 950 nm. Optical studies using the PhotonEtc. IR VIVO imaging system have demonstrated measurable fluorescence from 1050 nm to 1538 nm, with peak emission observed using the 1050 nm filter. Continuous excitation by an 808 nm laser over 5 min demonstrated that JAM317 is resistant to photobleaching. The synthesized fluorophores are modifiable and can be targeted to specific disease states or physical properties such as hydrophilicity may be altered through the introduction of different moieties. This study has demonstrated that DDAO-arachidonate has a promising TBR for imaging NSCLC in human tissue and new NIR-II fluorophores have been developed that may further improve the imaging of NSCLC. Abstract Body : Surgical margin assessment remains a critical step in the treatment of head and neck squamous cell carcinoma (HNSCC), where the extent of margin involvement determines postoperative treatment, and in turn influences patient prognosis. Cancer within 1 mm of the surface is defined as a "close" margin; within 1-5 mm, a "wide" margin. In either case, reoperation is carried out if possible. If not possible, patients with "close" margins require postoperative chemo-radiation, which substantially increases patient morbidity; whereas patients with only "wide" margins are given radiotherapy without the concomitant use of chemotherapy.1,2 Current margin assessment protocols require lengthy tissue processing in surgical pathology departments, where results often take more than 24 h to process, with only a small percentage of the margin being interrogated.

Fluorescence molecular imaging has proven to be a useful tool for intra-operative assessment of tumor margins. For example, in a study of HNC patients, the injection of cetuximab-IRDye800CW enabled real-time evaluation of resected margin status with a sensitivity and specificity of 100% and 91%, respectively.3 Although fluorescence intensity can indicate positive/negative status of margins, the signal alone does not provide any information of cancer depth; i.e. is the positive margin "close" or "wide". To address this, a ratiometric approach using angular domain fluorescence imaging is proposed. It is hypothesized that differences in depth sensitivity profile shapes of photons detected with an open aperture (conventional imaging) versus a narrow aperture (angular domain imaging) can be used to distinguish the depth from which the fluorescence signal comes from.

Monte Carlo simulations were conducted using the open-source MCmatlab4 where average soft biological tissue optical properties were assumed (µa = 0.4 cm-1, µs = 10 cm-1, g = 0.9). Fluorescence detection sensitivity profiles were modeled as an isotropically emitting point source focused at the tissue surface for open aperture detection, and as a pencil beam focused at half the depth for narrow aperture detection. Fluorescence uptake of tumors at varying depths was simulated for a 0.7 cm thick tissue, and open and narrow aperture wide-field reflectance images of the volume were generated using the sensitivity profiles [ Fig.1 

Results demonstrated the potential for ratiometric angular domain imaging to offer depth resolution and improved identification of both close and wide margins. While the inclusion closest to the surface was easily identified, neither open nor narrow aperture imaging alone was able to detect wide margins. However, after taking the ratio of the two images [ Fig. 1(d) ], the deeper tumor was delineated and relative signal intensity differences as a function of depth were maintained. With such information, a depth map [ Fig. 1 (e)] can be generated after thresholding to provide surgeons with rapid analysis of margin involvement to better guide postoperative treatment strategies. Although these simulations are simplified, the normalization of the ratiometric approach can account for common issues of fluorescence guided surgery such as heterogeneity of tissue optical properties, fluorescence distribution, and surgical artifacts (e.g. blood on tissue surfaces or imaging probes). Abstract Body : Chronic lung diseases, such as chronic obstructive pulmonary disease (COPD) and idiopathic pulmonary fibrosis (IPF), are major leading causes of death worldwide. Recently the usage of E-cigarettes has grown rapidly in the US contributing to the sharp increase of chronic lung disease. There is a pressing unmet medical need to develop noninvasive imaging methodologies and contrast agents to detect early stages of lung fibrosis, and stage fibrosis severity for progression of IPF and COPD to facilitate treatment. Here, we first demonstrate that overexpressed collagen 1 is an ideal molecular biomarker for detection, progression and monitoring of lung fibrosis in both COPD and IPF resulted lung damage. We then report the development of a new class of collagen targeted human protein-based Gd 3+ -based contrast agent, hProCA32.collagen. hProCA32.collagen exhibits strong affinity to collagen 1, high relativities per particle (r 1 and r 2 ) at both 1.4 and 7.0 T, and negligible immunogenicity. We have achieved the first detection of early stage lung fibrosis in both COPD and IPF animal models using hProCA32.collagen (pMRI) correlated with histological patterns and lung function analysis. In addition to early detection, hProCA32.collagen stages fibrosis progression in lung shown by receiver operating characteristic (ROC) curve. hProCA32.collagen mitigates metal toxicity due to its lower dosage, strong resistance to transmetalation and unprecedented metal selectivity for Gd 3+ over physiological metal ions. We anticipate hProCA32.collagen having strong translational potential in facilitating effective treatment to halt further chronic lung disease progression as well as reveal key factors that contribute to lung fibrosis including emerging public health issue associated with e-smoking. Abstract Body : Introduction Molecular imaging (MI) techniques offer unique functional and molecular information non-invasively at the cellular level [1] . While a variety of imaging modalities have been developed, they come with trade-offs between imaging depth, sensitivity, and/or spatial resolution. Hence, there is a need to develop a highly sensitive MI device that provides images with high spatial resolution and deep imaging depths without relying on ionizing radiation. Radio-frequency (RF)-Acoustic Tomography Molecular Imaging (RATMI) is an imaging technique that generates ultrasound images by irradiating tissue with non-ionizing RF fields for deeper penetration [2] [3] [4] . Despite the technical feasibility, the limited choices of biocompatible RF absorbing agents have hindered the use of RATMI [5] . The development and characterization of cancer-targeted RF nanodroplets have been recently studied which highlighted the ability to image the accumulation of the agent in prostate cancer with high specificity; however, the image reconstruction methods showed limitations with poor image quality. In this study, various image reconstruction methods were performed in order to enhance image quality and sensitivity. Materials and Methods The image reconstruction was performed on raw data sets acquired by the RATMI prototype system (ENDRA Life Sciences Inc.). The prototype is comprised of a pair of customized horn RF antennas (433 MHz) and an ultrasound linear-array transducer (Acuson L382) for generating and receiving the signals. The stage is attached to a rotational motor for generating tomographic images ( Figure 1 ). In this study, the sample stage was rotated 36 times with a 10-degree interval to cover the full tomographic angle. At each angle, 4,095 RF pulses were excited and the acquired acoustic signals were averaged for a better signal-to-noise ratio. Two saturated saline (35% of NaCl) phantoms were used for image analyses; (1) a needle (ID: 0.7 mm) placed at the isocenter of the field-of-view for assessing the spatial resolution and (2) two tubes (ID: 3 mm) placed 7.5 mm away from the center for evaluating the contrast. Four different image reconstructions were performed; (a) Default filtered back projection (FBP), (b) Delay-and-Sum (DAS), (c) Universal back-projection (UBP [6] ), and (d) Model-based algorithm (MBA [7] ). Algorithms (a), (b), and (c) are the most commonly used algorithms, while (d) accounts for the effects of acoustic absorption, detector element, impulse response, and directivity pattern. Quantitative image assessments were performed via a contrast-to-noise ratio (CNR) analysis and line profile assessment. The CNR was evaluated from the two-tube data set using Equation 1 (Persuasive data). Three regions-ofinterest (ROIs) with a diameter of 3 mm were selected; two ROIs for signal tubes ( Figure 2 ) and one for the background. In addition, a line profile was evaluated from the needle phantom image by drawing horizontal/vertical lines through the needle and calculating the full-width at half maximum (FWHM) of a gaussian fit. Results Two quantitative metrics were used to evaluate the image reconstruction performance; CNR and line profile ( Figure 2 ). The CNR for the MBA was 44.4±27.1% higher compared to other methods ( Figure 3 ). For the line profile analysis, the MBA showed significantly higher accuracy compared to other methods in the needle phantom image ( Figure 4 ). The FWHM of MBA needle image, 0.6 mm, showed significantly lower error, while other algorithms suffered from some blurring artifacts with FWHMs of 1.3 mm. These parameters may serve as a quantitative and intuitive parameter to assess the performance and quality of clinical investigations. Conclusion The image reconstruction methods were evaluated for the RATMI system. The preliminary results produced by MBA were promising in CNR and line profile assessments; however, further evaluations through a resolution phantom and small animal studies are needed to fully assess and compare the imaging performance. Abstract Body : Introduction: Tissue sodium concentration (TSC) is a sensitive indicator of disease. Changes in tissue sodium distribution are implicated in many diseases1. Sodium concentration increases in cancer cells as a result of aerobic glycolysis2. Longitudinal changes in TSC could potentially be a useful tool for understanding treatment response. Sodium has previously been imaged preclinically with magnetic resonance imaging (MRI) at high field strengths3. Since sodium has limited in vivo concentration, low gyromagnetic ratio, and quadrupolar relaxation (T1 = 10-40ms, T2,fast = 1-5ms, T2,slow = 15-30ms), adequate signal-tonoise ratio (SNR) is difficult to achieve1. The density-adapted three-dimensional projectionreconstruction (DA3DPR) sequence has been demonstrated to achieve improved sodium SNR compared to cartesian imaging sequences, and without exotic hardware requirements4. Improved sodium imaging adapted for preclinical studies at 3T lends significance to eventual translation of sodium imaging for clinical use. Hypothesis: Implementation of a DA3DPR sequence at 3T for preclinical sodium MRI will be able to detect changes in the TSC of rat glioma as a result of altered metabolism. Methods: DA3DPR imaging was optimized on a GE Discovery MR750 3.0T MRI. Performance was evaluated at 1-mm and 3-mm isotropic resolution (TE=0.5 ms, TR=100 ms, Taq = 8 ms, flip angle (FA)=90°, gradient slew rate=20 mT/m, isotropic FOV=80 mm). Sodium imaging was performed using a purpose-built, curved transmit/receive butterfly radiofrequency (RF) surface coil. A phantom of 50-mmol/L sodium (physiological sodium concentration) and 1% agarose was imaged to determine the coil sensitivity profile. Reference vials containing 2mL of 30-, 50-, 70-, and 100-mM concentrations of sodium were placed within the imaging area of the coil. Reproducible RF excitation of the imaging area was maintained using a reference vial containing 2-M sodium doped with 200 mM of the shift reagent Tm-DOTP. DA3DPR imaging at 3-mm resolution was further evaluated in an animal model. Stereotactic surgery was performed to implant 1´106 C6 glioma cells into the right hemisphere of a Wistar rat brain. Seven days following implantation, rats were imaged with DA3DPR sodium MRI (NEX = 10, total acquisition time = 12 mins) and 3D T2-weighted proton MRI (CUBE) every 3 days until experimental endpoint. Proton MRI was performed using a custom birdcage for small animal imaging (ID = 85.7mm). Regional intensity variations in the sodium imaging data due to RF inhomogeneity from the surface coil were corrected by normalization to the measured coil sensitivity profile. In vivo sodium images were manually co-registered to the 3D coil sensitivity data with 3D-slicer using fiducial markers visible in both imaging data sets. This yielded normalized in vivo maps of absolute TSC (in units of mM). Manual segmentation of the tumour and healthy brain were performed using ITK-SNAP to determine differences in TSC. Conclusions: A DA3DPR sequence for sodium imaging was successfully implemented with a 3mm isotropic resolution at 3T for preclinical imaging research. Our implementation can visualize regional changes in sodium with enough SNR (> 30) for assessment of brain tumours. TSC concentrations within the area of interest showed a standard deviation of ± 3.02-mM, demonstrating successful correction of surface coil related intensity variation. This correction was applied in vivo with longitudinal monitoring of TSC in rats implanted with C6 tumours showing increased sodium signal in areas of the tumour compared to the healthy brain. Future work will focus on increasing the study size as well as monitoring longitudinal changes of TSC in response to therapy. (1). Although studies with positron emission tomography (PET) have shown these drugs do target mGluR5 (1), mGluR5 expression has not been measured in FXS (2) . The goal of this study was to demonstrate the feasibility of measurements of mGluR5 expression in the living brain of humans with FXS. Methods Positron emission tomography (PET) with a high resolution research tomograph (HRRT) was conducted at for 90 minutes after the intravenous bolus injection of 185 megabequerels (MBq) (5 (3) to two men with the full mutation for fragile X syndrome (FXS) aged 24-27 (25.5+2.12) years, six men with ASD aged 18-23 (20+2.10) years, and three individuals (one man and two women) with typical development (TD) aged 19 to 24 (21+2.9) years (4). The nondisplaceable binding potential (BPND) (5) was calculated for relevant volumes of interest (VOIs) on PET images coregistered with magnetic resonance imagery (MRI) (3) . First, VOIs of cortical regions were obtained automatically by Freesurfer 6.0 (6, 7) and VOIs of subcortical regions were obtained automatically utilizing the subcortical segmentation tools of the software library of the Oxford Centre for fMRI of the Brain (8-10). We used cerebellar white matter (CbWM) for the reference region since this structure demonstrated negligible densities of mGluR5s (11). Then, VOIs were transferred from MRI to PET space according to MRI-to-PET coregistration parameters obtained with the coregistration module (12) of Statistical Parametric Mapping (SPM) (13), and applied to PET frames to obtain regional time activity (radioactivity) curves (TACs). Regional BPNDs were obtained by reference tissue graphical analysis (RTGA) (14) and the multilinear reference tissue method with two parameters (MRTM2) (15). Results In contrast to both the ASD and TD groups, BPNDs for the men with FXS were reduced in all VOIs The nondisplaceable binding potentials (BPNDs) (5) of [18F] FPEB by both reference tissue graphical analysis (RTGA) (14) and the multilinear reference tissue method with two parameters (MRTM2) (15) in each VOI of two men with FXS and ASD are greater than a standard deviation below the means of six men with autism spectrum disorder and three healthy people with typical development (3, 4) (Figure) . Reductions in the men with FXS were particularly marked in the limbic system and the striatum (Figure) . Conclusion This study supports the viability of PET to measure mGluR5 density in FXS (16). The results are confirmed through performance of a similar protocol on six participants with FXS (17). The marked reductions of mGluR5 density in the cortices and the limbic system support the finding that two distinct but interrelated social behavior abnormalities, one linked to impaired cognitive processes (delayed socialization) and the second one to disturbance in limbic circuits (avoidance), play a role in the development of ASD in boys with FXS (18). The protocol suggests that PET may be used to quantify a mGluR5 binding to confrim target engagement for clinical trials of novel glutamatergic agents for FXS. 

millicuries [mCi]) 3-[18F]fluoro-5-(2-pyridinylethynyl)benzonitrile ([18F]FPEB), a potent, selective mGluR5 inhibitor

Jae Yong Choi, Korea Institute Radiological and Medical Sciences, smhany@kirams.re.kr Category: Neuroscience Abstract Body : Purpose Although colony-stimulating factor 1 receptor (CSF-1R) inhibition has been shown to reduce neuroinflammation levels in the animal brain, it is unclear whether this therapeutic approach leads to the improvement of PD symptoms. Therefore, the purpose of the present study is to address the anti-inflammatory influence on behavioral and neuronal circuits. Methods In PD and PLX3397 treatment in PD groups, the rats (n=6) were pretreated with desipramine hydrochloride (12.5 mg/kg i.p.). After ketamine/xylazine anesthesia (40mg/kg and 5mg/kg, respectively), 6-OHDA (8 ug in 2uL/site in 0.2% ascorbic acid) was injected into the four coordinate sites in the right striatum. The inserted needle was withdrawn after 5 min, and the wound was sutured. In the sham group, the procedure was exact the same for the PD (n=6) except that the 2% ascorbic acid saline was injected into the striatum. PLX3397 treatment (Tx) in PD groups (n=6) was given to PLX3397 treatment for 21 days in the PD group. After the treatment of PLX3397, all groups underwent behavioral tests. Then, PET and MR images were acquired sequentially. Results Activated microglia depletion by the administration of CSF-1R inhibitor causes an improvement of motor function and depressive-like behavior. In addition, this therapeutic treatment generates partial recovery of dopaminergic and glutamatergic systems as well as neuroinflammation. In the immunoassay, we can observe the significant reduction of activated microglia in the drug treatment group. Conclusions In the present study, we demonstrated that microglia depletion suppressed the neuroinflammatory progress, and which has a beneficial effect on motor and non-motor symptoms of PD. Abstract Body : Purpose Alzheimer's disease (AD) is the most common cause of dementia in elderly individuals, which leads to problems including memory loss, cognitive deficits, and intellectual disabilities. The therapeutic drugs developed so far have various side effects such as dizziness, headache, and confusion. Therefore, it is necessary to develop alternative therapeutic treatments that are free from adverse effects. Taurine, 2-aminoethanesulfonic acid, is the second most abundant endogenous amino acid after glutamate, in the central nervous system. The chemical has nontoxic properties in the body, so taurine has been used in foods and drugs for the treatment of liver and heart diseases. In the present study, we investigate the effect of taurine supplementation in AD mice by glutamate PET. Materials Three groups of mice have used for these studies; the B6/SJL F1 hybrids (WT, n=5), 5xFAD mice (AD, n = 5), and 5xFAD mice that were orally administered the taurine as drinking water from 2 months to 9 months of age (ADTaurine, n=5). Taurine was administered at a dose of 1,000 mg/kg/day. At the time at 9 months of age, all groups underwent the glutamate PET study. Mice were anesthetized with 2.0% isoflurane, and 18F-FPEB (9.4 ± 1.2 MBq/200 uL) was injected through a tail vein. Simultaneously dynamic PET scan was performed for 60 min and images were reconstructed with the 3-dimensional ordered subset expectation maximization algorithm with 4 iterations (14 × 30 s, 3 × 60 s, 4 × 300 s, 3 × 600 s, totally 24 frames). The cortex, striatum, hippocampus, thalamus, and cerebellum were selected as volumes of interests (VOIs). The decay-corrected regional time-activity curves (TACs) were acquired from VOIs and normalized in units of standardized uptake value (SUV), which is calculated to normalize for differences in injected dose and body weight. Pharmacokinetic parameters (Cmax, Tmax, AUC) were estimated from TACs. In addition, the regional distribution volume ratio (DVR) was estimated using Logan graphical analysis. Results By visual inspection, the AD group showed significantly lower striatal and hippocampal 18F-FPEB uptake than those of the WT group. The ADtaurine group showed relatively higher uptakes than those of AD group. In AD group, all AUC values of the target regions were 27−38 % decrease compared with the corresponding values in WT mice. The taurine group showed a 37−46 % higher AUC value than the AD group. Cmax values of the target regions for WT were between 1.95 and 2.48 min, and the Cmax values for the AD were comparable (1.67-2.44 ). However, the Cmax values of ADtaurine increased relative to other groups (2.41-3.30 ). 18F-FPEP reached a maximum concentration at 3.25 min p.i in WT. However, the peak arrival times of the AD and ADtaurine groups were relatively fast (AD: 1.55-2.34 min, ADtaurine: 1.35-1.45 min). The mean DVR values for AD group were 29−42 % lower than those for WT. However, ADtaurine group showed 15−20 % higher DVR values compared to those for AD. Conclusions In summary, glutamate PET showed successfully differentiate the disease severity of AD mouse model. Moreover, the improvement of glutamate-related function in the brain by taurine was observed. From these perspectives, taurine caused glutamate related therapeutic effect in AD. Moreover, mGluR5 PET could be a useful tool to reflect the therapeutic effect of taurine in AD rodents, which is expected to be the basis for future studies for early diagnosis of AD. Abstract Body : Tumor organoids are self-organizing three-dimensional micro-tumors that are grown from patient-derived tumor tissue. They closely recapitulate the structural and functional tumor microenvironment of the original tumor, and therefore, function as a superior personalized cancer model1. Although emerging evidence indicates that organoids can be used to accurately predict drug responses in a personalized treatment setting, clinically relevant methods to perform high-throughput drug and immune-therapy screens are still missing. Here we present a proof of concept for using positron emission microscopy (o-PEM) to screen drug response of tumor organoids derived from head and neck cancer patients. Using 18-Fluorodeoxyglucose (FDG) as a radiotracer, we show that high-resolution images of these organoids can be captured rapidly revealing metabolic maps which in combination with brightfield imaging functionally resembles the PET/CT images from the clinic. Tumor organoids were fed with FDG and mounted on a thin CdWO4 scintillator plate. Scintillator flashes were directly imaged by a single-photon-sensitivity electron-multiplying charge-coupled devices (EMCCD) camera as previously described by Kim et al in their radioluminescence imaging platform2. The EMCCD signal was calibrated with known radioactivity of FDG for dose calculation or accurate quantification of PEM images. A reference fluorescence imaging with live/dead cell marker revealed that PEM imaging recognizes the viable region of a patient-derived tumor explant with high signal to noise ratio. Cisplatin treatment of tumor organoids caused significantly lower signal or metabolic activity in the PEM images, while organoids derived from cisplatin-resistant patients showed higher metabolic activity. The highest resolution of PEM (18µm) for organoid imaging was found to be 200 fold better than the standard PET images (5mm). Unlike a cell viability assay, this method can accurately identify the cell fate much early form its metabolic indication. This is the very same reason why FDG-PET has proved to be the most useful way to monitor therapeutic response in the patients. Microdosimetry calculation reveals the exact doses accumulated in each organoid. The quantification was performed with respect to a calibrated signal obtained from a known dose. The obtained time-dependent data falls very close to the radioactive decay curve of 18-F even up to a very low activity and camera signal, confirming the reliability of microdosimetry conversions. In this technique, the photon budget is only dependent on the radioactivity of 18-F, which is an energy-independent autonomous long term phenomenon and warrant the excess amount of photon yield in expense of imaging time or speed. Finally, the influx rate constant (Ki) of FDG into the tumor organoids was estimated with an approximation of the standard two-tissue compartmental model with irreversible uptake. The influx rate into the tumor organoids from multiple patients was found to be similar to the patient tumor from which the organoids were grown. Our data suggest that the metabolic heterogeneity of the patient tumor is significantly conserved in the organoid tumors in vitro and o-PEM can functionally recapitulate the original PET imaging of the patient performed in the clinic. In conclusion, due to the advantages of improved resolution and rapid data acquisition ability in a quantitative manner, this in-vitro PEM imaging platform has the potential to complement PET imaging in the clinic and emerge as the next-generation high-throughput tool for cancer precision therapy. Much remains unknown about the mechanism of action and the effects of the disease. Although the clinical manifestations are dominated by respiratory symptoms, other effects of the disease include dysfunction of other organs, such as cardiology [1, 2] , liver [3] , head [4] , and gonadal [5] . Previous reports mainly used different indicators to in-directly evaluate possible damages of different organs (such as blood pressure, CK-MB, and high-sensitivity cardiac troponin I for the cardiovascular system [1, 2] ; alanine aminotransferase and aspartate aminotransferase for liver [3] ; and luteinizing hormone for gonadal effects [5] ). Additional imaging techniques are needed to directly explore the details of these possible damages thus addressing the possible mechanisms. Herein, PET/MR hybrid scanner was used on a severe COVID-19 patient to investigate the effects of the virus infection on different organs (heart, spleen, head, liver, and lymph nodes). The patient underwent PET/MR scan using 18F-FDG as the PET tracer. T2WI and T1WI sequences were used for liver and spleen MR; T2-FLAIR and DTI(FA) were used for head MR; Cine, Black blood, and late gadolinium-enhanced sequences were used for cardiac MR. The detailed analyses of the MR, PET, and fusion images, were conducted. From the PET/MR images, no remarkable cardiovascular, spleen and brain damage was observed. While the uptake of 18F-FDG in liver and lymph nodes were increased. The incensement of 18F-FDG uptake in liver and lymph nodes could be originated from the inflammatory response. The absence of cardiovascular effects suggests that the distribution of ACE2 and the organs' damage are possibly not correlated in this case. Further studies about individual differences as well as the mechanisms of action are required. Continuous tracking of the patient is in progress to give more information. Abstract Body : Introduction: Breast cancer remains one of the most common cancers worldwide. When breast cancer results in brain metastases, this fate is typically fatal for patients [1] . In order to combat this, therapies need to be made to treat brain metastasis. While there are existing treatments such as surgery and radiation treatment, the life span of patients remains fairly low, with a median overall survival of 7 months [2] . Targeted therapeutics may not be able to cross the blood brain barrier (BBB), or have limited penetration, resulting in reduced efficacy in treating these metastases [3] . Therefore, it is important to examine potential therapeutic delivery vehicles that may cross the BBB, such as extracellular vesicles (EVs). EVs are small particles released from cells, and are a naturally occurring method of cell to cell communication that have been implicated in a number of biological processes, including metastasis [4] . EVs can be exploited to deliver a therapeutic agent to recipient cells, allowing them to be used as a potential treatment for various diseases including cancer [5] . They are also capable of crossing the BBB, which would allow them to deliver a therapeutic to brain metastases [5] [6] . However, before they can be used as such, it is important to understand the timing, localization, and accumulation of EVs during metastatic progression. Magnetic particle imaging (MPI) can aid in characterizing EVs in metastases after labeling them with superparamagnetic iron oxide nanoparticles (SPION). MPI is a new sensitive imaging method that is specific to superparamagnetic iron and provides quantitative data. Methods: Murine breast cancer (4T1) and macrophage (J774A.1) cells were labeled with SPIONs, using protamine-sulfate to enhance uptake. Following 24-hour labeling, the cells were washed to eliminate any extracellular iron and iron-labeled EVs (Fe-EVs) were collected by ultracentrifugation 24 hours later. Nanoparticle tracking analysis (NTA) provided information on size and a measure of concentration of Fe-EVs. Fe-EV pellets were imaged by MPI and iron content was quantified using a calibration curve. For in vivo experiments, Balb/C mice will receive an intracardiac injection to deliver 4T1-BR Fluc/GFP cells, a breast cancer cell line that forms brain metastases and stably expresses firefly luciferase and fluorescence; this allows for imaging of the metastatic sites using bioluminescence imaging (BLI). J774 or 4T1 Fe-EVs will then be injected into the mice at either early or late stages of metastatic growth, and subsequently imaged using MPI and BLI to determine timing and accumulation of EVs arriving to the metastatic site. Results: MPI was used to image J774 ( Fig 1A) and 4T1 (Fig 1B) Fe-EV pellets. A calibration curve (Fig 1C) was generated and was used to quantify the total amount of iron within each Fe-EV pellet. NTA identified the average size (107.3 nm & 98.5 nm) and concentration of the Fe-EVs (Fig 1D, E) , allowing for a measure of 213 and 222 attograms Fe/EV, for J774 or 4T1 Fe-EVs, respectively. With this extent of ironloading, a lower detection limit of 2.5x108 Fe-EVs could be identified using MPI, although high sensitivity imaging could reduce that number further. Future experiments will use MPI and BLI to identify Fe-EVs within brain metastases, in vivo. Imaging will be validated by histological examination of the brain. Conclusions: This multi-model imaging technique will demonstrate that EVs can cross the blood brain barrier and accumulate in metastatic sites. These findings will allow for further study of EVs for use as a therapeutic delivery vehicle to treat brain metastasis.

References: [1] 9E9 particles) collected after ultracentrifugation and resuspended in phosphate buffered saline. The white outline represents the axial view of the conical tube containing the pellet of Fe-EVs. A calibration curve was generated using the total MPI signal from a known amount of iron (D) . Particle concentration was calculated using NTA, and the size of these particles for (C) J774 Fe-EVs and (D) 4T1 Fe-EVs were similar to the average size of EVs. Abstract Body : Introduction Oncolytic virotherapy (OV) is a promising treatment for high mortality cancers [1] . Non-invasive monitoring of OV is essential for optimizing the clinical outcome and providing an improved understanding of the interactions between the virus and its tumor-host. Chemical exchange saturation transfer (CEST) MRI is a molecular imaging technique that may shed new light on OV, as it is capable of detecting protein concentration and pH changes. However, clinical translation of CEST methods has been hindered by the qualitative nature of the image contrast and the long image acquisition times. Objective The goal of this work was to develop a deep-learning-based CEST MR fingerprinting (MRF) method for quantitative and rapid multi-pool imaging of OV treatment response. The approach was evaluated in mice undergoing virotherapy treatment and translated to clinical scanners. Methods CEST-MRF Acquisition Two SE-EPI CEST-MRF acquisition protocols (105s each) were employed sequentially, designed for obtaining MT and amide proton exchange-rate and volumefraction maps. Both protocols varied the saturation power, but employed different saturation offsets (first: 6-14 ppm, second: 3.5 ppm). T1, T2, and B0 maps were acquired using variable repetition-time, multi-echo spin-echo, and water saturation shift referencing [2] , respectively. CEST-MRF Dictionary Generation The CEST-MRF signal trajectories for 184,800 and 70,224,000 multi-pool parameter combinations were synthesized for the two protocols using a numerical solution of the Bloch-McConnell equations [3] [4] . Deep Reconstruction Networks To avoid the exceedingly long dictionary matching-time required for conventional dot-product MRF [5] and to improve the multi-parameter reconstruction ability, image reconstruction was performed using a series of two Deep RecOnstruction NEtworks (DRONEs) [6] , trained on the synthesized data. The pixel-wise signal trajectories from the MT specific MRF schedule were input to the first DRONE, together with the water T1, T2 and B0 values. The two MT exchange parameter outputs, together with the water pool and B0 parameters were then input into the second DRONE, together with the pixel-wise signal trajectories from the amide-pool MRF schedule. In Vivo Imaging U87 tumors were implanted in the brain of 16 mice (25% used as control). The mice were imaged at 8-11 days after implantation, using a 7T preclinical MRI (Bruker, Germany). Next, a herpes simplex-derived oncolytic virus, NG34, was inoculated, and the mice were imaged 48hrs and 72hrs later. For clinical translation, the continuous wave saturation pulse was replaced by a train of spin-lock saturation pulses and the read-out was done using GRE-EPI. A healthy volunteer was recruited and imaged at 3T (Siemens Healthineers, Germany). Results Before virus inoculation, the semi-solid and amide proton concentrations were both decreased in the tumor, consistent with increased edema decreasing the protein concentrations. The tumor amide proton exchange-rate was increased, indicative of increased intracellular pH. Following OV, the core of the tumor presented significantly lower amide proton concentration and exchange-rate compared to the tumor rim and the contralateral region. Both effects are indicative of apoptosis as it is known to inhibit protein synthesis [7] and decrease cytosolic pH [8] . At baseline, control mice demonstrated similar MT and amide effects as OVtreated mice, however, no apoptotic process was evident. The preclinical MRI findings were in good agreement with the histology and immunohistochemistry findings (HSV antibodies, H&E, Caspase-3, Coomassie). The normal human subject parameter maps yielded MT volume fraction and exchange-rate values in good agreement with the literature [9] . The amide proton exchangerate was in good agreement with previous water exchange spectroscopy-based measurements [10] . The total reconstruction time for the 4 molecular maps using the neural-network scheme was 94 ms, compared to 2.3 hrs for conventional dot-product reconstruction. Conclusions The deep CEST-MRF technique successfully and rapidly quantified pH and molecular concentration changes, potentially serving as important biomarkers for OV-induced apoptosis.

References: [1] Russell, S. J., et al., Nature biotechnology 2012;30 (7) Abstract Body : Purpose: Molecular imaging and targeted theranostic agents are playing an increasing role in prostate cancer detection and therapy1. CD46 represents a novel prostate cancer (PCa) biomarker with high expression in aggressive and late stage PCa such as neuroendocrine prostate cancer2,3. ImmunoPET imaging is a noninvasive molecular imaging modality combined the excellent targeting specificity of antibodies or antibody fragments and the superior sensitivity and resolution of PET4,5. In this work, we developed a novel CD46-targeted positron emission tomography (PET) radiotracer, [89Zr]DFO-YS5, and evaluated its performance for immunoPET imaging in murine PCa models. Materials and methods: The radiolabeling of 89Zr oxalate to DFO-YS5 or control antibody was accomplished in a typical, two step procedure by first conjugating with desferoximine (DFO), and then subsequent chelation of the isotope. Mice bearing DU145 and 22Rv1 xenografts or LTL-331 patient derived xenografts were grown to at least 0.5 cc, and PET imaging was performed on a Siemens Inveon micro PET/CT system. Results: [89Zr]DFO-YS5 was isolated in 74 ± 11% (n = 6) yield based on starting 89Zr oxalate with molar activities ranging from 7.42 mCi/mg to 9.49 mCi/mg (274.2 MBq/mg -351.1 MBq/mg). Purity was greater than 95%. PET imaging and biodistribution analysis of [89Zr]DFO-YS5, [89Zr]DFO-IgG (control antibody) and 68Ga-PSMA-11 was performed in mice bearing DU145 (CD46+, PSMA-) and 22Rv1 (CD46+, PSMA+) xenografts. Both DU145 and 22Rv1 showed high uptake of [89Zr]DFO-YS5 (13.3±3.9% ID/gram and 11.2±2.5% ID/gram respectively, Figure C, E). As a comparison, control antibody [89Zr]DFO-IgG showed a low uptake in DU145 and 22Rv1 xenografts (Figure D, E) . In 68Ga-PSMA-11 imaging, the PSMA negative cell line DU145 showed very low uptake compared to the PSMA positive cell line 22Rv1, which demonstrated 4% ID/g uptake. Taken together, these data demonstrate the feasibility of imaging CD46 positive, PMSA negative tumors with [89Zr]DFO-YS5. PET/CT analysis of the LTL-331 patient derived xenograft model also demonstrated high tumoral uptake of [89Zr]DFO-YS5. Human equivalent dosimetry was performed in wild type male mice, demonstrating a radiation dose of 0.447 ± 0.043 mSv/MBq. Discussion: Taken together, the data presented herein demonstrate that [89Zr]DFO-YS5 is an effective radiopharmaceutical for imaging CD46 positive prostate cancer. CD46 is highly expressed in PCa, particularly in de-differentiated forms of prostate cancer including treatment emergent small cell/neuroendocrine prostate cancer, where expression is generally higher than PSMA. Theranostic targeting with prostate specific membrane antigen (PSMA) is highly promising for both imaging and therapy for prostate cancer. However, some PCa cells express low level of PSMA, which could not be detected through PSMA based imaging7. Our imaging study in the dual DU145 and 22Rv1 model mimics the situation seen in the clinic, with heterogeneous target expression. Thus, CD46 targeted imaging and therapy could be a complementary option for PSMA negative PCa. Conclusion: Our data demonstrate that [89Zr]DFO-YS5 imaging provides excellent imaging contrast with high uptake specific for CD46-expressing prostate cancer cells, including in PSMA negative tumors and patient derived xenografts. Overall, the novel radiotracer [89Zr]DFO-YS5 represents a promising candidate for clinical translation as an immunoPET agent for the detection of CD46-positive primary and metastatic prostate cancers. MBq/mg). The in vitro study has shown that [89Zr]DFO-YS5 had a high binding affinity for CD46 at 6.0 ± 0.6 nM. With the radiotracer in hand, a multi-time point micro PET/CT and biodistribution study was performed. Mice with CD46 positive DU145 xenograft model was imaged and their organ were harvested and measured at different time point. It showed a specific uptake at tumor site as early at 24 hours post injection. Biodistribution data revealed that the tumor uptake as measured in increased over time, measuring up to 18.8±11.3 % ID/gram at day 7. In contrast, the blood pool retention, which started at 10.8±1.2 %ID/gram at 24 hours, decreased to 3.8±1.6 %ID/gram % by 168h. The uptake at other organs remained below 5% ID/gram all the time, providing an excellent imaging contrast. To further evaluate the imaging ability of [89Zr]DFO-YS5, and compare it to the PSMA based prostate cancer imaging agent 68Ga-PSMA-11, [89Zr]DFO-YS5 or 68Ga-PSMA-11 were administered to athymic mice xenografted with 22Rv1 cells subcutaneously. Mice administered with [89Zr]DFO-YS5 were imaged 4 days post i.v. injection and then sacrificed for a biodistribution study. The probe could specifically localize at tumor site and had a low accumulation at other organs. The biodistribution study showed that the tumor uptake was 14 % ID/gram in average, the uptake at other organs was all below 5%. Mice administered with 68Ga-PSMA-11 showed high tumor, spleen, and especially kidney uptake. The biodistribution study showed 4% ID/gram uptake at tumor and over 120% ID/gram uptake at kidney because of its high expression level of PSMA. These results demonstrate that [89Zr]DFO-YS5 can specifically image CD46 positive prostate cancer with favorable imaging characteristics 68Ga-PSMA-11 in the 22Rv1 model. To validate [89Zr]DFO-YS5 that specifically targets CD46 positive tumor, and compare the uptake difference between CD46 positive tumor and CD46 negative tumor, a PET imaging and biodistribution study was performed in mice bearing CD46 positive DU145 and CD46 negative MC38 xenografts (see persuasive data figure). Mice were administered [89Zr]DFO-YS5, imaged at day 4 followed by a biodistribution study. The uptake at DU145 xenograft was 15 ± 5% ID/gram, almost three folds of the uptake at MC38. The DU145 tumor/muscle uptake ratio could reach 90, which is also almost 3 folds of MC38 tumor/muscle ratio, suggesting a good imaging contrast and high specificity for CD46-expressiong xenograft. Overall, [89Zr]DFO-YS5 showed high binding affinity to CD46, excellent imaging contrast against CD46 positive xenograft DU145 and 22Rv1 in vivo. The imaging was also verified by a control experiment using CD46 negative xenograft MC38. Taken together, our data demonstrated that [89Zr]DFO-YS5 can be specifically taken up by CD46 positive PCa xenograft, including in PSMA negative tumors, demonstrating strong potential for clinical translation. 

Intraoperative imaging technique to highlight peripheral nerves during cancer surgery using spectral reflectance Ngai Nick Alex Wong, The Hong Kong Polytechnic University, 18041854r@connect.polyu.hk

Abstract Body : Nerve preservation is vital for surgical treatment to reduce postoperative adverse effect. Accidental nerve transection during cancer surgery might result in significant complications such as chronic pain, sensory loss, or even ability loss such as erectile dysfunction and urinary incontinence after prostate cancer surgery, and abnormal vocal cord function after thyroid cancer surgery. Unfortunately, there is lack of reliable technique to identify peripheral nerve in the current operating room. Surgeons typically rely on appearance, tactile information, and invasive electromyography to distinguish nerves. Real-time visualization of nerves with high specificity and safety during surgery is critical to guide surgeon to avoid nerve damage or transection. Here, we develop a label-free imaging strategy to provide specific and safe intraoperative imaging guidance for detecting peripheral nerves. Focusing on strong wavelengthspecific reflectance from myelin sheaths surrounding nerve fiber by thin film interference phenomenon, we demonstrate myelin induced reflectance imaging can be useful to specifically visualize peripheral nerves during surgery without any exogenous label or other invasive procedures. Using this label-free technique, we visualized the reflectance signal change of nerve and non-nerve tissues at different wavelengths and different depths under ex vivo and in vivo conditions with multiphoton microscope equipped with wavelength tunable laser and reflectance detection filters. The detected patterns highlight several peripheral nerves with myelin showing strong reflectance at 820-900 nm over other non-nerve tissues including muscle, fat, and vein. Excellent nerve-to-muscle contrast was shown using spectral reflectance technique (contrast value of 5.1 at 850nm) compared to conventional nerve staining dye, fluoromyelin (contrast value of 1.9 at 850nm). Notably, evaluation of intraoperative imaging performance with multiple murine cancer models demonstrated significantly higher nerve-to-muscle contrast in specific wavelength regions in cancer models compared to normal mice. Imaging in breast cancer model depicted contrast value of 3.2 at 900nm and 4.3 at 960nm whereas normal model imaging showed 2.0 at 900nm and 2.3 at 960nm. This result implicates structural change of nerve by interaction with cancer cells resulting in enhanced myelin-induced reflectance signal which provides potential for clinical translation of our developed nerve identification technique in real cancer surgery. Intraoperative imaging in xenograft models of 4T1-lec2-RFP breast cancer and K562-GFP myelogenous leukemia showcased successful visualization of peripheral nerves under heterogenous cancer environment. Our novel technique provides high potential for a powerful navigation tool highlighting the exact location of peripheral nerves with high specificity and no toxicity to guide surgeons, which allows complete preservation of nerve structures. This enables minimally invasive surgery to improve overall patient outcome and to provide better quality of patient's life after cancer treatment. Abstract Body : Introduction: The immune system is complex, with different activities occurring at different anatomical locations. Autoreactive activated lymphocytes are responsible for pathology in diseases such as multiple sclerosis (MS) and promote the activation of additional leukocyte populations. Limiting the activity of these activated cells is the main goal of immunomodulatory therapies1,2. However, there is no strategy that specifically images immune cell activation as a whole. Positron emission tomography (PET) imaging can non-invasively image biochemical processes in vivo3. 18F-FAC is a deoxycytidine analogue PET radiotracer that measures deoxyribonucleoside salvage, a pathway that is enriched in activated immune cells4,5 and controlled by the rate-limiting enzyme deoxycytidine kinase (dCK). Objective: To characterize 18F-FAC PET across disease progression in experimental autoimmune encephalomyelitis (EAE), a mouse model of MS, to investigate whether 18F-FAC PET can function as a biomarker for an immunomodulatory therapy in EAE, and to determine whether the deoxyribonucleoside salvage pathway has functional relevance in immune activation. Methods: To induce EAE, 10-week-old female immunocompetent C57BL/6 were injected with myelin oligodendrocyte glycoprotein in Freund's complete adjuvant and pertussis toxin. To study 18F-FAC PET across disease progression, EAE and control mice were imaged on Day 7 (presymptomatic state), Day 14 (peak disease), Day 21 (stabilized state), and Day 28 (late disease) post-induction. Ex vivo biodistribution studies on perfused EAE mice were conducted to validate the image quantification. To determine whether 18F-FAC can be a biomarker for immunomodulatory treatments, mice were treated with fingolimod (0.5 mg/kg) or vehicle and imaged on Day 14. TRE-515 is a dCK inhibitor. To start to study the deoxyribonucleoside salvage pathway in EAE, we evaluated the pharmacodynamics of TRE-515 in EAE. C57BL/6 female mice were treated with TRE-515 (75 mg/kg) 16 and 4 hours before 18F-FAC PET imaging. For all imaging, mice were injected with ~40 µCi of 18F-FAC and imaged on the HiPET one-hour post-injection. Results: Immune activation differs in different organs across different time points in EAE and can be visualized using 18F-FAC PET. 18F-FAC accumulation was increased two-fold in the brain, spleen, and lymph nodes by day 7, even before clinical symptoms. This accumulation was maintained in the brain throughout day 14, and in the spleen and lymph nodes through day 28. Biodistribution studies confirmed the PET quantification results. Ex vivo studies of cells isolated from the brain showed that CD4 and CD11b cells accumulate the radiotracer at similar levels. 18F-FAC PET can be a biomarker to examine the efficacy of immunomodulatory treatments in EAE. Fingolimod is an immunomodulatory drug used in MS that sequesters leukocytes in the lymph nodes and spleen. Fingolimod significantly decreased brain 18F-FAC accumulation by two-fold in fingolimod compared to vehicle-treated EAE mice. Fingolimod had no effect on 18F-FAC accumulation in the spleen and lymph nodes. TRE-515, a small molecule inhibitor of dCK, blocks the deoxyribonucleoside salvage pathway in vivo. Mice treated with TRE-515 accumulated two-fold less 18F-FAC in the spleen than control treated mice, suggesting that TRE-515 has efficacy in vivo. Future studies will evaluate how TRE-515 effects immune cells in EAE mice. Innovation/Impact: These results are the first to show that 18F-FAC PET can be used as a biomarker for immune activation and assessing immunomodulatory therapies. This study is also the first to assess the role of the deoxyribonucleoside salvage pathway in immune activation. Understanding the biology behind 18F-FAC PET can enhance our understanding of the PET images. If translated into the clinic, patients with autoimmune diseases could benefit from using PET imaging to guide immunomodulatory treatments. KP46 did not deliver 68Ga to tumour (2.6±0.9 %ID/g) and most activity concentrated in liver (28.4±6.6% ID/g) and heart (16.9±2.1%ID/g). 68Ga acetate, on the other hand, showed tumour accumulation (5.8±2.4 %ID/g) probably by transferrin dependant mechanism. In group B, most activity remained in the large intestine (61.8 ±21.5%ID/g), with minimal trafficking of radioactivity to other tissues. The majority of radioactivity in tissue samples was in a hydrophilic form, i.e. no longer in the form of [68Ga]KP46. Adding therapeutic dose of KP46 to group C showed no trafficking change and most activity remained in the large intestines (56.5±21.4%ID/g). ICP-MS measurement of gallium-69 in tissue samples was consistent with 68Ga trafficking results. Conclusion: After oral administration of [68Ga]KP46 either as tracer or as bulk drug, gallium is not absorbed from the gut and is not delivered to tumour within the timeframe investigated. We suggest that in this short timescale, delivery of gallium to tumour is not a significant component of the mode of action of this drug. Delivery of 8-hydroxyquinoline could be investigated in the future as an alternative hypothesis for its mode of action. Abstract Body : Manganese porphyrins are known to have several therapeutic and imaging applications. Their superoxide dismutase activity makes them efficient tumour chemo/radiosensitisers -with several agents in clinical trials. [1] Additionally, their paramagnetism allows them to be used as MRI contrast agents. [2] Porphyrins also have a wellestablished affinity for lipid bilayers which also makes them candidates as cell and liposome labeling agents. It was hypothesised that metallation with the PET radionuclide 52Mn (t1/2 = 5.6 d) would allow long-term in vivo biodistribution studies of Mn-porphyrins as well, as a method to label and track cells and liposomes. Despite this, methods for fast and efficient radiolabelling of radiomanganese porphyrins are lacking. In this work, a radiosynthesis method for 52Mnporphyrins was designed, and the cell and liposome labelling ability of these agents were then tested. 52MnCl2 and the porphyrin ligands 1 -4 ( Fig. 1A) were produced as previously reported. [3, 4] Porphyrins 1 -4 were added to neutralised 52MnCl2 and heated at 165 oC for 1 h in a microwave (MW) synthesizer (Fig. 1A) at a porphyrin concentration of 0.6 -0.7 mM. These conditions were compared with heating at 70 oC using a heating block. The radiochemical yields (RCYs) were calculated by iTLC/TLC and HPLC, and the resulting 52Mn-porphyrin complexes characterised by comparison with their non-radioactive 55Mn counterparts and log P measurements. Following this, the porphyrin complexes 52Mn-(1 -4) were used to radiolabel empty DOXIL-like liposomes. 52Mn- (1 -4) were incubated with the liposomes at 50 oC for 30 min and purified by size-exclusion chromatography. Additionally, the porphyrin complexes 52Mn-(1 -2) were taken forward and used to radiolabel MDA-MB-231 cells in vitro which was directly compared with a previously reported direct cell labelling agent, 52Mn-oxine. [5] The MW radiosynthesis method allowed fast (1 h) and efficient radiolabelling with >95 % radiochemical yields with ligands 1, 2 & 4, and 85 ± 7 % for 52Mn- (3) . The RCY of 52Mn- (3) increased to 95 ± 2 % by using a ligand 3 concentration of 0.7 mM (Fig. 1B) . Conversely, non-MW heating at 70 oC for 1 h resulted in ca. 25 % RCY for 52Mn- (2 & 4) , with no conversion for 52Mn-(1 & 3) (Fig. 1B) . Whilst the RCYs increased after 24 h at 70 oC, none of the reactions reached completion. Formation of the 52Mn-complexes were confirmed with radio-HPLC (representative 52Mn-(1) trace in Fig 1C) , iTLC (Fig. 1D ) and log P. 52Mn-(1 -3) labelled empty DOXIL-like liposomes with high efficiencies of 78 ± 3 %, 86 ± 2 % and 74 ± 7 %, respectively (Fig. 1E ). 52Mn-(1 -2) radiolabelled MDA-MB-231 cells with moderate efficiencies of 10 ± 7 %LE and 12 ± 1 %LE, respectively (Fig. 1F ) whereas 52Mn-oxine had a higher LE of 42 ± 2% as previously reported. [5] However, the cellular retention of 52Mn after 24 h via 52Mn-oxine radiolabelling was significantly lower (14 ± 1 %), compared with 52Mn-(1) (32 ± 5 %) and 52Mn-(2) (45 ± 4 %) (Fig. 1G ). This suggests a distinct mechanism of trapping from the 52Mn-oxine methodology. In contrast to standard heating methods, MW heating allows the fast, efficient radiosynthesis of a variety of 52Mn-porphyrins with >95 % RCYs -avoiding the need for purification. This technique can be exploited for the radiolabeling and subsequent in vivo imaging of Mn-porphyrin therapeutics, as well as for the accurate in vivo quantification of Mn-porphyrin-based MRI agents. 52Mn-porphyrins also show promising cell and liposome labelling properties, with improved 52Mn cell retention over direct labelling methods using 52Mn-oxine. Abstract Body : Introduction: Pancreatic ductal adenocarcinoma (PDA) is a highly lethal disease with a 5-year survival rate only 8%. PDA is characterized with a dense stroma which underlies the chemoresistance and immune-suppressive microenvironment and pose the barrier for the treatment. DCE-MRI is a sensitive tool for the tumor microvascular perfusion/permeability assessment, has shown the potential for the PDA stroma-directed drug effect monitor1. The genetically engineered mouse (GEM) model of PDA recapitulates the human disease including the dense stroma, is a perfect model for therapeutic and biological investigation for PDA. However, the upper abdomen location of the pancreatic cancer makes it highly motion susceptible, Moreover, the fast respiration rate as well as the small blood vessel size, poses unique challenges for the DCE-MRI study in mice. In this study, we present a radial k-space sampled DCE-MRI protocol that achieves motion-robustness simultaneously with high temporal resolution for capturing the arterial input function (AIF) and spatial resolution of tumor DCE images . Methods: A genetically engineered mouse (GEM) that harbors Kras and p53 mutation in pancreas specific Cre allele, referred to as KPC model of PDA, was used in this study2. MRI was performed on 9.4T system 1. DCE series was acquired using multi-slice 2D cardiac gated saturation recovery GRE employing golden-angle (111.246°) radial k-space sampling (TR/TE= 2x cardiac period ~ 200/0.91 msec, FOV=38 x 38 mm2). Radial data were regridded to 256 x 256 matrix with or without employing k-space-weighted image contrast (KWIC) filter3,4 ( Fig 1A) . For KWIC reconstruction, 25 views encoding the center and 200 views encoding the outer-most regions of k-space region were used. Non-KWIC reconstruction utilized 200 views (according to Nyquist criterion) throughout the entire k-space. The AIF was extracted from the blood signal in heart left ventricle. Pixel-wise DCE series combined with T10 map and individual-AIF was least squares fit to the "shutter-speed" pharmacokinetic model to extract parameter maps including Ktrans. Results: The AIF extracted from fully sampled images (Fig 1 B -C) exhibits a blunted profile (Fig 1 F-G) due to relatively low temporal resolution, whereas AIF extracted from KWIC-reconstructed images (Fig 1 D-E) provide sufficient temporal resolution to fully characterize the initial rise of the AIF. . In the absence of respiratory gating, KWIC reconstruction resulted in motion-free high-resolution abdominal images including the PDA tumor (Fig 2A-C) . Selected DCE images reconstructed without KWIC (Fig 3C1-C5 )and with KWIC (Fig 3A1-A5 ) demonstrated minimal differences in image quality . The tumor ROI DCE signal time course resulted from the two reconstruction methods revealed that KWIC time course captures more time points during rapid wash-in phase of the contrast agent (Fig 3B, D) . Tumor T1, Ktrans and kep values (T1 = 2.5 ± 0.15 s, Ktrans = 0.09-0.33/min, and kep = 0.5-1.6 /min) obtained from parameter maps (Fig 3E, F , G) are comparable to previously reported values generated using Cartesian sampling methods at significantly lower spatial resolution1. Conclusion: Radial k-space sampling yields motion-free DCE images of PDA tumor in mice. The view sharing technique (KWIC) is highly compatible with radial data which is flexible in degrees of under sampling during image reconstruction to obtain optimal temporal resolution of AIF and spatial resolution of tumor DCE images. Acknowledgements This study was supported by U24CA231858 (Penn Quantitative Imaging Resource for Pancreatic Cancer), R21CA198563, R01CA211337 and P30-CA-016520-42. Overcoming this barrier has motivated the development of stroma-directed drugs to reduce or degrade the extracellular matrix components in PDA. The DCE-MRI has been proved to be an effective way to monitor the effect of this kind of drug3 while some questions need to be investigated more thoroughly to facilitate the clinical translation of this technique. In this study, we compared DCE-metric obtained from pharmacokinetic (PK) model in combination with individual-arterial input function (AIF) or a group-AIF for the ability to detect treatment effect. Corroborating imaging and immunostaining data, this study aims to provide rationale supporting choice of PK model and AIF approach which lead to quantitative imaging marker with optimal sensitivity and specificity for stroma-directed drug. Methods: The orthotopic PDA model was generated by injection 1.25×105 4662-KPC cells into the pancreas of syngeneic mice by a surgical procedure4. On a 9.4 system (Agilent) with a 12-cm gradient coil, DCE-MRI were performed at baseline and 24 hours after the PEGPH20(1mg/kg) or vehicle (VEH) treatment. The imaging protocol include the T10 map of tumor and blood (left ventricle) and DCE series3. The individual AIFs were extracted from the DCE series and the group-AIF was the average of 20 individual AIFs in 10 mice, two from each mouse obtained at baseline and post treatment. Pixel wise DCE data were fitted for three different PK models including Tofts (T), Modified-Tofts (M-T) and Shutter-Speed (SS) model using the least-squares method combined with the group or individual AIF 5-8. Pixel-wise parametric maps of Ktrans, kep, Ve and Vp were obtained as output. Responses to PEGPH20 was confirmed by reduction hyaluronan level by IHC compared to VEH-treated mice as reported earlier3. Results: While all mice were from the same strain and similar in age, individual AIFs exhibit a great difference in amplitude, leading to the highest SD at the peak point of group-AIF (Fig 1A) , the variation of AIF may result from physiological conditions at the time of study or occasionally imperfect bolus injection of contrast agent. Mean Ktrans of each tumor (Fig 1B) exhibit good correlation between individual versus group-AIF with Pearson coefficient r =0.86. For kep metric, individual-AIF allows both SS-and T-model to detect a significant increase of kep from baseline to post-PP, whereas using group-AIF, only SS model detects such increase (Fig 2A, D) . In comparison, for Ktrans metric, group-AIF combined with SS-model detected a significant increase of this metric from pre and post-treatment but individual-AIF is not able to do when combining with Tofts or SS PK model ( Fig 2B, E) . For the metric of %change of Ktrans, SS model combined with either individual-or group-AID can detect a significant difference between PP vs. VEH treatment (Fig 2C, F) . Notably, using the group-AIF, the M-T model revealed a moderate but significant increase of Vp after PP treatment ( Fig 3A) . Correspondingly, the "vascular lumen area" index obtained from CD31analysis exhibited an average of 48% increase after PP versus VEH treatment, although statistical significance was not reached due to small sample size (Fig 3B-3C) . Conclusion: SS model appears to be more sensitive than Tofts or M-T model for detection of treatment effect when applied with either individual-or group-AIF. Tofts model, however, is capable of detecting the treatment effect only when combined with Individual-AIF approach. Further validation of Vp metric is ongoing by IHC analysis. Acknowledgements: This study was supported by U24CA231858 (Penn Quantitative Imaging Resource for Pancreatic Cancer), R21CA198563, R01CA211337 and P30-CA-016520-42. Abstract Body : Objectives: To propose a way to predict the risk of breast cancer as early as possible, especially before a visible tumor appears. Methods: The parenchyma region was framed near to but not overlapping with the malignant tumor region delineated by professional radiologists. This region of interest was then used for the extraction of quantitative features, which were filtered by two methods consecutively: statistical analysis and the least absolute shrinkage and selection operator. The selected features were integrated to construct a radiomics signature, which was then evaluated by statistical analysis and support vector machine (SVM). Subsequently, a radiomics nomogram was established with this signature and "age", which was assessed by C-index, calibration curves and decision curves. Results: A radiomics signature, which consisted of 16 selected textural features, showed significantly differences (P 0.05 for the three cohorts, Hosmer-Lemeshow test) and clinically useful. Conclusion: The parenchyma region contains some lesion information before a visor tumor appears. Thus, it is possible to apply this information in breast cancer risk prediction. 

Jianbo Cao, University of Pennsylvania, caojianbocc@gmail.com

Abstract Body : Introduction Rapid respiration in mice and high magnetic fields of preclinical MRI scanners pose major challenges for quantitative diffusion weighted MRI (DWI) of the mouse abdomen. Two motion-resistant acquisition methods, radial spin-echo and EPI were evaluated for their effectiveness in mitigating motion and susceptibility artefacts in diffusionweighted images in a genetically engineered mouse (GEM) model of pancreatic ductal adenocarcinoma (PDA). The repeatability and accuracy in measuring diffusion coefficient were also assessed in a test-retest setting. Methods MRI was performed on a 4.7T horizontal bore system (Agilent) interfaced with a 12-cm gradient coil. Animals were anesthetized via 1.5-3% isoflurane, the temperature and respiration were monitored and maintained at the physiological level during the scan. Two phantoms containing water and butanol, respectively, were placed alongside the mouse for reference. Prospective respiratory gating was applied to both radially sampled diffusion weighted spin echo sequence(DW-SE-RAD) (TE = 30 msec, TR = 2-3 respiratory periods, slice thickness = 0.5 mm, FOV = 32 x 32 mm2, matrix = 64 read x 101 views, averages = 1, total acquisition time = 20-38 min, depending on respiratory rate and number of slices) and adiffusion-weighted spin-echo multi-shot EPI (DW-SE-EPI) sequence (TE = 26 msec, TR = 4 respiration periods, slice thickness = 1.5 mm, FOV = 32 x 32 mm2, matrix = 128 x 128, averages = 16, segments = 4, total acquisition time = 20-35 min) that spans the same tumor volume. Both sequences employ the same diffusion preparation parameters1: diffusion time ∆ = 14.4 msec, diffusion gradient duration, δ = 9 msec with five b-values (0.64, 535, 1071, 1478, 2141 mm2/s). All diffusion gradients were oriented along the diagonal direction; images were acquired with both positive and negative diffusion gradients. DW-SE-RAD and DW-SE-EPI protocols were also applied to a pair of tumor-bearing mice with and without prospective respiratory gating to assess the need for such gating. For each diffusion protocol, a test-retest study was performed in ten mice with 2-4 hours interval when mice were fully recovered from anesthesia. Apparent diffusion coefficient (ADC) was estimated in phantoms, tumours, and spinal muscles. The correlation of ADC between the two protocols was evaluated by GEE model in the R software; the reproducibility metrics include within-subject SD and CV and repeability coefficient (RC). Results Images from the DW-SE-EPI protocol without prospective gating were highly motion corrupted and meaningful ADC maps could not be generated. A comparison between DW-SE-RAD images generated with and without respiratory gating revealed streaking artifacts in non-gated images, resulting in increased noise in the ADC maps ( Figure 1 ). Therefore, respiratory gating was applied to both protocols in all studies reported below. The radial-DWI protocol produces artifact-free images over all b-values examined whereas ghosting is evident in the phase encoding direction (left to right) at all b-values and has comparable magnitude to the desired signal in the high b-value images. (Figure 2 ). ADC of water phantom estimated by Radial-DWI is consistent with well-known literature value3 whereas it is

Kwan Man, The University of Hong Kong, kwanman@hku.hk

Abstract Body : Background: Understanding of the mechanism of intrahepatic cholangiocarcinoma (iCCA) recurrence after curative surgery will be critical for development of prophylactic strategies. Regional immune-regulation in graft and tumor microenvironment may play a critical role of iCCA recurrence after curative surgery. Regulatory macrophages (Mregs) have potent immunosuppressive function. Currently, the existence and role of Mregs in iCCA remain unknown. Here, we aim to investigate the role and mechanism of Mregs on iCCA recurrence after curative surgical treatment. Methods: Gene expression data from our institute and TCGA were analysed to find out possible interaction between iCCA and monocytes. The serum CX3CL1 concentration and peripheral CX3CR1+ monocytes were assessed. A total of 75 patients with iCCA was included, and prospectively followed-up since 2001. The clinical significance of CX3CL1 expression and Mregs was evaluated by Cox proportional hazards regression model. The phenotypic analyses of Mregs from fresh iCCA tissue were conducted by flow cytometry. Results: 1. Higher serum CX3CL1 concentration and more peripheral CX3CR1+ monocytes were detected in post-transplantation patients (Fig.1 ). CX3CR1+ CD68+ macrophages could be observed in LT explant with iCCA. 2. Elevated CX3CL1 expression and more CX3CR1+ CD68+ macrophages in iCCA tissue were correlated with shorter disease-free survival (DFS) in iCCA patients (Fig.2) ; 3. In multivariate model, CX3CR1+ CD68+ macrophages were identified as independent prognostic factor for DFS (HR:3.400, 95%CI: 1.900-6.084, P Conclusion: CX3CL1/CX3CR1axis promoted iCCA recurrence through recruiting regulatory macrophages. Background: Understanding of the mechanism of intrahepatic cholangiocarcinoma (iCCA) recurrence after curative surgery will be critical for development of prophylactic strategies. Regional immune-regulation in graft and tumor microenvironment may play a critical role of iCCA recurrence after curative surgery. Regulatory macrophages (Mregs) have potent immunosuppressive function. Currently, the existence and role of Mregs in iCCA remain unknown. Here, we aim to investigate the role and mechanism of Mregs on iCCA recurrence after curative surgical treatment. Methods: Gene expression data from our institute and TCGA were analysed to find out possible interaction between iCCA and monocytes. The serum CX3CL1 concentration and peripheral CX3CR1+ monocytes were assessed. A total of 75 patients with iCCA was included, and prospectively followed-up since 2001. The clinical significance of CX3CL1 expression and Mregs was evaluated by Cox proportional hazards regression model. The phenotypic analyses of Mregs from fresh iCCA tissue were conducted by flow cytometry. Results: 1. Higher serum CX3CL1 concentration and more peripheral CX3CR1+ monocytes were detected in post-transplantation patients (Fig.1) . CX3CR1+ CD68+ macrophages could be observed in LT explant with iCCA. 2. Elevated CX3CL1 expression and more CX3CR1+ CD68+ macrophages in iCCA tissue were correlated with shorter disease-free survival (DFS) in iCCA patients (Fig.2) ; 3. In multivariate model, CX3CR1+ CD68+ macrophages were identified as independent prognostic factor for DFS (HR:3.400, 95%CI: 1.900-6.084, P Conclusion: CX3CL1/CX3CR1axis promoted iCCA recurrence through recruiting regulatory macrophages.

Abstract Body : I. Introduction Conventional MRI measurements in Multiple Sclerosis patients remain poorly correlated with disability and lack long-term prognostic value. A subsequent challenge is to identify the most reliable measure of myelin content in the white matter. Advanced MRI approaches (magnetization transfer, relaxometry...) are investigated ( 1). PET with repurposed amyloid tracers proved promising in translational studies (2) , and allowed the quantification of myelin changes in patients (3). Here, we used an EAE model in NHP to concurrently evaluate candidate MRI markers, and PET radiotracers, namely [11C]PiB and [18F] BF227. The MRI markers that are presented here are the T1w/T2w ratio (T1T2r), the Magnetization Transfer Ratio (MTR) and the inhomogeneous Magnetization Transfer Ratio (ihMTR). MT and ihMT are MRI techniques sensitive to the macromolecular content of biological tissues which has been proven to be strongly sensitive to myelin-rich content. II. Methods Experimental Autoimmune Encephalitis (EAE) was induced in one adult female primate (Macaca fascicularis) through intradermal immunizations with recombinant human myelin oligodendrocyte glycoprotein (rhMOG) and Freund's Incomplete Adjuvant (FIA) performed every 4 weeks. The animal underwent 3 simultaneous PET-MRI exams, under a Siemens Biograph mMR hybrid camera. PET consisted in a dynamic 60-min acquisition after [11C]PiB injection, immediately followed by 120-min acquisition after [18F] BF227 injection, while MRI included T1w, T2w, FLAIR, as well as MT_OFF, MT_ON and ihMT (7). For longitudinal analysis, T1w images were spatially normalized to a reference T1 template with corresponding atlas (4) . For visual inspection, T2w, FLAIR, MTs, ihMT and PET images were co-registered to FLAIR. MRI-derived images are computed according to the literature: T1T2r, MTR (6) , and ihMTR (7). WM lesion was manually segmented on FLAIR images. The contralateral region was defined as the symmetric of the segmented lesion on the reference T1 template images. Then, the contralateral region was co-registered back to the native space used for measurement. III. Results The animal developed motor symptoms (transient unilateral left paresis) twenty days after the second immunization, and was treated with dexamethasone for one week to prevent aggravation. Imaging was performed at D4, D11 and D39 after the appearance of symptoms. Paresis persisted until the final exam and sacrifice of the animal, though attenuated in frequency and intensity. Hyper-intense WM lesion was detected on the right hemisphere (see Figure) . WM lesion volume was measured and decreased over time: 1170mm3 (D4); 500mm3 (D11); 220mm3 (D39). Asymmetrical WM uptake was noticed on late PET images at D11 of both [11C]PiB and [18F] BF227. Left/Right asymmetry was also clearly visible on the MRderived images, T1T2r, MTR and ihMTR images. Lesion/contralateral ratio was measured in MR-derived images at D11: 0.67 (T1T2r), 0.56 (MTR), and 0.58 (ihMTR). These measured ratio are in accordance with the hyposignal observed in the WM lesion. IV. Conclusions This preliminary report on one animal suggests that the EAE NHP model is relevant for testing PET-MRI biomarkers of myelin changes over the course of the disease. The study design will allow a back-to-back comparison of the "candidate" [18F] BF227 (previously evaluated on a rat model (5)), against the "reference" radiotracer [11C]PiB. PET quantification will be performed as standardized uptake values (SUV) relatively to cerebellum after adequate definition of regions of interest (lesion core, border, and normal appearing white matter (NAWM). In parallel, a crossanalysis will be performed to identify the best myelin biomarkers combination to obtain a specific and sensitive PET-MRI biomarker to myelin content variation. Abstract Body : Chemiluminescence resonance energy transfer (CRET) refers to a non-radiative transfer mechanism between chemiluminescence (CL) donors and adjacent acceptors, usually in the nanoscale. Recently, reactive oxygen species (ROS) that are the representative hallmark of cancer were visualized by CRET-mediated molecular imaging [1] . In addition, there is no doubt about the resolution of CL imaging, responsible for high signal-to-noise ratio and superior specificity. Thus, CRET has received increasing for its applications to phototheranostics, particularly CRET-NPs have been extensively investigated for chemiluminescence imaging and photodynamic therapy (PDT). However, the short duration of their chemiluminescence signal and low therapeutic efficacy remain major limitations to translational application. Innovative strategies to achieve quantum yield-enhanced cancer phototheranostics should reflect the unique natures of CRET. Herein, we describe a new approach utilizing CRET-NPs capable of enhancing both photoacoustic (PA) and reactive oxygen species (ROS) quantum yield. CRET-NPs were prepared by physically encapsulating a CL donor, oxalate, into self-assembled NPs composed of biopolymer as the hydrophilic shell and chlorin e6 as the hydrophobic CL acceptor. Based on the hidden nature of oxalate and photosensitizer, we hypothesized that CRET-NPs could be longacting cancer phototheranostic agents, allowing effective PA imaging and PDT by the following feasible mechanisms: PA signal improvement via thermal expansion-induced vaporization (TEIV) and ROS quantum yield enhancement through both type I and II photochemical reactions, respectively. The CRET-NPs from this study facilitated the enhancement of PA quantum yield by TEIV, which is the most efficient mechanism of light-sound energy conversion, allowing for long-lasting PA imaging of H2O2-abundant tumor tissues. As a result, when CRET-NPs were intravenously injected into tumor-bearing mice, a strong PA signal was detected at the tumor site for at least 12 h. while CL signal lasted 10 min. Interestingly, there was no PA signal from CRET-NPs in normal liver tissue with a low level of ROS. These results suggesting that CRET-NPs allow long-term ROS imaging in a real-time and non-invasive manner. In addition to PA imaging, the CRET phenomenon of the NPs enhanced the ROS quantum yield of photosensitizer through both oxygen-independent type I and oxygen-dependent type II photochemical reactions. As a result, CRET-NPs effectively inhibited the growth of malignant tumors with complete regression in 60% of cases with only a single treatment. To the best of our knowledge, this is the first demonstration that CRET-NPs can generate an enhanced ROS quantum yield through both electron transfer and self-illumination. In summary, we developed CRET-NPs that maintain a stable nanostructure with high PA and ROS quantum yield for cancer phototheranostics. Quantum yield-enhanced CRET-NPs, resulting from the hidden nature of the CRET phenomenon, offer a new phototheranostic modality. Beyond cancer treatment, the findings of this study have broad implications for diagnosis and treatment of intractable diseases.

Anne Rix, RWTH Aachen International University, arix@ukaachen.de

Abstract Body : In preclinical imaging, contrast-enhanced ultrasound (CEUS) is an established technique to characterize the vascularization and expression of molecular markers in tumors. Next to that, US contrast agents can be used to intentionally permeabilize vessels during sonoporation1 or induce a vascular shutdown after inertial cavitation2. So far, no data are available to exclude an influence of diagnostic CEUS on tumor pathophysiology. Therefore, the present study should discover possible influences of diagnostic contrast-enhanced US on tumor pathophysiology and therapy outcome after antiangiogenic and antitumoral treatment in orthotopic triple negative breast cancer in mice. To achieve this, n = 100 female orthotopic 4T1tumor bearing BALB/c mice were allocated randomly to the following groups: no imaging, isoflurane anesthesia alone, ultrasound including a destructive pulse (destructive US), contrastenhanced ultrasound (non-destructive CEUS) or contrast-enhanced ultrasound with destruction of the contrast agent (destructive CEUS). Animals of each group were subdivided to either receive 10 mg/kg regorafenib or vehicle solution daily by oral gavage. On day 7, 10 and 14 after tumor cell injection, US measurements were performed using phospholipid microbubbles (MB) for non-destructive US or phospholipid MB targeting the vascular endothelial growth factor receptor-2 for destructive US. Furthermore, the impact of anesthesia and US imaging on animal welfare was evaluated twice a week by assessing heartrate, motor coordination and fecal corticosterone metabolites. The tumor size was assessed daily by caliper measurements. Before euthanization blood samples were collected and a complete blood count was performed. Fluoresecently labelled lectin was injected intravenously to assess perfused tumor vessels. Tumors were histologically characterized regarding vascularization and immune cell infiltration. Welfare assessment showed no impact of longitudinal ultrasound examinations on the animals' heartrate, motor coordination or fecal corticosterone metabolite levels. Histological characterization showed a significantly higher tumor vascularization and angiogenesis in vehicle treated animals examined with non-destructive CEUS (Fig. 1 A-C. ). Next to that, macrophage infiltration was significantly higher in tumors after non-destructive CEUS (Fig. 1 E-F.) , but no influence of the different imaging protocols on the tumor size could be detected. Both, nondestructive and destructive CEUS induced a systemic immune response of 4T1-tumor bearing mice represented by a reduced leukocyte count and spleen size. In conclusion, longitudinal US imaging did not increase the burden of the animals during the experiment. The significantly increased 4T1-tumor vascularization, perfusion, angiogenesis and immune cell infiltration after diagnostic CEUS resulted in a different interpretation of therapy outcome after regorafenib treatment. Furthermore, CEUS influenced the systemic immune reaction of 4T1-tumor bearing mice. These results show the need to carefully choose the right control groups to prevent unwanted falsification of study results and to improve the reproducibility of results between studies. Further investigations are ongoing regarding alterations of the tumor composition (e.g. collagen content, matrix metalloproteinases) and a second syngenic tumor model (CT26) will be included to explore if these changes are tumor type specific.

References: [1] Theek B, Baues M, Ojha T et. al 

Vertika Pathak, EXMI, vpathak@ukaachen.de

Abstract Body : Angiogenesis is one of the most studied hallmarks of cancer. Here, molecular ultrasound (US) with targeted microbubbles (MB) has shown potential in the early detection of malignant tumors as well as in the assessment of anti-angiogenic therapy response (1) . Magnetic Resonance Imaging (MRI) is another widely used imaging modality, which is highly desirable for molecular imaging of angiogenesis. However, MRI suffers from its inherently low sensitivity to probes. To circumvent the sensitivity issue of traditionally used contrast agents, nanoparticles loaded with greater payloads of gadolinium, or ultrasmall superparamagnetic iron oxide nanoparticles (USPIO) have been extensively developed. However, due to their small size, nanoparticles extravasate from the vessels and, their high unspecific uptake by the reticuloendothelial system makes it difficult to properly quantify the target-bound fraction. Therefore, to generate a potent intravascular MR contrast agent, USPIOs were loaded into the shell of micrometer-sized polymeric microbubbles. In this context, the location of the USPIOs at the air-liquid interface strongly enhances susceptibility effects, thus making the agent sensitively detectable in T2*W images (2) . As a proof of concept, we in this study investigated the capability of USPIO-MB as dual-modality 'molecular MRI and ultrasound contrast agents' to assess αvβ3-integrin expression at the neovasculature of murine breast carcinomas. USPIO-MB were synthesized as described previously (2), functionalized with c [RGDfK] , and scrambled c[RADfK] peptides. αvβ3-integrin-expressing 4T1 cells were incubated with the USPIO-MB for 8 minutes (in presence/absence of excess free-RGD), washed and embedded in gelatin phantoms. T2 relaxation times of the phantoms were obtained using a 7.0 T MRI system (Bio-Spec 70/20 USR, Bruker). Finally, an in vivo proof-of-principle study was performed in Balb/c mice bearing subcutaneous 4T1 tumours. 1x109 USPIO-MB were intravenously injected as a bolus and dynamic susceptibility contrast (DSC) scans with a three-shot EPI readout (TE/TR: 8/333 ms; FOV: 35x35; image size: 112x112; FA: 450) were acquired for 5 minutes. Targeted-US imaging was performed in the same set of animals that underwent MRI using a VEVO 3100 US system equipped with an MX-250 transducer (21MHz central frequency). After USPIO-MB administration, cine loops at 20 frames per second and 4% power were acquired for about a minute. Eight minutes post-USPIO-MB injection, a destructive pulse was applied to destroy USPIO-MB within the frame and immediately another 1500 frames were recorded to assess the unbound USPIO-MB fraction. Statistical analysis was performed using student multiple t-test and ANOVA in GraphPad Prism 8. USPIO-MB were characterized by narrow size distribution and high USPIOs encapsulation (Figure 2A-F) . Cells incubated with RGD-USPIO-MB showed higher R2 values than cells incubated with USPIO-MB. Adding free RGD significantly reduced the RGD-USPIO-MB binding ( Figure 2G) . The time-intensity curves obtained from DSC-MRI showed a prolonged signal loss in tumours due to RGD-USPIO-MB binding whereas both USPIO-MB and RAD-USPIO-MB only caused a transient signal loss due to MB passaging in the blood (Figure 1 ). The normalized signal intensity change before and after MB injection was quantified from T2W MR images. (Figure 2H, n=5) . Contrast-enhanced US results were in line with those from MRI showing a significantly higher binding of RGD-USPIO-MB than USPIO-MB or RAD-USPIO-MB to the tumor neovasculature ( Figure 2I, n=5) . In summary, we present the proof-of-principle for molecular MRI of tumor angiogenesis with targeted-USPIO-MB. From a translational point of view, our dual imaging methodology may be suitable for interventional procedures where the lesion is initially detected by MRI and the subsequent ultrasonic real-time feedback may be used to guide tissue biopsy or excision. Abstract Body : Introduction: Cirrhosis represents relevant social and public health diseases, with a progressively increasing incidence [1] . This scenario combines with the limited knowledge about the pathogenesis as well as with severe limitations in the diagnosis and treatment of the pathology. In this work we aimed to assess by non-invasive PET imaging some of the main pathophysiological processes involved in chronic liver disease, including angiogenesis, inflammation and metabolic disorders. For this purpose, three different radiotracers were employed: 68Ga-NOTAGA-RGD2 specific for angiogenesis [2] , [18F] Fluoro6TiaHeptadecanoic acid (FTHA) selective of fatty acid metabolism [3] , and 18FDG for the evaluation of glucose metabolism as surrogate of inflammation [4] . Methods: The tracers were tested in a control vs. cirrhotic rat model (control n=2; cirrhotic n=3). PET/CT scans were obtained from each rat employing the three different tracers, following the scheme: day 1 18F-FDG, day 2 18F-FTHA and day 3 68Ga-RGD2 plus ex vivo biodistribution. 18F-FTHA was synthetized based on previous bibliography [5] by azeotropic nucleophilic substitution of the tosylated precursor. The 68Ga-labeling of commercial NODAGA-RGD2 was optimized from previous works [6] concerning temperature, reaction time and radio-isotope concentration (20 nmol of NODAGA-RGD2, pH= 5.2, 7 mCi 68GaCl3, 15 min, RT). 68Ga was obtained from a 68Ge/68Ga-generator, based on nano-SnO2 and developed at CIEMAT [7] . 18FDG was commercially available from CuriumPharma. Cirrhosis was induced in Sprague-Dawley rats (male, 5 wk-o, 200g) by the administration of carbon tetrachloride (CCl4) by oral gavage twice a week for 12 weeks. Oral gavage with water (H2O) was given to control animals. Enzyme inducer for CCl4, Phenobarbital, was added to drinking water (35mg/dL) from 2 weeks before CCl4 administration until the end of the experiments [8] . PET/CT imaging was carried out in non-fasting rats at different time points according to the radiotracer employed (18FDG: 30 min post-injection/one 30-min static frame; 18F-FTHA: 0 min post.inj/dynamic 3x10min frames; 68Ga-NODAGA-RGD2: 0 min post.inj/dynamic 4x10min frames). On the CT images, cylindrical regions of interest (ROIs) were delimited in the three liver lobes and afterwards applied to the coregistered PET images to obtain the SUVmean. Ex vivo 68Ga-NODAGA-RGD2 biodistribution study was performed at the study endpoint, 40 min post injection. Results/Discussion: Cirrhosis led to different findings about the three pathophysiological processes evaluated. Confirming previous histological studies [9] , in vivo 18FDG-PET/CT imaging showed higher uptake in the cirrhotic model (0.83±0.05), as compared to control (0.60±0.04), probably reflecting an inflammatory response of the organ. Maximum uptake of 18F-FTHA was found at 10-20 min, but there were no apparent differences in fatty acid metabolism between control (4.48±1.29) and cirrhotic rats (3.74±1.47). Finally, 68Ga-NODAGA-RGD2 results showed an increase of tracer uptake in rats with cirrhosis (cirrhotic: 1.30±0.16 vs. control: 0.68±0.13), suggesting increased hepatic angiogenesis ( Figure 1A and B) .

These results were confirmed by ex vivo bioDi (control liver: 0.10±0.05; cirrhotic liver: 0.34±0.12 %ID/g) ( Figure 1C ). Conclusions: We have carried out a preliminary assessment of three of the main pathophysiological processes present in chronic liver disease. The increased 18FDG and 68Ga-NODAGA-RGD2 uptake in cirrhotic animals may indicate tissue inflammation together with angiogenesis, while the low difference in 18F-FTHA uptake suggests no significant changes in fatty acid metabolism. These non-invasive molecular imaging tools may provide a better insight into pathogenesis and may facilitate development of novel therapeutic strategies in cirrhosis. Acknowledgments: Authors thank Yolanda Sierra, Alexandra de Francisco and María de la Jara Felipe for their excellent work with animal preparation and imaging protocols. This work has been supported by "Diagnosis and treatment follow-up of severe Staphylococcal Infections with Anti-Staphylococcal antibodies and Immune-PET -Grant Fundación BBVA a Equipos de Investigación Científica 2018; Comunidad de Madrid, project "Y2018/NMT-4949 (NanoLiver-CM)" and "S2017/BMD-3867 (RENIM-CM)", co-funded by European Structural and Investment FUND.

References: [1] Abubakar, I. I., T. Tillmann, and A. Banerjee. "Global, regional, and national age-sex specific all-cause and cause-specific mortality (1), especially for triple negative breast cancer (TNBC), which lacks subtype-specific treatment. While TNBC patients initially respond to chemotherapy, majority of them develop drug resistant shortly after, leading to cancer progression. Potent inhibitors of TOPK (TOPKi) have been developed however substantial hematological toxicity were noted, which can be mitigated by liposomal delivery (2) . Meanwhile, TOPKi (e.g., OTS964) has been shown to sensitize drug-resistant glioma cells to temozolomide (3), however, their effect for chemoresistant TNBC has not been examined; furthermore, off-target effect and hematological toxicity induced by OTS964 require tumor-specific delivery platform (3). Here we formulate TOPKi into an ultra-pH sensitive nano delivery platform (pH-NP) (4) and demonstrate a remarkable tumor growth delay in paclitaxel-resistant TNBC by TOPKi-pH-NP. METHODS: Succinimidyl PEG-NHS (MW: 2000Da, NANOCS) in borate buffer solution at pH 8.5 was conjugated to pHsensitive peptide, KK-(GHFFH)3 reported earlier (4). The pH-sensitive nanoparticles encapsulating indocyanine green (ICG-pH-NP) or OTS964 (OTS-pH-NP) were manufactured similarly as described earlier (4) . A paclitaxel (Taxol)-resistant version was derived from human triple negative breast cancer line (HCC1806, ATCC) and 106 Taxol-resistant cells were inoculated into athymic nude mice. When xenografts reached ~200 mm3, mice were assigned randomly into one of the four groups: (1) OTS-pH-NP (10mg OTS964/kg, i.v.)+Taxol (10mg Taxol/kg, i.p.); (2) Taxol only (10mg/kg, i.p.); (3) Free OTS (10mg/kg, i.v.)+Taxol (10mg/kg, i.p.); (4) OTS-pH-NP only (10mg OTS964/kg, i.v.). OTS-pH-NP or Free OTS were treated at day 5, 7, 9, 12, 14, 16 while Taxol at day 0, 2, 5, 7, 9, 12, 14, 16 . Tumor size was measured by caliper. In separate cohort, tumor-bearing mice were injected with ICG-pH-NP or free ICG (at 2 and 10 mg ICG/kg, i.v.), NIR fluorescent imaging was performed (24 h post injection) on IVIS Lumina II (PerkinElmer) with ex/em =745/820~880nm. The ICG was extracted from tumor tissues and was quantified by HPLC (JASCO). RESULTS: Figure 1A -B show that pH-NP enhanced ICG accumulation in the tumor, owing to its ability to dissolve itself along the pH gradient (highàlow) thus releasing the cargo favorably at mild acidic extracellular pH of the tumor. In contrast, despite a large dose, free ICG did not penetrate in the tumor well and tumor concentration is less than 1/4th of that achieved by pH-NP (Fig 1B) . For this Taxol-resistant TNBC, it is not surprising that tumor growth was not delayed by Taxol only treatment ( Figure 1C ). In comparison, Taxol+free OTS moderately delayed tumor growth likely due to free OTS's unfavorable PK profile (data not shown) while OTS-pH-NP by itself also had moderate effect. OTS-pH-NP+Taxol treatment achieved a remarkable tumor regression (P=0.0004 compared to free OTS+ Taxol), suggesting that OTS964, when achieving effective concentration in tumor mediated by pH-NP, can overcome the drug-resistant TNBC to Taxol. CONCLUSION: OTS964 delivered by pH-NP was able to sensitize drug-resistant human TNBC to Taxol, leading to remarkable tumor regression. Such impact is mediated by pH-NP's ability to protect its cargos through the protein-rich plasma and release them into mildly acidic tumor extracellular space as it is confirmed by significantly higher tumor accumulation of ICG delivered by pH-NP. Abstract Body : The glioblastoma multiforme (GBM), known as human brain cancer (Class IV), is very aggressive and lethal in nature. The overall 90% of patients survival period extended up to only 14 months even though its early diagnosis or treatment with cutting edge technologies. The GBM does not have any forestanding boundary, which makes difficult to properly identify it from the normal brain tissues. Surgery is a current gold standard for GBM treatment though complete surgical resection of GBM tissues is practically impossible ascribe to their diffusive characteristics into circumvent normal brain tissues, thus resulting in high recurrence rate. Temozolomide chemotherapy is an additional treatment option, however, its efficacy is not satisfactory and therapeutic resistance is high because of molecularly heterogeneous malignancy of GBM. To address this issue, there is an urgent need to develop a sensitive imaging tool for accurate staging and prognosis of GBM and for image-guided surgery to show a clear margin of GBM in the surgery room. Previous studies already reported that the number of immune responsive microglias and macrophages infiltrating in glioblastoma tumor is surprisingly high, which comprises 30%-50% of the tumor mass which can be classified into M1 (inflammatory) and M2 (non-inflammatory) subsets. The M1 subset exhibits tumor suppressive indication promoting immune vigilance alternatively M2 subset shows tumor supportive functions with tumor progression, angiogenesis. The identification of M2 macrophages (tumor associated macrophages/TAM) has taken current spotlight in onco-immunotherapy for better cancer diagnosis and prognosis. In past few years there are extensive research has been carried out on small molecules and has proven that small molecules can successfully apprach to thair target across the physiological barrier. Here, we effort to localize M1 and M2 macrophages together by fluorescene molecular imaging technique to illucidaete circumference of glioblastoma apart from normal brain tissues which leads a surgical guidance to surgeon for maximum tumor resecetion and improving patient outcome. Elicited from this avidity, we have synthesized/selected M1 and M2 specific molecular probes for M1/M2 macrophage/microglia identification. In vitro we demonstrated on differentiated mouse bone marrow derived macrophage (BMDM) and mouse microglia cell line (BV2) to target M1/M2 probe. Differentiated M1 and M2 macrophages/microglias were characterized with CD86+ and CD206+staining, flow cytometry and qRT-PCR analysis.In vivo fluorence imaging on orthotopic U87-MG-Red Fluc GBM model exhibited an excellent localization of M2 macrophages/microglias at tumor site. In vivo Histophathological staining analysis also elucidated the presence of M1 and M2 macrophage/microglial localization around the GBM tumor (co-localized with GFAP+). Ex-vivo organ distribution imaging also indicate specificity of molecular probe with negligible non specific organ distribution. The presented molecular probes allows to elucidate the condition of GBM tumors with prominent boundaries which may provide new purview of GBM treatment. Abstract Body : Preeclampsia, a hypertensive disorder affecting 5-8% of pregnancies, is associated with placental ischemia and hypoxia during early development which causes systemic endothelial dysfunction responsible for the later onset of maternal symptoms. While potential therapies targeting cardiovascular dysfunction reduces maternal symptoms in preclinical models, the impact on in vivo placental function has not been demonstrated. Here we present multimodal imaging of in vivo placental function in the reduced uterine perfusion pressure (RUPP) rat model of preeclampsia. Using spectral photoacoustic (sPA) and contrast-enhanced ultrasound (CEUS) imaging, we demonstrate longitudinal measures of placental oxygenation and perfusion in response to tempol, a superoxide dismutase mimetic previously shown to improve the maternal symptoms of preeclampsia [1] . Timed-pregnant Sprague Dawley rats were administered tempol via drinking water beginning on gestational day (GD) 12. Imaging was performed every other day from GD14 to 18. Post-imaging on GD14, the RUPP procedure was implemented following established techniques [2] . sPA images were acquired at the relative optical absorption peaks of oxyhemoglobin and deoxyhemoglobin and the resulting photoacoustic signal was fit using a linear spectral unmixing algorithm to estimate the concentration of both chromophores in the image. Using co-registered B-mode US images of anatomy, the placenta was then manually selected and average oxygen saturation was calculated. To assess longitudinal changes in placental perfusion during normal pregnancy, nonlinear CEUS images were acquired on a separate cohort of untreated animals over a 10-minute period following a bolus injection of microbubble contrast agent. Conventionally, perfusion is quantified using a time-intensity curve (TIC) fit using the average CEUS signal in a region of interest. We implemented a pixel-wise TIC analysis in order to capture the complex flow profile in the placenta and surrounding tissue that's typically lost with conventional methods. Hemodynamic parameters related to blood flow and blood volume in the placenta were then calculated. Two days after the RUPP procedure, average placental oxygen saturation significantly decreased from 55% to 48%, while in normally developing placenta the oxygenation remains constant. Treatment of the pregnancy animal with tempol increased placental oxygen saturation in the RUPP by 6% -effectively restoring placental oxygenation to normal pregnant levels. This improved oxygenation was maintained through GD18 when tempol was also found to reduce mean arterial pressure in RUPP animals. A pixel-wise TIC analysis of CEUS images was used to assess longitudinal changes in placental perfusion during normal pregnancy. From GD14 to 18, relative blood flow and placental blood volume increased while mean transit time of microbubbles in the placenta decreased, demonstrating an increase in placental perfusion during normal pregnancy. We have demonstrated sPA and CEUS imaging of in vivo placental hypoxia and perfusion in response to therapeutic intervention for preeclampsia. Future work will adapt these multimodal imaging methods to include measures of vascular growth and provide a more complete understanding of overall placental function. Abstract Body : Skeletal muscle wasting is one of the hallmarks of cachexia, a syndrome that affects almost 80% of pancreatic cancer patient [1] . The syndrome is characterized by unintentional weight loss, loss of appetite, reduced physical ability and poor quality of life [2] . In a recent study, the role of mitochondrial dysfunction was identified in the amino acid derangement observed in cachectic muscle from a mouse model [3] . 1H magnetic resonance spectroscopy (MRS) studies have identified metabolic changes in muscle tissues associated with glioma and gastric cancer induced cachexia [4, 5] . Here we have investigated metabolic changes in muscle tissue occurring with pancreatic cancer-induced cachexia. Identification of metabolic changes in skeletal muscle may provide novel targets to treat this syndrome. Metabolic changes were quantified using high-resolution quantitative 1H MRS of muscle tissue obtained from normal, non-cachectic (Panc1) and cachectic (Pa04C) mice bearing pancreatic ductal adenocarcinoma (PDAC) xenografts. The Panc1 cell line was obtained from ATCC and the Pa04C cell line was kindly provided by Dr. Maitra. Tumors were generated by inoculating cancer cells in the right flank of six to eight-week-old male severe combined immunodeficient mice. Muscle tissue from the hind limb of euthanized normal (n=10), Pa04C (n=10) and Panc1 (n=9) tumor bearing mice, was excised once tumors were ~300 mm3, snap frozen and stored at -80°C prior to dual phase extraction. 1H MRS was performed on the water and lipid soluble phases. All 1H MR spectra with water suppression were acquired on a 750 MHz MR spectrometer using a single pulse sequence. All data analyses and quantification were performed with TOPSPIN 3.5 software. The presence of cachexia was confirmed by the significant weight loss observed in Pa04C tumor bearing mice, compared to Panc1 tumor bearing mice and normal mice. We identified a muscle metabolic signature in mice with and without cachexia inducing PDAC xenografts as summarized in Figure 1 . A significant decrease of alanine, glutathione, aspartate, creatine, glycine, lactate, glucose and an increase in isoleucine, acetate, phenylalanine and formate was detected in cachectic muscle tissue compared to control muscle tissue ( Figure  1A ). Previous studies have reported a decrease of glucose in muscle tissue of glioma induced cachexia and gastric cancer induced cachexia that indicated higher utilization of glucose by the tumor during cachexia [4, 5] . Reduced alanine in the cachectic group indicated an imbalance of systemic glucose hemostasis and amino acid metabolism [6] . Investigating the role of glycine could be important as supplementation of glycine may reduce muscle wasting [7] . We observed significant depletion of lipid (CH3, CH=CH) and triglycerides in cachectic muscle (Figure 1 B) that is consistent with a previous study [8] . Our results advance the understanding of changes in muscle metabolism with cachexia, and support investigating metabolic targets and biomarkers to reduce pancreatic cachexia-associated morbidity. Supported by NIH R01CA193365 and R35CA209960.

Swati Shah, National Institutes of Health , swati.shah@nih.gov

Abstract Body : Background The Ebola Virus (EBOV) is the causative agent of the Ebola virus disease (EVD) that is known to have a very high mortality rate. The EBOV has been shown to possess broad cell tropism and once inside the host, it infects multiple cell types. Previous studies have shown that early infection of macrophages, dendritic cells and monocytes are largely responsible for the systemic dissemination of the virus. EVD is largely characterized by lymphopenia, breach in vascular integrity, cytokine storm and multi-organ failure. Here, we measured translocator protein 18kDa (TSPO) binding by 18F-DPA714 PET imaging to evaluate peripheral disease progression in EBOV infected monkeys since TSPO is expressed in various immune cells at baseline and is generally upregulated in inflammatory conditions. Thus, correlating the changes in organ TSPO expression with disease markers can shed light on the underlying pathophysiology of EBOV infection. Methods We used 18F-DPA714, a PET ligand, to longitudinally quantify the peripheral TSPO binding in Rhesus macaques (n=10) at baseline and multiple time points after EBOV/Makona inoculation. We performed 40 min dynamic PET imaging of the chest and abdomen and used Logan plot (PMOD) with an image-derived blood input function (left ventricle) to measure lung, liver, spleen and bone marrow (BM) distribution volumes (Vt). Whole blood samples were collected during each imaging time point to measure various biomarkers such as plasma viral load, cytokine expression, differential blood counts, kidney and liver function panels. Post-mortem splenic tissue sections were immunostained for various markers of macrophages (CD68), T cells (CD3), B cells (CD20), TSPO, apoptosis markers (cleaved caspase 3(CC3) and cleaved Poly (ADP-ribose) polymerase 1 (PARP1)) and EBOV (vp40) by multiplex fluorescence immunohistochemistry (MF-IHC). Results Assessment of DPA714 Vt values over time (post-infection), using a mixed effect linear regression model, demonstrated a significant negative correlation of time with spleen and lung Vt and positive correlation with BM Vt. Multiple plasma cytokines including IFN-γ, MIP-1a, IL-6, IL-10, IL-1ra, IL-15 and IL-18 correlated negatively with lung/spleen Vt and positively with BM Vt (p Conclusions Our findings suggest a monocytic and lymphocytic depletion in the spleen, rather than immune activation and inflammatory reaction as would be expected in inflammatory conditions. Similarly, decreased uptake in lungs may reflect the depletion of alveolar macrophages, which could explain predisposition of EBOV patients to superimposed infections including pneumonia. However, increased uptake in the BM suggests hematopoietic activation in response to thrombocytopenia, monocyte depletion and neutrophilia. This is supported by BM Vt showing significant negative correlation with platelet counts and positive correlations with neutrophil percentages. Abstract Body : Introduction:Statins are low-cost cholesterol-depleting drugs used to treat patients with hypercholesterolemia. Preclinically, statins modulate caveolae-mediated endocytosis. Membrane receptors defined as tumor biomarkers and therapeutic targets are often internalized by an endocytic pathway. Indeed, receptor endocytosis and recycling are dynamic mechanisms that can affect receptor density at the cell surface. In therapies using monoclonal antibodies, a downregulation in receptor density at the cell surface decreases antibody binding to the extracellular domain of the membrane receptor. Here, we used immunoPET to demonstrate that statins can temporally modulate human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), and prostate-specific membrane antigen (PSMA) receptor density at the tumor cell surface for binding therapeutic monoclonal antibodies. Methods:Small-animal PET was used to study the binding of 89Zr-labeled antibodies in ectopic xenografts and patient-derived xenografts (PDXs). Lovastatin, simvastatin, and rosuvastatin were used to modulate caveolae. To retrospectively compare the non-statin vs.statin Kaplan-Meier survival curves, retrospective data from patients treated with trastuzumab were obtained from pharmacy records. Therapeutic assays were performed in PDXs to determine tumor populations that benefit from the combination of statin/antibody and to determine the mechanisms by which statins synergize with therapeutic monoclonal antibodies. Results: Using data from The Cancer Genome Atlas and the Cancer Cell Line Encyclopedia we observed a negative correlation at the protein level between the cholesterol-binding protein caveolin-1 (CAV1) and HER2, EGFR, or PSMA. In xenografts and patient samples, wefound that tumors with high CAV1 expression localized less receptor at the cell membrane and these features decreased antibody uptake and efficacy. In cultured cells and tumor xenografts, statins temporally depleted CAV1 expression in ways than enhanced tumors' avidity for anti-HER2, anti-EGFR, and anti-PSMA antibodies (Fig.  1A,B) . In HER2-and EGFR-expressing xenografts, treating mice with a statin accelerated and increased 89Zr-labeled antibody accumulation at the tumor site. The hydrophilic rosuvastatin demonstrated a lower ability to enhance antibody binding to tumors when compared with the lipophilic lovastatin and simvastatin. 89Zr-labeled huJ591 tumor accumulation was higher in PSMA-expressing tumors of statin-treated mice when compared with tumors of saline-treated mice at later time-points of antibody accumulation (24 h and 48 h), but the values were similar at 4 h and 8 h. These results suggest that treating mice with statins increases, but does not accelerate, anti-PSMA antibody accumulation. Anti-DLL3 antibody accumulation was similar in saline versus statin treated tumors. Retrospective data demonstrated that patients with HER2+/CAV1HIGH(29.4% of total HER2+gastric tumors) have a significantly worse overall survival than those expressing low CAV1 (Fig. 1C) . Kaplan-Meier analyses of statin use and HER2+gastric cancer disease outcome in patients treated with trastuzumab suggested that patients without statin treatment have worse survival than patients treated with a statin (Fig. 1C) .

Additionally, statins synergize with therapeutic antibodies to decrease oncogenic signaling pathways. Conclusions:Our data suggest that acute statin treatment with appropriate pharmacokinetics/pharmacodynamics are potential adjuvants for specific antibody-targeted therapies. Statins enhanced antibody accumulation while also enhancing target density at the cell membrane. In DLL3+tumors containing low levels of CAV1, statins did not alter antibody binding. Because lipophilic statins accumulate in cancer cells through passive diffusion, we observed higher antibody accumulation when compared with tumors from mice treated with rosuvastatin. Further investigations are necessary to define the statin doses necessary for tumoral CAV1 depletion, and the therapeutic value of combining cytotoxic antibodies with statins able to temporally stabilize receptor availability at the cell membrane. The translational significance of these findings lies in the potential of statins to modulate the heterogeneous staining of receptors at the cell membrane, a characteristic often associated with poor response in tumors to therapeutic antibodies. Abstract Body : Objectives: MUC16 ranks high among viable targets for the diagnosis and treatment of a class of ovarian and pancreatic adenocarcinomas. Aberrantly expressed O-linked glycans on mucins such as MUC16 are implicated in the biology that promotes the malignancy of tumors of epithelial origin. Such truncated MUC16 glycoforms are being investigated as potential targets for cancer treatment. We hypothesize that antibodies that bind to aberrantly expressed glycoforms of MUC16 can serve as a class of novel immunoPET tracers for noninvasive detection of ovarian and pancreatic tumors. Here, we utilized a humanized antibody, huAR9.6, that targets fully-glycosylated as well as hypoglycosylated isoforms of MUC16 to noninvasively image pancreatic and ovarian tumors via immunoPET. Methods: huAR9.6, was conjugated to desferrioxamine (DFO) and radiolabeled with zirconium-89 (89Zr) for immunoPET imaging. Radioligand binding assays were used to assess huAR9.6 binding to MUC16 on OVCAR3, SKOV3, Capan-2, and MiaPaCa-2 cell lines. OVCAR3 and Capan-2 cell lines were used as MUC16-positive ovarian and pancreatic cancer cell lines respectively, whereas SKOV3 and MiaPaCa-2 were used as MUC16-negative cell lines. Subcutaneous xenografts were developed by injecting the above-mentioned cell lines in the right flanks of athymic nude mice. Upon tumor development, mice were injected with [89Zr]Zr-DFO-huAR9.6 and serial PET imaging was performed at 24, 48, 72, 96, 120, and 144 h. Ex vivo biodistribution analyses were conducted at 144 h after injection of [89Zr]Zr-DFO-huAR9.6. Results: [89Zr]Zr-DFO-huAR9.6 was successfully synthesized with >99% radiochemical purity and high specific activity (~ 10 µCi/µg). Radioligand binding assays demonstrated specific binding of [89Zr]Zr-DFO-huAR9.6 to OVCAR3 and Capan-2 cell lines. The radiolabeled antibody showed >95% stability in human serum over a 7-day period. Significant tracer uptake was observed in Capan-2 tumors (~25% ID/g) whereas low uptake was shown in MiaPaCa-2 tumors (~5% ID/g). In vivo tumor uptake of [89Zr]Zr-DFO-huAR9.6 in Capan-2 tumors could be blocked by a co-injection of a 25-fold excess of unlabeled huAR9.6 antibody. Conclusion: Here, we successfully synthesized a 89Zr-labeled variant of the huAR9.6 antibody. Data from in vitro binding assays and in vivo PET imaging shows that radiolabeled huAR9.6 specifically binds to pancreatic cancer cells that express MUC16. Radioligand binding data also shows specific binding to ovarian cancer cells that are MUC16-positive. These findings demonstrate potential for using [89Zr]Zr-DFO-huAR9.6 as a non-invasive imaging agent to detect cancer of the pancreas and ovaries. Acknowledgments: This work was supported by the Emerson Collective Cancer Research Fund, OncoQuest Inc., and NIH R35 CA232130. Abstract Body : Introduction: Bronchopulmonary dysplasia is a common pulmonary complication in preterm infants and lung inflammation control by budesonide has been used to reduce the incidence and severity of bronchopulmonary dysplasia. For pharmacokinetic study of budesonide, 11-acetate 21 tosylate budesonide was prepared for F-18 radiolabelling and [18F] budesonide is used for imaging application. Methods: 11-acetate 21 tosylate budesonide was prepared by protecting the two hydroxyl groups of commercial budesonide with tosyl and acetyl group, sequentially. The tosylation of budesonide is performed by tosyl chloride using DMAP as an activating reagent and triethylamine as an HCl scavenger. Then, the acetylation for another hydroxyl group is performed by acetic anhydride to have 11-acetate 21 tosylate budesonide. For radiolabeling, [18F] fluoride was trapped on an anion exchange cartridge and eluted using kryptofix-2.2.2. After the drying procedure, [18F] fluoride was reacted with 11acetate 21 tosylate budesonide at 110°C for 10 min in anhydrous dimethyl fluoride via an aromatic nucleophilic substitution, in which the tosyl leaving group is substituted with [18F] fluoride. After passing through a Sep-Pak C-18 to remove the free [18F] fluoride and followed by hydrolysis with NaOMe/methanol, [18F] budesonide was synthesized. The radiolabeled products were analyzed by radio-HPLC using LichroCART column (5mm x 25 cmx 10 mm) and eluted with 50% ethanol for 30 min at a flow rate of 0.5mL/min. The intratracheal pharmacokinetic study in SD rats was measured by PET/CT. Results: The 11-acetate-21- [18F] budesonide showed coupling peaks at retention times of approximately 21 min and 22 min, which is because commercial budesonide has mixed budesonide R(+) and S(-) isomers. However, after de-acetylation, the epimers of [18F] budesonide were eluted at 25 min as a single peak. In animal study, [18F] budesonide was strongly detected in the trachea at 15 min after intratracheal injection, but rapidly metabolized to liver. With surfactant supplement in [18F] budesonide, [18F] budesonide can steadily stay in lung for more than 60 min. Conclusion: 11-acetate 21 tosylate budesonide and [18F] budesonide were successfully prepared and purified. PET/CT images showed that surfactant supplement is necessary for [18F] budesonide to have better distribution in lung for more than 60 min. Abstract Body : Cancers metastasize, making the disease difficult to effectively detect and treat. Metastatic cancer cells uniquely overexpress a class of enzymes called matrix metalloproteinases (MMPs). MMP9 expression and activity is correlated with metastasis and tumor spread in breast and prostate cancer. Using enzyme-cleavable peptide sequences, we will be able to harness these MMPs for use against the disease. My central hypothesis is that radiolabeled enzyme-responsive peptides will provide a means to image metastatic cancer characterized by elevated MMP expression. Our laboratory has developed enzyme-responsive self-assembling peptide amphiphiles (ERSAPAs), that adopt a micelle structure, and switch conformation to a fiber when cleaved by enzymes. Short hydrophobic peptide fibers have been demonstrated to adhere to tumor tissue. We propose to tether a radioactive iodine molecule to an ERSAPA, permitting it to travel freely in the body, and when it arrives at a site of MMP9-expressing cancer tissue, the resulting cleavage and fiber formation will "anchor" the radio-label at the cancer site. Using established peptide design rules, we have synthesized peptides with three distinct regions: (i) a hydrophobic fiber-forming sequence, (ii) a hydrophilic sequence that adjusts the amphiphilic balance of the peptide to favor micelle formation, and (iii) an MMP9-cleavable sequence separating the hydrophilic and hydrophobic regions. Our negative control consists of a micelleforming peptide sequence in which the MMP-cleavable sequence is scrambled, preventing cleavage. Our positive control is a micelle-forming, cleavable peptide that does not have a radioligand attached. We have used atomic force microscopy (AFM) and fluorescence experiments to determine peptide assembly and nanostructure formation. Liquid chromatography -mass spectrometry (LCMS) has been used to determine cleavage of the peptide by the enzyme. Thus far, we have demonstrated micelle formation of our peptides, MMP-cleavage of our positive control peptide, and effective "cold" iodine coupling to the peptides. Remaining work involves demonstrating cleavage and micelle -to-fiber transition of the iodine-labeled peptide. Following this, we will assess the tumor-targeting effectiveness using radioactive "hot" iodine in murine models of MMP9 -expressing cancer. This technology will provide a nuclear imaging agent established from an enzyme-responsive nanoparticle, and will be the first imaging nanoparticle to undergo a programmed morphological switch. Funder Acknowledgement: This work is supported by the National Science Foundation CREST Center for Interface Design and Engineered Assembly of Low Dimensional systems (IDEALS), NSF grant number HRD-1547830. Abstract Body : Objectives: Pancreatic cancer (PC) remains the 4th leading cause of cancer death; therefore, there is a clinically unmet need for novel therapeutics and diagnostic markers to treat this devastating disease. Physicians often rely on biopsy or CT for diagnosis, but more specific protein biomarkers are highly desired to assess the stage and severity of PC in a noninvasive manner. Serum biomarkers such as CA19.9 are of particular interest as they are commonly elevated in PC but have exhibited suboptimal performance in the clinic. MUC5AC has emerged as a useful serum biomarker that is specific for PC vs. inflammation. We developed RA96, an anti-MUC5AC antibody, to gauge its utility in PC diagnosis through immunohistochemical (IHC) analysis and whole-body PET/CT imaging in PC. Methods: We determined the utility of RA96 for MUC5AC IHC on clinical PC and pre-clinical PC by staining on tissue microarrays containing 300 cores of normal ducts and PanINs. These samples originated from 21 disease-free pancreata and 81 resected pancreata of different neoplasms, 30 alcoholic chronic pancreatitis and 10 autoimmune pancreatitis specimens, and 48 pancreatic ductal adenocarcinomas (PDACs). Immunohistochemical staining was performed with RA96 and the intensity of the reactions was scored as mild, moderate, or strong (score of 1, 2, or 3, respectively) by an experienced pathologist. Finally, we radiolabeled RA96 with zirconium-89 to assess its applications as a whole-body PET radiotracer for MUC5AC quantification in preclinical models of PC. RA96 and isotype-matched IgG were functionalized with pisothiocyanatobenzyl-desferrioxamine (DFO-Bn-NCS) and radiolabeled with zirconium-89 and characterized via instant thin-layer chromatography. Serial PET/CT imaging was performed in MUC5AC+ capan-2 xenografts. Results: Immunohistochemical staining with RA96 distinguished chronic pancreatitis (CP), PanIN, and varying grades of PDAC in clinical samples. The specific activity and molar activity for both [89Zr]Zr-DFO-RA96 and [89Zr]Zr-DFO-IgG was 0.3-0.4 MBq/mg and 45-55 GBq/mmol, respectively. Serial PET imaging with [89Zr]Zr-DFO-RA96 shows significant uptake in MUC5AC+ capan-2 tumors (14.6 ± 1.5 %ID/g) compared to the blocking group (6.6 ± 1.5 %ID/g) and [89Zr]Zr-DFO-IgG control (8.8 ± 0.3 %ID/g), with a P Conclusions: Our study demonstrates that RA96 can differentiate between inflammation and PC, improving the fidelity of PC diagnosis. Our immuno-PET tracer [89Zr]Zr-DFO-RA96 shows specific detection of MUC5AC+ tumors in vivo, highlighting the utility of MUC5AC targeting for diagnosis of PC. We expect that immuno-PET targeting MUC5AC has high potential to be clinically useful for early detection of PC. Acknowledgments: We would like to acknowledge financial support from NIH grant R35 CA232130-01A1 and the MSK Center for Molecular Imaging and Nanotechnology Tow Fellowship. Abstract Body : Abnormal vasculature in solid tumours often results in poor oxygenation, a phenomenon that is known as tumour hypoxia. Hypoxic cells are resistant to chemotherapy, radiation therapy and immunotherapy, and thus contribute to treatment failure and cancer progression. Therefore, detection of tumour hypoxia is crucial for making appropriate therapy choices and monitoring treatment response. Nitroimidazoles (NI) have demonstrated a significant potential in hypoxia diagnosis because they get selectively activated in hypoxic cells and react with proteins to form drug-protein adducts. Pimonidazole, a commercially available 2-NI derivative, is used in immunohistochemistry (IHC) to detect hypoxic regions in tissue biopsies. Here, we report the synthesis and characterization of a novel 2-NI-derived hypoxia mapping agent named Azido Conjugated Nitroimidazole (ACN). Instead of immunodetection, our method utilizes copper (I) catalyzed "click chemistry". Using a fluorophore conjugated alkyne, we have been able to detect the hypoxia-selective entrapment of ACN in cells and in mouse tumour models. By combining the hypoxic affinity of NIs with the ease and versatility of click chemistry, we have developed a histological tool that bypasses the necessity of antibodies to facilitate a rapid detection of hypoxia. Our approach offers an easier and less expensive alternative to Pimonidazole IHC, while still maintaining high fidelity and precision. A direct comparison of the two methods shows that ACN click chemistry is equally efficient at detecting histological hypoxia as Pimonidazole IHC. The modular nature of the reaction also allowed us to use ACN and a biotin labelled alkyne to pull down its potential protein targets, which were then characterized using liquid chromatography-coupled mass spectrometry (LC-MS/MS). A total of 62 candidate proteins were identified, and we analyzed them using various bioinformatic platforms (PANTHER, IPA) for biological significance. ACN target proteins appeared to be predominantly cytoplasmic (~71%) with only ~21% being nuclear proteins. Canonical pathways analysis suggests that most of these proteins are involved in stress and glucose metabolism pathways. We studied the effects of ACN binding to one of its top protein targets, glyceraldehyde-3-phosphate dehydrogenase (GAPDH). While ACN, at IC50 concentration, does not affect GAPDH protein levels or cellular localization, it significantly reduces GAPDH enzymatic activity under hypoxia. This information is valuable because NIs have long been explored for hypoxia directed cancer therapy. Given the therapeutic potential of these compounds, our study provides insights into the molecular mechanism of NI drugs. Abstract Body : Background: 111In-Oxine has been successfully used for imaging of different types of white blood cells and immune cells using SPECT over decades. It was also previously used to monitor in vivo trafficking of dendritic cells [1] . Use of 89Zr-Oxine provides advantages of quantitative approach to the immune cell imaging using PET, as it was reported for lymphocytes. In the current study we demonstrate feasibility of use of 89Zr-Oxine for stable ex vivo labeling of dendritic cells for PET imaging. Materials/Methods: porcine dendritic cells were obtained by apheresis of monocytes from the pigs. 89Zr labeling of Oxine was performed, as previously described [2] with 98% radiochemical purity. In in vitro testing a range of 89Zr-Oxine doses from 101 to 2*102 uCi/106 cells were tested with incubation time from 30 to 60 minutes, at room temperature in PBS. Cell viability at 0, 4, 24-and 72-hours following incubation was tested. The labeling efficiency and label stability over time was evaluated. The most optimal labeling condition was used to pre-label 1.2*108 cells 1 hour prior to the administration to the animal (according to IACUC approved protocol). PET images were acquired at 4, 24 and 72 hours to evaluate the stability of 89Zr-Oxine labeling in vivo. Results: In vitro, the labeling efficiency for 30-and 60-minute incubation did not differ significantly (54% and 62%, respectively). In the range of tested radiotracer concentrations, no significant difference of labeling efficiency was observed (p>0.05). Loss of 89Zr-Oxine in medium was observed from 8.1+/-2.7% at 24 hours to 17.3+/-5.6% at 72 hours. No significant differences in cell viability was observed at 24 and 72 hours for all tested radiotracer concentrations and exposition times ranging from 74% to 88% (p>0.1). For improved sensitivity of in vivo visualization, the maximal dose and labeling time values were selected. At the time of in vivo experiment, 89Zr-Oxine labeling efficiency was 58% with 77% viability. 3.1mCi total activity of 89Zr-Oxine was administered. PET imaging demonstrated expected biodistribution of the ex vivo labeled cells to the pulmonary vasculature as early as 4 hours following administration (SUVmax 43.3). Minimal radiotracer uptake was observed in the liver at 4, 24 and 72 hours, where excretion of free 89Zr released from the cells could have been expected (SUVmax 5.2, 3.3, and 2.8, respectively). Conclusions: 89Zr-Oxine ex vivo labeling of porcine dendritic cells can be performed with reliable efficiency and stability in the dose range applicable for the in vivo PET studies. 89Zr-Oxine labeling appears stable over time and does not affect the viability of the labeled cells. In vivo PET imaging demonstrates localization of reinfused dendritic cells labeled ex vivo with 89Zr-Oxine without significant in vivo loss of the label. This approach appears feasible for quantitative PET imaging of dendritic cells in vivo in the future studies. Support: research is supported through NIH HLT32. (1); Ag(I) was chosen in order to increase fluorescence intensity of NCs (2); and Fe ions served as precursors for the formation of a potential contrast agent being employed in magnetic resonance imaging (MRI) (3) . Different ratios of these ions used during the synthesis of NCs substantially influenced NCs fluorescence intensity, as well as the position of the characteristic fluorescence peak. Decrease of the fluorescence intensity was observed depending on the ratios of metal ions as demonstrated in Figure 1 . Moreover, it was revealed that NCs fluorescence intensity is a nonmonotonic function of time if the samples were not dialysed. It was tested that our NCs can be used for MRI if the concentration of incorporated Fe ions is high enough. To further evaluate whether [11C]mHED PET may be useful in basic research with mouse models of heart diseases, we investigated by PET the specificity of [11C]mHED for NET in the mouse heart and characterized the kinetics of [11C]mHED uptake by compartmental modelling. Methods Female FVB/N mice (20-24 g) were assigned to 3 groups, (i) baseline scan (n = 5), (ii) scan with NET blocking by desipramine (20 mg/kg intraperitoneally 10 min before [11C]mHED administration; n = 6), and (iii) baseline scan with simultaneous recording of the blood coincidences by a femoral arterio-venous shunt guided through a blood coincidence counter (n = 3). [11C]mHED was synthesized by direct N-methylation of metaraminol with [11C]methyl triflate and injected intravenously (3-14 MBq; 1.5-62 nmol/kg combined mHED and metaraminol), 1 min before (baseline, desipramine blocking) or 1 min after (kinetics) scan start. PET was acquired for 60 min. Standardized uptake values (SUV) of left ventricle wall averaged from 1 to 30 or 1 to 60 min were plotted in a dose-response curve against the logarithmic injected dose as nmol mHED plus residual metaraminol plus desipramine (if applicable) to determine the dose at half maximal NET saturation and the maximally tolerated dose for linear (dose-independent) kinetics. PET data with blood coincidences were fit to one-and two-tissue compartment models. PET data from the blocking group were analysed using one of the blood coincidence curves from the kinetic modelling group. Results Desipramine reduced the SUV(1-30min) of left ventricle wall from 4.4 (upper plateau of dose response curve; SE 0.18) to 1.0 (lower plateau, SE 0.13; Figure) . Half-saturation of NET uptake was fit to 88.3 nmol/kg combined mHED and metaraminol (SE 30.3; Figure) , in agreement with earlier findings in mice (ref. 1) and rats (ref. 2) . This suggests linear kinetics up to a dose of 10 nmol/kg combined mHED and metaraminol. PET data of the kinetic modelling group (1.5-2.0 nmol/kg tracer and metaraminol) followed two-tissue compartment models. Blocking by desipramine reduced K1, the transfer clearance parameter from blood to the region of interest and had no effect on the other fit parameters in serial twotissue compartment models. This indicated that NET transport is represented by K1 in our models. K1 determined from the 3 scans with blood coincidence data was 0.947±0.146 mL/cm3/min (input function corresponding to parent tracer in plasma). The fit partial volume of blood in the region of interest was 0.46±0.035. This high number is in agreement with the high fraction of blood in the region of interest due to myocardial motion and partial volume effects. According to the model, this partial volume in addition includes tracer in extracellular space with rapid exchange with blood. SUV(1-30min) correlated with K1 (p n = 14), indicating that

[11C]mHED SUV may be used as a surrogate parameter for NET function in mouse. Conclusion [11C]mHED is NET-specific in mouse heart and follows a two-tissue compartment model. NET function is represented by K1 in serial two-tissue compartment models and SUV averaged from 1-30 or 1-60 min correlates well with K1, suggesting SUV as a parameter for estimating NET function in the mouse heart. We suggest [11C]mHED PET as a useful method to address sympathetic innervation in the mouse heart in basic research. The combined dose of tracer and precursor should be kept below 10 nmol/kg. Abstract Body : Metastatic breast cancer is a leading cause of death among women in the United States, and approximately 30% of initially node-negative breast cancer patients will develop metastatic lesions. The 5-year survival rate for metastatic breast cancer remains near 27 % due to the complexity and challenges associated with treating this stage of the disease.1 Consequently, significant research efforts have been trained at interrogating the underlying biology of metastasis formation and growth in an effort to develop therapeutic strategies that can more effectively combat metastatic cancers.2-5 While the specific processes involved in metastatic dissemination are not fully understood, the recruitment of hematopoietic stem cells with high expression of the vascular endothelial growth factor receptor 1 (VEGFR-1) has been implicated in early stages of metastasis formation.6 Herein, we propose the use of an 18F sitespecifically labeled, single-chain version of VEGF121 re-engineered to selectively bind to VEGFR-1 (scVR1), as a positron emission tomography (PET) imaging agent to non-invasively image early stage metastases. The scVR1 was radiolabeled with 18F via a biorthogonal click reaction between site-specifically trans-cyclooctene functionalized scVR1 and an Al18F labeled tetrazine-NODA (1,4,7-triazacyclononane-1,4-diiacetic acid). The [18F] AlF-NODA-scVR1 was purified using a PD10 column and subsequently analyzed on HPLC to determine radiochemical purity. Animal experiments were performed in 6-8 week old female BALB/c mice, bearing orthotopic primary 4T1 breast tumors or 4T1 metastatic lesions. The [18F] AlF-NODA-scVR1 tracer was administered via tail vein injection, PET imaging and ex vivo analysis was performed 2 hours post-injection. The [18F] AlF-NODA-scVR1 was prepared with a 98.2 ± 1.5 % radiochemical purity and specific activity of 7.5 ± 1.2 GBq/µmol. The specific binding of scVR1 to VEGFR-1 was confirmed via bead-based assay. The ex vivo biodistribution showed tumor uptake of 2.75 ± 0.30 %ID/g and was readily observable in PET images. Metastasis formation was detected with [18F] AlF-NODA-scVR1 tracer showing colocalization with bioluminescent imaging as well as ex vivo autoradiography and immunofluorescent staining of VEGFR-1. In addition, the [18F] AlF-NODA-scVR1 tracer uptake in metastatic lesions correlated with VEGFR-1 expression. Systemic treatment strategies such as chemotherapy and endocrine therapy have improved outcomes when performed in patients early in progression and metastatic dissemination. However, the ability to image the initial stages of metastatic development and progression remains elusive. A PET tracer that could provide clinicians a means of imaging and diagnosing metastatic disease could further improve survival rates for metastatic breast cancers. The [18F] AlF-NODA-scVR1 PET tracer was developed and its selectivity for the VEGFR-1 receptor was established. In addition, the in vivo PET/CT and ex vivo analyses support the potential of VEGFR-1 specific tracers to image early stages of metastatic tumor formation and growth. While further development of [18F] AlF-NODA-scVR1 is necessary to improve its performance, the results presented herein provide support for its prospective implementation in clinical applications. The [18F] AlF-NODA-scVR1 PET tracer offers a promising tool that could help physicians make informed decisions when determining if a patient requires adjuvant or neoadjuvant therapies. [18F] AlF-NODA-scVR1. B) Bioluminescent imaging of 4T1luc cells in chest cavity associated with lung metastases. C) Ex vivo analysis of lung tissue including autoradiography, hematoxylin and eosin staining, and immunofluorescence staining, VEGFR-1 (red), CD-31 (green), and 4′,6-diamidino-2-phenylindole (DAPI) (blue). D) Correlation between relative autoradiography signal intensity and VEGFR-1+ cells in lung metastases (P < 0.05). E) Correlation between relative autoradiography signal intensity and VEGFR-1+ cells in surrounding lung tissue of metastatic mice (P < 0.05). F) Percentage of CD31+ cells in metastatic tumors, surrounding lung tissue in metastatic mice, and healthy lung tissue. G) Percentage of VEGFR-1+ cells in metastatic tumors, surrounding lung tissue in metastatic mice, and healthy lung tissue. Abstract Body : In this study, we evaluate a novel small molecule choline kinase inhibitor, JAS239, for its toxicity in mice, ability to delineate tumor margins in vivo, and measure ChoKα expression in spontaneous canine lung adenocarcinomas. Our goal is to translate JAS239 to the veterinary clinic for intraoperative tumor imaging of spontaneous canine patient lung adenocarcinomas. Choline Kinase α (ChoKα) is an enzyme overexpressed in 39% of human breast and 60% of human lung tumors. It is associated with an aggressive phenotype, high histological tumor grade, and poor clinical outcome in many human cancers1-3. ChoKα catalyzes the phosphorylation of choline to generate phosphocholine, a precursor of the cell membrane phospholipid, phosphatidylcholine (PtdCho)4. A novel small-molecule near infrared (NIR) fluorescent ChoKα inhibitor, JAS239, has been developed in our lab (excitation: 742 nm, emission: 775 nm). In the NIR window (700-900 nm), light absorption by water, hemoglobin, and tissue is reduced, decreasing light scattering and allowing for better depth penetration and tumor margin differentiation5-7. JAS239 binds to the active site of ChoKα, competitively inhibiting phosphocholine production, and thus can be used as a targeted imaging agent for intraoperative tumor resection guidance in ChoKα-overexpressing tumors. Tumor resection continues to be the most effective approach to cure patients with local disease. Surgeons typically use visual inspection and finger palpation to define solid tumor margins8. However, this approach is often insufficient for the detection of residual disease surrounding the wound bed, thus leading to local recurrence in up to 40% of patients and a significantly reduced 5-year survival9,10. Clinically-approved intraoperative tumor imaging fluorophores, such as indocyanine green (ICG), have shown promise in detecting tumor margins and residual tumor during surgery, however, they are limited by their lack of specificity between tumorigenic cells and inflammation, relying predominantly on leaky vasculature to localize to the tumor. Toxicity studies showed that there were no hematological, biochemical, or histopathological differences when mice were treated with 50x the imaging dose of JAS239 (five times the imaging dose of 40 nmol for 10 days). This study provides the data necessary for a clinical toxicology study in non-patient canines. In mice, detection of tumor margins was complicated byvariable background fluorescence in different models of cancer, while JAS239 tumor localization and fluorescence remains relatively similar. We hypothesize that this could be due to variable ChoKα expression in the muscle tissue of different mouse strains. Nude mice exhibited the least amount of fluorescent background signal, while DBA/2 mice had large fluorescent background signal when compared to tumor signal. In preparation for translation to canines, we are accumulating a panel of patient-derived canine lung tumors and histologically staining them for ChoKα expression. Spontaneous lung cancers from canine patients express high levels of ChoK and are suitable candidates for JAS239 fluorescence-guided surgery.

Edwin Pratt, Memorial Sloan Kettering Cancer Center, pratte@mskcc.org

Abstract Body : Trametinib (MEKINIST™) is an extremely potent allosteric inhibitor of MEK1/2 that has been approved for treatment of metastatic melanoma and anaplastic thyroid cancer in patients with confirmed BRAFV600E/K mutations. Though highly efficacious, adverse side effects including skin, gastrointestinal and hepatic toxicity, are dose limiting and can lead to treatment termination. Development of a non-invasive tool to visualize and quantify the delivery and distribution of trametinib (either as single agent or in combination with other therapeutics) to tumors and organs would be very helpful in assessing therapeutic index, personalizing individual dose and potentially predict resistance to therapy. To address these issues, we have developed a radiolabeled trametinib and evaluated the in vitro and in vivo properties. 123I-, 124I-and 131Itrametinib, pure tracer analogs to trametinib, were synthesized in >95% purity from a boronatotrametinib precursor with average yields of 69.7% and >100GBq/µmol specific activity. Overall, 124I-trametinib uptake in a panel of cancer cell lines can be blocked with cold trametinib along with other Type-III MEK inhibitors confirming specificity of the radiotracer in vitro and in vivo. 124I-Trametinib was taken up at higher rates in KRAS and BRAF mutant cell lines compared to wild type KRAS cancer cell lines. In vivo, biodistribution revealed high uptake in the liver 2 hours post injection followed by clearance through the gastrointestinal tract over 4 days. Importantly, higher than expected uptake was observed in the lung and heart for up to 24 hours. Peak uptake in the skin and gastrointestinal tract was observed between 6 and 24 hours in c57bl/6j mice while in B16F10 melanoma bearing mice peak tumor concentrations were achieved between 24 to 48 hours. Tumor uptake at 48 hours peaked at 4 % injected dose per gram while tumor to muscle or skin ratios were relatively low, peaking at 3.4-to 8.1-fold by 72 hours, respectively. Biodistribution of 124I-trametinib was significantly reduced in mice on trametinib therapy providing a quantitative method to observe MEK inhibition or expression in vivo. 124I-trametinib serves as a tool for dose personalization in vivo instead of the current fixed dosage scheme and when combined with radiomic data monitor emergence of therapy resistance. In addition, the production of iodinated trametinib affords researchers the ability to measure changes in MEK during therapy as well as drug distribution for improved drug delivery studies. iodination. Radiosynthesis and purification was complete within 1.5 hours with a radiochemical yield (RCY) of 70% for 124I-trametinib with a specific activity >100GBq/µmol. B) HPLC purification shows majority of radioiodinated product is distinct from boronato-precursor and identical to cold trametinib. C) Trametinib is a type-III MEK allosteric inhibitor designed to reversibly inhibit ATP from phosphorylating MEK1/2. 124I-trametinib addition to cell lines expressing reveal a nearly ten-fold range of uptake after 2h, all of which are blocked with 1µg cold trametinib. Human KRAS mutant (G12V, G12C or G12D) or BRAFV600E cell lines retained a higher percentage of 124I-trametinib with the double KRAS BRAF mutant having the highest uptake differentiating KRAS and BRAF mutant tissue from normal. Murine cancer cell lines were all similarly avid for 124I-trametinib except Raw264.7 representing normal KRAS and BRAF and lower uptake. 

Ryan Davis, University of California Davis, rydavis@ucdavis.edu

Abstract Body : Introduction: The integrin αvβ6 is a cell surface receptor that is undetectable in healthy adult tissue but is over expressed in many cancers and has been proven a good target for positron emission tomography (PET) imaging of metastatic disease with a fluorine-18 radiolabeled peptide [18F] avb6-BP.1-3 The integrin avb6 binding peptide (avb6-BP)3 however is rapidly cleared from the blood, thus, the use of a serum albumin binding moiety (ABM) to increase blood half-life and tumor accumulation4,5 was explored. We compared two ABMs, Evans blue (EB) and p-iodophenylbutyryl (IP) for their ability to enhance the pharmacokinetic profile of avb6-BP. The ABM-peptides were synthesized with the DOTA chelator for radiolabeling with copper-64 to yield [64Cu]Cu DOTA-EB-avb6-BP ( Abstract Body : Objectives: The integrin αvβ6 is a cell surface receptor that is undetectable in healthy adult tissue but is overexpressed in several carcinomas and an indicator of poor prognosis [1] [2] [3] [4] . Although 64Cu-labeled peptide-based positron emission tomography (PET) imaging agents targeting αvβ6 have been reported, their high kidney accumulation and limited circulation half-life remain a challenge [5, 6] . The incorporation of non-covalent albumin-binding motifs has shown to increase the circulation time of radiopharmaceuticals, resulting in higher uptake in target tissue, and, in some cases, reduced kidney uptake [7] [8] [9] . Here we evaluated the effect of incorporating an albumin-binding motif (ABM) into the αvβ6-BP (binding peptide). Methods: Peptides were synthesized without (1) and with (2) an iodophenyl-based ABM using Fmoc solid-phase chemistry and radiolabeled with copper-64 to yield [64Cu]1 and [64Cu]2 respectively. The affinity of natCu1 and natCu2 for αvβ6 was assessed by ELISA.

[64Cu]1 and [64Cu]2 were evaluated in vitro (cell binding, internalization) in the DX3puroβ6 (αvβ6 +) and DX3puro (αvβ6 -), and the pancreatic BxPC-3 (αvβ6 +) cells, in an albumin binding assay, and for serum stability. In vivo (PET/CT imaging) and biodistribution studies were done in mice bearing either the paired DX3puroβ6/DX3puro or BxPC-3 xenograft tumors. Conclusions: Incorporation of an ABM in the αvβ6-binding peptide ([64Cu]2) resulted in improved pharmacokinetics, including increased uptake in αvβ6 + tumors, increased circulation time and improved tumor/off target contrast. Acknowledgements: This work was funded by NIH NCI RO1CA199725 Figure. Reconstructed PET/CT maximum intensity projections (MIPs) of mice bearing DX3puroβ6 (αvβ6 +, green arrow) and DX3puro (αvβ6 -, red arrow) or BxPC-3 (αvβ6 +, magenta arrow) tumors using [64Cu]1 (without AB group) and [64Cu]2 (with AB group). Mice were anesthetized using 2-3% isoflurane in medical grade oxygen and received tail vein injections of the radiotracer (200-250 μCi) formulated in saline. Static 15-minute scans were acquired at 4 and 24 h p.i.; 30-minute and 60-minute scans were acquired at 48 h and 72 h, respectively. PET data are shown in color scale, CT data in gray. Abstract Body : Introduction: Fibroblasts are versatile cells that produce several extracellular matrix (ECM) proteins such as collagen 1 and extradomain A containing fibronectin, as well as degradative enzymes such as matrix metalloproteinases1. Fibroblasts play an important role in wound healing, and in different pathological conditions such as the extensive fibrosis observed in chronic conditions, and in cancer progression. We recently identified increased cancer associated fibroblasts (CAFs) in more metastatic prostate cancers2. To understand the influence of hypoxia that is frequently observed in prostate cancer in modifying fibroblast metabolism, here we characterized human prostate normal and cancer associated fibroblast metabolites under normoxic and hypoxic conditions. Methods: Experiments were performed using human prostate fibroblasts (WPMY-1, ATCC, Manassas, VA) and human prostate cancer associated fibroblasts (PCAFs, Asterand Bioscience, Detroit, MI). WPMY-1 were derived from stromal cells from the peripheral zone of the histologically normal adult prostate3. PCAFs were obtained from an adenocarcinoma of the prostate gland. In order to induce hypoxia, WPMY-1 and PCAF cells were incubated for 48h under hypoxic conditions (0% O2) using an induction chamber where the oxygen was displaced by 100% N2 at the start of incubation. For high resolution 1H MRS, cell extracts were obtained using a dual-phase extraction method4. High-resolution 1H MR spectra were recorded on a Bruker Biospin Avance-III 750 MHz NMR spectrometer. 1H MR spectra were manually phased and automated baseline corrected using TOPSPIN 3.2 software. Integrals of the metabolites of interest were determined and normalized to the TSP (trimethylsilylpropanoic acid) reference and the number of cells. Metabolites were estimated from at least four experimental samples. Statistical significance was evaluated using the Student t test. Results: Metabolic profiles were significantly different between WPMY-1 and PCAFs under normoxic conditions, as shown in the heat map in Figure 1a , with significant differences observed in several water-soluble metabolites. On the other hand, other than a slight elevation in the levels of sphingomyelin and the total lipid content, lipid profiles were similar in WPMY-1 and PCAFs under normoxia. WMPY-1s responded to hypoxia by significantly increasing several metabolites although the lipids remained unaffected, as indicated in Figure 1b (Persuasive data Figure 1 ). PCAFs, on the other hand, responded to hypoxia by significantly increasing the lipid content, whereas only a few metabolites showed changes, such as an increase in phosphocholine and a decrease in glutamine, as shown in Figure 1c (Persuasive data Figure 2 ). Discussion: These results are especially relevant in the view of recent studies reporting that, in the context of cancer metastasis, cancer cells do not behave differently but the fibroblasts respond differently to cancer cells5. Here we found that the metabolic response to hypoxia of normal (WMPY-1) and cancer associated (PCAF) fibroblasts was very different. This is important, as these two cell populations are spatially localized in different regions. Normal fibroblasts are usually found at the interface of the tumor and adjacent normal tissue and are associated with HIF expression6. We previously reported that hypoxia triggered significantly faster degradation of the extracellular matrix by fibroblasts suggesting that under hypoxic conditions these fibroblast may be assisting invasion and facilitating metastasis. PCAFs are located within the tumor. The metabolic changes observed in PCAFs in response to hypoxia suggest that they may be playing an important role in immune suppression. Different studies reported that lipid accumulation was found to skew tumor infiltrating myeloid cells towards immunosuppressive and anti-inflammatory phenotypes7,8. These results support targeting fibroblasts within and at the periphery of prostate cancers to improve survival. Metabolic or lipid inhibitors can provide treatment strategies to target normal fibroblasts and PCAFs. Acknowledgements: This work was supported by NIH R35CA209960 and R01CA82337. JPT was funded by Fundación Alonso Martín Escudero y MSCA. (1) (2) (3) (4) impact BOLD fMRI signal intensity (i.e., the BOLD mechanism) of a human subject. We hypnotized that for a patient with Arteriovenous Malformations (AVM) the distance of the lesion to the BOLD signal is one of the most important parameters on the BOLD mechanism. Hence in this investigation, we applied BOLD task fMRI and resting state fMRI modalities to AVM patients and healthy human subjects for understanding the impact from the distance on the BOLD mechanisms. Retrospectively we studied brain MRI data of 8 AVM patients with their lesions in or adjacent to left frontal or temporal lobes, and 7 age and gender matched healthy human subjects. The MRI data were from T1-weighted images, task BOLD fMRI from the right hand only finger tapping task and resting state BOLD fMRI (rsfMRI). The task and resting state fMRI data's group and individual analyses were performed by using FSL (6) and AFNI (7). We also did a t-test for determining the statistically significant difference between AVM and normal subjects in motor function. A 2 tails pair T-Test was applied, and a P value is less than 0.05 to be considered statistically significant. Results: In Figure 1 , we showed the task fMRI BOLD, and the resting state BOLD co-registered on the T1 weighted image for an AVM case (the top row) and a normal subject (the bottom row).The averages and standard deviations (STDs) of the BOLD signals (i.e., r values) in the AVM and non-AVM (NAVM) sides for these 8 patients and the signals in two brain hemispheres for the 7 normal (Nor) subjects were measured and listed in the Table 1 , and the P values for the signals of the AVM to NAVM sides for the patients and for the left to right brain hemispheres of the NOR were also calculated by the T-Test and listed in the table. Also the distances between AVMs to the PMC fMRI signals for the patients were included in the table. The correlation plot and linear fitting for the fMRI BOLD ratios (AVM/NAVM) VS. the distance (mm) based on the data in the Table 1 was plotted in the Figure 2 . The images from the group analyses for the task BOLD (the top raw) and resting state (the bottom row) BOLD images of the AVM and NOR subjects groups were shown in the Figure 3 , in which the time course and the power spectrum for each image were also demonstrated. In Table 2 we listed the Z (i.e., the signal) values from the group analyses for the AVMs' task BOLD (i.e., fMRI) and resting state BOLD (i.e., rsfMRI) and also for the Normals' task BOLD and resting state BOLD. Figure 2 demonstrated that with the lesion distance increasing the BOLD fMRI signal ratio (AVM/NAVM) increased, and the linear correlation was significant (r>95%). This suggested that there was dependency between the lesion distance and the task BOLD functional signal of the AVM patients. The BOLD fMRI signal in the AVM side seems to be reduced linearly with the distance due to AVM's stealing effect on CBF to the functional areas. From the results of the group analyses, the differences between the fMRI and rsfMRI BOLD for the patients were demonstrated from the BOLD ratios, and corresponding timing courses and power spectra. The group analyses results implied that the AVM has great impact on the BOLD responses resulting likely from its disruption of the blood supply to the functional areas.

References: References: [1] Buxton, RB, et. al., Neuroimage, 2004 , 23 Suppl 1:S220-33. [2] RW, COX, Computers and Biomedical Research, 1996, 162-173. [3] Smith, SM., et. al., NeuroImage, 2004, 208-219. [4] Hou, BL, et. al., Neuroimage, 2006, 489-497. [5] The AVM Study Group, N Engl J Med, 1999, 1812-1818. [6]Smith, SM., et. al., NeuroImage, 2004, 208-219. [7] COX, RW, Computers and Biomedical Research, 1996, 162-173. Table 1 :The averages and standard deviations (STDs) of the BOLD signals (i.e., r values) in the AVM and non-AVM (NAVM) sides for these 8 patients with the distances between AVMs to the PMC fMRI signals, and the signals in two brain hemispheres for the 7 normal (Nor) subjects. Figure 2 : The correlation plot and linear fitting for the fMRI BOLD ratios (AVM/NAVM) VS. the distance (mm). Figure 3 : The images from the group analyses for the task BOLD (the top raw) and resting state (the bottom row) BOLD images of the AVM and NOR subjects groups. Table 2 : The Z (i.e., the signal) values from the group analyses for the AVMs' task BOLD (i.e., fMRI) and resting state BOLD (i.e., rsfMRI) and also for the Normals' task BOLD and resting state BOLD.

First Name: Bob Last Name: Hou Email: bhou@hsc.wvu.edu

Carmen Azevedo, Stanford University, ecarmenazevedo@gmail.com

Abstract Body : Objectives: Multiple sclerosis (MS) affects over two million people globally1 and costs millions of dollars in lost income and healthcare. Several disease modifying therapies (DMTs) exist for certain types of MS but not all patients respond due to disease heterogeneity. One such DMT that has shown dramatic benefits for some MS patients targets B lymphocytes, but there is currently no way to predict response. Positron emission tomography (PET) imaging of B cells could aid in selecting patients for anti-B cell therapy and monitor their response in real time. The aim of this study was to create the first anti-human CD19 antibody (mAb) PET tracer and assess its utility for detecting B cell infiltration into the CNS of humanized CD19 transgenic mice (hCD19-tg) induced with experimental autoimmune encephalomyelitis (EAE). Methods: A mutated variant of inebilizumab (B cell DMT in clinical trials) was radiolabeled; this mAb lacks the ability to deplete B cells due to mutations at its FcRγ binding site while maintaining high affinity for CD19.2 Female hCD19-tg C57/BL6J mice age 9 -13 weeks were induced with EAE through inoculation of myelin oligodendrocyte glycoprotein (MOG1-125) and observed daily for signs of paralysis from day 8 onward. On day 12 post inoculation, 3.0 ± 0.5 MBq [64Cu]hCD19-mAb was injected intravenously into 12 mice (n = 8 EAE, n = 4 naïve). PET/CT imaging was performed on all mice 19 -20 hours post injection. Percent of injected dose per gram (%ID/g) was determined by gamma counting of perfused organs of interest and PET signal was normalized to heart %ID/g. Autoradiography (ARG) was performed to obtain high resolution images of tracer distribution in brain regions, expressed as mean pixel intensity/decay corrected dose. Results: Radiolabeling of DOTA-conjugated mAb (2 -3 DOTA/mAb, MALDI-MS) was achieved by incubating 20 µg mAb with 37 MBq [64Cu]CuCl3 at 38 ˚C for 30 minutes, producing a radiotracer with high radiochemical purity (>99%, ITLC) and molar activity of 2 MBq/μg. The findings from PET, ARG, and biodistribution provide evidence that [64Cu]hCD19 can detect B cell infiltration into the CNS of EAE mice. Specifically, EAE mice had significantly higher PET signal in whole brain, medulla, and cerebellum, (Fig.1a , d; WB p Conclusion: The first human CD19-mAb was successfully radiolabeled and demonstrated its ability to detect B cells in humanized EAE mice. This new tracer represents a rapidly translatable tool for predicting and monitoring response to MS therapy. [64Cu]hCD19 PET of hCD19-tg mice reveals increased tracer binding in the brain of EAE vs naïve mice, shown by representative sagittal PET/CT brain images taken 19 h post injection (a). Tracer signal in whole brain was calculated by coregistering a mouse brain atlas to PET/CT images (d, 0.160 ± 0.0156 vs 0.1265 ± 0.01603; p nissl stained brain tissue (b, overlay). Increased accumulation of tracer in brain was verified by ex vivo gamma counting and quantification of autoradiography images (e, brain stem 25.35 ± 6.308 vs 17.64 ± 2.457 MPI/dose, p p p p ex vivo gamma counting (f, TSC 1.104 ± 0.333 vs 0.2715 ± 0.0343 %ID/g, p p ***p-value p-value p-value < 0 .05 

Veronica Nagle, Memorial Sloan Kettering Cancer Center, naglev@mskcc.org

Abstract Body : Objectives: Glioblastoma Multiforme (GBM) is the most common malignant brain tumor in adults. Patients with GBM suffer from inevitable relapse and a median survival of less than two years1. Immunotherapeutic agents have shown antitumor activity against a variety of human cancers, including melanoma, non-small cell lung cancer, as well as brain metastases from these tumor types. However, GBMs have failed all immunotherapeutic phase 3 clinical trials, likely due to the inability to stratify and monitor patients in an immunologically informed manner2. Cytotoxic CD8+ T cells mitigate the therapeutic response to the majority of FDA approved immunotherapeutics and are currently indiscernible from tumor tissue in brain tumors with the current imaging modalities. We assessed a human-CD8+ T cell-targeted minibody positron emission tomography (PET) tracer (IAB22M2C) to monitor systemic and brain parenchymal cytotoxic T cell infiltrate in GBM-bearing humanized immune system (HIS) mice. Methods: Anti-Human-CD8a minibody was conjugated to 2-S-(4-Isothiocyanatobenzyl)-1,4,7triazacyclononane-1,4,7-triacetic acid (p-SCN-NOTA) and radiolabeled with 64-Copper (64Cu) for positron emission tomography (PET). To generate HIS mice, 107 HLA-typed PBMCs from a single donor (Stem Cell Technologies) were injected into the tail vein of 5-8 week old female NOG CIEA (Taconic). To generate GBM PDX bearing HIS mice, 5-8 week old female NOG CIEA (Taconic) were IC implanted with PDX cells (provided by the MSK Brain Tumor Center). Three weeks following tumor implantation, mice were humanized with PBMCs with HLA partially matched to the major loci. Radiotracer was injected by tail vein and 15 min static PET/CT images were acquired at 1, 4, 12, and 24 h using a dual microPET/CT scanner (Inveon, Siemens). T cells were quantified by multi-channel flow cytometry (spleen) and immunohistochemistry (brain). High resolution autoradiography was performed on PBS perfused frozen brains. Results: Here, we show [64Cu]Cu-NOTA-anti-CD8 is able to quantitatively detect systemic human CD8+ T cell burden as well as CD8+ tumor-infiltrating T cells in an orthotopic GBM PDX HIS mouse model. Further, we demonstrate the variability of donor-dependent human T cell infiltrate in the peripheral blood mononuclear cell (PBMC) HIS mouse model and show that [64Cu]Cu-NOTA-anti-CD8 uptake correlates with donor-dependent T cell infiltrate in a orthotopic GBM patient-derived xenograft (PDX). Conclusion: These data indicate the capacity of [64Cu]Cu-NOTA-anti-CD8 to non-invasively monitor peripheral and brain tumor human CD8+ T cell burden with remarkable specificity. The anti-CD8 tracer IAB22M2C has passed phase 1 clinical trials in peripheral solid tumors3 and with the data herein, we propose the rapid clinical translation of this radiotracer for utility in human GBMs as well as patients suffering other brain tumors. The validation of this tracer as a tool to monitor GBM CD8+ tumor infiltrating lymphocytes will enable clinicians to non-invasively evaluate T cell responses in brain tumors, leading to both better designed and immunologically informed clinical trials.

Ksenia Lisova, Crump Institute for Molecular Imaging,UCLA, ksenlisan@gmail.com Category: New Chemistry, Biology & Bioengineering Abstract Body : Introduction: Conventional automated radiosynthesizers are generally optimized for producing large batches of PET tracers, while small portions needed for a typical preclinical imaging cannot be economically produced with such techniques. In this work we describe the first reported microvolume procedure for production of injectable [18F] Florbetaben [2] ([18F]FBB) for use in imaging of β-amyloid -a hallmark of Alzheimer's disease. Methods: Typically, production of [18F] FBB starts with a 2-step synthesis using an N-Boc-protected precursor (Fig. 1A) , followed by semi-preparative HPLC purification and reformulation by solidphase extraction (SPE) [1] , [3] . Reagent amounts in the conventional method were scaled down 90-fold to implement the synthesis in droplet format on Teflon-coated silicon chips [4] . In addition, semi-prep HPLC was replaced with an analytical scale method, and an automated SPE system based on a miniature cartridge was developed to formulate into small volumes (Fig. 1B) . Several parameters of the fluorination step were optimized (temperature, base amount, precursor amount), resulting in the following procedure. Aqueous [18F] fluoride was azeotropically dried with K2CO3/K222 (275/383 nmol) and then reacted with the Boc-protected precursor (80 nmol) in 10 μL DMSO at 130°C for 5 min. The resulting intermediate was deprotected with HCl at 90°C for 3 min and recovered from the chip in aqueous acetonitrile solution (80 µL of 1:1 v/v H2O:MeCN). The crude product was purified via analytical scale HPLC (Kinetex 5 µm C18 column, 250 x 4.6 mm; 1:1 v/v MeCN:20 mM phosphate buffer (pH 7.2), 1.5 mL/min; 254 nm). Formulation was performed using a miniature cartridge (fabricated by packing frits and C18 resin into a short length of tubing) installed into a custom automated formulation system driven by a LabView program to carry out the following procedure. The HPLC purified fraction was diluted with water and trapped on the cartridge. The cartridge was next washed with 10 mL DI water, after which the product was eluted with 150 µL of EtOH into amber vial with 180 µL PEG 400 and 670 µL DI water with 33 mg of sodium ascorbate. Results: Starting with 270 ± 100 MBq (n=3) of [18F] Fluoride, the method affords formulated [18F] FBB with RCY of 49 ± 3% (n=3) and >98% radiochemical purity and molar activity of 170 ± 50 GBq/μmol (n=3) after 55 min synthesis time. The miniature C18 cartridge enables efficient low-volume elution with EtOH and a small final volume (1.0 mL), which would not be feasible using standard cartridges, but ensures sufficient probe concentration for preclinical imaging even for small batches. Conclusion: This work describes an novel economic approach to produce large or small quantities of [18F] FBB for preclinical applications. 1 mL containing 77 ± 35 MBq (n=3) of formulated tracer was obtained in 55 min with very high molar activity. This method provides very high yields and molar activities compared to reported literature methods, and requires less space and shielding and significantly reduced amount of reagents. Abstract Body : Glioblastoma (GBM) is the most common and aggressive malignant primary brain tumor with extremely poor prognosis1. Treatment options for GBM remain limited, as systemically delivering effective dosage of therapeutics to GBM is challenging due to the physiological barriers, such as the blood-brain barrier (BBB) or blood-tumor barrier (BTB), blocking the delivery of most chemotherapy agents2. While theranostic nanoparticles offer advantages in drug delivery, including long blood circulation time, tumor targeted specificity and carrying water insoluble drugs, most nanoparticles, typically with core sizes over 10 nm, have not been successfully applied to intracranial brain tumors3. In this abstract, we present the initial investigation of using sub-5 nm ultrafine iron oxide nanoparticles (uIONPs) with the core diameter of 3 nm to deliver a water insoluble chemotherapy agent SN38 to intracranial U87MG glioblastomas in mice. SN38 is the highly potent yet hydrophobic active metabolite of topoisomerase I inhibitor Irinotecan, and can be uploaded into the hydrophobic blocks of (polyethylene glycol)-block-(allyl glycidyl ether) (PEG-b-AGE) coating polymer of uIONPs in tetrahydrofuran (THF) upon mixing. Transferring the mixture of SN38 loaded PEG-b-AGE and uIONP to the aqueous phase promotes the hydrophobic moiety of PEG-b-AGE polymer to collapse onto the surface of uIONP, thus encapsulating SN38 molecules (uIONP/SN38). The loading efficiency of SN38 on the reported uIONP was determined to be 48.7 ± 6.4% (SN38/Fe wt%) by measuring the UV absorbance of SN38. Cytotoxicity of uIONP/SN38 towards U87MG glioblastoma cells was observed in the in vitro assay with IC50 of ~ 25 nM (SN38 dosage) comparing to ~ 2 µM obtained from free SN38. uIONP/SN38 was then functionalized with cyclo(Arg-Gly-Asp-d-Phe-Cys) (RDG) peptide as the ligand to actively target tumor integrins αvβ3 for delivering SN38 to the intracranial tumor of mouse model. RGD-conjugated uIONP/SN38 was labeled with near infrared dye NIR830 (NIR830-RGD-uIONP/SN38) for both MRI and optical imaging to track the delivery and intratumoral distribution of uIONP/SN38. The results showed accumulation of NIR830-RGD-uIONP in the intracranial tumors as strong NIR signals were observed from ex vivo imaging of collected tumors at 12 hours after intravenous (i.v.) injection. The immunofluorescence staining of collected brain tissue sections using antibody for αvβ3 integrins revealed the co-localized uIONPs with αvβ3 integrins antibody, indicating targeting of NIR830-RGD-uIONP/SN38 to the αvβ3 integrin over expressed U87MG glioblastoma. To study the efficacy of RGD-uIONP/SN38 in treating GBM, 24 mice bearing intracranial U87MG glioblastomas were randomly divided into four groups (n = 6/group) to receive different treatments, i.e., RGD-uIONP/SN38 as the treatment group (group 1), RGD-uIONP vehicle only (group 2), SN38 formulated with Cremophor EL (group 3) as a free drug control, and no treatment (group 4). RGD-uIONP/SN38 and SN38/Cremophor EL were given to mice every 3 days at the SN38 dosage of 5 mg/kg body weight 10 days after tumor implantation with tumor growth monitored by Gd contrast enhanced MRI. RGD-uIONP was given to mice at the Fe dosage equivalent to RGD-uIONP/SN38 in group 1. Longitudinal monitoring of tumor growth was performed after each administration of treatment. The results showed that mice receiving αvβ3 tumor integrin targeted RGD-uIONP/SN38 exhibited longer survival (12.7 ± 2.3 days) than those treated with SN38/Cremophor EL, which did not demonstrate significant improvement in survival (7.5 ± 3.8 days, P > 0.05) over the non-treated control (7.3 ± 0.5 days) . In summary, sub-5 nm uIONP with PEG-b-AGE coating demonstrated a high loading efficiency for highly potent yet hydrophobic small molecule drug SN38, and ~80-fold higher cytotoxicity towards glioblastoma cells than free SN38 as shown in in vitro assay. In an intracranial U87MG glioblastoma mouse model, tumor-integrin-targeted uIONP/SN38 conjugated with RGD can be delivered to the intracranial tumors to achieve efficacious treatment. Abstract Body : Current approaches to biomedical image quantification and analysis of Magnetic Particle Imaging (MPI) are thwarted by the lack of reliable, standardized methods of segmentation of a region of interest (ROI) due to inability of various human raters to determine threshold regarding the integrity of nanoparticle signal from background in images and the high degree of inter-rater variability and resulting selection bias from human error1,2,3,4. Furthermore, there are minimal methods for rapid and adequate prediction of Total Iron Value (TIV), which can be indicative of cellular nanoparticle uptake/content, from an ROI within an MPI Image scan5,6. This calls for the development of artificial intelligence (AI) systems for analysis of MPI. Here, we utilize a canonical algorithm in the domain of unsupervised machine learning, known as K-means++7, 8, 9, 10 , in order to precisely segment the regions of interest (ROI) of raw MPI image data and preform subsequent analysis of TIV. We have developed the algorithm to preform iron prediction tasks through a novel AI generated standard curve method, in which several fiducial markers within a scan are segmented by the algorithm, and their total pixel intensity sum and corresponding total iron amounts are used to generate a standard curve through linear regression, from which the total iron amount of an unknown ROI is predicted. For initial parameter optimization of the algorithm, 3D-printed phantoms of several sizes were generated in two shapes-Circle and a letter "S", and were injected with varying amounts of VivoTrax superparamagnetic iron oxide nanoparticles (SPIONs) and imaged with the Momentum MPI scanner (Magnetic Insight) to generate data on which the algorithm can be calibrated. From this, we observed a linear correlation between MPI signal intensity and iron amount within the phantoms, so the standard curve (SC) model was elected as a reliable model for total iron amount prediction and was used as the algorithm for quantification and predictive analysis of the segmented ROI throughout the rest of the study. In order to provide a robust domain of data to the algorithm and evaluate its performance in multiple MPI scenarios, we generated in vitro, in vivo, and ex vivo models of human pancreatic islets and applied the AI algorithm to these datasets in order to gain insight into segmentation and iron prediction on these images. In vitro models included labeling islets with VivoTrax SPIONs and imaging the cells in varying amounts. Immunofluorescence staining and fluorescence microscopy of labeled cells showed insulin expression and uptake of dextran coated VivoTrax nanoparticles by the islet cell clusters (Fig 1. A-D) . In vivo mouse models were generated through transplantation of increasing numbers of the labeled human islets under the kidney capsule of NOD/scid mice and mice were scanned in 2D using the Momentum MPI scanner (Magnetic Insight). A 3D MPI scan was co-registered with CT to determine correct anatomical location of islet cell transplant under kidney capsule of mice (Fig 1. E-H) . Ex Vivo data was collected from excised mouse kidneys which were imaged with the MPI scanner. The proposed K-means++ algorithm and SC model were then applied to the MPI data for segmentation, quantification and TIV analysis of the transplanted labeled islets (Fig 1. I & J) . Through statistical evaluation of the algorithm performance via intraclass correlation coefficient validation on these datasets, we observed the ability of the K-means++ based machine learning model employed in this study to provide a novel, standardized method of segmentation, quantification, and analysis of MPI scans of transplanted human islets across a wide domain of data. This will provide a powerful tool for MPI segmentation and ROI analysis for applications in cell-based therapy including islet transplantation, stem cell transplantation, CAR T cell therapy, et al. in vivo. 

Neha Antil, Stanford University, drantil7@stanford.edu

Abstract Body : Purpose: Ovarian cancer (OC) is the fifth leading cause of cancer-related deaths in women and the most lethal gynecologic cancer. Ovarian cancer usually remains asymptomatic until advanced stages, but is curable when discovered early. Thus it is crucial to diagnose the disease at early stage for proper patient management. Conventional transvaginal ultrasound (US) imaging suffers from limited sensitivity and specificity in the detection of ovarian cancer. Adding molecularly targeted ligands to contrast microbubbles enables US to detect, characterize and monitor ovarian cancer at a molecular level. Using kinase insert domain receptor (KDR)targeted ultrasound molecular imaging (USMI) in this clinical study, we sought to correlate USMI signal of a variety of ovarian lesions with histology and immunohistochemistry (IHC) staining for KDR. Material and Methods: Ten women (mean age 55.0±14.6 years, range 28 to 76 years) with complex ovarian lesions were prospectively recruited in this HIPAA-compliant IRB-approved study after meeting eligibility criteria and obtaining informed consent. USMI was performed on clinical EPIQ 7 US scanner using endovaginal broadband curved array transducer (C10-3v). The complex ovarian lesion was first identified on B-Mode US and a plane showing a solid component (target lesion) was then selected for further evaluation. While imaging in dual mode with B-mode and contrast mode side-by-side, clinical grade KDR-targeted microbubbles (MBKDR, BR55, Bracco, 0.03 mL/kg of body weight) were manually injected over 10 seconds followed by 10 mL saline flush. Imaging was obtained starting with initial 45 seconds acquisition to capture the wash-in phase of MBKDR, followed by 10 second acquisition every 2 mins until 30 minutes post injection. Blood pressure, ECG, oxygen levels, heart rate, CBC, and metabolic panel were obtained before and after MBKDR administration. Surgery for these complex ovarian lesions was performed within 2-week interval following imaging (5.6±5.2 days, range 1-14 days) after USMI. Tumor regions overlying the imaged target lesions were marked by the surgeon for accurate correlation on histology. For USMI analysis, USMI stationary signal was categorized into 1: no, 2: weak and 3: strong signal, which was then correlated with ex-vivo KDR signal expression along with other marker expression like CD31, B7H3 and prostatespecific membrane antigen (PSMA). Results: All patients tolerated MBKDR without adverse events. 7 patients were included in the data analysis, surgery was not performed in one patient; ex vivo analysis was not yet completed in one patient; one lesion was intraoperatively found outside a severely atrophic ovary in one patient. In all 7 patients, successful intravenous administration of MBKDR was confirmed with enhancement of the solid component of lesions during wash-in phase of MBKDR within the first 45 seconds. 3 out of 7 lesions were benign (43%), and 4 lesions were borderline (low malignant potential) tumors (57%) on histology. The enhancement observed in the 7 lesions ranged from none (n=2) over weak (n=3) to strong (n=2), respectively. In all 7 patients the imaged target lesions could be identified on gross specimens and processed for IHC. In-vivo USMI signal showed correlations with ex-vivo IHC for KDR in 86% (6/7 lesions), whereas B7H3 and PSMA expression was observed in 57% (4/7) 43% (3/7) lesions, respectively. Conclusion: KDR-targeted USMI is feasible and safe, and enables noninvasive detection of KDR expression in patients with ovarian lesions. All imaged target lesions could be identified on gross specimens for IHC work-up and in-vivo USMI signals matched with ex-vivo KDR expression in the majority of cases and warrants further exploration. USMI has potential to be used with other types of US MBs targeted at novel biomarkers such as B7H3 and PSMA. Abstract Body : We have developed near-infrared (NIR) luciferin analogues that exhibit firefly bioluminescence, including compounds dubbed "TokeOni" and "seMpai," which are now commercially available. In this conference, I will present the characteristics of NIR luciferin analogues. Firefly bioluminescence, a process whereby light is emitted with high efficiency, has found widespread use in life science applications. Notably, NIR light is suitable for use in in vivo imaging because deep tissues are highly permeable to light in the relevant wavelength range (650-900 nm). However, the wavelength of light emitted via firefly bioluminescence is usually yellow-green (560 nm). We performed studies on the structure-activity relationship of compounds producing firefly bioluminescence in order to synthesize luciferin analogues emitting NIR light. Thus, AkaLumine, a luciferin analogue that produces light at 675 nm wavelength, was developed. This compound was shown to yield useful results when applied to the in vivo imaging of mice. We have improved the characteristics of AkaLumine to optimize this compound for use in in vivo imaging, thus obtaining the compounds dubbed TokeOni and seMpai. Recently, Akaluc, an artificial enzyme that specifically catalyzes the light-producing oxidation of TokeOni, has been developed (Iwano S., et al., Science, 2018, 359, 935-939) . As a result, it became possible to in vivo imaging for larger animals than mice like micro pig and marmoset. In this way, bioluminescence imaging technology is undergoing constant innovation, and we continue to develop NIR luciferin analogues that can be used as imaging tools. In this conference, I will present the characteristics of recently developed NIR luciferin analogues.

References: Iwano S., et al., Science, 2018 , 359, 935-939 Kuchimaru, T. et al. Nat. Commun. 2016 , 7, 11856. Saito, R. et al. BCSJ, 2019 Full Name of Abstract's 1st Author : Nobuo Kitada Abstract Body : Magnetic particle imaging (MPI) is a tracer imaging modality that detects superparamagnetic iron oxide nanoparticles (SPIOs). MPI has shown promise for sensitive (~200 cell), long term tracking of stem cells [1, 2] . White blood cell (WBC) tracking using scintigraphy has traditionally allowed for the clinical diagnosis of inflammation and is ordered in cases of fever-of-unknown-origin [3, 4] . Recently, antibody based immunotherapeutic approaches have been developed to selectively modulate immune cells, e.g. for cancer treatment [5] . Tracking WBCs in the tumor microenvironment, especially specific phenotypes, has proved crucial to monitoring and predicting outcomes of immunotherapy [6] . Objective: We developed the first in vivo approach for the in situ labelling and tracking of neutrophils with MPI for a radiation-free alternative to traditional WBC scans. The Ly6 family of glycosylphosphatidylinositol-linked proteins has been implicated in neutrophil migration and recruitment. Anti-Ly6 antibodies have been used at high concentration in mice to deplete neutrophils from circulation [5] . Fluorescent conjugated Anti-Ly6 antibodies have also been used to track neutrophils [7] . We utilized Anti-Ly6G antibodies conjugated to SPIOs for in situ labelling of WBCs in mice for MPI-based WBC tracking. Methods: Inflammation was induced in a C57BL6 mouse (7-8 wks) by injecting lipopolysaccharide (LPS) (O111:B4, 50 ug) in the right thigh. The biodistribution was imaged after i.v. tail vein administration of Anti-Ly6G antibody (IgG1, REA526 clone, Miltenyi Biotec, GmBH) SPIOs (6-6.5 mg Fe/kg, ~40 ug of protein/mouse) using a 6.3 T/m field-free line MPI scanner (field-of-view = 10.1 × 4.7 cm2, tscan = 15 min). The SPIOs were also injected in another healthy C57BL6 mouse and scanned. The inflamed mouse had neutrophil activity validated with a bioluminescence scan using an i.p. injection of luminol (XenoLight RediJect Inflammation Probe, Perkin Elmer) [8] , imaged in an IVIS Lumina (Xenogen, Perkin Elmer). A high boost-filter was applied to the MPI scans to improve conspicuity. Separately, the Anti-Ly6G-Ab-SPIOs were physically and magnetically characterized using a vibrating sample magnetometer (Lakeshore Cryotronics), transmission electron microscope (JEOL, 80 KeV) and arbitrary wave relaxometer [9] , and compared to Vivotrax (Magnetic Insight), a commercial formulation of ferucarbotran used as a common MPI tracer, to assess magnetic properties and imaging performance. Results: SPIO characterization is shown in Fig 1A-F . The Anti-Ly6G-Ab-SPIOs were larger than than VivoTrax (Fig 1A-B) and showed improved performance in both resolution and signal per g Fe as compared to VivoTrax (Fig 1C-E) . Notably, the resolution of the Anti-Ly6G-Ab-SPIOs was better than Vivotrax (FWHMAnti-Ly6G = 8.8 mT, FWHMVivoTrax = 10.4 mT) despite the Anti-Ly6G-Ab-SPIO's saturation magnetization, an oft-cited metric for high-performing MPI tracers [10], being lower (Msat,Anti-Ly6G = 44 emu/g, Msat,VivoTrax = 131 emu/g), Moreover, the Ab-SPIOs show minimal change in sensitivity in whole blood versus in saline as compared to Vivotrax's variation in sensitivity (Fig 1F) . In the in vivo experiments, MPI images taken 24-hr post i.v. injection of SPIOs (Fig 1G) showed Anti-Ly6G Ab-SPIO distribution in organs of the reticuloendothelial system (liver, spleen, as well as cranial, lower limb, and pelvic bone marrow) in healthy mice. These results concur with the delayed distribution of labelled WBCs in In111 scans shown in Fig 1J [11] . Fig 1H is an MPI image in which the mouse's right flank is inflamed, and tracer accumulation at the inflamed site appears with high contrast. This image and inflammation were validated with our luminol-based images, showing the myeloperoxidase activity in the inflamed part (Fig 1K) . Conclusion: We performed the first WBC tracking MPI study targeting a surface specific ligand on neutrophils and observed homing of labelled neutrophils to inflamed sites, and demonstrated antibody conjugated SPIOs' superior MPI behaviour as compared to VivoTrax. This zero-radiation approach can be invaluable for in situ labelled WBC-MPI scans for inflammation and infection and for optimizing antibody-based cancer immunotherapies. The magnetization curve of VT and the Anti-Ly6G SPIOs as measured in a vibrating sample magnetometer was fit to a log-normal diameter distribution (Anti-Ly6G SPIO diameter mean 12.6 nm, with log-normal standard deviation of 1.1, and VT diameter mean 8.6, log-normal standard deviation of 14) Notably, VT has a saturation magnetization that is 3 times that of the Anti-Ly6G particles. We used a homebuilt relaxometer to measure the MPI point spread function (PSF) of (D) VT and (E) Anti-Ly6G SPIOs in saline and in blood, normalized to the maximum MPI signal per g Fe from both particles. Note the solvent-dependent blur in the PSF of VT in comparison unchanging behaviour of the Anti-Ly6G SPIOs, as well the Anti-Ly6G SPIO's superior full width half max and signal. (F) A standard curve of MPI signal was taken versus iron concentration for all particles in mouse blood and in saline. Both particles show linearity with concentration (R2 > 0.99). VT showed significant (p 0.5) MPI images of in situ labeled neutrophils using anti-Ly6G iron oxide tracers and reference images from other modalities. (G) The tracer signal reveals the distribution of neutrophils in the myeloid cells of the bone marrow, and the RES organs of liver and spleen. (H) MPI images of a C57BL6 mouse with inflamed right flank. Images 24-hours post anti-Ly6G tracer administration revealed the site of inflammation with high contrast -note the asymmetric signal in the flank of the mouse. (J) A healthy human WBC scan for comparison with (G) -note the uptake in the organs of the reticuloendothelial system (bone marrow, liver, spleen) (adapted from Love and Palestro J. Nucl. Med. Tech. 2004) (K) A bioluminescence image of an inflamed mouse with an i.p injection of luminol, a marker of neutrophil myeloperoxidase activity, co-registered to a representative x-ray. Note the strong luminescence in the inflamed right flank, similar to (H). 

Rebecca Faresjö, Uppsala University, rebecca.faresjo@gmail.com

Abstract Body : Introduction: If engineered to enter the brain, antibodies with high specificity to intra-brain proteins, for example amyloid-beta (Aβ), can be used in positron emission tomography (PET) to study diseases of the brain, such as Alzheimer's disease. Previously we have developed bispecific antibodies, utilising the mouse transferrin receptor as a transport shuttle, with high brain uptake1. These bispecific antibodies are large constructs with long residence time in blood, which is not optimal for PET. Here we have developed a smaller bispecific antibody-construct with improved faster blood clearance. However, knowledge of brain clearance time is lacking. The aim was to study brain pharmacokinetics of bispecific antibody-based ligands. Methods: Bispecific antibody ligands mAb3D6-scFv8D3 (210 kDA) and di-scFv3D6-8D3 (58 kDa) were expressed in Expi293-cells and purified by affinity chromatography. The ligands were 125I-labelled and each of them administered i.v. to wild-type mice (C57/Bl6). Brains were isolated 2h-24h post-injection, ligand uptake was measured with a gamma counter, then frozen. Brains were sectioned sagittaly and exposed to a phosphor screen for 7 days. Results: [125I]mAb3D6-scFv8D3 displayed slow kinetics, brain concentrations were high and stable for 8 h before elimination became apparent, while the smaller [125I]di-scFv3D6-8D3 started to be cleared from the brain earlier, at 4 h post administration. Retention and elimination phases of [125I]mAb3D6-scFv8D3 were confirmed by autoradiography (Fig. 1 ). Brain-to-blood ratio was 1.5-fold higher 12h after administration of [125I]di-scFv3D6-8D3 than for [125I]mAb3D6-scFv8D3 and the difference increased with time. Conclusions: Nonspecifically bound small non-IgG constructs (e.g. di-scFv3D6-8D3) clear faster from brain than larger conventional IgG-like constructs. This is likely to improve the specific-to-nonspecific signal, potentially allowing for fluorine-18 radiolabeling. (1). Early treatment can mitigate biofilmassociated recalcitrance to antibiotics (2) ; however, this relies on accurate detection, and there is no imaging technique to detect biofilms in vivo. Through optical-based screens, we identified a small, lipophile-enhanced peptide, (4Iph)(f)LPNSNHIKQGL (HN17), that targets the biofilm matrix that makes up ~80% of biofilm volume (3). We previously established that HN17-Cy5 rapidly penetrated P. aeruginosa biofilms and maintained biofilm selectively in epithelial cell infection models. Here, we expanded in vitro testing of HN17-Cy5 using a human primary neutrophil infection model and evaluated HN17-Cy5 in vivo in a biofilm-infected wound model. Materials and Methods: HN17 was synthesized by solid state methods at a purity of >90% and labeled with Cy5. Imaging flow cytometry was used to examine HN17-Cy5 interaction with human primary neutrophils infected or not with GFP-expressing P. aeruginosa (Pa; strain PAO1). Neutrophils were infected with Pa at an MOI of 5:1, washed, and incubated with HN17-Cy5 (2 μM) or PBS as a control for 20 min at 37°C with gently rocking. Following 2 additional washes, neutrophils were fixed and analyzed on an ImageStreamX Mark II (Amnis Corporation). Data for 10,000 cells were recorded for each sample with live/dead stain, GFP, and HN17-Cy5 fluorescence analyzed for each cell along with 60x brightfield images. Two replicate experiments were conducted using neutrophils from two different donors. Data were analyzed using IDEAS v6.0 software. In vivo biofilm targeting of HN17-Cy5 was evaluated using a biofilm-infected wound model. Briefly, wounds of 4 mice were infected with an inoculum of 107 cells consisting of a 50:50 mixture of lux-tagged and TdTomato-expressing Pa. 24 h after wound inoculation, mice were administered 60 nmoles of HN17-Cy5 via the tail vein. Control groups include n=3 mice that were mock infected in both wounds and injected with 60 nmoles of HN17-Cy5 and n=2 mice that were infected with Pa but injected with PBS. At 6 and 18 h after probe injection, live whole-animal bioluminescence and fluorescence imaging was performed on an IVIS Lumina II optical imaging system to assess bacterial burden and HN17-Cy5 distribution. Region of interest analysis was performed using Living Image 4.4 software (Caliper Life Sciences, Hopkinton, MA). Following imaging, mice were euthanized and tissues harvested, fixed, and sectioned. Confocal laser scanning microscopy (CLSM) of the sectioned tissue was performed on the Olympus Fluoview 3000 confocal microscope using a 60x oil objective (N.A. 1.4). Images were processed using the Olympus Fluoview software. Results: HN17-Cy5 targeting, as measured by mean fluorescence intensity, was significantly higher in the Painfected compared to uninfected neutrophils (p = 0.008). Discrete punctate spots of HN17-Cy5 fluorescence were apparent in the vast majority of infected neutrophils, though intracellular colocalization between HN17-Cy5 and GFP was lo< ( < 0 .2). In vivo, wound ROI analysis showed that HN17-Cy5 localized to biofilm-infected wounds 3.2 times greater than surrounding tissue and showed significantly higher uptake in biofilm-infected wounds compared to uninfected wounds at 18 h<(p < 0 .001). Ex vivo CLSM imaging of wound tissue revealed punctate HN17-Cy5 signals that co-localized with P. aeruginosa. Minimal Cy5 fluorescence was observed in uninfected wound tissue. Conclusion: Bacteria encased within a self-produced matrix known as biofilms cause devastating, chronic infections. Through optical screens, we identified a small peptide (HN17) that specifically targeted P. aeruginosa biofilms in mammalian cell infection models and showed enhanced accumulation in biofilm-infected tissues in vivo. HN17 represents a novel approach to imaging biofilms and additional studies are warranted to determine pathogen range and targeting mechanism. . ROI analysis showed that HN17-Cy5 localized to biofilm-infected wounds 3.2 times greater than surrounding tissue and showed significantly higher uptake in biofilm-infected wounds compared to uninfected wounds (C; p < 0 .001). Ex vivo confocal microscopy imaging showing HN17-Cy5 (D3) co-localized with TdTomatoexpressing P. aeruginosa biofilms (D2) while not tracking with host cells (D1; stained with DAPI). Merged channels are shown in D4. (PEPT) is based on the accurate localisation of a single labelled particle using a pair of back-to-back γ-rays generated by the annihilation of a positron.1 Our aim is to explore the use of PEPT for biomedical applications, however, there is a lack of research in particulates that allow the high specific activities (0.5 -2 MBq/particle) in the required size range ( < 5 0 µm).2 Whilst the radiolabelling of macroscopic materials provide isolated particles with high activity, the decrease in particle size brings associated a significant reduction in the activity per particle. Most importantly, there are no optimised protocols to isolate a single radiolabelled microparticle below 50 µm. In this work, a series of SiO2 particles ranging from 3 µm to 3 mm were synthesised, radiolabelled with 68Ga, and evaluated as potential PEPT radiotracers. Moreover, flow cytometry was evaluated for particle isolation. Hypothesis: Considering that the radiolabelling of SiO2 particles with 68Ga is a straightforward and fast protocol with high radiolabelling yields,3 it seems a suitable candidate to render tracers with the radioactivity per particle required for PEPT. Flow cytometry has potential as the isolation/purification technique of PEPT tracers due to its ability to isolate single cells of the same range of size. Methods: SiO2 microparticles (3 µm) were synthesised using a brief modification of the Stöber method.4 SiO2 particles of 150 -250 µm (Sigma-Aldrich) and 1 -3 mm (VWR chemicals) were also used for 68Ga labelling. Radiolabelling: Aliquots of 68GaCl3 were adjusted to pH = 5.2 with HEPES buffer. The particles suspended at different concentrations were added over the 68Ga solution. The mixture was heated at 90ºC for 30 min. Isolation: 3 µm particles at 0.25 mg/mL were isolated using a FACS MelodyTM. Sorting was carried out using a 6-well plate and a microcentrifuge tube holder independently. Results: Radiolabelling yields of 3 µm particles were very high (96%) even at a low concentration such as 2 µg/mL (Figure 1a ). 1 -3 mm and 150 -250 µm particles demonstrated high radiolabelling yields, 84 -86% (Figure 1b) . Strating from approximately 90 MBq, remarkable specific activities of 8.2 ± 1.0 MBq/particle was obtained for 1 -3 mm SiO2 and 1.4 ± 0.3 MBq/particle for 150 -250 µm SiO2 (Figure 1c ). 1 -3 mm particles showed a high radiochemical stability in highly ionic aqueous solutions of 99.0 ± 0.5% at 5 min postincubation, remaining almost identical after 2 hours (99.2 ± 0.5%). Moreover, 150 -250 µm particles demonstrated a radiochemical stability of 95.1 ± 0.4% after 5 min of incubation increasing to 99.3 ± 0.6% for 2 hours (Figure 1d ). Finally, FACS sorting was used to isolate 3 µm SiO2 particles. The images showed good isolation capabilities for single particles, whilst aggregation was evident for 20 and 50 particles after sorting in a 6-well plate (Figure 1e ). To solve this issue, sorting was carried out in a microcentrifuge tube with 20 µL of H2O. After setting up a 10 particles sorting, images showed no aggregation with 8 spherical particles of homogeneous shape and size (Figure 1f ). Conclusions: 68Ga-SiO2 particles have been evaluated as potential PEPT tracers. The larger SiO2 particles showed unprecedent specific activities suitable for PEPT applications. For the small particles, specific activities are close to those required and optimisation of the radiochemistry protocol should allow for improvements. Noteworthy, a fresh generator could render required activities. Preliminary results show the potential of FACS for the isolation of single particles. , 1993, 326, 592-607. 2. D. J. Parker and X. Fan, Particuology, 2008, 6, 16-23. 3 Figure 1 . a) Radiolabelling yields of 3 µm 68Ga-SiO2 particles at different particle concentrations (mg SiO2/mL, b) Radiolabelling yield of 1-3 mm 68Ga-SiO2 and 150-250 µm 68Ga-SiO2 particles after three different repetitions, c) Specific activity (MBq/particle) for 1-3 mm 68Ga-SiO2 and 150-250 µm 68Ga-SiO2 particles after three different repetitions, d) Radiochemical stability of 1-3 mm 68Ga-SiO2 and 150-250 µm 68Ga-SiO2 particles from 5 min to 120 min after heavy stirring in 5% NaCl, e) Optical microscope images taken after the sorting of 1, 20 and 50 SiO2 (3 µm) particles in a 6-well plate, f) Optical microscope image of the Neubauer chamber after the sorting of 10 SiO2 (3 µm) particles. Abstract Body : Over the last twenty years, immunoconjugates have become essential therapeutic and diagnostic tools. However, many antibody-drug conjugates (ADCs) and radioimmunoconjugates are still synthesized via the indiscriminate attachment of cargos (e.g. chelators, fluorophores, or toxins) to lysine residues, a remarkably imprecise approach that limits their overall efficacy. Indeed, these non-site-specific synthetic methods not only produce heterogeneous mixtures of immunoconjugates but also yield constructs with suboptimal immunoreactivity and in vivo performance. To overcome these problems, several "site-specific" bioconjugation strategies have been developed. The most established of these is the use of maleimide-based bifunctional probes for conjugations to cysteine residues. However, the resultant thioether linkages are unstable in vivo both spontaneously (via the retro-Michael reaction) as well as in the presence of competing thiols. In the context of radioimmunoconjugates, this can result in the detachment of radionuclides from the immunoglobulin vector, thereby reducing target-to-background activity concentration ratios and increasing radiation doses to healthy tissues. Herein, we present the synthesis, characterization, and evaluation of DiPODS, a novel bioconjugation reagent containing two oxadiazolyl methyl sulfone moieties. DiPODS irreversibly forms covalent bonds with two thiols, thus facilitating the re-bridging of reduced disulfide linkages in antibodies, antibody fragments, and proteins. DiPODS was synthesized from commercially available starting materials in 8 steps. Subsequently, a fluorescein-labeled derivative of the reagent -DiPODS-FITC -was created to interrogate the performance of DiPODS for the bioconjugation of both isotype-control and HER2-targeting Fab fragments. Ultimately, the modification of a HER2-targeting Fab with DiPODS-FITC produced an immunoconjugate with in vitro performance superior to an analogous construct created using a traditional, lysine-based approach to bioconjugation.

A) Schematic site-specific bioconjugation of a bifunctional DiPODSbearing reagent to proteins via the re-bridging of disulfide linkages. B) Flow cytometry analysis of HER2-positive BT474 human breast cancer cells stained with (left) a site-specifically modified FabHER2-DiPODS-FITC) synthesized using DiPODS and (right) a non-sitespecifically modified HER2-targeting immunoconjugate (FabHER2-Lys-FITC) synthesized using a traditional lysine-based conjugation method.

First Name: GUILLAUME (MPI) is an imaging modality that directly detects the nonlinear response of magnetic nanoparticles (MNPs). Spatial encoding is realized by saturating MNPs almost everywhere except in the vicinity of a special point called the field-free point (FFP) using a static magnetic field [1] . Recently, It has been shown that the sensitivity of MPI can be significantly improved using a simultaneous encoding scheme by scanning the FOV with a field-free line (FFL) instead of FFP [2] [3] .We have developed the MPI with FFL device that can change magnetic field gradient using a neodymium magnet and an iron yoke (Fig.1a) . We study of the influence magnetic field gradient of FFL on imaging quality. Methods: The magnet for FFL was designed using magnetic field analysis so that magnetic field gradient changed from 1.0 to 3.0T/m while the distance of the pair of magnets (Fig.1b ).An oscillating magnetic field was generated using an excitation coil (pair of solenoid coils). The frequency and peak-to-peak strength of the oscillating magnetic field were taken as 500 Hz and 30mT, respectively. The signal generated by MNPs was received by a gradiometer coil and the third harmonic signal was amplified by a lock-in amplifier. We used Resovist® (Fujifilm RI Pharma) as MNPs. Resovist® is an organ-specific contrast agent for magnetic resonance imaging and consists of superparamagnetic iron oxide (magnetite, γ-Fe2O3) coated with carboxydextran. The spatial resolution by the gradient magnetic field strength was evaluated using a cylindrical sample of φ5 mm and a volume of 0.15mL (Fig.1c) . The spatial resolution was quantified by the full-width-at-half-maximum (FWHM) of the signal intensity. Results: Fig.1d shows the measurement results of the signal intensity distribution with various magnetic field gradient. It is shown that the higher magnetic field gradient, the better the spatial resolution. It is also shown that the signal intensity decreases as magnetic field gradient increases (Fig.1e ). This indicates that the region of magnetically saturated MNPs increases with increasing magnetic field gradient. We will discuss the relationship between magnetic field gradient distribution of FFL and MPI signal. This research was supported by AMED under Grant Number JP19hm0102073h0001 ID: GA377 PSMA-targeted photodynamic therapy in surgical prostate tumour samples Melline Schilham, Radboud University Medical Center, melline.schilham@gmail.com

Abstract Body : Introduction Complete surgical resection of the primary tumour is one of the most important aspects of curative therapy in prostate cancer (PCa) patients. Incomplete resection is associated with recurrent disease and worse prognosis (1) . However, too extensive resection may cause damage to vital structures potentially causing post-surgical complications such as incontinence and impotence (2, 3) . These surgical challenges emphasize the significance of improved intraoperative visualisation of tumour margins and adjuvant ablative therapies. Photodynamic agents have the potential of combining both fluorescence imaging for imageguided surgery as well as ablation of cancer cells induced by photodynamic therapy (PDT). Labelling prostate specific membrane antigen (PSMA) ligands with a photosensitiser may thus enable intraoperative tumour detection, delineation and tumour-targeted PDT (tPDT) (4) . Previously, the PSMA-targeting tPDT ligand PSMA-N064 was developed and evaluated. To make the next translational step before clinical implementation, the present study aims to evaluate the potential for tPDT in an ex vivo incubation study on surgically obtained human PCa samples. Methods The PSMA-N064 peptide consists of a PSMA binding motif, a DOTA chelator and the photosensitizer IRDye700DX. In this study, PSMA-N064 is used to investigate the therapeutic effect of tPDT on human tumour and normal prostate tissue. We aim to include 25 PCa patients scheduled for radical prostatectomy for obtaining prostate tissue samples. The samples of the first ten patients will be used for protocol optimisation. Immediately after resection of the prostate, four samples are taken from the tumour and one from contralateral normal prostate tissue. The samples are incubated with 0.05 nmol/ml of the ligand in binding buffer (or binding buffer only, for the control) for 2-4 hours at 37 degrees Celsius. After washing, fluorescence flatbed scanning is used to evaluate tissue distribution of the ligand. Three out of five samples are illuminated with 690nm light (50-150J/cm2 at a fluence rate of 450 mW/cm2 ) using a light emitting diode. 4-24 hours later, the samples are fixated in 4% formalin, embedded in paraffin, sectioned (4µm slices) and stained (HE and PSMA expression). The therapeutic effect of tPDT on cells is evaluated by immunohistochemistry for cleaved-caspase-3 and γ-H2AX to assess apoptosis and double-strand DNA breaks (DSB), respectively. Viability of the fully treated tumour sample (tPDT-tumour) is compared to the fully treated normal prostate sample (tPDT-normal) and control tumour samples that receive either NIR-light irradiation only (control-1), ligand incubation only (control-2) or no treatment at all (control-3). Results Thus far, we obtained and treated samples from two patients. Macroscopic fluorescence imaging showed favourable accumulation of the ligand in tumour tissue samples (mean fluorescence intensity (MFI) 65296 ± 51284) compared to normal prostate tissue samples (MFI 5391 ± 2860), indicating a specific uptake of the ligand in tumour tissue. Preliminary results of cleavedcaspase-3 and γ-H2AX stainings suggested more profound apoptosis and DSB in the treated tumour samples compared to controls. However, NIR-light alone also seemed to increase apoptosis and DSB compared with the non-treated tumour sample (Figure 1 ). Additional apoptosis assays (e.g. TUNEL) will be performed and these results will be used to further optimise the protocol, i.e. duration of ligand incubation, washing steps, fluence and fluence rate of light delivery. After the optimal protocol has been determined, we will include and analyse samples from 15 additional patients. The results of this evaluation will be presented on the WMIC-2020 upon acceptation. Conclusion Although further evaluation of the results has to be completed, preliminary results of our study suggest an effect of tumour-targeted PDT using IRDye700DX-conjugated PSMA-N064 on human PCa-samples. In the future, this ligand may be of value in intraoperative tumour detection and tPDT by eradication of unresectable tumour rests and thus improving surgical outcomes. Abstract Body : The individual prognosis of a cancer patient is, much more than by characteristics of the primary tumour, determined by the fact whether of not metastases have already manifested. Treatment of patients with metastatic disease is considerably more challenging than local tumour therapy. Metastatic spread is not a random process. The distribution of metastases cannot be explained by mechanical properties such as vessel size or integrity alone but is specific for each tumour type. Tumours, by release of cytokines and other factors including exosomes, induce a premetastatic niche in the target tissue of metastasis. The premetastatic niche constitutes an envionment that can ideally host a circulating tumour cell. Protects the cell from anti-tumour immunity and provides it with growth factors. The major players in the niche are immune cells of various sort. And while tumour cells differ wildly, these immune cells are virtually the same across all different cancer types. The key to universal visualisation of the niche in order to unravel the process of metastatic spread for interference at a time, tumour cell seeding has not even happened is thus to address the immune cell components in the niche. Monocyte-specific, antibody-based imaging agents enabled for imaging of tumourinduced immune cell activity and accumulation in different models of metastatic cancer. The effects of the different tumours on target-tissue of metastasis reflected the individual malignant potential and allowed for prediction of the subsequent pattern of metastatic spread. Using this model system, different tumour-host communication media could be assessed for their respective effect. In this context, the distribution and biological effect of exosomes during premetastatic niche establishment could be shown. Imaging premetastatic tissue priming bears the potential to understand how tumours pave the way for systemic disease. Ultimately, these methods can help to make cancer controllable on a long term basis. Abstract Body : Background:Cyclooxygenase-2 (COX-2) is an inducible enzyme that catalyzes the production of prostaglandin E2 (PGE2) as a cellular response to inflammation. COX-2 overexpression has a pleiotropic and multifaceted role in inflammation and carcinogenesis. Upregulation of COX-2/PGE2 is associated with poor prognosis in breast cancer and other malignancies [1] . We previously found that COX-2 downregulation resulted in a significant decrease of collagen 1 (Col1 fibers), numbers of cancer associated fibroblasts (CAFs), and metastasis, whereas COX-2 overexpression resulted in a significant increase of Col1 fibers, CAFs, and metastasis [2] . Our purpose in this study was to determine if COX-2 driven changes in the ECM, including changes in Col1 fibers and their patterns, altered ECM stiffness. Altered ECM stiffness can provide mechanical cues to cancer cells through mechanotransduction pathways to modify migration and invasion. We used our established model of SUM-149 TNBC cells lentivirally transduced to overexpress COX-2 to characterize, for the first time, the effects of COX-2 overexpression on ECM stiffness using a novel microindentation assay. In parallel we are evaluating molecular changes in ECM proteins and proteases -such as fibronectin, kalinin, ADAM, ADAMTS1 -that actively play a dynamic role in the formation and degradation of ECM using immunohistochemistry (IHC), western blot analysis, and ELISA. Methods: Establishment of triple negative SUM-149 human breast cancer cells that stably overexpress COX-2 (SUM-149-COX-2) or with an empty vector (SUM-149-EV) was achieved through cloning and construction of a lentivirus vector expressing the COX-2 gene [2] . These cells were inoculated in the mammary fat pad of severe combined immunodeficient (SCID) mice. Nine weeks post-inoculation, excised xenograft tumors were measured using iNano nanoindenter (Nanomechanics Inc.) to measure tissue elastic modulus (i.e. stiffness). Downstream analyses of collected biological specimens including immunoblotting, immunohistochemistry, serum ELISA assay, and microscopy is ongoing. Indentation results were processed into resulting photomaps and heatmaps of indented regions and tissue stiffness, respectively. Results: Representative photomap and heatmap of tissue stiffness measured in a SUM-149-COX-2 tumor are shown in Figure 1A . Values for multiple indentation points measured per individual tumor and data summarized from elastic modulus-probed tumor tissues of SUM149-EV Control (n=5) vs. SUM-149-COX-2 (n=5) are displayed in Figure 1B . Discussion: Our indentation data identified, for the first time, a reduction in tumor stiffness with COX-2 overexpression. We are currently evaluating the relationship between differences in tumor tissue stiffness with COX-2 overexpression to molecular changes in ECM proteins such as fibronectin, collagen, kalinin, and ECM proteases such as ADAM and ADAMTS to further understand the role of COX-2 in altering the ECM and tissue stiffness. These insights may lead to an expanded understanding of the role of COX-2 and inflammation in cancer aggression. Acknowledgements: This work was supported by Supported by NIH R35CA209960 and R01CA82337. Abstract Body : Background/objective: Preclinical computed tomography (CT) is a widely used in vivo imaging technique providing anatomical information which is expanding its usefulness as a quantitative imaging tool. The expansion of utilizing CT as a dedicated system and with PET, SPECT or optical will aid with the interpretion and quantitative analysis of multi-modality imaging data sets. For example, cardiovascular research evaluating calcium deposits in vessels use CT Hounsfield Units (HU) to assist in quantifying the level of atherosclerotic plaque burden and vulnerability. Reconstructing CT images using different filters and parameters to improve spatial resolution and signal-to noise ratio is standard procedure. This study focuses on the potential impact reconstruction filters have on measured HU values (high or low pass frequency filters). Materials and Methods: A CT image was acquired of the air/water quality control and the tissue equivalent material (TEM) phantoms on the Mediso nanoPET/CT using a low dose protocol (tube voltage of 25kVp, exposure of 170ms and 360 projections, dose Results: As expected, data revealed slight variations across the different applied filters when other parameters remained constant. The greatest HU percent difference measured was between the high pass and low pass filters (Blackman vs Ram-Lak) of 15% in muscle HU. Otherwise the biases remained Conclusion: Measured HU results for muscle (soft tissue) were the most impacted by varying the reconstruction parameters. Choosing suitable CT reconstruction parameters requires going beyond spatial resolution and noise trade-offs to include considering potential quantitative biases in measured HU. Thus, the choices of reconstruction filtering warrants revisiting not only for improved signal-to-noise ratio but also for HU quantification. (1). A major immune response occurs that leads to the biodegradation of the implanted hydrogel and a reconstitution of brain tissue ensues (2) . To improve our understanding of the spatio-temporal dynamics of the immune cell infiltration into ECM hydrogel, we used 19F MR imaging to trace immune cells that were systemically labeled using perfluorcarbon (PFC) nanoemulsions. METHODS: Rats underwent middle cerebral artery occlusion to induce a stroke that lead to tissue cavitation. To tag circulating immune cells, PFC nanoemulsions (CelSense) were injected through the tail vein 12 days following stroke. On day 13 post-MCAo, a preimplantation 19F and T2-weighted MRI scans (9.4T) were acquired to visualize a baseline distribution of immune cells, as well as to verify the injection of PFC (i.e. blood vessels containing 19F signal). On day 14, ECM (4mg/mL) was implanted into the stroke cavity. On day 15, 24 hours post-implantation 19F and T2-weighted images were acquired to investigate the invasion and distribution of immune cells in vivo. Straight after MRI, animals were perfusionfixed for histological analyses. Fluorescently activated cell sorting (FACS) was performed to determine the phenotype of immune cells labeled in the blood. RESULTS: FACS indicated that of all PFC-labeled cells, 48% were macrophages, 40% neutrophils and 5% lymphocytes. Baseline 19F images revealed no infiltration of labeled immune cells into the stroke-damaged brain, but verified that PFC was circulating with a strong 19F signal present within major blood vessels ( Figure 1 ). 24 hours post-implantation, a strong 19F signal was evident within the lesion cavity implanted with ECM hydrogel, as well as the peri-infarct parenchyma. Moreover, PFC-labeled cells also produced a strong signal in the surgical wound on top of the head, providing a positive control for the host immune response to tissue damage. Histology further confirmed these in vivo results with 97% of macrophages in the ECM hydrogel containing PFC and 80% of neutrophils. No lymphocytes were present in the ECM hydrogel 24 hours post-implantation These results further indicate that the main immune response to ECM hydrogel implantation in the acute phase is mediated by invasion of peripheral immune cells into the CNS, rather than through a recruitment of local microglia. DISCUSSION & CONCLUSIONS: The spatio-temporal dynamics of immune cell invasion into ECM hydrogel after implantation into stroke are difficult to capture using histological methods. However, we here demonstrated that 19F MR imaging can potentially provide this type of information to improve our understanding of the contribution of immune response to hydrogel biodegradation and tissue restoration.

Aixia Sun, Michigan state university, sunaixia@msu.edu Category: Systemic Diseases (Kidney, Liver and Pancreas) Abstract Body : INTRODUCTION Transplantation of stem cell-derived islet organoids is a promising approach for Type 1 Diabetes (T1D) treatment. However, there is an absence of appropriate imaging techniques available to determine graft outcome after transplantation [1] [2] [3] [4] . Magnetic particle imaging (MPI) is a novel imaging modality that directly detects the superparamagnetic properties of iron oxide nanoparticles (SPIONs), and is a specific, sensitive, and linearly quantitative technique. Here, we demonstrate the feasibility of in vivo tracking of transplanted stem cell-derived islet organoids using MPI [5] [6] [7] . METHODS Human iPSCs were differentiated to islet organoids according to the established protocol [8] . Islet organoids were generated after 21 days of differentiation and verified using immunofluorescence staining of islet cell markers followed by labeling with dextran coated SPIONs (VivoTrax, Magnetic Insight Inc., Alameda, CA, USA, 560 µg/mL) for 48 hours at 37oC. Cell labeling efficiency was examined by anti-dextran staining. Islet organoids phantoms, comprising of different numbers of labeled islet organoids (25, 50, 100, 200 or 400, n = 2), were imaged using a MOMENTUM imager (Magnetic Insight Inc., Alameda, CA). The iron content of samples was calculated from the quantification of MPI image and verified using the inductively coupled plasma (ICP) mass spectrometry analysis. Next, 800 SPIONs labeled organoids were transplanted into the left kidney capsule of NOD/scid mice (n = 3, Jackson Laboratory, Bar Harbor, ME). Mice were examined by 3D MPI image at 1, 7, 14, 28, and 42 days post transplantation. Quantitative assessment of the islet organoids was performed using MPI image intensity calibrated against a fiducial marker of known iron content. Before euthanasia serum, the left kidney was collected from the mice and processed for human insulin ELISA and immunofluorescence staining of dextran in labeled islets organoids. RESULTS Immunofluorescence staining of islet organoids expressed islet cell markers including insulin and glucagon after 21 days of differentiation (Fig.  1A&B) . Ninety nine percent of the cells in islet organoids were labeled with SPIONs as confirmed by anti-dextran antibody staining (Fig. 1C) . Image analysis of labeled islet organoid phantoms revealed a direct linear correlation between the iron content and the number of islet organoids (R2 = 0.997, P < 0 .0001), and this was consistent with the result of the ICP analysis (Fig. 1D) . Strong MPI signals were detected under the kidney capsule on the first day of imaging in all recipients; this represented the labeled islet organoids (Fig. 1E) . During the course of the study the signal from 3D images gradually decreased, as expected. Immunofluorescence staining of the kidney sections showed the presence of SPIONs was under the left kidney capsule (Fig.  1F) . CONCLUSION MPI can longitudinally monitor islet organoids labeled with SPIONs post transplantation and provide quantitative information of their presence in vivo. Figure 1 A. 3D differentiation of human pluripotent stem cells produce islet organoids. B. Immunofluorescence staining islet organoids for insulin and glucagon, both of them were expressed in the islet organoid (green, insulin; red, glucagon; blue, DAPI; bar =20μm). C. Since SPIONs coated with dextran, we stained labeled islet organoids for dextran. Labeled islet organoids showed an effective accumulation of SPIONs (green, dextran; blue, DAPI; bar =10μm). D. MPI image indicated the signal intensity was increased as the increasing number of islet organoids. The quantification of MPI signal manifested a direct linear correlation with the number of islet organoids (R2 = 0.997, P < 0 .0001). E. 3D MPI longitudinal monitoring of transplanted islet organoids in vivo. Signal was detected under the left kidney capsule and MPI image intensity of transplanted organoids gradually decreased over a period of 42 days post transplantation. F. Immunofluorescence staining the kidney section for dextran, the green florescence represents SPIONs were present under the left kidney capsule indicated transplanted the organoids (green, dextran; blue, DAPI; bar=20μm). In order to assess bacterial colonization and its inflammation counterparts, classical whole-body [18F] -FDG PET-CT was performed 45min after injecting intravenously 5MBq/kg (0.135mCi/kg) of [18F] -FDG at baseline, 8 days and 30 days post baboon exposure to B. pertussis clinical strain by intranasal and intra-tracheal routes. For optical in vivo studies, GFPexpressing B. pertussis (derived from the previous clinical strain) was inoculated in young baboons. Besides, monoclonal anti-HLA-DR-AF647 antibody was administered by topical application in the trachea to specifically target and label APCs. FCFM, coupled with bronchoscopy, was performed in the lower respiratory tract weekly post-infection. Bacterial colonization was quantified using GFP positive signal area detected by FCFM. Results/Discussion: Animals exposed with clinical B. pertussis strain harbor significant lung lesions and FDG uptake at early stage of the clinical disease (D8), especially in the right lung, with remission observed by day 30 (Figure 1 ). Animals infected with a GFP-expressing B. pertussis strain developed the classical clinical symptoms for whooping cough as previously described in baboons infected with wild-type strains5,6. Furthermore, in vivo FCFM coupled with bronchoscopy allowed us to detect and quantify bacterial colonization kinetics and APC interactions in the lower respiratory tract of young baboons after B. pertussis-GFP infection in baboons. Ex vivo co-culture analyses also confirmed the interactions between B. pertussis and APCs in lungs explants. We were also able to detect by FCFM bacterial tracheal colonization in animals challenged by intranasal route only providing new insights concerning bacterial localization after natural infection. Conclusion: This approach combining PET-CT and fluorescence imaging will then be a useful tool to describe the mechanisms of action of the bacteria both at macroscopic and cellular resolution during infection to develop more effective vaccines against pertussis. However, to better characterize B. pertussis colonization at wholebody scale, more specific PET radiotracer than FDG (eg. Antibodies, bacterial substrate analog) must be used in the future.

http://www.who.int/immunization/monitoring_surveillance/burden/vpd/surveillance_type/passiv e/pertussis_standards/en/ (2018). 3 Abstract Body : Introduction Antibodies against neuronal receptors and synaptic proteins are associated with a group of ill-defined CNS autoimmune diseases named autoimmune encephalitis (AE), characterized by an abrupt onset of seizures, movement disorders, and psychiatric symptoms. Basal ganglia encephalitis (BGE), representing a subset of AE syndromes, can be triggered in children by repeated Group A Streptococcus (GAS) infections that lead to neuropsychiatric symptoms termed pediatric autoimmune neuropsychiatric disorders associated with streptococcus (PANDAS). Antibodies that are produced as a result of GAS infection crossreact with host neuronal proteins and cause a neuroinflammatory response. We have previously shown that multiple GAS infections of mice induce migration of Th17 lymphocytes from the nose into the brain, producing blood-brain barrier (BBB) breakdown, microglia activation, extravasation of autoantibodies into the CNS and loss of excitatory synapses of the olfactory bulb (OB). Prior PET studies in humans have also implicated microglia activation in the basal ganglia in PANDAS. 11C-ER176 is a third generation PET radio ligand for 18 kDa translocator protein (TSPO) -a marker of activated microglia, and so far has not been used in an animal model of post-infectious BGE. We sought to evaluate the ability of 11C-ER176 to detect differences in TSPO binding in our animal model and correlate it to microglia activation by immunofluorescence. Methods Fifteen four-week-old C57BL/6 mice were intranasally infected weekly with either PBS (control group) or GAS-2W streptococci (treated group) for five weeks. Twenty-four hours after the last inoculation, mice were injected intravenously with 11C-ER176 (103.12 ± 20.97 uCi) and PET images were acquired 0-60 minutes post-injection. Images were co-registered and normalized to standard T2 MRI using PMOD 3.6. Standardized uptake values ratios (SUVR) were calculated using the rostral thalamus (rThal) as a "pseudo-reference" region. Statistical analysis was performed using frames from 30-60 min post-injection. Microglia activation was examined by immunofluorescence for Iba1 and CD68 in serial sections from the olfactory bulb in treated and untreated mice of the same strain who did not undergo PET. Results GAS treated mice (n=8, 55.25 ± 5.0 days, weight 16.12 ± 2.42 g, 7 female) showed greater binding of 11C-ER176 than the PBS group (n=11, 58.81 ± 5.59 days, weight 19 ± 2.14g, 4 female) in the olfactory bulb (22.4%, p=0.004), and in the amygdala (12.9%, p=0.023). Sexcorrected data showed similar results. The data were in agreement with immunofluorescence results in which there was a two-fold increase in CD68+ Iba1+ activated microglia in the olfactory bulbs of GAS treated mice compared to control mice. Conclusion GAS-treated C57BL/6 mice show increased binding of 11C-ER176 to TSPO in olfactory bulb and amygdala, suggesting an active neuroinflammatory process. Our findings agree with histological evidence of a leaky blood-brain barrier, serum IgG deposition in the olfactory bulb and amygdala and microglial activation in the olfactory bulb in the same mouse model for post-infectious BGE.

Jiadong Pang, Sun Yat-sen University Cancer Center, pangjd@sysucc.org.cn

Abstract Body : OBJECTIVE Lung cancer is among the most prevalent and the deadliest cancers worldwide, over 80% of which belongs to non-small cell lung cancer (NSCLC) with poor prognoses. NSCLC would usually respond to ALK inhibitors, but unfortunately resistance almost definitely develops during the treatment process. In addition, efficient delivery of targeted therapeutics to tumor lesions remains challenging. In order to fight against both ALK wild type and resistance in the treatment, we aim to synthesize a series of disulfiram derivatives that allow in situ self-assembly of nanoparticles (NPs) effectively encapsulating ceritinib, as screened from a panel of ALK inhibitors. Non-invasive inhalation enables rapid distribution and high enrichment of NPs in the tumor lesions to achieve targeted treatment. Meanwhile, the formation of ditiocarb-copper complex (GnD-Cu2+) from the disulfiram derivatives via glutathione cleavage in tumor provides a viable alternative ALK-independent tumor-suppressing pathway and thus tackles ALK resistance and wild type. Intriguingly, the residues are primarily excreted by renal clearance, which minimizes systemic toxicity. The blood circulation half-life by intravenous injection and inhalation is also investigated and compared. In summary, this new class of disulfiram derivatives is a possible solution to treat NSCLC. METHODS A click chemistry-based synthesis of disulfiram with glucosamine (GnDs) was proposed. Chemical structures of the final products were confirmed using fourier transform infra-red and 1H nuclear magnetic resonance. Thermal stability, thermal decomposition and critical micelle concentration of GnDs were measured. Sizes of GnD NPs encapsulated with ceritinib were studied using dynamic light scattering and TEM. The robustness of the optimal GnD was verified by different drying and storage methods. The cytotoxicity of GnDs, GnDs-Cu2+, GnDs/ceritinib was accessed in ALK mutant H3122 and wild-type A549 cells using MTT. The mechanism of GnD degradation forming GnD-Cu2+ and its effect in H3122 cells were also studied. Biodistribution and renal clearance following inhalation were confirmed by imaging in two types of lung cancer models. H&E staining and ALK immunofluorescence staining were performed to scrutinize distribution of GnD in tumors. The blood half-life of GnD complex was tracked and calculated from periodical imaging. RESUTLS GnDs can effectively encapsulate ceritinib into NPs. The NPs retains a high stability over 10 days in PBS, after several cycles of lyophilization, and by other standard drying including rotary evaporation and vacuum drying. Sustained drug release can be maintained for days. GnDs in the presence of Cu2+ and ceritinib have shown synergistic anti-tumor effect in H3122 cells, while only the former exhibits an appreciable inhibition in A549 cells. GnD bond-breaking is induced by excessive glutathione in tumor cells, leading to formation of disulfiram residues complexing excessive Cu2+ in tumors from blood, which reduces the level of P-glycoprotein and sensitize H3122 cells. GnD/Cy5.5-COOH demonstrates renal excretion and long blood circulation with primary RES distribution. Interestingly, GnD from inhalation with only a proportion entering systemic absorption presents a different half-life in both phases of distribution and elimination in contrast to intravenous injection. CONCLUSION We demonstrate the chemical synthesis of a series of disulfiram derivatives that actively deliver ALK inhibitors for targeted therapy against NSCLC, while serving as a secondary ALK-indepedent therapeutic option by itself via forming cytotoxic ditiocarb-copper complex. The monosaccharide functionality mediates its efficient uptake in cancer cells. Its long blood circulation, extended pulmonary residence, and rapid renal clearance potentiate GnDs as promising nanotherapeutic agents. GnDs-Cu2+ formed in situ in cells can rescue ALK resistance. Furthermore, the inhalable nature endows major advantages in terms of noninvasiveness, repetitive administration, and expeditious accessibility to multiple lung cancer lesions and lobes at the same time. Its low toxicity and minimal off-target accumulation can potentially meet the unmet needs in clinic practice. Figure Caption: a Schematic of the mode of action of the inhalable GnDs/ceritinib for damaging tumor tissue. Inhalation of GnDs/ceritinib enables targeted delivery of the drug in lung cancer and protects the drug from being metabolized. b Structure of GnDs. c Critical micelle concentration (CMC) of G11D by pyrene fluorescence. Inset shows G11D could self assemble into nanostructures. d Images of G6D, G11D and G18D loading Cy5.5-COOH from CHCl3 after shaking 1 h. Inset shows dye was encased in a hydrophobic cavity. e Cumulative release of Cy5.5-COOH from GnD NPs under pH 7.4 at 1, 2, 3, 4, 5 and 6 d. Sustained drug release can be maintained for days. f Size distribution, PDI, and surface charge of G11D, and G11D/ceritinib by dynamic light scattering (DLS) and TEM images of G11D, and G11D/ceritinib. g Size distribution, PDI, and surface charge of G11D after different drying methods or different freezedrying times. The nanotherapeutic formulation retains a high stability with reproducible hydrodynamic sizes after a number of cycles of lyophilization, and by other standard pharmaceutical drying including rotary evaporation and vacuum drying. h Diameter change of G11D at pH 7.4 in 10% PBS (4 °C) over time. i TEM images of G11D at pH 7.4 in 10% FBS (37 °C). j Cell survival rate of G11D, G11D-Cu, G11D/ceritinib and G11D/ceritinib-Cu in H3122 cells and A549 cells by MTT. GnDs in the presence of Cu2+ and ceritinib have jointly shown synergistic anti-cancer effect in H3122 cells, while only the former exhibits an appreciable inhibition of proliferation in A549 cells. k Cu content of A549 and H3122 cells in tumor tissue or dish and Cu content of A549 and H3122 cells at 1, 2, 3, and 4 d after the addition of whole blood. Inset shows Cu in blood accumulated in tumor tissue. l Images of GnDs and GnDs-Cu2+ degradating by 1 uM GSH after shaking 1 h. GnD bond cleavage is induced in situ by GSH, leading to formation of disulfiram residues complexing excessive Cu2+ in tumors m GSH content of H3122 cells 48 h after the addition of GnDs-Cu2+ and GnDs/ceritinib. GnDs-Cu2+ reduce the content of GSH in tumor cells. n Western blot detection of P-gp expression level in H3122 after treatment for 24 h. o Cell survival rate of 24 h treated H3122 in 0.5 uM ceritinib 48 h. GnDs-Cu2+ reduces the expression level of P-glycoprotein and sensitize ALK-resistant H3122 cells. p H3122 cells were pre-treated with EIPA, genistein or chlorpromazine for 1 h and exposed to 20 uM GnDs/Cy5.5-COOH for 1 h. q Confocal images of H3122 cells after the addition of G11D/Cy5.5-COOH for 1 h and 2 h. r G11D delivers the drug into tumor tissue and be excreted mainly through kidney. The H3122 lung tumor model was established by intravenously (i.v. injecting 1 × 106 H3122 cells) or in situ (injecting 5 × 105 H3122 cells) into Nu/Nu mice. On day 7, after confirming development of multifocal metastases in both lungs, the mice were inhalation of G11D/Cy5.5-COOH. Cy5.5-COOH served as a control of G11D/Cy5.5-COOH. Live imaging photos of nude mice inhaled with G11D/Cy5.5-COOH or Cy5.5-COOH. s Luminescence images and fluorescent images of nude mice organs with G11D/Cy5.5-COOH. t Confocal images of lung tissue section from sacrificed mice treated with HE staining and ALK immunofluorescence staining. u Ffluorescent images of blood from C57 mice at different time. v Fluorescent intensity and calculation of half-life by fitting curve. G11D from inhalation with only a proportion entering systemic absorption presents a different half-life in both phases of distribution and elimination in contrast to intravenous injection. 

Wenshe Sun, Shanghai Jiao Tong University, weihaihahaha@163.com

Abstract Body : Introduction Tumor immunosuppressive microenvironment is a great obstacle for tumor immune therapy [1, 2] . Artificially modulating the type, density and location of immune cells within the tumor microenvironment could restrain tumor growth and promote current immunotherapy efficiency. Herein, an iron oxide nanoparticles based 'immune-guide' are fabricated by functionalization of the nanoparticles with hyaluronic acid (HA), an extracellular matrix components that can target various CD44-overexpressed tumors and mediate adhesion and migration of multiple types of immune cells [3] . Its capability to recruit immune cells (T cell and B cell) to tumor sites and regulate the proportion of different original tumor-related macrophages (TAMs), and thus to inhibit tumor growth and facilitate cancer immunotherapy were carefully evaluated. Materials and methods The nano guide was prepared by first modification of Fe3O4 with poly dopamine and then functionalization with hyaluronic acid (HA-PDA@Fe3O4). Its specificity for CD44 overexpressed tumor cells and capability of recruiting immune cells and reprograming macrophage were examined in vitro. Its tumor targeting efficiency was evaluated by magnetic resonance imaging (MRI) and photoacoustic imaging (PAI). The immune cell recruitment and TAM conversion efficiency of HA-PDA@Fe3O4 were examined by flow cytometry after the tumor bearing mice were i.v. injected with the particles. Inhibition of the primary tumors and its metastases were also examined by treating the mice with HA-PDA@Fe3O4 alone or in combination with immune checkpoint inhibitors. Results and discussion HA-PDA@Fe3O4 assembled into cluster with hydrodynamic diameter about 200 nm. T2 relaxivity (r2) of the clusters was 34 s-1mM-1. PDA coating improved the photothermal conversion performance of Fe3O4. HA-PDA@Fe3O4 could specifically target CD44overexpressed breast cancer, recruit immune cells and tilt M2 macrophage to M1 phenotype. After intravenous injection, HA-PDA@Fe3O4 could effectively accumulated in tumor, and compared with the control group, the proportion of total immune cells increased from 9% to 14%. Among which, the most obviously change was T cells, which increased from 7% to 21%, followed by B cells, which increased from 1% to 4%, whereas, the proportion of tissue-resident macrophage decreased dramatically, from 23% to 4%. As a result, tumor growth and metastases in lung were inhibited after the tumor-bearing mice were treated with HA-PDA@Fe3O4 alone, and completely eliminated in combination with immune checkpoint inhibitor PD-1and CD47. Conclusion: HA-PDA@Fe3O4 could penetrate and accumulate in the depth of tumor, and increase tumor inflammation, turning tumor into immune hot phenotype though recruit large amounts of T cells. Moreover, HA-PDA@Fe3O4 could also regulate the proportion of different original TAMs. Consequently, the biocompatible HA-PDA@Fe3O4 nanoparticles could synchronize facilitate T cell and macrophage based immunotherapy to efficiently inhibit tumors. More importantly, HA-PDA@Fe3O4 could also disturb pre-metastatic niches to prevent lung metastasis. Our study implicated HA-PDA@Fe3O4-based immunotherapy has tremendous potential for clinical diagnostic and therapeutic applications. Acknowledgements: This study was supported by the National Natural Science Foundation of China (81571729), Innovation Research Plan supported by Shanghai Municipal Education Commission (ZXWF082101) and Med-Engineering Crossing Foundation from Shanghai Jiao Tong University (YG2017ZD05). 

Wing Sum Cherry Lau, The Hong Kong Polytechnic University, cherrylauws@yahoo.com

Abstract Body : BACKGROUND. Potentials of gold nanoparticles mediating photothermal therapy have expanded options for treating triple-negative breast cancer (TNBC) with enhanced curative efficacy. Gold nanostars demonstrate superior photothermal conversion efficiency for ablating tumor cells such as ovarian cancer and breast cancer, yet the non-invasive monitoring of gold nanostars in tumor at therapeutic dose remain unclear. AIM. This research aims at assessing the temporal uptake and distribution of PEGylated gold nanostars (pAuNSs) in TNBC tumour by photoacoustic imaging (PAI), and thus optimizing the pAuNSs-mediated photothermal effect. METHODS. In vivo photoacoustic imaging of 4T1 orthotopic mouse model was conducted using the Multispectral Optoacoustic Tomography (MSOT) system, at a series of time intervals upon intravenous administration of pAuNSs (1, 2, 4, 24, 48 h) . Laser exposure of the tumor was carried out subsequent to the injection of pAuNSs, using a 808 nm laser (1 W/cm2) irradiating the tumor for 3 min twice and a 2 min rest in between, while the surface temperature of tumor was recorded by thermal imaging camera. RESULTS. Post-injection signals of pAuNSs in tumor elevated significantly and remained similarly prominent during 2-24h. Signal intensity in tumor reached a four-fold increment at 24h comparing with the pre-injection value. The tumor-tomuscle ratio obtained from MSOT signal quantification was analysed and displayed a close correlation with biodistribution data. The photothermal conversion effect was compared between laser irradiation at 2h or 24h post-injection. The tumor temperature reached up to 60℃ and 55℃ respectively, whereas the laser-only group increased from 31℃ to 37℃. The pAuNSs accumulated at these two timepoints effectively converted laser energy into heat generation in the tumor, at temperature that is sufficient for thermally ablating tumor cells by inducing cell death. CONCLUSION. PAI successfully tracked the pAuNS uptake in TNBC tumor both quantitatively and qualitatively, which helped identifying the optimal photothermal therapy window and predicting the anti-tumor effect. Not only does this theranostic nanoplatform achieve image-guided cancer therapy, but it may also facilitate the prognosis of TNBC patients. Abstract Body : Phase-changing nanodroplets are nanosized constructs with perfluorocarbon liquid cores that can be remotely vaporized to form gaseous microbubbles and expel their contents. This process can generate shock waves, shear forces, and microstreams in the interstitial tumor space, enhancing intratumoral drug delivery1. Their nanometric size brings several advantages over microbubbles, namely the ability to extravasate within the tumor tissue, and longer in vivo circulation times. After vaporization, the presence of microbubbles within the interstitial tumor space allows for ultrasound images of the agents to be obtained. Therefore, linking the ultrasound contrast and the delivery of therapeutics enables image-guided therapy with high temporal and spatial resolution and good sensitivity. This study focuses on the development and functionality of double-drug loaded nanodroplets towards ultrasound-guided co-delivery of doxorubicin and paclitaxel for combination therapy. As combination therapy was shown to be more effective for various types of cancer chemotherapy regimens2-4, dual-loaded nanodroplets were developed for this project. The benefits of co-encapsulation include the normalization of the pharmacokinetics of the two agents, paclitaxel and doxorubicin, that have different solubility and hydrophilicity properties. Moreover, the ratio between the two can be controlled from the design stage. Phase-changing nanodroplets (DDDs) were obtained by a double emulsion technique. The core emulsion consisted of aqueous droplets of doxorubicin dispersed in liquid perfluoropentane, while the shell was made up from a lipid layer intertwined with paclitaxel molecules and a NIR dye to allow for optical droplet vaporization (ODV). In vitro testing focused on determining the droplets' size and stability by DLS and optimizing their composition. The ultrasound contrast capabilities of the droplets were followed by imaging with a L22-8 Verasonics ultrasound transducer (18 MHz, 128 elements) before and after triggering events. For ODV a pulsed Phocus Opotek Mobile HE tunable laser system, operating at 1064 nm and 0.38 J/cm2 was used. The activation depth of the droplets was investigated by using fresh chicken breast tissue of increasing thickness on top of droplets-containing polyacrylamide phantoms. For estimating the biological effects of DDDs, MCF 7 breast cancer cells and FaDu head and neck cancer cells were incubated with different formulations: free drugs, empty droplets and loaded droplets and results for all the groups were compared. Dual-drug droplets, containing paclitaxel (encapsulation efficiency 78%) and doxorubicin (encapsulation efficiency 60%), that can release their load and turn into microbubbles upon exposure to pulsed NIR light ( Fig 1A) were synthesized. Their mean size was measured at 321 ± 10 nm. Phantom studies showed the localized triggering of droplets' vaporization, proven by ultrasound signal from microbubbles (Fig 1B) , as well as a linear relation between the activation and the relative concentration of droplets in phantoms (R2 = 0.97). The highest depth at which these nanodroplets can be vaporized by NIR laser was shown to be in the range of 18-19 mm. Cell studies indicated that activated droplets have the highest cytotoxicity compared to empty droplets, free drug formulations, and inactivated droplets with statistical differences between all the groups, while laser exposure alone did not account for a significant viability decrease ( Fig  1C) . The study investigating the timing of activation with respect to the incubation time revealed that for 30 minutes or more of incubation, the activation of droplets had a much higher effect on cell viability than the 10 minutes timeframe. Overall, the results indicate a promising and versatile platform, that could be used for combination therapy of several types of solid tumors. In vivo validation studies are ongoing. Nanoparticle-based combination therapy toward overcoming drug resistance in cancer. Biochem. Pharmacol. 83, 1104 -1111 . 4. Xu, X., Ho, W., Zhang, X., Bertrand, N. & Farokhzad, O. Cancer nanomedicine: from targeted delivery to combination therapy. Trends Mol. Med. 21, 223-232 (2015) .

Image/ Figure Caption: Fig 1. A. Schematic of nanodroplets' ODV; B. Differential ultrasound image of phantoms with DDDs activated by pulsed NIR light within the laser spot (marked with the white circle); C. MTT results of cells exposed to different formulations of loaded droplets or free drugs. 

Michael Phipps, City University of New York, mphipps@gradcenter.cuny.edu

Abstract Body : Background/Objectives: The membrane-bound glycoprotein complex known as tumor associated glycoprotein 72 (TAG-72) is overexpressed in many adenocarcinomas including colorectal carcinoma [1] . The CC49 monoclonal antibody (also known as minretumomab) binds TAG-72 and was one of the first radiolabeled antibodies to reach clinical trials [2, 3] . Deglycosylated TAG-72 does not bind to CC49, suggesting that the epitope recognized by CC49 is likely a glucan. Desferrioxamine B (DFO) is an acyclic hexadentate siderophore ligand that has been used in various work to create 89Zr based imaging agents. 3,4,3-(LI-1,2-HOPO) (HOPO) is an acyclic octadentate hydroxypyridonate based chelator that has been demonstrated as an effective chelator for hard ions and has garnered interest for use in imaging and therapy, and in the treatment of heavy metal toxicity by chelation therapy [4, 5] . HOPO has demonstrated superior in vivo stability with 89Zr over DFO, where some demetallation of the osteophilic 89Zr from DFO resulted in bone uptake several times greater than [89Zr]-HOPO [4] . Positron emission tomography (PET) images taken using [89Zr]-HOPOtrastuzumab in breast cancer models using BT474 xenografted mice, and it exhibited high in vivo stability [4] . Therefore, 89Zr-CC49 constructs may be effective imaging agents for cancers overexpressing TAG-72. Our investigation created such constructs and evaluated their stability and their binding to TAG-72 in vitro. Methods: Bifunctional DFO-NCS and HOPO-NCS were used to synthesize DFO-CC49 and HOPO-CC49 by bioconjugation to lysine residues on CC49. Each ligand-antibody construct was radiolabeled with ≥ 100 uCi 89Zr. The radiolabeling reaction was purified by elution from a G-25 column with phosphate buffered saline (PBS) solution. Site specific binding of CC49 was demonstrated by a blocking assay in vitro using TAG-72 expressing LS174T cells. Stability studies in PBS, serum, and whole blood will be performed to evaluate and compare [89Zr]-DFO-CC49 and [89Zr]-HOPO-CC49. Results: Both DFO-CC49 and HOPO-CC49 exhibited comparable labeling with 89Zr at 37 oC with > 90% radiochemical yield after 1 hour. Purification resulted > 98% radiochemical purity and a specific activity of 3-4 mCi/mg mAb. The cell block assay verified site specific binding of CC49. A preliminary stability study with [89Zr]-HOPO-CC49 in PBS at 37 oC demonstrated 89% stability after 7 d. Conclusions/Outlook: We have synthesized HOPO and DFO conjugated CC49 constructs and are evaluating the stability of their 89Zr-complexes in vitro.

References: [1] Scott, A. M., et al. Nat. Rev. Cancer, 2012, 12, 278-287 [2] Sheer, D.G., et al. Cancer Res., 1988 , 48, 6811-6818 [3] Forero, A., et al. Cancer Biother. Radio., 2005 

Jung Sun Yoo, The Hong Kong Polytechnic University, jungsun.yoo1@gmail.com Category: New Chemistry, Biology & Bioengineering Abstract Body : Terahertz waves (0.1 THz ~ 10 THz) have attracted great interest with recent technological advances and are now the subject of interdisciplinary research. A variety of applications have shown initial promise for their biomedical use, particularly in broadband Terahertz based imaging including noninvasive or intraoperative cancer imaging and skin and cornea hydration sensing. However, efforts to reveal the Terahertz-induced effects at cellular and molecular levels, especially how biological objects interact with Terahertz wave having specific frequency and intensity, is scarce and remains challenging. Recent progress in Terahertz spectroscopy have shown that many important biomolecules have intrinsic vibrational resonances in the Terahertz range, implicating that appropriate Terahertz irradiation may direct biochemical reactions and biological energy transport [1, 2, 3] . In addition, mild inflammation response in mouse skin under Terahertz irradiation was reported recently, suggesting possibility of immune response mediated skin regeneration [4] . Herein, biological effects of Terahertz irradiation for skin regeneration and hair growth were investigated to develop a novel therapeutic technique for dermatological diseases. Hair loss, or alopecia, is a markedly increasing dermatological disease in aging society, involving an annual global market revenue of US$85 billion and a growth rate of 5%. Unfortunately, currently available therapeutic drugs such as Minoxidil and Finasteride have either low efficacy or significant adverse effects. Focusing on capability of Terahertz wave for nonionizing local irradiation and deep penetration to cover a whole hair follicle, skin exposure to Terahertz (0.2 THz) was identified as an effective tool to promote cell proliferation and to activate hair stem cells. T-ray induced cell proliferation using Ki-67 marker was observed in the different aged mice (4-and 6-week-old balb/c male mice, each 3-4 mice per experimental group). Immunofluorescence staining and statistical analysis showed significantly increased expression of proliferation marker, Ki-67 over hair follicles and neighbouring skin regions in the T-ray illuminated group (0.2 THz wave for 10 min) compared to the room-light illuminated control group. Interestingly, elevated enhancement was reported in 6-week old mice (more telogen and categen phases) compared to 4-week old mice (more anagen phases). This result indicates T-ray can promote proliferation of hair follicle cell components especially in the resting or regression phases of hair cycle. Hair stem cell regeneration was also measured with various hair stem cell markers, CD34 (bulge stem cell marker), β-catenin (Wnt/βcatenin activity), P-cadherin (hair germ stem cell marker) in 6-week old balb/c mice. All three markers were upregulated within hair follicles, but not in the extra-follicular dermal macroenvironment. Particularly increased expression of β-catenin proves T-ray's regeneration capability because it is well reported that cyclic activation of hair stem cell involves periodic βcatenin activity and the eventual anagen initiation requires activation of Wnt/β-catenin [5] . Altogether, we provide, for the first time, a powerful strategy to treat hair loss with high efficacy and no side effect using a new promising frequency regime, i.e., Terahertz wave. Such Terahertz medicine will eventually revolutionize current clinical practice in dermatological treatment to improve overall patient outcome and reduce health care cost. Figure Caption: (A-F) Immunofluorescence analysis using markers for bulge stem cells (CD34, green in A and B), progency cells (P-cadherin, red in E and F) and Wnt/β-catenin activity (β-catenin, red in C and D). Active fluorescence signals (arrows) highlighting hair regeneration is found in the hair follicle inner layers of the THz-irradiated mice (0.2THz, 10 min). Scale bars, 100 μm. (G) Quantification of regeneration marker expression. All data were averaged from independent experiments (n = 3 to 4 mice per group). Statistical analysis was performed by unpaired t-test. (* p < 0.05, ** p < 0.01, *** p < 0.005). Error bars, mean ± s.d. 

Lucia Le Roux, MD Anderson Cancer Center, lucia.leroux@mdanderson.org

Abstract Body : Background: Prospective studies in cancer patients indicate that 68% of patients suffer from chemotherapy-induced peripheral neuropathy (CIPN) [1] . This debilitating condition is a leading cause of dose reduction and discontinuation and can cause long-term disability. The exact molecular mechanism for chemotherapy induced axonal dysfunction is not known, but there is a growing body of evidence that altered microtubule function leading to axonal transport defects could be one of the causes for the development of neuropathic disease [2] . Axonal degeneration after administration of chemotherapeutic agents has been shown in several human and animal studies [3] . Oxaliplatin also causes axonal degeneration and causes CIPN in about 70% of patients exceeding accumulative doses of 540 mg/m2 [4] . We hypothesized that loss of axonal transport is an early event in the development of neuropathy. Histone deacetylase 6 (HDAC6) is a cytoplasmic class II HDAC with preference towards non-histone proteins, as alpha-tubulin. Selective HDAC6 inhibition and HDAC6 genetic deletion both prevent CIPN [5, 6] . Inhibition of HDAC6 increases tubulin acetylation, which has been linked to improved axonal transport using in vitro model systems. However, direct evidence that axonal transport is improved by HDAC6 inhibition following chemotherapy is lacking. The goal of this study is to use an in vivo neural imaging probe as a readout for axonal transport, to assess the effect of HDAC6 inhibition by the selective HDAC6 inhibitor ACY-1083 on axonal transport upon oxaliplatin induced neuropathy. Methods: We used a 2x2 factorial design: oxaliplatin (30mg/kg IP cumulative dose over 2 weeks) versus vehicle x ACY-1083 (30mg/kg IP daily for 2 weeks after completion of the oxaliplatin treatment) versus vehicle on 20 female SKH1 mice with 4 treatment groups for 5 mice per group. A fluorescently labeled neural imaging probe based on non-toxic, Tetanus Toxin C (TTc-790) was injected in the calf muscle of each animal followed by in vivo imaging after injection. In vivo whole animal imaging of the spine was done at baseline, after Oxaliplatin treatment, after ACY-1083 treatment and at the conclusion of the study, after 2 weeks of no treatment. Quantitative assessment of axonal transport was done by measuring TTc-790 uptake with a Region of Interest (ROI) over the spine at the thoracic-lumbar junction, 60 minutes after injection. Results: Qualitative in vivo neural imaging results show a reduction in the amount of TTc-790 transport over the spine for the oxaliplatin animals compared to baseline, with preservation of transport in the other groups, including the ACY-1083-oxaliplatin co-treatment group. There was no significant difference in TTc-790 transport between groups at baseline. At the conclusion of the study, after 2 weeks of no treatment, a main effect of oxaliplatin was observed, such that oxaliplatin significantly reduced TTc-790 transport F (1, 15) = 6.48, after 2 weeks of no treatment (p Conclusion: TTc-790 transport in vivo can be successfully used to track the effect of neuroprotective agents on chemotherapy-induced deficits in axonal transport. In this study the HDAC6 inhibitor ACY-1083 showed imaging evidence of efficacy as a neural protectant for oxaliplatin-induced CIPN. In this study a 2x2 factorial design was used with oxaliplatin (30mg/kg IP cumulative dose over 2 weeks) versus vehicle x ACY-1083 (30mg/kg IP daily for 2 weeks after completion of the oxaliplatin treatment) versus vehicle on 20 female SKH1 mice with 4 treatment groups for 5 mice per group. B) Qualitative in vivo neural imaging results show a reduction in the amount of TTc-790 transport over the spine for the oxaliplatin animals compared to baseline, with preservation of transport in the cotreatment control, ACY-1083, and the Oxaliplatin + ACY-1083 groups. C) Quantitative assessment of axonal transport expressed as average radiance (p/s/cm2/sr) was done by measuring TTc-790 uptake with a region of interest (ROI) over the spine at the thoracic-lumbar junction at 60 minutes after injection. Oxaliplatin treatment significantly reduced TTc-790 transport compared to controls and compared to the oxaliplatin + neuroprotection group F (1, 15) = 6.48, at the conclusion of the study after 2 weeks of no treatment (p < 0 .05). The ACY-1083 treated animals had significantly more TTc-790 axonal transport than the oxaliplatin treated group (P=0.03) and showed no significant difference from the control or ACY-1083 alone group. Image analysis was used to quantitatively compare control and treated groups, followed by 2way ANOVA and Tukey's corrected multiple comparisons in GraphPad Prism. Abstract Body : Conventional diagnostic magnetic resonance imaging (MRI) techniques have focused on the improvement of the spatial resolution by using high magnetic fields (1-7T) . High field allows the visualization of small tumour mass but lacks to give a precise evaluation of tumour grading and metastatic potential. Recently, [1, 2] we showed that the intracellular water lifetime represents a hallmark of tumour tissue cells status that can be easily monitored by measuring T1 at different and relatively low magnetic field strengths, ranging from 0.2 to 200 mT. A fast exchange through cell membranes indicates a high metabolic rate and thus a high activity of the tumor cells as also reported by Springer C Jr in many studies [3] . Thus it is possible to measure the high metabolic pressure by an enhance water exchange with the exterior of the cell. Therefore, intracellular water lifetime can be considered an important tumour biomarker directly depending on the rate of cell proliferation, cell migration and in responding to external stimuli as hypoxia or extracellular acidosis. 1/T1 relaxometric profiles are acquired on Fast Field Cycling relaxometers able to switch the magnetic field between different field strengths, during the measurement procedure. A field cycle overcomes the problem of the low sensitivity at low magnetic fields and allows rapid acquisition. The currently available relaxometers are designed for liquid samples measurements. Therefore, in order to host a mouse, the commercially available relaxometer was modified with the implementation of a 40 mm 0.5T Field Cycling magnet and a dedicated 11mm solenoid detection coil placed around the mouse's leg where is located the tumour xenograft prepared with mouse mammary adenocarcinoma cells, namely TSA, 4T1, 168Farn, injected in the leg muscle. These cell display different characteristics in terms of aggressiveness and metastatic potential (i.e. 168FarnAnother interesting application in which the low field relaxation can give fundamental information is the therapeutic outcome. Currently tumour responses to therapy are monitored primarily by imaging evaluating essentially the decrease of tumor size. This approach, however, lacks sensitivity and can only give a delayed indication of a positive response to treatment. In our study, we propose the use of FFC-NMR to provide relevant information about response to treatment by monitoring changes of water exchange rates through cell membranes that are directly dependent on the metabolism alterations caused by the chemo-or radio-therapy. Even though FFC-NMR instrumentationis not endowed with spatial resolution, the fundamental knowledge obtained in this study can enable new diagnostic opportunities in oncology that were previously unrecognized and are potentially transferable to the two prototype humanwhole-body-sized FFC-MRI scanners recently built at Aberdeen University by Lurie and co-workers. Pilot studies performed on these FFC-MRI scanners have already demonstrated the potential use of FFC-MRI in a range of several pathologies such us musculoskeletal and cardiovascular diseases. Figure Caption: NMRD profiles of the tumour tissues grown on hind limbs: 4T1 (triangles), TS/A, (squares) and 168FARN (circles) acquired 11, 13, and 25 days after intramuscular injection, respectively. R1tum is the averaged relaxation rate normalized to the tumour mass fraction compared to the whole hind limb. Error bars report the standard deviation (SD). Abstract Body : Molecular imaging can be performed through introductions of probing agents for specific molecular targets, it can also be achieved through untargeted measurements of endogenous disease metabolomic markers. Metabolomic imaging produces a disease distribution map and can be measured with localized magnetic resonance spectroscopy (MRS) or mass spectrometry imaging (MSI).. 7T MRS-based Metabolomic Imaging of Human Prostates. Twenty-two prostatectomy-removed whole prostates with biopsy-proven PCa were analyzed with a 7T human MRI scanner. Prostates were immersed in 70% D2O to reduce magnetic susceptibility effects during measurements of T2-weighted (T2w)MRI and multi-voxel (mv)MRS of the middle transverse cross-sectional plane, which consists 16x16 voxels of a thickness of 3mm. Following MR measurements, prostates were sectioned, fixed, wholemounted, H&E stained, and evaluated with traditional histopathology. The mvMRS data were processed using an in-house developed MATLAB-based program. After correction of the phase and baseline, the program performed an automatic curve-fitting with Lorentzian-Gaussian lineshapes. These 22 analyzed prostates were randomly divided into training and testing cohorts with 11 cases in each cohort. Through co-registrations of T2wMRI with the histological wholemounted images, and based on the results of histological identification of PCa lesions, metabolomic profiles for identification of PCa lesions were generated from the mvMRS results of the training cohort. Coefficients associated with these metabolomic profiles were applied onto the mvMRS results of the testing cohort and resulted in observations of statistically significant differentiations of PCa lesions for the latter, as shown in Fig. 1A , for means and standard errors calculated based on individual mvMRS voxels. Examples from training (1B) and testing (1C) cohorts are also illustrated in the figure. Mapping Human Lung Cancer Metabolomics with MSI. Our laboratory has engaged in the discovery and interrogation of LuCa metabolomic markers with intact tissue MRS analyses on paired LuCa tissue and serum samples obtained from LuCa patients. The initial feasibility study group of tissue MSI from our MRS studied tissueserum pairs included six LuCa patients of both squamous cell carcinoma and adenocarcinoma, as well as sera from three controls. Samples were embedded, frozen sectioned into 12 um thickness, mounted on indium tin oxide-coated microscope slides, and measured on a MALDI-TOF/TOF mass spectrometer. Our previous MRS metabolomic analyses of LuCa tissue-serum pairs identified a group of cellular metabolites as potential markers, as shown in Table 1 . Here, we interrogate these markers further with tissue MSI and serum MS. Examples of our resulting MALDI MSI are shown in Fig. 2 . Data in this figure illustrate that LuCa and necrosis identified by histology in Case A (M, 72yo , SCC, Stage I) presented increased metabolic intensities for both glutamine (Gln) and phosphocholine (PChol), which were not seen with necrosis alone (B, F, 72yo, Adeno, Stage I). Of note, the interpretation of PChol increase requires caution, as an increase in collagen can also result in the PChol increase as seen in (C, F, 60yo, SCC, Stage I) .

However, in that case, the increase in Gln is almost absent. This observation further emphasizes the importance of conducting metabolomic evaluations, rather than merely observing individual metabolites. In conclusion, our results presented here demonstrate the potential of utilizing endogenous disease metabolomic markers for molecular imaging through MRS and MSI. While the former can be explored for in vivo non-invasive metabolomic disease evaluations, the latter can provide metabolomic characterizations at close to single cell level for molecular pathology examinations. Abstract Body : Introduction: Down Syndrome (DS) or trisomy 21 is the most common genetic cause of mild to moderate intellectual disability in the world. Defects in myelination process can be associated to developmental delays and intellectual disability (1, 2) . Molecular neuroimaging allows in vivo and noninvasive studies of cellular and molecular processes, and therefore, it is an important tool to understand the cells changing over time. Positron Emission Tomography (PET) imaging using Pittsburgh Compound B labelled with carbon-11 (11C-PIB) is traditionally used for β-amyloid plaque quantification, however, in the last decade has also been used for myelin quantification (3) (4) (5) . Objective: The aim of this study was to evaluate myelination progression in Ts65Dn young mice, model of Down syndrome, by 11C-PIB PET imaging Methods: Transgenic mice (TS65Dn) and its wild type (WT), were evaluated with 11C-PIB PET imaging in 2 different ages: 2-month-old (WT = 31 animals and Ts65Dn = 16 animals) and 5-month-old (WT = 7 animals and Ts65Dn = 6 animals). 11C-PIB (± 15 MBq) was injected intravenously and 30 min after tracer injection a static PET image of 20 min was acquired in a PET scanner for small animals (β-cube, Molecubes, Belgium) with the animal anesthetized with isoflurane 2-3% in oxygen. Images were quantified by PMOD™ software (Switzerland) using T1 MRI as brain template and the data was expressed in SUV (Standardized Uptake Value) mean ± standard deviation in different brain regions: striatum, cortex, hippocampus, thalamus, cerebellum, brainstem and midbrain. Group comparisons were analyzed by 2-way ANOVA (GraphPad Prism version 6). Statistical differences were considered significant when P≤0.05. Results: All analyzed brain regions, in the WT group, showed 11C-PIB uptake increase between overtime . In the Ts65Dn animals there was an increase in tracer uptake, however this increase was not statistically significant between 2 and 5-month-old data. The 11C-PIB uptake (SUV) for the WT 2 month-old group was 0.065±0.027 for the striatum, 0.072±0.029 for the cortex, 0.068±0.028 for the hippocampus, 0.064±0.026 for the thalamus, 0.072±0.033 for the cerebellum, 0.086±0.037 for the brainstem and 0.075±0.031 for the midbrain. Comparing with WT 5 monthold, the uptake was 0.108±0.039 (p=0.0057), 0.120±0.040 (p= 0.0031), 0.113±0.037 (p=0.0045), 0.108±0.044 (p=0.0055), 0.121±0.040 (p=0.0057), 0.133±0.052 (p=0.0322) and 0.131±0.062 (p=0.0039) in the respective brain regions. For the Down syndrome model with 2-month-old, the 11C-PIB SUV was 0.064±0.034 for the striatum, 0.067±0.035 for the cortex, 0.066±0.034 for the hippocampus, 0.061±0.035 for the thalamus, 0.069±0.039 for the cerebellum, 0.081±0.045 for the brainstem and 0.071±0.040 for the midbrain and in the 5-month-age was 0.091±0.020 (p=0.2714), 0.094±0.015 (p=0.2900), 0.096±0.020 (p=0.1777), 0.082±0.019 (p=0.4962), 0.094±0.010 (p=0.4518), 0.103±0.020 (p=0.6739) and 0.094±0.023 (p=0.5785) in the respective brain regions. Conclusion: 11C-PIB PET imaging was able to detect differences in myelin content of different brain regions of WT and Ts65Dn animals. In the WT group was possible to detect increase in tracer uptake, meaning increase of myelin content between 2-month of age to 5-month of age. The increase in the Ts65Dn was not statically significant, meaning a slower development in myelin content in this animal model of Down syndrome what can be related to brain development. Abstract Body : Background Background ONM-100, a polymeric micelle covalently conjugated to indocyanine green (ICG), is being developed for intraoperative detection of solid cancers. ONM-100 micelles dissociate in the acidic environment of solid tumors via an ultra pH-sensitive binary activation mechanism, resulting in ICG fluorescence. As low pH is a hallmark of all solid tumors, ONM-100 has the potential to be a broadly indicated, tumor-agnostic imaging agent. A 30 patient Phase 1 study was completed in breast (BC), head and neck (HNSCC), colorectal (CRC) and esophageal (EC) cancers, where a single intravenous (IV) dose (0.1 -1.2 mg/kg) of ONM-100 was administered 24 hours before surgery. ONM-100 was well tolerated with no drug related safety issues, fluoresced in all four tumor types, and detected histology-confirmed positive margins and occult disease missed by SOC surgery in 30% of the patients (WMIC 2019, oral presentation). This ongoing Phase 2a study (NCT03735680) optimizes imaging performance of ONM-100 over a range of doses and imaging schedules and explores potential for ONM-100 guided surgery in new tumor types and using additional FDA-approved imaging devices. Methods This is a non-randomized, open-label, multi-center 2 part study. The study is designed to be adaptive in terms of selection of dose, imaging schedule and tumor type based on continuous review and analysis of imaging and safety data. All patients receive a single IV dose (0.5 -7mg/kg) of ONM-100 the same day as, or previous to the day of surgery. Part 1 comprises up to 6 dose Cohorts of 3 patients each with confirmed biopsy of BC or HNSCC. In Part 2, up to 3 dose/schedule Groups are verified in multiple tumor types (prostate (PrC), ovarian (OC), nonsmall cell lung (NSCLC), urothelial (UC) cancers, CRC, HNSCC, BC). ONM-100 fluorescence is imaged intraoperatively before and after tumor excision, in the specimens on the operating room back table, and during pathology workflow. ONM-100 fluorescence is evaluated in terms of tumor visualization, tumor to background ratio, diagnostic performance vis-à-vis standard pathology and detection of positive margins or occult disease missed by SOC surgery. Results In this ongoing study, 16 patients have been enrolled in 3 Cohorts in Part 1 (4 BC, 5 HNSCC) and 2 Groups in Part 2 (6 PrC, 1 OC) at two US sites. Here, we show for the first time, ONM-100 imaging feasibility in PrC during robotic prostatectomy using the da Vinci Xi Firefly™. ONM-100 fluorescence was detected through the prostatic capsule, and tissue slice fluorescence corresponded to pathologically confirmed tumor regions (Figure 1 ). ONM-100 imaging intraoperatively detected a clinically suspicious bladder neck margin in the wound bed of a prostatectomy patient. The excised margin was histologically confirmed to have tumor present. These data also demonstrated feasibility to image another new tumor type -ovarian cancer. All 4 enrolled tumor types were successfully imaged using the Visionsense™ camera. Pharmacokinetic data demonstrated dose proportionality across the dose ranges studied in Phase 1 and Phase 2 (0.1 -3 mg/kg). ONM-100 was well tolerated at all dose levels. Strong fluorescence signals were observed even within few hours after dosing, which combined with a half life > 55 hours increases the potential for a wide imaging window. Conclusion ONM-100 continues to exhibit favorable safety and PK profiles even at doses 3X higher than in Phase 1. The study extends the feasibility of detecting tumor positive margins in real-time to prostatectomies. The tumor and camera agnostic imaging is further confirmed in 2 additional tumor types (total 6) and 2 additional imaging devices (total 4) attesting to the robustness of the imaging agent and the mechanism of action, independent of the metabolic phenotype of the tumor. White light (A) and near-infrared (B) intraoperative in vivo imaging of the prostate in a patient undergoing robotic prostatectomy using the da Vinci Xi Firefly system. Ex vivo tissue slabs are processed to formalin fixed paraffin embedded (FFPE) blocks, cut, stained with hemtoxylin and eosin (H&E) and analyzed for tumor (C) . Annotations of tumor areas within the H&E (C, blue outlines) are superimposed on fresh tissue slab NIR images (D) previously captured using the Li-COR PearlÒ Trilogy imager. Areas of prominent fluorescence are found to highly correlate with histologically confirmed areas of tumor Abstract Body : Introduction. The medial temporal lobe has been identified by modern neuroscience to be the locus of learning and memory formation, and many neurodegenerative conditions wreak havoc on cognitive ability because of their damage to these structures. Thus, this region of the brain has been the focus of much investigation; however, there remain many unanswered questions about its molecular physiology even in healthy individuals. Glutamate is the primary excitatory neurotransmitter in the mammalian central nervous system and implicated in almost all neurophysiological processes; accordingly, dysregulation of its metabolism has been observed in many disease states. Like many small molecules of interest, glutamate remains challenging to measure non-invasively in vivo. While 1HMRS, biochemical and histological measurements have endeavored to quantify the amounts of glutamate present in human and animal brain tissue, spatially-resolved detection of any differences in glutamate concentration between tissue types or anatomic regions of living brain has remained elusive. Glutamateweighted Chemical Exchange Saturation Transfer (gluCEST) magnetic resonance imaging (MRI) is nearly unique in its ability to provide non-invasive, spatially resolved measurements of glutamate in vivo. We recently developed and improved correction for B1 inhomogeneity of gluCEST images of the human brain which enabled truly volumetric (3D k-space) imaging, even in regions of the brain where B1 inhomogeneity is problematic. In this work, using 3D gluCEST imaging [1, 2] at 7T with this improved post-processing strategy [3] , glutamate-weighted MR images of a slab which includes the medial temporal lobe were generated and then analyzed by registration with a high-resolution hippocampal subfield segmentation. Hypothesis. Based on the general awareness of finely specific physiology and function in the medial temporal lobe [e.g. 4,5] and published findings on the presence of glutamate in rodent brains [6, 7, 8] , we hypothesized that there may be subtle, as of yet unmeasured physiological differences in glutamate concentration between various subfields of the hippocampus and neighboring structures of the human medial temporal lobe. Methods. Images of 13 healthy subjects ages 21-69 were obtained on a Siemens 7.0T Terra outfitted with a single volume transmit/32 channel receive phased array head coil. Post-processing was performed using in-house code written in MATLAB and MEX (Matlab executable). Segmentation of the medial temporal lobe was performed using the Magdeburg 7T Atlas available in the Distributed Segmentation Service of ITK-SNAP [9, 10, 11] , with T1 and T2-weighted structural images as input. Registration and calculation of ROI averages was then performed using the c3d package accompanying ITK-SNAP. Results. Our results suggest that we are indeed able to detect gluCEST differences in tissue types in vivo; namely, an elevated amount of glutamate in the dentate gyrus in comparison with the other medial temporal lobe structures (which are all considered gray matter). This corroborates the findings in the preclinical works referenced; such a result has not previously been demonstrated in living human beings, for lack of technology capable of measuring it.

Although these preliminary results demonstrate the potential of 3D gluCEST MRI in measuring substructural glutamate differences in human medial temporal lobe, the method is equally applicable to measuring glutamate changes in the other parts of the brain. There are several points of validation needed, specific to the gluCEST acquisition and in particular the method for correcting for B1 inhomogeneity, which is an essential step in the processing of gluCEST raw data to generate CEST maps. Namely, we sought to demonstrate that the findings to date are not an artefact of extreme values or user-set thresholds, of the saturation B1 distribution of the head coil, nor of our method for correcting for it. Please see "Persuasive Data" for figures addressing these points. on this axis is 7-13%, calculated using B0 and B1 corrected negativenormalized CEST asymmetry; the center line of the boxplot represents the subject with the median value for that segment, with edges of the colored bars being first and third quartile (standard boxplot structure). The dentate gyrus has a distribution that is clearly shifted higher than the other nearby segments. Both the median and the mean (not shown) ROI-averaged values for the dentate gyrus (bilateral) are 11%, while the next highest gluCEST segments are CA1 and the hippocampal tail, at only 10.5%. It should be noted that the apparently lower values in the more inferior subfields are likely to be an artefact of the B1 distribution, whereas we do not

Melissa Lucero, University of Illinois at Urbana-Champaign, MLucero3@illinois.edu Category: New Chemistry, Biology & Bioengineering Abstract Body : Personalized medicine promises to improve treatment outcomes by matching patients with therapy. A recent approach towards this goal involves employing companion diagnostics (CDx), which are tests that can identify the presence of a biomarker critical to the activation a drug. However, all FDA-approved CDx are limited to the in vitro setting which does not account for factors that may influence efficacy such as the influence of the tissue microenvironment. Technologies such as PA imaging can overcome this limitation by providing information that can account for changes in the body during disease progression. Here, we have developed the first PA imaging-based companion diagnostic (PACDx) specifically for identifying elevated concentrations of glutathione (GSH) in cancer. GSH plays essential biological roles, however it has also been associated with a variety of pathologies such as cancer. For instance, GSH was reported to be 4-fold higher in non-small cell lung carcinoma (NSCLC) compared to healthy lung tissue in humans. With this information in mind, we chose to develop a CDx assay for lung cancer based on GSH detection. Since GSH is the most abundant thiol in the body it is typically not considered an ideal target for cancer detection or treatment. Current sensing approaches that use thiol-disulfide exchange, Michael additions, or nucleophilic aromatic substitution reaction (SNAr) are too reactive to distinguish levels between cancer and healthy tissue. To address this issue, we took a physical organic approach to home into a reactivity that provided a dose-dependent response in the 1 -10 mM range. Importantly, it is widely accepted that cancer cells contain GSH closer to the 10 mM range, while healthy cells are closer to 1 mM. By generating Hammett, Eyring, and Brønsted plots, we were able to tune the reactivity of the most commonly used 2,4-dinitrobenzenesulfonyl sensor (fully reacts with 100 µM GSH) to a sensor that can accurately distinguish GSH concentrations. Further in vitro studies revealed that PACDx had good selectivity over various biologically relevant analytes including metals, amino acids, reactive oxygen and nitrogen species, and other thiols. Following cellular evaluation studies with PACDx, we ported this chemistry into a prodrug (PARx) which will essentially use PA imaging to track the delivery of a chemotherapeutic into lung cancer tissue. Via both local and systemic routes, PARx produced an enhanced PA signal in the tumor compared to a control and successfully inhibited tumor growth without causing toxicity in vital organs. In a blinded study, we further demonstrate the utility of PACDx to stratify groups of cancer types with varying GSH concentrations. Moreover, we demonstrated the high selectivity of PARx to lung cancer as it did not inhibit tumor growth in mice carrying tumor types (i.e., brain) with low GSH concentrations. Overall, we envision that this work will have potential applications in personalized therapy and drug development. Abstract Body : Introduction Photoacoustic (PA) imaging is an emerging modality, usually combined with standard ultrasound imaging. The method is based on the photoacoustic effect. When light interacts with a material, part of the energy is absorbed and partially transformed into heat, which leads to thermal expansion and formation of acoustic waves detectable by an ultrasound transducer. Although PA imaging can be used natively, substantially higher number of applications uses contrast agents. Among them, polypyrrole nanoparticles represent an interesting option [1] . The aim of the study was to reveal, how the nanoparticle size affects PA signal. Materials and Methods We prepared uniform polypyrrole nanoparticles (PPy) with narrow size distributions via water-based redox precipitation polymerization technique of pyrrole monomer in the presence of iron(III) chloride (nPy:nOxi, 1:2.3) as an oxidant. To stabilize growing nanoparticles in reaction mixtures, non-ionic poly(vinylpyrrolide) (PVP, Mn = 40 000) was used as a stabilizer. The size, morphology and dispersity were directly evaluated from microphotographs obtained by a transmission electron microscope (FEI-TEM, Tecnai G2 Spirit, Oregon, USA) and analyzed with ImageJ software [2] . Suspensions of PPy samples were diluted to 0.2, 0.5 1, 2, 3, 4, 5 mg/mL, inserted into silicon tubes submerged in water and tested as a photoacoustic contrast agent in vitro by using an imaging platform Vevo 3100/LAZR-X (Fujifilm VisualSonics, Amsterdam, Netherlands) combining ultrasound and photoacoustic imaging (excitation wavelength range 680 -970 nm and 1200 -2000 nm). Results and Discussion Various reaction conditions in the presence of PVP led to formation of almost uniform nanoparticles with size range from 48 to 130 nm. The TEM images showed regularly spherical nanoparticles with narrow size distribution in all prepared samples. A strong dependence of the signal on particle size was observed. A relative PA signal recalculated per one particle is shown in the Figure. Both dependences on the particle diameter (A) and particle crosssection area (B) are presented. Particle size is important from two perspectives. First, the size may affect cytotoxicity, safety in vivo, NP distribution in the organism, and also elimination routes. The reasonable size ranges between tens to few hundreds of nanometers, which was fulfilled with the investigated set of nanoparticle suspensions. Secondly, particle size strongly influences signal intensity. If the thickness of the nanoparticle layer penetrated by the excitation light is substantially smaller than the core diameter (which may be fulfilled in the diameter range we used in our experiment), heat absorption is proportional rather to particle cross section [3] (or irradiated particle surface). Our data (Fig. B) indicated linear dependence of the signal on particle cross section, which was anticipated. The measured dependence also proved, that suspensions containing smaller particles (with the same solid content in the suspension) were more efficient than suspensions with bigger nanoparticles. Conclusion Polypyrrole nanoparticles represent a promising substance for photoacoustic imaging. Their physical properties, which determine photoacoustic performance, can be tuned by varying of polymerization conditions during NP preparation. Knowledge of the chemical structure and its relation to their physical properties may help to optimize nanoparticle preparation for in vivo imaging in terms of biocompatibility and high photoacoustic efficiency. Acknowledgement The work was supported by Czech Science Foundation -(grant No. 18-05200S), Ministry of Education, Youth and Sports of the Czech Republic (Czech-BioImaging LM2015062), National Sustainability Program II (Project BIOCEV-FAR LQ1604). The facility infrastructure was supported by European Abstract Body : Variable domains of heavy chain only antibodies (VHHs) are interesting molecules for tumor imaging and therapy, mainly because of their small size and fast clearance, leading to efficient tissue penetration, homogeneous tissue distribution and good tumor-tobackground ratios at early time points after injection. VHHs can be conjugated to both a diagnostic probe and a therapeutic compound, turning them into valuable theranostic agents. The most commonly used method is conjugation to naturally occurring lysines, however this random conjugation approach can affect the binding affinity of the VHH (1). Furthermore there is no control over the stoichiometry. In the current study, we employed an unpaired cysteine for sitespecific conjugation using a dichlorotetrazine platform. We site-specifically equipped anti-EGFR VHH 7D12 with both the chelator DTPA and the photosensitizer IRDye700DX, and characterized the properties of this conjugate preclinically. 7D12 with a C-terminal unpaired cysteine was reduced with tris-(2-carboxyethyl)phosphine and conjugated to BCN-PEG2maleimide. The bimodal probe DTPA-tetrazine-IRDye700DX was synthesized according to methods described in (2) and conjugated to 7D12-BCN, yielding DTPA-IRDye700DX-7D12 ( Figure 1A ). The conjugate was characterized with RP-HPLC-HRMS and UV-visible spectroscopy. Fluorescence quantum yields and generation of molecular singlet oxygen upon illumination were determined. DTPA-IRDye700DX-7D12 was labeled with indium-111 and stability of labeling in PBS and human serum was verified with instant thin layer chromatography. Subsequently, binding and internalization and the half maximal inhibitory concentration (IC50) of the parent 7D12 and DTPA-IRDye700DX-7D12 were determined using EGFR overexpressing A431 cells. Furthermore, A431 cells were incubated with various concentrations of DTPA-IRDye700DX-7D12 (0-100 nM) and illuminated with various doses and dose rates of 690 nm light using a light emitting diode (30-90 J/cm2 and 50-200 mW/cm2, respectively). Cell viability was determined using a luminescence-based assay (CellTiter-Glo®). The bimodal DTPA-Tz-IRDye700DX probe was obtained with an overall yield of 18%. 7D12-BCN and DTPA-IRDye700DX-7D12 were retrieved with a protein recovery of 74% and 43%, and a degree of labeling of 1 and 0.56, respectively. Spectral properties of the IRDye700DX were retained upon conjugation (excmax 689 nm, emmax 698 nm). Fluorescence quantum yield of the conjugate was slightly lower compared to that of free IRDye700DX (0.12 vs 0.29, respectively), as was the singlet oxygen generation upon illumination. The conjugate could be labeled with indium-111 and labeling was stable in both PBS and human serum for up to 24 hours. [111In]In-DTPA-IRDye700DX-7D12 bound to A431 cells, and the absence of binding in presence of an excess of unlabeled 7D12 confirmed EGFR specificity. The tracer was efficiently internalized, with 8.8 ± 0.7% of added activity found in the intracellular compartment after 4 hours of incubation at 37°C. IC50 concentrations of native 7D12 and conjugate were similar (22.8 and 18.1 nM, respectively; Figure 1B) , indicating that conjugation of the bimodal probe does not interfere with target binding. A431 cells were effectively killed upon illumination after incubation with DTPA-IRDye700DX-7D12 ( Figure 1C ). In conclusion, the dichlorotetrazine platform offers a feasible method for site-specific dual-labeling of VHH 7D12, without altering binding affinity. Currently, the compound is characterized further in vivo, by performing biodistribution, SPECT/CT imaging and photodynamic therapy studies in a murine subcutaneous A431 model. In the future, the compound could be used for both EGFR-targeted nuclear imaging and photodynamic therapy, in a theranostic approach. The described conjugation method can be applied to other VHHs and small proteins with introduced unpaired cysteines (e.g. DARPins, affibodies). Furthermore, the flexibility of the approach makes it easy to vary the nature of the probes if necessary.

Stefanie Rosenhain, Institute for Experimental Molecular Imaging, srosenhain@ukaachen.de

Abstract Body : Introduction: Fluorescence-mediated computed tomography (FLT-CT) enables to non-invasively assess the biodistribution and tumor accumulation of near-infrared fluorescently (NIRF) labeled drug carriers and thus, plays an important role in preclinical cancer research. However, NIRF dyes are actively taken up by transporting proteins and can affect the biodistribution pattern of the drugs. In this context, organic anion transporting proteins (OATP) recently gained attention, because they are involved in the uptake of fluorescence dyes. It is proven that not only hepatocytes but also most cancer cells overexpress different OATPs. Since this can cause false-positive results, especially in tumor and liver tissue, it is important to understand OATP-mediated transport of fluorescence dyes and their influence on biodistribution and tumor accumulation. Material/Methods: OATP-mediated uptake of the free dyes AlexaFluor750 (AF750), Cy7, DY-730, DY-736, DY-750, ICG, IR-783, IRDye 750, and 800 CW was investigated by FLT-CT phantom and absorption measurements. OATP-mediated uptake of the dyes in A431, HepG2, and HUVECs (control) were investigated at 4°C and 37°C (20 µM; 2h) using fluorescence microscopy and flow cytometry. Inhibition of the OATPs was performed using Rifampicin (20µM). OATP expression (SLC01A2, SLC01B1, SLC01B3, and SLC02B1) was determined by qPCR. Effects on biodistribution and tumor accumulation of free AF750, Cy7, DY750, ICG (analysis is pending) and IR-783 were analyzed in five A431 tumorbearing athymic nude mice. After injection of the dyes (2 nmol/animal), mice were measured by FLT-CT at 0h, 3h, 6h, 24h, 48h, 72h, and 96h. After 96h, excised organs were imaged by 2D fluorescence mode. Further in vitro and in vivo investigation of poly(N-(2hydroxypropyl)methacrylamide) (pHPMA) labeled with the above-mentioned NIRF dyes are ongoing. Results: qPCR demonstrated a higher expression of OATP1A2 and OATP1B1 in both A431 and HepG2. Microscopic analysis revealed no uptake of AF750 and Cy7 into A431, in contrast to the uptake of DY-750 and IR-783. These finding were also confirmed in vivo. After injection of AF750, no tumor signal but a strong bladder signal along with a fast clearance from the whole body until 6h post-injection can be seen. Fluorescence intensities of Cy7 were strong both in liver and gut at early points in time. Most of the dye is cleared not later than 48h post injection via the hepatobiliary elimination route, although some signal remained in the kidneys. IR-783 was showing a strong tumor accumulation even at 96h. These findings were supported by the ex vivo results. The preliminary results indicate a possible impact of NIRF dyes on biodistribution assessment of fluorescence-tagged molecules. Especially IR-783 shows a strong tumor signal, possibly influencing tumor accumulation of molecules labeled with IR-783. Conclusion: A comprehensive knowledge of the transport processes of NIRF dyes which are mainly OATP-mediated but not exclusively will increase proper fluorescence labeling strategies without distorting the pharmacokinetic properties and can optimize the specificity of favorable targeting. Further studies are needed to investigate the biodistribution of fluorescence-tagged molecules and the role of OATP-mediated transport in vitro and in vivo. After injection of IR-783, there is a strong tumor accumulation that last until 96h p.i. AF750 exhibit a fast clearance from the whole body which is evidenced by the strong bladder signals. Cy7 depicted strong fluorescence intensities both in liver and gut at early points in time. Although some signal remains in the kidneys, most of Cy7 is hepatobiliary cleared not later than 48h p.i. DY750 shows indistinct signals in the liver, but is completely cleared after 24h with no remaining signal in the organs 96h post injection. Further fluorescence intensity quantification needs to be performed. After injection of AF750, no tumor signal but a strong bladder signal along with. IR-783 was showing a strong tumor accumulation even at 96h. These findings were supported by the in vitro and ex vivo results. Abstract Body : Ischemic stroke is a sudden neurological disease caused by the obstruction of cerebral blood flow. It represents the third cause of death and the first cause of acquired handicap among adults in industrialized countries. The use of advanced medical imaging techniques in acute setting, such as computed tomography (CT), and perfusion-and diffusion-weighted magnetic resonance imaging (MRI), has recently significantly improved treatment decisionmaking for acute ischemic stroke.1 Smart optimization of clinical imaging protocols, in combination with novel treatment strategies, has enabled extension of the treatment window up to 24h for some patients, while it was for decades limited to 4.5h post stroke onset.2 However, the information provided by perfusion and/or diffusion imaging in stroke patients has some limits. It does not directly inform on the status of vascular inflammation for example. Yet, inflammation is known to play an important role in ischemic stroke damage, suspected to remain once treatment restored brain vessel patency.3 This preclinical research investigated the efficacy of a molecular MRI approach to reveal vascular inflammation in ischemic stroke. The dynamics of vessel flow, tissue perfusion, and vascular inflammation and their correlation to brain lesion growth were studied in a transient middle cerebral artery occlusion (tMCAO) rat model with a preclinical 9.4 T MRI scanner (Varian Inc.). MRI was performed at 1h, 6h, 24h and 96h post reperfusion. Microparticles of iron oxide targeted to vascular cell adhesion molecule 1 (MPIO-αVCAM-1) were injected intravenously to reveal endothelial cell activation on susceptibilityweighted images (SWI). Bolus injections of Gadolinium-based contrast agent (Gadovist®) were used for perfusion-weighted imaging (PWI). Time-of-flight weighted sequences were used for magnetic resonance angiography (MRA). Brain lesions were measured on apparent diffusion coefficient (ADC) maps for early time points (1h and 6h post reperfusion) and on T2 weighted images for later time points (24h and 96h post reperfusion). Twenty-four hours post reperfusion, the intravenous injection of microparticles of iron oxide functionalized with anti-VCAM-1 antibody (MPIO-αVCAM-1) induced localized negative signals in the susceptibility-weighted images (SWI) (cf. figure) . The method enabled high resolution mapping of VCAM-1 expression. It revealed an area of vascular inflammation that appeared around the core of the lesion. We detected the strongest VCAM-1 expression 6h post reperfusion, whereas no signal was measured 1h post reperfusion (p The strategy of molecular MRI with VCAM-1-specific iron oxide contrast agent enabled in vivo detection of the dynamic vascular inflammation process involved after abrupt reperfusion in ischemic stroke. Implementation of this technique in clinical settings may provide complementary information on stroke patient status and could help to guide therapeutic decision-making for emergency radiologist. Abstract Body : Overexpression of Human EGF Receptor 2 (HER2) in cancer tumors is a marker of aggressive metastatic disease and poor prognosis. At the same time, it presents a target for precision therapy such as trastuzumab (TZM). Anti-HER2 humanized monoclonal antibody trastuzumab (TZM) has been successfully used clinically over the last two decades. However, not all eligible patients benefit from this therapy due to either primary or acquired TZM resistance1,2. Further investigation of TZM-HER2 binding, internalization and trafficking/degradation in cancer cells in vitro and in vivo may provide better understanding of this chemoresistance. In particular, fluorescence lifetime Forster resonance energy transfer imaging (FL-FRET) offers a unique approach to monitor TZM-HER2 binding followed by their uptake into target cells via the reduction of donor fluorophore lifetime3,4. Since this approach does not rely on intensity of contrast agent, it provides an efficient method to quantify ligandreceptor binding and internalization in cancer cells with an additional benefit of sensing intracellular or intra-tumoral microenvironment heterogeneity. We validated quantification of target engagement both in vitro and in vivo using FL-FRET in transferrin-transferrin receptor system5,6 In this study, we report a multiscale quantification of TZM-HER2 engagement in AU565 breast cancer cells and tumor xenografts using FL FRET imaging and phasor analysis approach. First, we characterized TZM-AF700 uptake and its relation to HER2 expression in AU565 cells using confocal microscopy. Further, we have quantified the dimerization of HER2 via NIR TZM FLIM FRET in vitro microscopy. Extensive analysis confirmed high specificity and efficiency of TZM FRET signal. Interestingly, we observed a significant heterogeneity of FRET within the cells: the highest TZM FRET levels occurred at the plasma membrane, whereas a decreasing gradient of donor lifetime reduction was registered as the TZM-HER2 complexes undergo internalization and intracellular endocytic trafficking. Phasor analysis7 confirmed these observations qualitatively and quantitatively. These results suggest that TZM-HER2 complexes undergo significant conformational changes, including possible dissociation or degradation upon TZM-HER2 binding and trafficking. In addition, we performed FL-FRET imaging of TZM AF700/AF750 internalization in AU565 tumor xenografts in live intact animals. We successfully quantified the fraction of dimerized HER2 receptors bound by TZM and internalized in the tumors, livers and bladders. Importantly, immunohistological analysis of tumors showed significant upregulation of HER2 and HER3 expression in the tumors of mice injected with TZM compared to the untreated ones, indicating that TZM modulates their heterodimerization and signaling. Altogether, our results demonstrate that FL-FRET imaging robustly and directly reports on TZM-HER2 engagement in multiscale cancer models. Figure 1 : In vivo specificity of targeting engineered exosomes. (A) In vivo SPECT/CT images (coronal view) after 3 hrs of intravenous injection showed significant accumulation of M2-targeting exo in tumor, lung, spleen, lymph node, and bones. 111In-oxinelabeled non-targeting exosomes (HEK293 exo) and CD206-positive M2-macrophage targeting exosomes (M2-targeting exo) were injected into the 4T1 tumor-bearing mice. One group was treated with Clophosome® to deplete macrophages. Yellow and green arrows denote lymph node and bone metastasis, respectively. (B) 3D surface images showing M2-targeting exo are profoundly distributed in both lung and tumor areas compared to the group injected with HEK293 exo and pre-treated with Clophosome®. The yellow arrow indicates the tumor center. Abstract Body : Water-soluble vitamin H (biotin) is essential for cellular growth, development and well-being. Biotin (vitamin B7) is a cofactor for five carboxylases involved in gluconeogenesis and catabolism of amino acids and fatty acid biosynthesis [1] . Biotin is taken up by cells via the sodium-dependent vitamin transporter (SMVT), expressed in the cytoplasm and mitochondrial membranes. SMVT is present in the gastrointestinal tract, liver, kidneys, retina, heart, brain and skin. Biotin conjugates have been radiolabelled using PET and single-photon emission computed tomography (SPECT) radionuclides such as 64Cu, 18F, 111In, 125I, 131I, 68Ga and 99mTc [2] [3] [4] . These "biotin-inspired" imaging probes have been used in clinical and preclinical settings for infection and tumour imaging. However, there is no report of studies of radiolabeled isotopologue of biotin itself using non-invasive PET imaging modality in vivo to date. In this work, we have used a new radiochemical reaction [5] to achieve the first synthesis of carbon-11 radiolabelled biotin ([11C]biotin, Figure 1A ) as an in vivo PET imaging agent for probing biotin trafficking. The urea moiety of biotin was radiolabelled using cyclotron-produced carbon dioxide ([11C]CO2) and diamino biotin via a Mitsunobu reaction. Total synthesis time including HPLC purification was 32 ± 1 minutes from end of [11C]CO2 delivery. The amount of [11C]biotin obtained was 352 ± 38 MBq in 4-5 mL PBS with 2.5% ethanol in injectable solution starting from 5.9 ± 0.3 GBq of cyclotron produced ([11C]CO2 with an isolated radiochemical yield of 19 ± 2%, radiochemical purity >99%, molar activity of 7 ± 1 GBq/μmol at end of [11C]CO2 delivery. To examine [11C]biotin trafficking in vivo, the radiotracer was administered intravenously (IV) in healthy anaesthetised mice placed in a high-resolution microPET scanner. Sixteen mice were divided into three groups: group A1 (female, vehicle IV injection 10 minutes before [11C]biotin IV injection, Figure 1B ), group A2 (male, vehicle IV injection 10 minutes before [11C]biotin IV injection, Figure 1B ) and group A3 (female, biotin 5 mg/Kg IV injection 10 minutes before [11C]biotin IV injection, Figure 1B ). Dynamic PET image data were acquired for 60 minutes. Preclinical experiments showed that [11C]biotin administered intravenously is distributed in the liver, kidneys, retina, heart, brain and brown adipose tissue (BAT) consistent with known in vivo biotin transporter expression ( Figure 1C) . A significant decrease of [11C] biotin uptake in the biotin-challenged group (group A3) compared to the control group A1 was observed in liver, brain and BAT with a remarkable increase in [11C]biotin excretion ( Figure 1D ). In conclusion, by incorporating a positron emission tomography (PET) radioisotope (carbon-11) in the structure of biotin ([11C]biotin), we have provided a qualitative and quantitative method for non-invasive evaluation of real-time and whole-body pharmacokinetic of [11C]biotin in living organism (animals) using PET imaging. Potential applications of [11C]biotin include the diagnosis and localisation of bacterial infection and cancer via a directlabelling or pre-targeting strategy. Firstly, the uptake of biotin by E. Coli is highly efficient.

[11C]Biotin might have prospects for imaging bacterial infection compared with the "biotininspired" radiotracers in which the carboxylic group is modified [6] . Secondly, various aggressive cancer cell lines such as ovarian, leukemia, mastocytoma, colon, breast renal, and lung cancer cell lines overexpress SMVT. The use of [11C] biotin in cancer might be useful for cancer diagnosis/management and understanding the role of SMVT in tumorigenesis. Alongside the potential of [11C]biotin as an imaging biomarker in pathological conditions, its clinical translation will allow the study of whole-body biotin trafficking in healthy humans.

References: [1] Said, H. M. Biotin: biochemical, physiological and clinical aspects. Subcell. Biochem. 56, 1-19, doi:10.1007 Biochem. 56, 1-19, doi:10. /978-94-007-2199 Biochem. 56, 1-19, doi:10. -9_1 (2012 . [2] Ren, W. X. et al. Recent development of biotin conjugation in biological imaging, sensing, and target delivery. Chem. Commun. (Camb) 51, 10403-10418, doi:10.1039 /c5cc03075g (2015 . [3] Lewis, M. R. et al. In vivo evaluation of pretargeted 64Cu for tumor imaging and therapy. J. Nucl. Med. 44, 1284 -1292 . Abstract Body : Background: Fibrosis is a fundamental process involved in healing and remodelling during heart failure (HF). Positron emission tomography (PET) with fluoroprolines 1 is a potential tool for investigating active collagen biosynthesis in the myocardium. In order to test the utility of 18F-fluoroprolines as markers of collagen biosynthesis, animal models of cardiovascular fibrosis need to be developed and characterised. It was hypothesised that rats treated with angiotensin II (AngII) would develop myocardial fibrosis and could be used for testing PET radiotracers targeting collagen biosynthesis. This study aims to investigate the development of myocardial fibrosis in AngII treated rats using ex vivo imaging and biochemistry techniques. Methods: Adult male Sprague-Dawley rats (13-15 weeks, n=6-9) were administered AngII (250 or 500ng/kg/min, s.c.) or saline (control) for 4 weeks (using osmotic mini-pumps) to induce left ventricular hypertrophy and diffuse fibrosis. Blood pressure was measured during this period by tail cuff plethysmography. Ex vivo assessment of myocardial collagen content (Picrosirius red (PSR) stain; hydroxyproline (HP) assay) were undertaken in tissues isolated at the end of the treatment period. Furthermore, immunofluorescence was performed to investigate the presence of myocardial inflammation using macrophage and endothelial cell markers (CD68 and isolectin, respectively). Results: Systolic blood pressure was increased after treatment with 250 or 500ng/kg/min AngII, for 2weeks (p Conclusions: Infusion of AngII for 4 weeks results in the development of a promising diffuse fibrosis model able to mimic hypertensive HF. Results of the HP assay suggest potential for success using fluoroproline PET radiotracers in detecting active collagen biosynthesis in this model. This potential will be assessed by exposing the high dose (4 weeks of 500 ng/kg/min AngII) group to PET/MRI using 18F-fluoroprolines.

References: 1 Geisler, fluoro-proline: radiosynthesis, biological evaluation and results in humans using PET. Curr Radiopharm 7, 123-132 (2014) . Figure B . Hydroxyproline (HP) assay results showing increased HP content in high dose angiotensin II treated rats compared to controls. Figure C . Example images of Picrosirius red staining of a heart section from all treatment groups, red areas labelling collagen deposition. Comparison of all three groups showed increased fibrosis in the high dose group only. Figure D . Example images of the same hearts (adjacent sections to PSR) stained for DAPI (blue-nuclei) and CD68 (red-total macrophages), global analysis of CD68 % area showed no differences between groups. Figure E . The same sections as in , global analysis showed no differences between groups. Figure F . Graph shows the comparison of PSR % area measurements in perivascular areas of the myocardium, suggesting increased collagen deposition with high dose treatment. Figure G . Graph shows CD68 % area values of vascular areas in the myocardium compared between the three groups, suggesting an increase in macrophages with low dose treatment. Abstract Body : The present study aims at developing an innovative class of MRI contrast agents for Fast Field Cycling-MRI applications. They represent a completely new class of contrast agents, able to generate 14N-Quadrupolar Peaks (QPs), i.e. a relaxation enhancement at magnetic fields where the proton NMR frequency and the 14N nuclear quadrupole resonance frequency coincide [1] . Their detection requires the acquisition of images at variable magnetic field strength as provided by the innovative Fast Field Cycling MRI (FFC-MRI) scanners (Fig.  1 ). The QP from the contrast agents has to fall at frequencies well distinguishable from those associated with the amidic peptide bonds from endogenous proteins, which occur at 0.7, 2.1 and 2.8 MHz. The first generation of QP contrast agents contain a polymeric form of histidine, which imidazole groups produce a characteristic enhancement at about 1.3-1.4 MHz [2] . The QPs are detectable only when the contrast agent is in a gel-or solid-like form. Since the maintenance of the immobilized state depends on the pH, the contrast agents may report on tissue pH changes, that can be associated to the occurrence of a pathological condition, such as cancer, or to cellular apoptosis/necrosis. These innovative contrast agents are suitable for the design of nano-and micro-devices, such as scaffolds for tissue engineering in the field of regenerative medicine. They represent a possible answer to the current medical needs, such as an almost complete lack of methods for the rapid, non-invasive and repeated monitoring of tissue implants. Furthermore, new methods are needed to monitor cell status and polymer degradation under physiological conditions (temperature, saline medium, pH, enzymes etc.) thus allowing the physician to control, in real time, the status of the implanted scaffold. Herein, we present the preparation, relaxometric characterization, preliminary in vitro and in vivo data of porous poly-histidine-PLGA scaffolds suitable as QPs contrast agent for FFC-MRI, a new technique that introduces the strength of the applied magnetic field as an entirely new dimension into MRI, which opens new avenues for non-invasive imaging technologies with human applications.

Agne Stadulyte, University of Edinburgh, a.stadulyte@sms.ed.ac.uk

Abstract Body : Introduction: One of the most widely studied biomarkers of inflammation in the field of molecular imaging is the 18kDa translocator protein (TSPO). Various TSPO Positron Emission Tomography (PET) radiotracers have been developed for preclinical and clinical use. Unfortunately, these have suboptimal properties, including the binding to human tissue dependent on the rs6971 polymorphism [1] and high non-specific and non-displaceable binding in vivo [2] [3] [4] . Recently, the first fluorine-18 labelled PET TSPO radiotracer, [18F] LW223, with binding to human tissue independent of the rs6971 polymorphism [5] , has been developed, representing the resolution of a major limitation to clinical adoption of TSPO PET imaging. We hypothesise that [18F] LW223 will also have low non-specific and non-displaceable binding in vivo, which would resolve the second bottle neck in the field of TSPO PET imaging. This study aims to investigate the specific binding and equilibrium dissociation constant (Kd) of [18F] LW223. Furthermore it aims to report the mass dose limits (5% occupancy D5 and 10% occupancy D10) of [18F] LW223 in mice. Methods: [18F] LW223 was injected to male CD57Bl/6J mice (age=14.1±4.7 weeks, weight =28.4±3.1 g; mean±SD, n=16) via femoral or tail vein (8.1±3.1 MBq, 100-200 µL) either on its own (baseline molar activity of 84.8±47, mean±SD, n=3) or with varying doses of non-radioactive LW223 (0.0028±0.0008 mg/kg -'Dose 1', n=6; 0.016 mg/kg -'Dose 2', n=1; 0.22±0.04 mg/kg -'Dose 3', n=3; or 0.55±0.09 mg/kg -'Dose 4', n=3). Dynamic PET scans were performed over 2h. Standard uptake values (SUVs) of brain, heart, lungs, liver, spleen, marrow, gall bladder, urinary bladder intestines, skeletal muscle and blood (vena cava) were measured. Averaged SUV at 60-120 minutes post- [18F] LW223 injection were compared for each organ across groups and used to calculate % specific binding (%SB) for plotting of saturation curves and subsequent derivation of constants using one site specific binding equation. Results/Discussion: In vivo dynamic PET imaging showed rapid uptake of [18F] LW223 in the mouse organs with known TSPO expression. The results also suggested that [18F] LW223 was mainly excreted via hepatobiliary system, with only negligible levels excreted via renal system. Significant homologous blocking with doses 3 and 4 was measured in the brain, heart and lungs (p Conclusion: [18F] LW223 has nanomolar affinity for TSPO in vivo in the mouse and low non-specific binding across all target organs, therefore holds huge potential for TSPO PET imaging of sites throughout the body. Mass dose limits demonstrate feasible use of radiotracer in small animals and easy translation to humans. Abstract Body : Introduction: Nonalcoholic steatohepatitis (NASH), a severe form of nonalcoholic fatty liver disease (NAFLD), is characterized on liver histology by lobular inflammation, hepatocyte ballooning and more progressive liver fibrosis1. The presence and severity of liver fibrosis in NASH patients determines the long-term prognosis2. Activated hepatic stellate cells (HSCs) are the major cellular source of myofibroblasts, which are responsible for producing excess extracellular matrix and the central driver of liver fibrosis3. Integrin avb3 is emerging as a potential antifibrotic target and an imaging marker in liver fibrosis because of its upregulated expression in activated HSCs4. Our aim was to test whether 18F-FPPRGD2, a PET molecular imaging tracer for targeting integrin avb3 expressed on activated hepatic stellate cells, could noninvasively assess the target engagement by the novel avb1, avb3, avb6 integrin antagonist IDL-2965 in a choline-deficient, L-amino aciddefined, high-fat diet (CDAHFD) NASH mouse model. Method: C57Bl/6 male mice were fed with CDAHFD for either 2, 6, 10, or 14 weeks to induce different stages of fibrosis (n = 6/time point). Age-matched mice (n = 6/time point) fed on normal chow served as control. All mice were imaged by 4.7 Tesla MRI scanner (Bruker, Billerica MA) with a Bruker PET insert on the first day of each time point for 90 minutes after intravenously injection of 100-150 mL of 18F-FPPRGD2 tracer solution (100-200 mCi); PET/MRI was repeated in CDAHFD mice on the following day one hour post oral gavage of IDL-2965 (10 mg/Kg). Tissues were harvested for qRT-PCR, hydroxyproline assay, hematoxylin and eosin (H&E), and Sirius red staining at the end of the study. PET signal was quantified by volume of interest (1 cm3) in liver avoiding large hepatic vessels. Statistical analysis was tested for all measurements among groups using one-way ANOVA with two-tailed distribution by GraphPad Prism 7 software. Results: PET quantification showed tracer uptake in the liver increased with the duration of CDAHFD up to 10 weeks (Figure 1 ). At each time point, the uptake of tracer in NASH liver was statistically higher than that in age-matched control mice (week 2: 0.55±0.12 %ID vs. 0.26±0.05 %ID, P a-SMA (marker of activated HSCs), Col1a1, integrin αv and β3 showed increasing upregulation as the disease progressed and reached statistical significance at the advanced stage of NASH at week 14 ( Figure 4) . Conclusion: 18F-FPPRGD2 PET/MRI noninvasively detects increased expression of integrin αvβ3 with progression of liver fibrosis in the CDAHFD-induced NASH model and shows potential to assess target engagement of a novel therapeutic. Acknowledgements: This work was supported by funding from the National Institutes of Health (DK121789, DK104956, OD023503, OD010650), the Shanghai Municipal Planning Commission of Science and Research Fund (201640143) , and Indalo Therapeutics.

Shatadru Chakravarty, Michigan State University, shatadru@msu.edu

Abstract Body : Objective: X-ray Computed Tomography (CT) can potentially transform into a robust molecular imaging platform fueled by recent developments in contrast agent (CA) research. [1] [2] [3] We have previously reported TaOx nanocrystals (NCs) with the highest Ta content reported till date (78%) as CAs for CT imaging. [4] In this work, the in vivo performance of this unique NC construct has been evaluated for contrast enhanced micro-CT imaging of the breast ductal tree in rodents. Briefly, we investigated an improved formulation of the TaOx NCs in ethanol as a high-resolution contrast agent to monitor in vivo filling of breast ductal trees and ablative effects of ethanol (EtOH) by CT imaging. Our hypothesis is that the local killing of mammary epithelial cells will be as effective as prophylactic mastectomy in preventing breast cancer, but with minimal side effects. In addition, we have used ethyl-cellulose (EC) as a gelling agent to further minimize collateral tissue damage. These TaOx NCs will also extensively evaluated for their in vivo toxicity in rats. Methods, Results and Discussion: We first synthesized the highly hydrophilic, homogenously sized (9-12 nm) TaOx NCs using a previously reported procedure. [4] In a recent study, we have demonstrated the ability of TaOx NCs to be administered locally and reveal in exquisite detail the continuous non-anastomosed branched structure of a murine ductal tree. We have also shown that intraductal (ID) injection of a 70% EtOH solution is effective at locally ablating mammary epithelial cells with limited collateral tissue damage and at preventing tumor formation in an aggressive and multifocal mouse model of breast cancer. [5] To advance the efficacy and prognosis of this treatment regimen, we designed an improved formulation of TaOx NCs with higher solubility in EtOH. Interestingly, EC has been used clinically to improve delivery of EtOH and to limit EtOH diffusion from the intended target area in the treatment of venous malformations. [6, 7] Thus, to improve the tolerability of the TaOx formulation, EC was added to it as a gelling agent. ID injection of 70% EtOH/3% EC in nontransgenic mice and rats achieved the same epithelial ablation rate as 70% EtOH alone, while significantly minimizing collateral tissue damage. The TaOx NCs within this novel formulation served to evaluate any local or systemic toxicity of this new imaging agent. These data demonstrate the compatibility, safety, and stability of this refined EtOH formulation and support its further investigation in cancer-prone rodent and larger animal models as an innovative primary intervention strategy for BC prevention. Conclusion: We report the in vivo performance of the highly hydrophilic TaOx NCs formulations in saline and EtOH/EC via two exploratory experiments. Taken in unison, our work demonstrates the CT imaging application, theranostic potential and safety of TaOx NCs.

Yuanxue Gao, ZheJiang University, 11818370@zju.edu.cn

Abstract Body : Objective: Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) are a promising cell source for regenerating injured heart tissue. Tracking the fate of implanted hiPSC-CMs allows us to understand the mechanisms underlying the functional improvement in the infarcted heart and accelerate the clinical translation of hiPSC-CM therapies. This study aims to investigate the posttransplantation behaviour and metabolic changes after transplantation of hiPSC-CMs following acute myocardial infarction (AMI). Methods: Genome editing with CRISPR/Cas 9 technology was used to infuse a triple-fusion (TF) reporter gene into the AAVS1 locus of human urinary induced pluripotent stem cell (TF-hUiPSCs) which stably express monomeric red fluorescent protein for fluorescence, firefly luciferase for bioluminescence imaging (BLI), and herpes simplex virus thymidine kinase for positron emission tomography (PET) imaging. TF-hUiPSCs were induced to differentiate into cardiomyocytes (TF-hUiCMs) and then were processed using a pro-survival cocktail protocol to enhance engraftment post transplantation. AMI was surgically induced in rats by ligation of the left anterior descending artery, TF-hUiCMs (1.0 × 107 per rat) in a volume of 100 µl or an equal volume of PBS was injected directly into the border zone of the infarcted myocardium immediately after ligation. Transplanted cells were detected by BLI at day 4 and day 35. The dynamic metabolic change was monitored by [18F] -FDG PET imaging. Results: In the rat model of acute myocardial infarction, bioluminescence imaging tracked the TF-hUiCMs noninvasively and the signal showed an increasing trend from day 4 to 35 post injection. [18F] -FDG microPET imaging showed an enhancement in myocardial glucose uptake in the peri-infarct zone at day 35 after TF-hUiCMs transplantation. Conclusion: Our study demonstrated that TF-hUiCMs engrafted and proliferated in the infarcted heart and enhanced cardiomyocytes glucose metabolism under ischaemic conditions. BLI and PET imaging could provide biological information of transplanted stem cells in myocardial infarction. Abstract Body : Background: Loss of synapses and damage of neural circuits in the cerebral cortex are the fundamental causes of stroke-induced neurological deficits in adult humans. Stem cell-based therapies for synapse replacement and circuit repair after stroke require generation of specific neuronal subtypes and maintainence of excitatory and inhibitory balance closely resembling neuronal networks in vivo. In order to restore injured synapses and circuits more effectively, we used the human cortical progenitors (hCPs) derived from human induced pluripotent stem cells (iPSCs) to generate both excitatory and inhibitory neurons resembling the composition of the human cortex. Methods: Focal cortical infarct in rats was induced by the photothrombotic occlusion. Magnetic resonance imaging (MRI) was used to identify the exact location and size of the infarct, which provided a precise guidance for transplantation of hCPs to peri-stroke cortical area. Then, positron emission tomography (PET) imaging was utilized to monitor the metabolic changes of infarcted cortex after hCPs transplantation. Meanwhile, in vivo bioluminescence imaging (BLI) was used to evaluate the survival, proliferation and distribution of hCPs and their progeny cells. Moreover, we used immunofluorescence, immunoelectron microscopy and patch-clamp to evaluate the differentiation and maturation of hCPs after transplantation. Virus-based anterograde monosynaptic tracing technique was used to confirm the establishment of direct synaptic connections between progeny cells of hCPs and ventral thalamus of rat brain. To investigate the long-term effects of stem cell transplantation on brain function, we used EEG to monitor the epileptic discharge activity in the peri-infarct areas for up to 12 months after stem cell transplantation. Results: 18F-FDG PET imaging showed that the glucose metabolism began to improve at the 1st week after hCPs transplantation (L/H=0.80), and the maximum improvement was achieved by the 4th week (L/H=0.90). BLI results showed that the number of hCPs increased gradually with the time of transplantation within 5 weeks after transplantation. At 2 months after transplantation, immunofluorescence, immunoelectron microscopy and patch-clamp results confirmed that hCPs successfully differentiated into all six layers of cortical neurons, and the progeny neurons formed direct synaptic connections with host neurons and obtained mature electrophysiological characteristics. Moreover, the virus-based anterograde monosynaptic tracing result demonstrated that the grafted neurons receive direct synaptic inputs from the ventral thalamus of the rat brain within 3 months after transplantation. The EEG data showed that the probability of epilepsy in the hCPs transplantation group was 1/8 (12.5%), while that in the control group was 3/5 (60%), which indicated that hCPs transplantation has the potential to prevent epilepsy after focal cortical stroke. At 12 months after stroke, there was no tumor formation in immunocompromised rats. Conclusions: Our findings show, for the first time, that human iPSCs-derived hCPs could largely prevent chronic cortical seizures after stroke. Transplanted hCPs could differentiate into mature neuronal cells, regenerate new synapses and restore circuitry, and thereby contribute to functional repair at later time points after intracortical transplantation in a rat stroke model. Abstract Body : Among first world nations, Singapore has one of the highest rates of diabetes related lower extremity amputation [1] . According to the national health survey 2010, the prevalence of diabetes in Singapore has increased from 8.2% in the year 2004 to 11.3% in the year 2010 and is likely to increase to 15.9% by the year 2050 [2, 3] . This has caused a huge burden including medical expenses and productivity loss on Singapore's economy. Chronic diseases like diabetes or peripheral vascular disease cause compromised lower limb vascularization impacting perfusion and tissue oxygenation. Current assessment tools for tissue oxygenation like monofilament test, ankle brachial index (ABI), toe brachial index (TBI) are subjective, time-consuming, and allow for local tissue quantification. Also, it has been studied that in the presence of other artery complications, these methods are inaccurate for diabetic foot ulcer screening [4] . Herein, we report the screening of fifteen (15) patients with diabetes in a currently ongoing clinical study using Reflect RSTM, a spatial frequency domain imaging (SFDI) based system to evaluate tissue health status in terms of perfusion and oxygenation. SFDI is a non-contact, non-invasive optical imaging technique that allows for large area (20cm x 15cm) measurement of tissue chromophores and physiological parameters such as melanin, oxyhemoglobin (HbO2), deoxy-hemoglobin (HbR), oxygen saturation (StO2) and total haemoglobin (HbT) up to ~3 mm below the skin surface. With the help of depth-resolved model analysis, SFDI evaluates tissue haemoglobin in the papillary (micro-circulation, HbT1) and reticular dermis (macro-circulation, HbT2) respectively [5] . From the analysis of acquired data, we observed that patients with diabetes show poor microcirculation in the papillary dermis (HbT1) of plantar foot as diabetes is known to affect the microcirculation due to diabetic neuropathy. One of the diabetic patients with foot ulcer (DFU) on the right hallux indicated lower microcirculation values than other diabetic subjects with no ulcer and elevated StO2 values all over both plantar feet. The lower values of micro and macrocirculation may be due to the arteriopathic condition supplying less blood throughout the feet. The reduced HbT1 and elevated StO2 could be attributed to arteriovenous shunting [5] resulting in under-extraction of oxygen and may help identify ulceration risk that other tools cannot identify. With the help of these imaging metrics, we envision that these objective measures augment the poorly reproducible and variable clinical parameters that are currently used to risk stratify patients with diabetes ensuring timely interventions and treatments delivered to those at most risk. [1-13C] BHB to a mixture of BDH enzyme and NAD, production of HP [1-13C] acetoacetate was observed at 11.7T (C) Time course of substrate and product signal from acquisition in B (D) 13C thermal NMR spectrum of brain tissue snap-frozen immediately after injection of 13C BHB, showing the resonances of [1-13C] BHB and [1-13C] AcAc. (E-J) Characterization of serum BHB and brain BDH activity in healthy and hAPP-J20 mice. In healthy mice, (E) serum BHB and (F) cerebral BDH activity were both significantly elevated following a ketogenic diet (Student's t-test). (G) No correlation was observed between these two parameters (Pearson correlation). Similarly in hAPP-J20 mice, (H) serum BHB and (I) cerebral BDH activity were both significantly elevated following a ketogenic diet. (J) No correlation was observed between these two parameters. *p≤0.05, ***p (K) Following injection of HP [1-13C] BHB into an anesthetized control mouse, non-localized dynamic 13C spectra were acquired using a 3T Bruker system, showing the presence of [1-13C] BHB (181 ppm) in the brain (T2 image inset). On summing the spectra, HP acetoacetate was observed (175.5ppm). Abstract Body : Introduction: Intra-operative management of surgical margins in patients diagnosed with solid tumors remains challenging as surgeons continue to rely primarily on visual and tactile information. Currently, no reliable, clinically-approved method supporting real-time intra-operative decision making exists. Fluorescence-guided surgery using tumor-specific compounds is an interesting clinical tool for aiding in real-time evaluation of the margin status. ONM-100 is a novel, pH-activatable micelle encapsulating the fluorescent imaging agent, Indocyanine Green (ICG). ONM-100 exhibits a binary off/on mechanism that allows for specific fluorescence activation in the acidotic tumor microenvironment. This study investigates a standardized, specimen-driven, fluorescence imaging framework using ONM-100 as a clinical decision-making tool during the surgical treatment of head and neck squamous cell carcinoma (HNSCC) and breast cancer (BC). Material and Methods: 13 patients with histologically confirmed head and neck squamous cell carcinoma (n=9) or breast cancer (n=4) were included in this phase I study (Netherlands National Trial Register 7085). ONM-100 was administered intravenously 24 ± 8 hours before surgery. Fluorescence images of the excised surgical specimen and the surgical cavity were analyzed. Fluorescent lesions with a Tumor-to-Background Ratio (TBR) > 1.5, either on the surgical specimen or in the surgical cavity, were considered to be fluorescence-positive. In vivo imaging was performed with the NovaDaq Spy Elite and the ex-vivo imaging with the Pearl-trilogy. Fluorescence positive specimens were collected and submitted for standard of care histopathological evaluation. Results: In total, nine histologically confirmed tumor-positive surgical margins were discovered immediately following primary tumor excision using ONM-100 fluorescence-guided intra-operative imaging, six in HNSCC patients and three in BC patients. Postoperative analysis showed a median (±IQR) TBR of the fluorescent lesions at the margin of resection to be 3.36±1.62 and 3.8±1.8 for HNSCC and BC, respectively. Three additional fluorescent lesions in the surgical cavity (2 HNSCC, 1 BC) were biopsied, and showed occult carcinoma, severe dysplasia and false-positive detection of muscle tissue. Conclusion: Our fluorescence framework using a novel, pH-activatable, fluorescent imaging agent enables reliable and adequate surgical decision making in real-time. Using a fluorescence TBR cutoff value of 1.5, we demonstrate a strong correlation of fluorescence to tumor-positive resection margins across 13 patients and two tumor types. The binary mechanism of ONM-100 allows for sharp tumor delineation in HNSSC and BC, and therefore provides the surgeon with a highly sensitive clinical tool for real-time margin assessment. In current standard of care, intra-operative surgical margin analysis is performed on frozen sections harvested from the specimen and not from the woundbed, because this is considered the most reliable approach. The fluorescent framework allows detailed specimen driven analysis as well as surgical woundbed analysis, which in this study revealed additional lesions in the woundbed. There are three possible scenarios identified in this study following the detection of a fluorescent spot in the ex vivo specimen or the surgical woundbed. Fluorescence spots 1) can be resected with adequate margin without compromising vital structures, 2) an identified fluorescent spot cannot be resected or 3) new spots outside the identified tumor field are identified in the mucosa or deep margin (in transit metastasis). Situations 1 and 3 occurred during this study. In this fluorescence guided decision making, 8 out of 9 spots would have been re-resected in real time, demonstrating the potential clinical benefit of this fluorescent framework. In this study, we will first show that by using Multi-frequency Excitation Wideband EPI (MEWB-EPI), this technique can improve temporal resolution by two folds on the current MR imager without upgrading its hardware. This advantage can instead tradeoff to have higher signal to ratio image by averaging 2 fasten MR acquisitions while remain the same acquisition time. In this paper, we employ this higher SNR images to improve the accuracy of diffusion study analysis to achieve more consistent FA and accurate fiber tracking. Introduction: The selection of parameter is a tradeoff between time and image quality. Recently, simultaneous multiband studies have opened a new horizon to shorten the imaging time; and among them, ME-wideband MRI and SE-wideband MRI are two extraordinary examples. [1] [2] In this study, we apply this novel acquisition method, "Multi-frequency Excitation Wideband EPI (MEWB-EPI) on the 7T Bruker imager, and show that this technique can easily accelerate 2fold temporal resolution. We can average these ultrafast images to achieve better SNR in the same imaging time.

Material and Method The MEWB-EPI sequence was implemented in both EPI and EPI DTI sequence to assess the technique for the anatomy and diffusion weight imaging of rat brain. All acquisitions were obtained from 7T MRI (Bruker, Ettlingen, Germany) system with MEWB-EPI (W, wideband acceleration factor, = 2) 2X Temporal Resolution Spin-Echo EPI Image

The Spin-Echo EPI images are the following parameters: FOV=2.5x2.5cm2, 1mm thickness, Matrix Size= 96x96, TR/TE = 5000/70 ms, NEX = 40. The two set images are the same spatial resolution: 260um2. Scan time for standard EPI and WB-EPI are 43m20s & 21m40s, half of the original time respectively. 2X Average Number Diffusion Application The spin echo diffusion imaging is combined with 1 null-image and 12-diretion diffusion gradients. The parameters of the diffusion study are as follows: FOV=2.5x2.5cm2, 1mm thickness, TR/TE = 2500/21 ms, Δ/δ = 8/3ms, b-value = 1500s/mm2, Average Number = 40/80 respectively. Since 2X ME-Wideband technique is 2 times faster, we can take this advantage to imaging an 80 average number instead of the original 40 average number imaging, while maintaining the same scan time in 43min 20s. FA analysis and fiber track MR images and analysis were implemented in MATLAB (MathWorks, Natick, U.S.A.). The diffusion tracking are reconstructed by DSI Studio (http://dsistudio.labsolver.org). To evaluate the reproducibility of FA performance, the same DTI sequence are repeated twice in this study. This Statistics of this difference of FA value can provide as a DTI performance information. Result and Discussion:

In the MEWB-EPI 2x temporal resolution experiment, the acquisition time of EPI sequence is reduced from 43min 20s to 21min 40s. The signal to noise, contrast and details of this 2X fast image kept well at the equal quality of the original one (Fig.1a, 1b) . In the 2X average number diffusion experiment, SNR of DTI null images can be increased from 76.2±8.2 to 105.2±5.8 (Fig.1c, 1d) , which match the 1.414 theoretical average gain. Its associated calculated FA images accordingly have better tissue delineation (Fig.1e, 1f ) and its error of FA values is evidently reduced in Fig.1g . And this enhanced reconstructed Corpus callosum (Fig.1h, 1i ) also clearly shows a smooth and stable fiber tracking.

Conclusions:

The purpose of this study is to show the feasibility of accurate diffusion studies by using the MEWB-EPI technique. Our results show, this novel technique could easily accelerate the temporal resolution and also can take the scan speed as a tradeoff into its associated 1.414 SNR gain for a 2X faster sequence. In general, user can use the speed gain to either benefit on either speed gain or on anatomical details gain. In the MEWB-EPI 2x temporal resolution experiment, the acquisition time of EPI sequence is reduced from 43min 20s to 21min 40s. The signal to. noise, contrast and details of this 2X fast image kept well at the equal quality of the original one (Fig.1a, 1b) . In the 2X average number diffusion experiment, SNR of DTI null images can be increased from 76.2±8.2 to 105.2±5.8 (Fig.1c, 1d) , which match the 1.414 theoretical average gain. Its associated calculated FA images accordingly have better tissue delineation (Fig.1e, 1f ) and its error of FA values is evidently reduced in Fig.1g . And this enhanced reconstructed Corpus callosum (Fig.1h, 1i ) also clearly shows a smooth and stable fiber tracking to reveal fine fiber for better 3D brain mapping in the future. Abstract Body : Magnetic resonance imaging (MRI) is a powerful non-invasive medical imaging technique that produces high spatial and temporal resolution functional 3D images anatomy and physiological functions.1 Working on the principles of NMR, MRI offers clear images by monitoring the differences between longitudinal (T1) and transverse (T2) relaxation times of water protons in the body. Different tissues in the body present different T1 and T2 values due to the differential water content. MRI can detect these intrinsic differences and produce a unique visual distinction between normal and abnormal tissues. To enhance the MR signal contrast agents (CAs) are employed. Gadolinium-based MRI T1 contrast agents (GdCAs) are the current gold standard for contrast enhanced medical applications in radiology.2 However, due to the low kinetic stability of acyclic GdCAs in vivo, a red flag has been raised for most of the clinical GdCAs due to agravated nephrogenic systemic fibrosis (NSF)3,4 and detection of Gd3+ deposition in the brain upon recurrent injection of GdCAs.5 Due to these concerns, the European Medicines Agency (EMA) had suspended the marketing authorizations of four acyclic GdCAs such as gadodiamide, (Omniscan), gadoversetamide (OptiMARK), gadopentetate dimeglumine (Magnevist) and gadobenic acid (Multihance-specifically restricted "to be used in liver scans due to they are taken up in the liver and meet an important diagnostic need").6,7 The US -Food and Drug Administration (FDA) also advised against the use of linear acyclic GdCAs in renal compromised patients.6,8 To overcome these hurdles, the development of Gd3+-free MR contrast agents are critical in radiology. To tackle this, manganese-based contrast agents (MnCAs) are proposed as safer alternatives9 to GdCAs. Mn2+ has five unpaired d electrons, long electronic relaxation time and rapid water exchange. These properties are ideal for CAs capable of ennhancing T1.10 Manganese is a biogenic and vital metal that is found in all tissues (mean serum concentration is 0.5-1.2 μg/L) and is required for bone development, neuronal health and several physiological processes including cellular (mitochondrial) functions.11 Despite the increased interest in MnCAs development, only, Manganese(II) chloride (LumenHanceTM), is currently used for animal MRI studies.12 Therefore, there is a significant interest in the development in next-generation MnCAs, with high thermodynamic stability, kinetic inertness, and fast clearance from the body. In this work, we have employed for the first time a single pot template synthesis strategy to develop a manganese-based MRI contrast agent MnLMe. MnLMe has been fully chemically characterized via, X-ray and NMR, determination of the hydrated states and stability and relaxometry has been performed. MnLMe displays relaxivity, r1 = 4.2 ± 0.2 mM-1s-1 (400 MHz, 298 K) and 4.9 ± 0.1 mM-1s-1 (64 MHz, 298 K) which is superior to the several reported acyclic MnCAs. The stability of the MnLMe has been investigated by zinc transmetallation, via time-dependent relaxivity and absorption spectral titration methods. The MnLMe displays remarkable kinetic inertness even in the presence of Zn2+ ions. MnLMe shows binding to BSA (Ka = 4.16 × 103 M-1) with an enhanced r1 = 21.5 mM-1 s- 1 (20 MHz, 298 K, pH 7.4) . In vivo MRI studies in healthy mice show MnLMe (n =5) has a very different biodistribution to the GdCA Magnevist (n =5), with greater excretion form the liver and prolonged blood half-life, as observed by the continuous high ΔSI in the left ventricle of the heart. These data indicate that MnLMe interacts with albumin and other proteins throughout the imaging window, confirming the binding to serum albumin and potential for use as a blood pool agent.

Imaging glucose-stimulated zinc secretion from the prostate and pancreas using a Mn(II)-based sensor Andre Martins, Werner Siemens Imaging Center, andre.martins@med.uni-tuebingen.de Category: Systemic Diseases (Kidney, Liver and Pancreas) Abstract Body : The divalent zinc ion (Zn2+) plays an essential structural and catalytic role in many metalloenzymes and zinc finger proteins but also plays a key role in the secretory function of many tissues including the brain, pancreas, mammary glands, and prostate.1 Dysregulation in tissue Zn2+ has been associated with diseases such as Alzheimer's disease, diabetes, transient neonatal zinc deficiency, and prostate cancer. For instance, the healthy human prostate has been reported to have the highest levels of zinc while malignant prostate cells undergo a metabolic transformation that results in reduced accumulation of Zn(II).2 Pancreatic b-cells also store insulin in granules in crystalline form containing molecules of pro-insulin with Zn2+ ions. Upon release of zinc by glucose stimulation, the local concentration of Zn2+ ions around β-cells (pancreas) and epithelial cells (prostate acinar glands), transiently rises to ∼0.5 mM.1 This process is known as glucose-stimulated zinc secretion (GSZS). Imaging GSZS from secretory tissues has proven useful at assessing organ function and health. So far, the only successful probes to detect zinc secretion by MRI in vivo are gadolinium-based sensors.3 In this work we introduce a manganese-based zinc sensor and show that pancreatic and prostatic zinc detection is not compromised when using Mn instead of Gd for imaging GSZS in vivo. Given the current concerns associated with Gd deposition after repeated usage of some Gd-based MRI contrast agents, an effective manganese alternative for the MRI detection of GSZS has been developed herein ( Figure 1 ). In the absence of Zn2+, the newly synthesized MnL zinc sensor has little to no affinity for human serum albumin (HSA) but after binding to 1 equivalent of Zn2+, the resulting complexes then bind to HSA with strong zinc binding (KD(Zn) = 90 nM). This in turn results in a remarkable increase in r1 relaxivity (~430%). MnL showed also superior kinetic inertness toward transmetallation by Zn2+ and transchelation by HSA. This sensor is excreted via hepatobiliary pathways and as an intact complex via renal filtration; as observed by LC-MS of urine. Control experiments showed marginal enhancement in the prostate and pancreas when no D-glucose was co-injected with MnL. In turn, when glucose was co-injected with MnL both the pancreas and the prostate showed superior T1 contrast when compared to the respective controls. Data suggested that MnL is uniquely sensitive to zinc. Tissue measurements of Mn content by ICP-AES together with imaging shows that after 15 minutes post-injection, MnL is found primarily in the kidneys, liver, pancreas, and prostate. After 90 minutes, the Mn content was greatly diminished in the kidneys and liver tissues but accumulated in the spleen and in the prostate. The later observation may reflect Mn in the prostatic urethra during excretion of the agent. In summary, a zinc-responsive Mn-based MRI contrast agent has been used to image GSZS in mice, similar to that reported previously using Gd-based agents. 3-6 The tissue Mn content showed that the MnL chelate is excreted through hepatobiliary pathways and intact through renal filtration. These data suggest that a Mn-based zinc sensor is a reliable alternative to Gd agents for detection of GSZS by MRI.7 ID: GA501 Intraoperative tissue characterisation during robotic surgery using fluorescencebased smart forceps -development and evaluation in large animal models Matthias van Oosterom, Leiden University Medical Center, m.n.van_oosterom@lumc.nl

Abstract Body : Aim: Robot-assisted surgery has become an established technique for many minimal invasive laparoscopic procedures. With the increase in surgical precision, there is a great potential to specifically integrate molecular image-guided surgery concepts within this robotic workflow, further improving treatment precision: radical excision of diseased tissues with minimal damage to healthy tissues. An example of this is the recently introduction DROP-IN gamma probe, a small and tethered detection modality that integrates radioguided surgery in the robotic workflow, optimally exploiting the enhanced manoeuvrability in the robotic setup [1, 2, 3] . Next to the integration of radioguidance, fluorescence guidance is also increasingly being used during robotic surgery (e.g. lymphatic mapping and angiography using indocyanine green (ICG)). In addition to the limitations in manoeuvrability of the current fluorescence laparoscopes, frequent use of fluorescence imaging may restrict the surgical workflow: within laparoscopic surgery the overall visualization of the patient anatomy (e.g. bleedings) during fluorescence imaging has proven inferior to white light imaging. Therefore, implementation of fluorescence imaging often requires the operating surgeon to pause his/her resection, interrupting the surgical workflow. Furthermore, this means that, during the actual resection under white light, fluorescent tracers are not detected. For that reason, we developed and investigated small click-on fluorescence detectors that transform standard robotic surgical instruments into 'smart' fluorescence imaging devices (SmartForceps), monitoring the presence of fluorescent tracers in every tissue grasped during the surgical procedure. Materials and Methods: A click-on fibrebased fluorescence detector was engineered for the ProGrasp forceps of a da Vinci surgical robotic system, capable of detecting ICG. Translational investigation of this concept was evaluated in: 1) phantoms; 2) robotic surgery in porcine models (n=5), investigating lymph node localisation and angiography settings; and 3) ex vivo human specimens from the prostate cancer sentinel node procedure (n=2; using the hybrid tracer ICG-99mTc-nanocolloid). Results: The smart robotic instrument, with the detectors clicked-on, was small enough to be introduced through the standard 12 mm trocars. After introduction in the porcine abdomen, the fluorescence detectors did not hinder the forceps manoeuvrability or grasping ability. As confirmed with Firefly fluorescence imaging, fluorescent counts were only detected in the tracer containing tissues. The porcine surgery demonstrated that grasping with the SmartForceps of the different tissues allowed for the identification of vascularization (kidney, bowel and bladder) and lymphatic nodes. The ex vivo human tissue evaluation indicated that the SmartForceps could distinguish sentinel lymph nodes from non-sentinel lymph nodes (confirmed with Firefly fluorescence and gamma probe investigation). Conclusion: This study describes the successful introduction of using click-on fluorescence-based sensing modules within the surgical robotic workflow, converting regular forceps into so-called SmartForceps. This concept allows for fluorescence detection during white light resections. This might open the way for various biosensing applications, where the surgical instruments themselves are used to characterise the molecular aspects of tissues handled during surgery. Abstract Body : Aim: Robot-assisted laparoscopic surgery is claiming an increasing share of the minimal invasive interventions performed. With the successful in-human introduction of the DROP-IN gamma probe, radioguided surgery has become compatible with the surgical robotic setting (e.g. sentinel lymph node biopsy or even PSMA-targeted dissections for prostate cancer) [1, 2, 3] . As a next step in (molecular-based) precision surgery, 'GPS-like' navigation of surgical laparoscopic detection modalities using preoperative patient scans has been suggested [4] . However, the tool-tracking needed for such a surgical navigation setup is complex in a dynamic laparoscopic workflow. In an attempt to integrate surgical navigation with the DROP-IN modality, a fluorescence-based video-tracking setup was developed and evaluated in the robotic surgical setting. Materials and Methods: The Firefly fluorescence laparoscope, integrated in the da Vinci surgical platform and capable of multispectral fluorescence imaging of fluorescein (lex = 488 nm, lem = 515 nm) and indocyanine green (ICG; lex = 800 nm, lem = 820 nm), was used [5] . Computer vision video-processing algorithms were engineered to recognize and segment fluorescent tracking-markers on the DROP-IN probe housing. Based on this information, the dedicated algorithms were capable to calculate in real-time the relative position and orientation (pose) of the probe with respect to the robotic fluorescence laparoscope. These tracking algorithms were incorporated in a SPECT-based surgical navigation device, capable to also track the pose of the laparoscope itself and the patient, using near-infrared optical tracking. Using a torso phantom for robotic surgery, this enabled navigation of the DROP-IN probe towards radioactive lesions (marked with 99mTc as identified on preoperative SPECT/CT imaging). Functioning of the navigated DROP-IN probe concept was further investigated with robotic surgery in porcine models (n=4). Results: Exploiting the multispectral fluorescence imaging capabilities, the fluorescein containing DROP-IN markers could be separated from ICG-positive lesions based on the differing emission wavelengths. Even under surgical conditions, preserving a direct line-of-sight with respect to these fluorescent markers allowed for intraoperative pose tracking of the DROP-IN modality with respect to tissue lesions. This enabled the augmented reality visualization of the real-time distance between the lesion targets and the DROP-IN probe, as visualized in the robotic console. Once integrated with SPECT/CT navigation, accurate guidance of the DROP-IN probe relative to preoperatively marked lesions became possible. In this setting, the real-time acoustic and numeric feedback, as delivered by the DROP-IN gamma probe, confirmed that the lesion localization was performed successfully. Conclusion: Using (multi-spectral) fluorescence imaging, we have demonstrated that future usage of the DROP-IN probe modality might be enriched with surgical navigation towards target-tissues as marked in preoperative images. With that a next step has been taken towards (molecular-based) imageguided surgery in the robotic setting.

Imaging tumor hypersialylation using metabolic labelling Flaviu Bulat, Cancer Research UK, flaviu.bulat@cruk.cam.ac.uk

Abstract Body : Introduction: Aberrant cell surface glycosylation has been described as one of the key hallmarks of cancer, with hypersialylation the most commonly observed alteration in cancer.1 Levels of sialylation can be monitored by incorporating unnatural sugars bearing bioorthogonal chemical reporters. Monitoring hypersialylation could provide an insight into tumor progression, proliferation and ultimately could potentially be used for monitoring treatment response.2,3 We describe here in vivo imaging of hypersialylation using a novel triacetylated cyclopropene mannosamine (Ac3ManNCyoc) derivative. Methods: Mannosamine was functionalised using a carbamate-linked methyl cyclopropene to yield an unnatural sugar (ManNCyoc) that can be metabolised by cancer cells. A panel of sugars was synthesised with different degrees of acetylation (1 to 4), which we expected to show different diffusion rates across the cell membrane and different solubilities. A 'click' chemistry fluorescent probe, tetrazine-PEG11-DyLight800 was synthesised and used to label the methyl cyclopropene moiety and sugar incorporation levels were assessed in vitro using flow cytometry. One candidate was selected for in vivo imaging as this gave the highest signal-to-baseline ratio (SBR) on flow cytometry. In vivo fluorescence imaging was performed in Colo205 xenografts. Results: Varying the levels of sugar acetylation resulted in significant differences in cell labelling indicating that there was an optimum level of acetylation that balanced sugar solubility and membrane penetration. Flow cytometry (Fig 1B) showed that of the 4-sugar panel, the triacetylated sugar Ac3ManNCyoc gave the highest SBR (8.7) . The fluorescently labelled tetrazine-PEG11-DyLight800 probe showed a favourable biodistribution profile, with quick renal clearance and minimal organ retention in the xenograft-bearing mice, at 24 h following administration. Mice treated with Ac3ManNCyoc and the tetrazine probe (+/+); Fig 1A) showed a 1.6-fold increase in fluorescence, compared to vehicle treated animals (-/+); Fig 1A & 1C, n = 2). Ex vivo fluorescence imaging of tumors showed an average 2.9-fold increase in fluorescence between control and sugar treated lesions (n = 2, Fig 1D) . Conclusions: Ac3ManNCyoc is a novel candidate for probing hypersialylation by metabolic labelling and subsequent imaging. The triacetylated sugar represents the optimal compromise between sugar solubility and membrane penetration.

vivo imaging with GdOA showed significant (P ) increases in ΔR1 in the cortex and medulla in KO mice (n=14) when compared to controls (n=12). However, the non-specific binding probe Gd-DOTA showed no significant differences in ΔR1 between KO and control groups. The GdOA induced ΔR1 changes were highly correlated to ex vivo measures of Gd and hydroxyproline in the renal cortex and medulla, indicating the specificity of GdOA to detect and stage fibrosis. Conclusion We demonstrated that GdOA enhanced molecular MR detects and quantifies renal fibrosis in a mouse model of AS. GdOA MRI may enable monitoring of disease progression and the effectiveness of potential new treatments for chronic kidney disease. NIH 

Annasofia Anemone, University of Turin, annasofiaantonia.anemone@unito.it

Abstract Body : Introduction Native glucose and the non-metabolizable derivative 3OMG have been proposed as MRI-CEST contrast agents for tumor detection and as likely alternatives to FDG-PET [1, 2] . Although several studies have investigated their contrast efficiency at several magnetic fields in preclinical and clinical MRI studies, both molecules have never been compared in a strict and robust way to assess their CEST contrast efficiency and pH dependence for elucidating their contrast capabilities at two clinical magnetics fields (3T and 7T). This study investigated the effect of pH, saturation power level (B1) and magnetic field strength (B0) on the MRI-CEST contrast generated by D-glucose [3] and 3-O-Methyl-D-glucose (3OMG) [4] with the aim of determining the in vivo detection limits of the two agents in the glucoCEST procedure in a murine melanoma tumor model after single bolus injection. Methods The experiments were conducted by comparing D-Glucose or 3OMG (20mM) in phosphate buffered solutions at different pH values on two preclinical MRI scanners working at 7T and and at 3T. In vivo GlucoCEST images were obtained after injection of Glucose or 3OMG (at two different doses of 1.5 and 3g/Kg) and with different administration routes (intravenous or oral) on subcutaneous B16-F10 melanoma bearing mice. The CEST signal was recorder over a period of 30 min after the injection; CEST contrast effects and percentage of tumor detection (fraction pixel) were compared between the two molecules. Results Glucose showed a marked pH dependent CEST contrast with higher ST% values at more acidic pH values (Figure 1a ; 7T, 37°C). Conversely, 3OMG CEST contrast reached higher ST% effect at neutral pH values (ca. 35% between 7.2and 6.8 pH, Figure 1b ) with a constant reduction towards acidic pH values (25% of ST effect at pH 6.0). At the lower magnetic field (Figure 1b and c) , glucose showed a slightly higher CEST detectability in comparison to 3OMG. The opposite pH dependence of the CEST response that was observed for D-Glucose and 3OMG are maintained independently of the applied B1 saturation power level. Surprisingly, the glucoCEST contrast enhancement in the tumour regions following intravenous administration (at the of dose 3 g/kg b.w) showed similar contrast between glucose and the non-metabolizable 3OMG (around 1.5-2% at 3T and 2-4% at 7T) (Figure1). Conclusion Despite a marked different pH dependence of the CEST contrast observed in vitro for glucose and 3OMG, when injected at the same dose by i.v. in tumor bearing mice a comparable contrast was observed. No clear advantages were observed for the non-metabolizable glucose derivative 3OMG at both ultra-high field (7T) and high-field (3T). Both molecules showed a prolonged CEST enhancement up to the first 30 min following the administration. This work was supported by the European Union's Horizon 2020 research and innovation programme 

A "smart" paraCEST agent for imaging pH James Soundrarajan, UT Southwestern Medical Center, james.ratnakar@utsouthwestern.edu Category: New Chemistry, Biology & Bioengineering Abstract Body : Paramagnetic lanthanide-based DOTA-tetraamide complexes provide contrast in MRI by chemical exchange saturation transfer (paraCEST).1 The CEST signal is often quite sensitive to the pH which makes them well suited for the measuring pH in vivo. The in vivo sensitivity reduces due to the T2 through a T2ex mechanism caused by the presence of a Ln3+bound inner sphere water.2 This effect is eliminated by having no Ln3+-bound inner sphere water, which is achieved by having three bulky coordinating methylphosphonate pendant arms (1) , and a Ln3+-bound-OH group for the paraCEST signal (Figure1a). The protonation of the non-coordinating phosphonate oxygen atoms shows pH sensitivity to the chemical shift. Yb3+ was chosen because it induces large paramagnetic shifts while minimizing line-broadening effects. The 1H NMR spectrum of Yb (1) shows 27 sharp resonances consistent with a ligand structure having C1 symmetry and a single twisted square antiprism (TSAP) isomer. The broad highly-shifted resonance at 133.6 ppm is assigned to the hydroxypropyl (-OH) proton on the fourth pendant arm which is well outside the typical tissue magnetization transfer (MT) window. The 31P NMR spectrum of Yb (1) shows three resonances, the chemical shift of each was highly pH dependent. A simultaneous fit of those data to a three-step equilibrium model gave three pKa values: log K1 = 7.0, log K2 = 5.8 and log K3 = 4.2. The CEST spectrum of the Yb(1) complex in water at 25°C, displays a single highly shifted CEST peak reflecting proton exchange between the Yb3+-coordinated -OH group and bulk water protons at different pH, and also shows the complex exists as a single coordination isomer in solution (Figure 1b) . CEST images of phantoms at different pH were collected and a false color pH map of each phantom was constructed (Figure 1c ). The pH value estimated from the frequency of the CEST peak showed good agreement with the solution pH measured using a glass electrode (Figure 1d ). The strong, sharp CEST signal of Yb(1) was not observed in kidney images during infusion of the agent but the characteristic CEST signal of Yb(1) appeared in images of the bladder only late during the imaging experiment. These preliminary studies show that Yb(1) complex is cleared by renal filtration but the process is much slower than expected for a molecule of this size. To examine this unexpected observation in more detail, the Gd(1) complex was prepared and dynamic T1weighted images verified that the agent clears only slowly from blood. Further investigations of the origin of this effect are in progress.

Kaiyuan Ni, Massachusetts Institute of Technology, kaiyuann@mit.edu Category: New Chemistry, Biology & Bioengineering Abstract Body : Checkpoint blockade immunotherapy (CBI) affords systemic and durable antitumor immunity by targeting T cell regulatory pathway, yet fails on immunosuppressive tumors with inadequate T cell infiltration. Radiotherapy as local immunomodulatory effects kills tumor cells in an immunogenic mode and alters tumor microenvironment to synergize with CBI. To maximize anti-tumor efficacy as well as immunoadjuvant effect of radiotherapy, intratumorally enriched high-Z radiosensitizers are clinically investigated to enlarge therapeutic index. We developed nanoscale metal-organic frameworks (nMOFs) as next-generation radiosensitizers. Compared to nanoparticle-based radiosensitizer such as HfO2, Hf-based nMOF effectively amplifies energy deposition with ordered structure and facilitates diffusion of reactive species with pores. By crystal engineering, we systemically compared nMOFs with different Hfoxo clusters to find Hf12-oxo a better X-ray absorber. By molecular engineering, we employed photosensitizing porphyrin-based linker to elicit a new therapeutic modality, radiotherapyradiodynamic therapy (RT-RDT). Upon low-dose X-ray, Hf12-oxo clusters not only absorb incident energy to generate hydroxyl radical via radiolysis but also scatter secondary energy to proximate linker to produce singlet oxygen. Combination of nMOF-mediated RT or RT-RDT with CBI not only eradicated primary tumor, but also extended local therapeutic efficacy to distant tumor via abscopal effect, suggesting a strong anti-tumor immunity induced by nMOF technology.

References: 1. Nanoscale metal-organic frameworks enhance radiotherapy to potentiate checkpoint blockade immunotherapy. Nature Communications 2018, 9 (1), 1-12. 2. Ultrathin metal-organic layer-mediated radiotherapy-radiodynamic therapy enhances immunotherapy of metastatic cancers. Matter 2019, 1 (5) Compared to solid NPs, lattice assembled with ultrasmall metal clusters amplifies secondary energy deposition and facilitates ROS generation and diffusion for better radiosensitization effect. By tuning metal-oxo clusters via crystal engineering, functionalizing organic bridging linkers via molecular engineering, and incorporating immunotherapeutics via immunoengineering, nMOF-mediated immunoadjuvant treatment amplifies local inflammation to potentiate systemic anti-tumor immunity when combined with immune checkpoint blockade. Abstract Body : Background & Motivation: Hyperpolarized (HP) 13C MRI enables in vivo quantitative dynamic imaging of enzyme-catalyzed cellular metabolism, with clinical applications in prostate, brain, heart, and kidney.1-4 However, the unique dual blood supply of the liver (via the hepatic artery and portal vein) complicates quantitative analyses of both the delivery of injected HP substrates and the timing of subsequent observed metabolism. Decomposing the pyruvate delivery into the contributions of each blood vessel would allow for improved real-time acquisition triggering and kinetic modeling accuracy by providing perfusion information. Here, we investigated the distinct vascular contributions in the human liver and assessed the effects on observed HP 13C signals in the liver. Methods & Results: Two healthy subjects (Healthy #1 and #2), one subject with metastatic pancreatic cancer (Cancer #1), and one subject with metastatic breast cancer (Cancer #2) underwent HP 13C MRI on a clinical 3 T MR scanner with a 1-, 8-, or 16-channel surface coil/clamshell transmitter combination with an injection of 0.43 mL/kg of 250 mM HP [1-13C] pyruvate. Axial T1-weighted spoiled gradientecho anatomical scans were acquired for blood vessel registration. [1-13C] pyruvate, and downstream metabolites [1-13C] lactate, [1-13C] alanine, and [13C]bicarbonate signals were acquired every 3 s using a specialized multi-slice 13C echo planar imaging acquisition with constant flip angles (10-20° for pyruvate, 20-30° for lactate).2 Total scan time was 1 minute with in-plane voxels of 1.0x1.0 to 2.0x2.0 cm, slice thicknesses of 2.0-3.0 cm, and 3-5 slices. Dynamic curves were summed, interpolated, and smoothed using a moving average to obtain a total 13C signal dynamic curve for selected voxels in the aorta, the inferior vena cava (IVC), the portal vein (PV), healthy liver (or normal appearing in cancer subjects), and liver metastasis. The total 13C dynamic curves for subject Healthy #2 are shown in Figure 1a . The peak arrival times from the start of acquisition for each anatomical structure and subject are shown in Figure 1b . The cancer voxel for subject Cancer #2 is not displayed as the voxel was within an unintended saturation band due to the single channel surface coil's inhomogeneous excitation profile. Overall, we observed a differential arrival of [1-13C] pyruvate signal, first in the aorta followed by the IVC, PV, and lastly the healthy liver and liver metastasis, matching physiologic expectations as the IVC rapidly receives venous return from the kidneys while the PV is delayed as it receives blood that passes first through the GI tract. The healthy subjects showed a consistent peak arrival time relative to the aorta in the PV (10-12 s) and healthy liver (14-17 s). A different pattern was observed in both cancer subjects; a shorter peak arrival time relative to the aorta was observed in the PV (3.4-4.5 s) and normal appearing liver (4.5-6 s). These consistencies within subject groups suggest that real-time triggering would be advantageous for future scans. Figure 1c shows the lactate/pyruvate ratios for each anatomical structure and subject except for subject Cancer #2 because of poor lactate SNR across all voxels. On the time scale of our experiment, lactate in the IVC predominantly reflects that produced by

Oluwatosin Ibhagui, Georgia State Uuniversity, oodubade1@student.gsu.edu Category: Immunology: Inflammation & Infection Abstract Body : Fibrosis; the scarring and thickening of tissue resulting from organ injury is characterized by overexpression of deposition of Extracellular matrix proteins such as collagen and accounts for an uprising burden of mortality and morbidity worldwide. Liver (liver fibrosis) and lung (idiopathic pulmonary fibrosis (IPF) and chronic obstructive pulmonary disease (COPD)) are among the more common site for fibrotic diseases with tendency to progress to liver and lung failure thereby necessitating the need for transplant. Despite major advancement in clinical research, Early diagnosis and precise noninvasive detection of these diseases remain a critical unmet medical need. In this study, we report the development and optimization of our novel collagen-targeted human protein-based magnetic resonance imaging contrast agent (hProCA32.collagen) in early diagnosis, progression and monitoring of fibrosis in liver and lung. hProCA32.collagen exhibits strong affinity to collagen 1 of Kd of 1uM. It has high relativities per particle (r 1 and r 2 ) at both 1.4 and 7.0 T that enable us to perform dual imaging for diffusive fibrosis detection. It also possesses strong metal stability, serum stability and unprecedented metal selectivity for Gd3+ over physiological metal ions. We then report the detection of liver fibrosis induced by non-alcoholic fatty liver disease and more aggressive model utilizing human Comparative Gene Identification-58 knock out (CGI-58 KO) mouse liver fibrosis mouse model for both early and late stage (Ishak 3 and 5 of 6). In addition, we further report the quantitative detection of early fibrosis in a diabetic mouse model by ProCA32.collagen and pMRI. We report distinct vascular architecture changes associated with portal hypertension and intrahepatic angiogenesis during late-stage fibrosis and molecular signatures of different liver fibrosis. We further report the detection of lung fibrosis in our developed mouse model by ProCA32.collagen 1 and pMRI. Pathological analysis verified our noninvasive detection and correlation of histopathological and radiological pattern of usual interstitial pneumonia (UIP) characterized by honeycombing and heterogeneous architecture and collagen accumulation in the lungs. Furthermore, we have achieved the first MRI detection of lung fibrosis of Scnn1b-Tg+ mouse upon exposure to E-cigarettes. We show that the pattern observed for IPF both in histology and MR imaging is markedly different from e-cigarette induced lung fibrosing for COP-disease. We anticipate hProCA32.collagen having strong clinical translational potential in facilitating effective treatment to prevent disease progression as well as reveal key factors that contribute to organ fibrosis including emerging public health issue associated acute viral infection such as COVID-19.

Connor Barth, Oregon Health & Science University, connorwbarth@gmail.com

Abstract Body : Nerve damage plagues surgical outcomes, significantly affecting post-surgical quality of life. Intraoperative nerve detection is completed using anatomical knowledge and conventional white light visualization when possible. However, neuroanatomy is varilable between patients especially in injured and diseased states that would be present during surgery. Additionally, nerves are typically protected deep within the tissue, so that white light visualization is largely impossible. Fluorescence image-guided surgery offers a potential means for enhanced intraoperative nerve identification and preservation. To date, a variety of nerve specific fluorophores have been tested in preclinical models, however a clinically approved nerve-specific contrast agent does not yet exist. Synthesis and functional validation of a nervespecific library has resulted in the development of near infrared (NIR) nerve-specific fluorophores for translation to first-in-human studies using the clinical infrastructure of NIR fluorescence guided surgery (FGS) systems.1 Further synthetic tuning of the most promising fluorophores has generated improved brightness, solubility, and safety. Increased brightness levels have resulted in 2-3x increase in nerve signal-to-background ratio (SBR). Improved solubility has enable water-soluble nerve-specific fluorophores with dose-limiting toxicity (DLT) values improved 5-10x. Single-dose toxicity testing including blood marker analysis and 14-day monitoring has facilitated definition of the no observable adverse effect level (NOAEL) and informed plans for upcoming investigational new drug (IND) enabling studies. Pharmacodynamics studies have been completed to identify the optimal imaging time point and dose, where nerve contrast is generated within 15 minutes and remains present for up to 4 hours following systemic administration. Clinically relevant, minimally invasive surgical procedures guided by NIR nerve-specific fluorescence have been completed in large animal models using the da Vinci surgical robot, demonstrating efficacy in surgically relevant models using existing clinical FGS system intrastructure. Given the interplay between recent synthetic tuning efforts and safety and pharmacology testing, we anticipate translation to the clinic within the next two years. Abstract Body : Objective: Patients with differentiated thyroid cancer of the same risk stratification have very different initial treatment effects, especially for high-risk patients. This study aims to explore the influencing factors of high-risk DTC patients after initial treatment, in order to better guide clinical, Provide a basis for formulating individualized treatment and follow-up strategies. Methods: A retrospective analysis of DTC patients who were present in our hospital from January 2015 to January 2018 was performed to exclude patients with positive TgAb and incomplete medical records. A total of 217 high-risk patients were included in this study, including 61 males and 156 females. Cases; average age (44.81±12.13) years old. According to the criteria of the 2015ATA guidelines for the classification of efficacy, the patients were divided into excellent response group (ER group), uncertain response group (IDR group), and biochemical insufficiency response group (BIR group) according to the data measured under the stimulation state of the patient half a year after surgery. The structural incomplete response group (SIR group) used the c2 test or Fisher's exact probability method to compare the differences between the four treatment groups of gender, tumor multifocality, TNM stage, and N stage; the Kruskal-Wallis rank sum test was used to compare the age and the largest tumor The difference in diameter, stimulating Tg (ps-Tg) level before iodine treatment and postoperative iodine treatment dose among the 4 therapeutic groups. The receiver operating characteristic curve (ROC curve) of the relationship between ps-Tg and tumor diameter and ER was established. Use ordered logistics regression to analyze the independent influencing factors of patients' therapeutic efficacy. Results: There were 92 patients (42.4%) in the ER group, 61 patients (28.1%) in the IDR group, 38 patients (17.5%) in the BIR group, and 26 patients (12.0%) in the SIR group. Univariate analysis showed: gender (c2 = 13.426, P = 0.005), tumor diameter (H = 21.368, P = 0.000), tumor multifocality (c2 = 15.510, P＜0.001), N stage (c2 = 46.819, P＜0.001), Ps-Tg level (H=3597.206, P＜0.001) significantly affects the classification of patients' efficacy after initial treatment. The difference between ps-Tg and tumor diameter cutoff values obtained by ROC curve for ER and non-ER are 9.92ng/ml and 1.45cm, the area under the curve (AUC) is 0.876 and 0.654, and the sensitivities are 69.5% and 61.1%, respectively. The specificity was 95.1% and 66.7%, the negative predictive value was 70.2% and 60.1%, and the positive predictive value was 91.5% and 60.1%, respectively. Taking the excellent response group as a reference, ordered multivariate regression analysis showed that N1b stage, ps-Tg level, and tumor diameter are independent factors that affect the therapeutic effect. Patients with ps-Tg>9.92ng/ml had a 36.307 times worse degree of efficacy than patients with ps-Tg≤9.92ng/ml; patients with N1b stage had a poorer degree of efficacy 2.210 times that of patients with N0 stage; tumors with diameters >1.45cm The degree of poor efficacy of the patient was 1.978 times that of patients with tumor diameter ≤1.45cm. Conclusion: High-risk patients still have nearly half of the patients who can achieve excellent response after six months of initial treatment. Patients with N0 staging, ps-Tg < 9 .92ng/ml, and tumor diameter < 1 .45cm have better curative effect after initial treatment. For this part of patients, 131I treatment dose and follow-up intensity can be appropriately reduced. Patients with ≥9.92ng/ml and cancer lesion diameter ≥1.45cm should appropriately increase the treatment dose of 131I and the intensity of TSH inhibition and increase the intensity of follow-up to detect metastatic lesions as early as possible. , concerns have been raised regarding the assessment of these injuries in discharged patients. We aim to identify residual lesions throughout the whole body by 18F-FDG PET/CT in these subjects. Methods: Patients who were discharged for COVID-19 at our hospital over a period of one week from 5th Feb to 12th Feb 2020, were recruited. Diagnosis and treatment protocol for COVID-19 were followed according to the WHO interim guidance [1] and the guidance from China [2] . The lung function tests were performed one month after discharge, and PET/CT scans were carried out two to three months after discharge. Patients had negative nucleic acid test for SARS-CoV-2 within three days before PET/CT scans. A total of twenty-five organs, including digestive system, urinary system, reproductive system, and cardiovascular system, were included in our study. To reflect the SUVmax of each organ more objectively, we randomly obtained five SUVmaxe values from each organ, and the mean of the biggest three SUVmaxe was recorded for this organ. In addition, if a definite or suspicious lesion can be seen in an organ, SUVmax values were mainly collected from there. The definitions of SUVmax thresholds in normal organs were followed the criteria reported by Tan et al. [3] and Engel et al. [4] , above which the organ was considered abnormal uptake. Results: Seven patients were recruited, which included four cases of severe and three of mild pneumonia. The baseline clinical characteristics between severe and mild groups were well balanced. The mean time from discharge to the date of lung function test was 36.4 days, and the mean time from discharge to the date of PET/CT scan was 59.0 days. In the follow-up examinations, only one patient suffered an increased 18F-FDG uptake of lower lobe of bilateral lungs (SUVmax, 2.37± 0.06 [L] and 2.41± 0.07 [R]; Figure 1 ) and a slightly impaired of pulmonary function (i.e. FVC [75.7%], DLCO [68%], and DLCO/VA [76%]), which recovered normally three months after discharge. In the other six patients, both the lung function tests and the 18F-FDG uptake of bilateral lungs were normal (Figure 1 ). Further analysis revealed that no significant difference was observed in 18F-FDG uptake and lung function between severe and mild groups (p > 0.05, Table 1 ). Moreover, no increase of 18F-FDG uptake was observed in any organs tested, such as stomach, colorectum, myocardium, liver, kidney, spleen, pancreas, testis, prostate, bladder, and so on, for all patients. Overall, there were no significant difference in 18F-FDG uptake among the lungs and extrapulmonary organs between severe and mild group (p > 0.05, Table 1 ). Conclusion: Lung function and 18F-FDG uptake can be quickly recovered two to three months after discharge among COVID-19 survivors regardless of pneumonia severity. Our findings will provide new insights into the recovery of recent functional injury of various organs in the whole body among COVID-19 survivors, which may not only help to relieve patients' panic of pneumonia-related organs injury, but also could help physicians to formulate an optimized follow-up healthcare strategy.

Micron scale direct conversion X-ray detector for phase contrast X-ray imaging in biomedical applications Karim Karim, KA Imaging, kkarim@kaimaging.com

Abstract Body : When conventional x-ray radiography presents inadequate absorption-contrast, higher sensitivity can be achieved using phase-contrast methods. The implementation of phasecontrast x-ray imaging using propagation-based techniques requires stringent spatial resolution requirements that necessitate lengthy propagation distances and inefficient scintillator-based detectors. Thus, imaging throughput is limited, and the absorbed dose by the sample can be unacceptable for radiation sensitive life science and biomedical applications. This work develops a hybrid direct X-ray conversion amorphous selenium and complementary metal-oxidesemiconductor detector technology that offers a unique combination of high spatial resolution and quantum efficiency for hard x-rays. A semiconductor fabrication process was developed for large area compatible vertical detector integration by back-end processing. Characterization of signal and noise performance using Fourier-based methods was performed by modulation transfer function, noise power spectrum, and detective quantum efficiency experiments using radiography and microfocus x-ray sources. The measured spatial resolution at each stage of detector development was one of the highest, if not the highest reported for hard x-rays. In fact, charge carrier spreading from x-ray interactions with amorphous selenium was shown physically larger than the pixel pitch for the first time. A simultaneous factor of three improvement in quantum efficiency was achieved compared to scintillator-based detectors, despite the detector being a relatively unoptimized prototype. Lastly, fast propagation-based phase-contrast x-ray imaging in compact geometries is demonstrated using a conventional low power microfocus source and the phase-contrast technique was applied to a number of biological samples including mouse organs, seeds, and insects. The imaging results from this research suggest that hybrid semiconductor technology offers the potential to fill the large performance deficit in high spatial resolution scintillator-based detectors for phase-contrast X-ray imaging and to even enable high speed dynamic phase contrast X-ray imaging for in situ characterization of biological samples.

Allison Solanki, OHSU, solanki@ohsu.edu

Abstract Body : In the era of targeted cancer therapy, genetic screening typically motivates treatment recommendations; however, gene profiling doesn't guarantee initial, or lasting, therapeutic efficacy. To ensure a drug is effective, it is important to consider that a therapeutic must first engage with its target(s) in the complex disease setting, where efficacious therapy is dictated by the duration, completeness and heterogeneity of drug target engagement (DTE)1, 2. Indeed, quantification of local cellular drug distribution should play a key role in assessing and predicting patients' treatment success; yet despite being an integral piece of the therapeutic puzzle, such analysis is not routine. Instead, standard pharmacokinetic or biochemical drug screening tools are typically bulk in nature (e.g., plasma analysis, western blot)3 and inherently limited in their utility for understanding heterogeneous diseases and mechanisms of initial or acquired therapy resistance. In response to this challenge, we have developed a novel imaging platform to quantify drug target engagement termed intracellular paired agent imaging (iPAI)4, 5. PAI utilizes untargeted imaging agents to correct for nonspecific uptake of spectrally distinct targeted agents, providing quantitative assessment of receptor density. PAI has recently been extended to small molecule therapeutics using fluorescently-labeled targeted and untargeted drug derivatives, including kinase inhibitors (KIs) in the epidermal growth factor receptor (EGFR) signaling pathway. Notably, although this technique relies on fluorescently-labeled drugs for quantification, all treatment is completed with the parent drug, classifying iPAI as a label-free method permitting assessment of the interaction between the parent drug and its native target. In preliminary work, we have designed, synthesized and performed in vitro validation on an iPAI reagent toolbox containing the EGFR KI erlotinib (Erl), in addition to downstream EGFR pathway-targeted drugs: buparlisib (PI3K), selumetinib (MEK1/2, ERK1/2), MK2206 (AKT1/2/3), and LY3214996 (ERK1/2). Evaluation of this suite of drugs is particularly valuable, since it is largely understood that monotherapy will not provide a durable, long-term response, but instead could create an actionable window to exploit collateral sensitivities to other drugs generated by therapeutically-induced selective pressure within the existing or parallel pathways. Targeted and untargeted behavior of each iPAI pair was quantified and evaluated using a suite of four functional assays, including (1) competitive binding with the parent drug, (2) fluorescence imaging vs. the parent fluorophore and vs. target-specific antibodies, which were compared to (3) quantitative flow cytometry to enumerate protein expression per cell and (4) tandem liquid chromatography mass spectrometry (LCMS) measured uptake and stability. Initial screening on positively and negatively protein expressing cell lines showed substantial non-specific accumulation of the targeted iPAI agent that was recapitulated by the untargeted iPAI agent. Calculated ratiometric DTE images (where targeted signal is subtracted from the untargeted signal, and divided by the untargeted signal) revealed an expression pattern that was spatially matched to antibody staining, representing the number of drug targets available for binding.

Minocycline treatment during adolescence prevents structural and oxidative stress brain abnormalities at adulthood: a functional and structural imaging study in the Poly I:C rat model of schizophrenia Diego Romero-Miguel, Fundación para la Investigación Biomédica del Hospital Gregorio Marañón Biomedical Imaging and Instrumentation Group (BiiG), ldromero@hggm.es Category: Neuroscience Abstract Body : Introduction: Minocycline (MIN) is a tetracycline with antioxidant, antiinflamatory and neuroprotective properties. Given the likely involvement of inflammation and oxidative stress (IOS) in schizophrenia, MIN has been proposed as a potential adjuvant treatment in this pathology 1, 2 . Thus, we aim to evaluate whether MIN treatment during adolescence in the PolyI:C maternal immune stimulation (MIS) animal model of schizophrenia could prevent the brain physiological and behavioral deficits described at adulthood. To our knowledge, this is the first work to tackle this topic using positron emission tomography (PET) with 2-deoxy-2- [18F] fluoro-D-glucose (FDG), together with MRI voxel-based morphometry (VBM), adapted to rodents. We also evaluated improvements in IOS and behavior deficits as outcomes. Methods: In gestational day 15, PolyI:C (4 mg/kg) or saline (VH) were injected to pregnant Wistar rats. 93 male offspring received MIN (30 mg/kg) or saline (Sal) daily from postnatal day (PND) 35-49. Four groups were evaluated attending to the study factors: MIS condition (Saline, MIS) and treatment (VH, MIN). At PND70, rats were submitted to prepulse inhibition test (PPI). FDG-PET and T2-weighted MRI brain studies were performed at adulthood (PND120) and analyzed by means of SPM12. IOS markers (iNOS, COX2, GPx, CAT, SOD, MDA, NRF2, KEAP1, HO1, NQO1, GSH free , GSH total and GSSG) were evaluated by western blot, enzymatic activity and concentration assays in frozen tissue of prefrontal cortex, hippocampus, caudate-putamen, and amygdala. Results: MIS-offspring showed PPI deficits compared with VH-offspring but MIN treatment did not fully prevent this behavioral deficit. In addition, MIN prevented the volumetric abnormalities in the third ventricle but not in the hippocampus in MIS-offspring. Also, MIN reduced brain metabolism in the cerebellum and increased FDG uptake in the nucleus accumbens. Besides, MIN reduced the expression of iNOS in PFC and caudate putamen and increased the activity/concentration of KEAP1 in PFC, HO1 and NQO1 in amygdala, HO1 in caudate-putamen and HO1 and NQO1 in hippocampus. Conclusions: Our study demonstrates that a MIN treatment during adolescence partially counteracts volumetric abnormalities and IOS deficits via iNOS and Nrf2-ARE pathways in the MIS model, increasing the expression of cytoprotective enzymes. However, MIN treatment during this peripubertal stage does not prevent sensorimotor gating deficit. Therefore, despite not preventing all the pathological levels assessed, MIN effectivity highlights the usefulness of anti-IOS compounds to halt the disease course at early stages. Acknowledgements: This study was supported by the Ministerio de Ciencia, Innovación y Universidades, ISCIII (PI14/00860, PI18/01691, PI17/01766, CPII14/00005) and Consejería de Educación e Investigación, Comunidad de Madrid (grants PEJD-2018-PRE/BMD-7899, PEJ-2017-TL/BMD-7385) cofinanced by European Regional Development Fund (ERDF), "A way of making Europe",

Cynthia Schreiber, University of Notre Dame, cschrei2@nd.edu Category: New Chemistry, Biology & Bioengineering Abstract Body : The near-infrared window of fluorescent heptamethine cyanine dyes greatly facilitates biological imaging because there is deep penetration of the light and negligible background fluorescence. But dye instability, aggregation, and poor pharmacokinetics are current drawbacks that limit performance and the scope of possible applications. All these limitations are simultaneously overcome with a new molecular design strategy that produces a charge balanced and sterically shielded fluorochrome. The key design feature is a meso-Aryl group that simultaneously projects two shielding arms directly over each face of a linear heptamethine polyene. Cell and mouse imaging experiments compared a shielded heptamethine cyanine dye (and several peptide and antibody bioconjugates) to benchmark heptamethine dyes and found that the shielded systems possess an unsurpassed combination of photophysical, physiochemical and biodistribution properties that greatly enhance bioimaging performance.

Image/Figure Caption: Just like a superhero, an ultrastable shielded heptamethine cyanine dye uses its two strong arms to ward off self-aggregation and non-specific biological interactions. Yet the arms are short enough to allow dye-labeled bioconjugates to selectively target cell receptors for high-contrast and photon-intense microscopy or tumor imaging in living subjects. Abstract Body : Introduction MRI-CEST pH imaging has shown a great utility in characterizing tumor acidosis and in assessing tumor pH heterogeneity [1] . However, when pushing the spatial or temporal resolution of the CEST-MRI images, the acquired low signal-to-noise ratio (SNR) images may significantly affect the accuracy of the tumor pH quantification. This work aims to investigate a new denoising approach for MRI-CEST images in simulated and real situations, to compare with different denoising methods and to validate in vivo its robustness to improve CEST contrast quantification and pH measurements. Methods Synthetic CEST data for the Iopamidol contrast agent were contaminated with two different levels of Rician noise. In vitro data were imaged with a Bruker 7T MRI scanner for Iopamidol solution titrated at different concentrations and pH values. Moreover, in vivo data were acquired upon iopamidol injection in a prostate orthotopic tumor murine model and the CEST contrast was measured as the difference between pre-and post-injection images and exploited for pH calculation. A novel denoising technique based on the combination between Non-Local Mean and Anisotropic diffusion Tensor (NLmCED) [2] was evaluated on simulated data corrupted by noise and validated by comparing with the ground truth. Further evaluation was performed with both in vitro and in vivo data in order to improve the CEST contrast quantification and pH measurements and the robustness and accuracy of the proposed approach was compared with conventional denoising methods: Gaussian filter and Cubic Smoothing Splines [3] . Results/Discussion As quantitative measures, both PSNR and the similarity between the reference and the denoised image (SSIM index) were calculated on the denoised synthetic data at two different level of noise ( Fig.1 a-b) . The NLmCED method exceed the other techniques by providing highest quantitative measures for the two pools of iopamidol. For in vitro data, comparable performance of the NLmCED was obtained at different iopamidol concentrations and pH values ( Fig.1 c-d) . The obtained results for denoising in vivo data validated the performance of the NLmCED algorithm which succeeded to increase the fraction of pixels in the region of interest (ROI) (Fig.1 i-k) by surpassing both the Gaussian filter and the Cubic Smoothing Spline method, improving both the contrast detection and pH measurement ( Fig.1 e-h) . Conclusion The proposed denoising method is expected to have a great utility for robust and accurate tumor pH imaging and quantification with the MRI-CEST technique following Iopamidol injection. Acknowledgement This work was supported by grants from the Associazione italiana Ricerca Cancro (AIRC MFAG #20153).

Janet Eary, National Cancer Institute, janet.eary@nih.gov Category: Immunology: Inflammation & Infection Abstract Body : Inflammation is a normal process in our body; acute inflammation acts to suppress infections and support wound healing, while chronic inflammation can lead to cancer. Chronic inflammation often does not exhibit the classic signs and symptoms of inflammation and is thus often neglected until a disease like cancer is clinically apparent. The biological understanding of the transition from acute to chronic inflammation is important as it can provide an opportunity for clinical intervention and in monitoring disease progression. Current clinical evaluation for inflammatory status in an individual patient relies on immunological tests of biological fluids and pathological analyses of biopsied tissue samples. Molecular imaging techniques that can track the changes of molecular signatures, immune cell trafficking, metabolic, and functional parameters are essential for obtaining a better understanding of the role that inflammation plays in the development and progression of cancer. It can also help us understand the biology of cancer development and progression, while providing insights into new targets for cancer treatment. NCI is interested in providing a pathway for adoption of such molecular imaging techniques for the understanding of cancer inflammation. Abstract Body : Electro-cardiac mapping is a smart technique to visualize the cardiac muscle functionality and status of heart function. In case of incurable or unmanageable conditions of ischemic heart disease or heart failure, stem cell therapy is good option as optimal delivery of stem cells to revive the cardiac viability. Stem cell delivery is given as intramyocardial, epicardial and intracoronary injections. In present time, intramyocardial NOGA system is popular choice. The NOGA system delineates the cardiac target sites for stem cell therapy and NOGA system navigates the cardiac recovery and its rate of deployment. The system visualizes 3D heart endocardial surface using spatial, mechanical and electrophysilogical information of major left ventricle apex region undergoing necrosis, ischemia spreading over normal surface from base to apex, inferior to anterior, septal to lateral border zone regions (see the Figure) . The NOGA system distinguishes infarct regions (low linear local shortening and low unipolar voltage) from ischemic regions(low linear local shortening and preserved unipolar voltage). The approaches of catherization in NOGA system are transfemoral and transradial guidewires. Still investigations and clinical trials are in progress. In future, clinical success will decide the safe application of NOGA system.

M. Zubair Iqbal, Zhejiang Sci-Tech University, zubair@zstu.edu.cn

Abstract Body : Abstract Magnetic resonance imaging (MRI), a sophisticated promising threedimensional tomographic noninvasive diagnostic technique, has intrinsic advantage in safety when compared with radiotracer and optical imaging modalities. However, MRI contrast agents are less sensitive than complexes used in other imaging techniques and toxicity issue still endures in nanoparticle-based MRI-T1 contrast agents. Therefore, demand for nontoxic novel (T1&T2) T1-weighted MRI potential candidate with ultrasensitive imaging and advanced functionality is very high. In this research, silica coated ultra small monodispersed superparamagnetic iron oxide nanoparticles were synthesized via thermal decomposition which demonstrated high performance T1-weighted MRI contrast agent for heart, liver, kidney and bladder. Advanced characterization techniques were used to investigate the crystal structure, morphological evaluation, concentration of iron, size distribution, active modes and magnetization of as-synthesized nanoparticles. Transmission electron microscopy (TEM) results have illustrated that the diameter of SPIONPs was in the range of 4nm and the average size of Fe3O4@SiO2 was about 30nm~40nm. X-ray diffraction (XRD) and Raman spectroscopy analyses revealed the purity in phase of the prepared SPIONPs. These magnetite nanoparticles exhibited weak magnetic moment at room temperature because of spin-canting effect which escorted high positive signal enhancement ability. MCF-7 and HeLa cell viabilities experiments demonstrated good biocompatibility of the SPIONPs. In addition, the first ever study has been made on T1 MRI contrast agent of silica coated ultra small (4nm-sized) magnetite nanoparticles exhibited a good r1 relaxivity of 1.2 and low r2/r1 ratio of 6.5, attributed to low magnetization, large surface area and 5 unpaired valence electrons on the surface of Fe3+. In vivo T1-weighted MR imaging of heart, liver, kidney and bladder after intravenous injection of nanoparticles further verified the high sensitivity and biocompatibility of as-synthesized magnetite nanoparticles. These results reveal silica coated SPIONPs as a promising candidate for T1 contrast agent with extraordinary capability to enhance MR images.

In vivo MRI-CEST tumor pH imaging can detect early resistance to proton pump inhibitors in human prostate cancer murine models Pietro Irrera, Moffitt Cancer Center, pietro.irrera@moffitt.org

Abstract Body : INTRODUCTION Tumor lesions have common characteristics among different type of cancer like genetic mutations, altered metabolism, hypoxia and extracellular acidosis. Because survival in the tumor microenvironment depends on pH homeostasis, interference with pH regulating systems is considered a new therapeutic strategy. Therefore, novel imaging approaches are urgently needed for non-invasively assessing the efficacy of novel anticancer therapies. Recent pre-clinical investigations reported promising results with proton pump inhibitors (PPIs) treatments in tumor therapy, both individually and combined with a chemotherapeutic drug [1] [2] [3] . Aim of this work was to evaluate MRI-CEST (chemical exchange saturation transfer) tumor pH imaging to assess treatment response to esomeprazole, a proton pump inhibitor (PPI) and to investigate the efficacy of this new therapeutic strategy in vivo. MATERIALS AND METHOD Cell viability and pH measurements were performed in vitro for the following drugs: lansoprazole, esomeprazole (V-ATPases), cariporide and amiloride (NHE1). Expression of V-ATP and NHE1 was quantified by RT-PCR and WB in vitro. For in vivo experiments, athymic nude mice were inoculated subcutaneously with human prostate PC3 or DU145 cells and treated with esomeprazole (dose: 2.5 mg/kg) for two weeks and tumor growth was monitored. A fast-multi-slice CEST sequence was used for pH imaging following iopamidol injection. MRI-CEST tumor pH imaging and FDG-PET were perfomed one week after treatment for assessing early treatment response by measuring tumor acidosis and glucose uptake. RESULTS AND DISCUSSION Only esomeprazole and amiloride treatments reported a dosedependent and marked alkalinisation of the extracellular environment together with a dosedependent cell death in in vitro studies. Despite a potent effect of esomeprazole in vitro, in vivo studies showed no reduction in tumor growth for both the two prostate tumor murine models after two weeks of treatment ( Fig.1 a,c) . MRI pH imaging detected similar baseline tumor pH acidosis between PC3 and DU145 models. Interestingly, as early as one week after treatment, tumor pH imaging demonstrated no differences in tumor acidosis between treated and control mice (Fig.1 b,d) . FDG-PET imaging confirmed the observed findings showing no differences in the glucose uptake levels for PC3 mice between treated and untreated groups ( Fig.1 e) . CONCLUSION Esomeprazole provided a marked pH alteration in vitro in both human prostate cell lines but it was not effective in reducing tumor growth rate in vivo. MRI-CEST tumor pH imaging showed, as early as one week after the treatment, no effects in tumor pH values for both the prostate murine models. Therefore, this approach can be used to detect upfront resistance to PPIs for validating novel therapeutic strategies that are effective only in in vitro conditions but not in vivo. Acknowledgements: This work was supported by the Compagnia San Paolo (Regione Piemonte, project grant #CSTO165925) and by Associazione Italiana Ricerca Cancro (AIRC MFAG #20153) ID: GA573 Synthesizing and functionalizing of X-ray luminescence nanoparticles and studying x-ray stimulated behavior of Caenorhabditis elegans for optogenetic studies Meenakshi Ranasinghe, Clemson University, mranasi@g.clemson.edu

Abstract Body : Rare-earth doped phosphors have interesting optical properties depending on their size, crystal structure and type of dopant. They are used in bioimaging, chemical sensing and radiology imaging screens. We are interested in synthesizing x-ray excited nanophosphors for a non-invasive, in vivo optogenetic neuron simulation method. However, it is challenging to synthesize monodisperse nano-size particles that emit bright visible light photons when irradiated with X-ray. We are synthesizing Eu-doped NaGdF4 (NaGdF4: Eu) nanoparticles using a citrate method which involves lanthanides-citrate complex formation followed by nucleation and growth upon the addition of NaF. Then, the nanoparticles are annealed at high temperature (up to 1000 °C) to increase the emission intensity by removing defect sites that can quench luminescence. To prevent nanoparticle sintering during annealing we encapsulated NaGdF4: Eu nanoparticle in a silica shell. The synthesized NaGdF4: Eu nanoparticles are characterized using dynamic light scattering, powder x-ray diffraction, transmission electron microscopy and x-ray luminescence spectroscopy. The TEM images confirm that the size of the nanoparticles is around 100 nm. According to the x-ray luminescence spectroscopy measurements, at low and high dopant levels, they show low emission intensity, likely due to lack of luminescent centers and self-quenching, respectively. Hydrothermal treatment increased the emission intensity by a factor of 2-3 and annealing without a silica shell increased the emission intensity by another factor of 5 which resulted in sintered particles, negating their bio-application. However, this may have potential to form thin films which use in solar cells and fluorescence screens. Formation of silica shell resulted in a decrease in emission intensity due to boundary defects and small molecules (H2O and CO2) trap in the shell. After annealing silica-coated NaGdF4: Eu, there is no large increase in intensity compared to pristine nanoparticles. Annealing at 350-450 °C resulted in porous NaGdF4: Eu coated with silica which may have potential drug delivery applications. At high temperatures, silica-coated NaGdF4: Eu change its structure to Eu doped sodium gadolinium silicate. We functionalized NaGdF4: Eu@SiO2 nanoparticles with Biotin and confirmed attaching to streptavidin by performing in vitro study using streptavidin-coated microbeads. The second part of the project is to study radiation-induced behavior of C. elegans. They are widely used as unique model-organism in neuronal networks, cancer studies, aging studying, etc. Here, we are interested in the radiation-induced locomotory behavior of C. elegans. Our study shows that they show x-ray induced locomotory movements and their LITE-1 photoreceptor is responsible for the observation. In the future, we plan to optimize the synthesis, annealing, and surface passivation protocols to obtain bright monodisperse nanophosphors. We will also functionalize them and test their application as in situ light sources for sensing and actuating in vivo using C. elegans is an animal model. Further, we will study the mechanism of x-ray induced behavior of C. elegans.

Xuandi Hou, The Hong Kong Polytechnic University, xuandi.hou@connect.polyu.hk

Abstract Body : Techniques for noninvasive and remote brain stimulation have long been desired for both probing brain functions and treating brain dysfunctions. Potential technologies candidate for brain stimulation include optical, magnetic, electric, and acoustic brain stimulation [1] [2] [3] [4] [5] . Of these, ultrasound brain stimulation is considered among the most promising technologies since ultrasound can non-invasively penetrate through the skull into deep brain structures with high spatiotemporal resolution [6] . Recently, ultrasound neuromodulation has gained increasing attention, although the spatial resolution of ultrasound is sufficient for regionspecific brain stimulation, it lacks cellular and molecular specificity. Here, we developed an ultrasonic mechanogenetic tool to manipulate the neuronal activity and signaling with excellent precision by introducing gas-filled nanostructures, gas vesicles (GVs). GVs have unique size which can induce highly localized ultrasonic pressure gradient, unique microstreaming and shear stress in low intensity ultrasound field [7, 8] . In this study, we customized low intensity low frequency pulsed ultrasound stimulation system and a high-speed fluorescence microscopy was employed for calcium imaging. Neuronal calcium concentration was monitored simultaneously as a readout under ultrasonic stimulation. We hypothesized that the GVs could oscillate and consequently trigger widespread and reversible firing of neurons when exposed to ultrasound field. We showed that GVs mediated ultrasound can directly activate cultured neurons and initiate calcium influx ( Figure 1 ). Further, the induced calcium response could be significantly suppressed by inhibiting the mechanosensitive ion channels with ruthenium red (RR), indicating that these ion channels mediated this kind of neuron activation. To manipulate target cells activities, we further sensitized the neurons by overexpression of MscL, a mechanosensitive ion channel. It showed that MscL-neurons could be activated under much lower ultrasound intensity and c-Fos expression significantly increased in the MscL-transfected cells, compared with the control. Furthermore, the parameters of ultrasound excitation we used do not affect cell viability or cell membrane integrity and temperature change. Altogether, our findings demonstrated that under low intensity ultrasound irradiation, local ultrasound pressure will be significantly intensified where the oscillating GVs are placed and this GVs mediated ultrasound system can activate mechanosensitive ion channels and control calcium release in neurons. Moreover, the activity of mechanosensitive ion channels such as MscL stimulated by ultrasound is an important contributor to its ability to stimulate specific cells in vitro.

Quanxiang Xian, The Hong Kong Polytechnic University, 18072003r@connect.polyu.hk

Abstract Body : Manipulating specific targeted neural activity by physical intervention is a powerful method to gain causal insight into brain functions and treat brains disorders. Ultrasound stimulation is a promising modality for probing brain function and treating brain diseases. It can be non-invasively steered and focused into mm-scale regions across the human skull, facilitated to produce controlled modulation of neuronal activity. However, it still remains challenging to stimulate well-defined neurons in a desired brain area without significant surgical invasion. Here, we introduce a non-invasive and selective neural excitation strategy using ultrasound, through the specific activation a mutant of the large-conductance mechanosensitive channel. Methods we used included ultrasound stimulation, cell culture, viral transduction, patch clamp, calcium imaging, virus injection, immunocytochemical staining, Electromyography (EMG). In the vitro experiment, we first tested Low-intensity ultrasound can stimulate Ca2+ influx in 293T cells expressing MscL-G22S. Next, we tested the effect of ultrasound on cortical neurons collected from the embryonic mouse brain (E16) to test the feasibility of selectively stimulating neurons by ultrasound. Our data indicate that the expression of MscL-G22S can significantly reduce the ultrasound intensity required to cause Ca2+ influx in cortical neurons. Then test our setup in vivo. Ultrasound stimulation (recognized by c-Fos expression) of the neuron-induced cortical area in the mouse brain, the data showed that ultrasound treatment induced significantly higher expression of c-Fos in MscL-expressing cortical neuron group compared to the non-ultrasound and control virus cortex treated with ultrasound groups. A similar comparison of sham-injected mice showed that there was no significant difference in c-Fos expression in the cortex of mice treated with ultrasound or untreated, and judging by body weight, there was no significant health difference in the groups of these mice. Therefore, we indicated that the application of ultrasound in the mouse cortex can induce neuronal activity, but the induced MscL expression makes this effect many times stronger. We also used EMG methods to test the effect of low-intensity ultrasound stimulation different parameters on motor cortex response characteristics. This experiment illustrated the effect of mechanosensitive channel (CamKII-EYFP mice and CamKII-mscl-EYFP mice) on the twitch response to sonication. The data showed that the successful rate in mscl group was significant higher than control group when the mice were treated with 0.05 MPa, 0.15 MPa, 0.30 MPa, 0.35 MPa, 0.5 ultrasound stimulation. The ultrasound stimulation parameters correspond to stimulation success rages in mscl group. We next tested the spatial selectivity of the ultrasound stimulation method by using the MscL-G22S virus to sensitize neurons in deeper regions of the brain, the dorsomedial striatum (DMS). Increased c-Fos expression was also observed in neurons expressing the control virus, however, c-Fos of MscL-EYFP neurons was also significantly higher than this group (MscL-EYFP 72.1 c-Fos+ cells/slices compared to 37.9 Ctrl-EYFP), which shows that MscL can effectively sensitize cells to ultrasound stimulation. Thus, we demonstrate an effective non-invasive approach for activating neurons in the intact brain using

Catharine Brady, University of Illinois at Urbana-Champaign, cjbrady2@illinois.edu Category: New Chemistry, Biology & Bioengineering Abstract Body : Reactive oxygen species (ROS), such as singlet oxygen (1O2) and superoxide (O2•-), are transient and chemically reactive molecules often generated in the mitochondria. At endogenous concentrations, ROS have been shown to play a role in cellular signaling and redox homeostasis, but increased levels are associated with oxidative stress, can cause downstream damage to biomolecules, and can result in cellular apoptosis and necrosis. Photodynamic therapy takes advantage of this reactivity to kill diseased cells through the use of photosensitizers that generate ROS from molecular oxygen upon light activation. While this therapeutic technique has shown potential, there are few clinically approved photosensitizers which demonstrate little structural diversity because it is difficult to predict photosensitizing ability based on chemical features. This highlights the need for new photosensitizer platforms with improved properties for in vivo applications. To address this, a novel near-infrared (NIR) small molecule photosensitizer with integrated photoacoustic imaging readout was developed, optimized, and evaluated. Various modifications to the hemicyanine dye (HD) scaffold were synthesized and assessed for their photosensitizing, photoacoustic, and general photochemical properties. It was demonstrated that a selenium-containing HD dye showed increased photosensitization capability, and red shifted absorbance and emission maxima. The same wavelength light for singlet oxygen generation was shown to be able to be used for photoacoustic imaging which can allow for in vivo tracking of the photosensitizer at centimeter depths. Further work will include the incorporation of specific disease targeting groups for to help eliminate off-target effects. Abstract Body : The preferred imaging modality for the detection of tumor extent and treatment response to neo-adjuvant and adjuvant medical therapies involves generation of high-resolution 3D anatomic and dynamic MRI images without exposing patients to ionizing radiation. We have established a series of robust, sensitive protein MRI contrast agents to longitudinally quantify and multi-dimensionally map the expression and distribution of diagnostic tumor biomarkers and tumor microenvironment biomarkers, including collagen1, CXCR4 and PDL-1. Here we report the optimization of production and manufacturing of multiple biomarker targeted contrast agents for collagen1, CXCR4, PDL-1, integrin, HER2, and EGFR, exhibiting high stability, reproducibility and scale up productivity. These developed biomarker targeted protein contrast agents exhibit 10-to 50-fold increases in both r1 and r2 relaxivities, compared to clinicallyapproved Gd3+ contrast agents, resulting in exceptional imaging capability that can discern heterogeneous tissue signals via a dual MR imaging methodology. These contrast agents have also demonstrated strong metal binding stability for Gd3+ and significant metal selectivity over physiological ions. The ability of these agents to bind biomarkers CXCR4, collagen, HER2, and PDL1, has also been optimized, and includes the capability to detect overexpressed biomarkers in TME of primary cancers and liver metastasis from pancreatic cancer PDAC, breast cancer, and uveal melanoma. We have further demonstrated their in vivo capability to visualize heterogeneous expression of biomarkers in TME using several mouse models. Our studies in precision imaging with unprecedented sensitivity and accuracy address major medical gaps, including early detection of small lesions and heterogeneity of multiple biomarker expression, especially for high risk patients, and offer means to monitor dynamic changes of biomarkers and the tumor environment, both during disease progression and following treatment with applications in image-guided interventions.

References: [1] Y. Fan, et al. Modulation of Intracellular Oxygen Pressure by Dual-Drug References: (1)

Effect of Behavioral Testing on Spine Density of Basal Dendrites in the CA1 Region of the Hippocampus Modulated

Analysis of White Blood Cell Counts in Mice after Gamma-or Proton-Radiation Exposure

Sorbitol dehydrogenase of Aspergillus niger, SdhA, is part of the oxido-reductive D-galactose pathway and essential for D-sorbitol catabolism

Fluorescent probes for sensing and imaging

New strategies for fluorescent probe design in medical diagnostic imaging

In vivo near-infrared fluorescence imaging

A Glowing Trajectory between Bio-and Chemiluminescence: From Luciferin-Based Probes to Triggerable Dioxetanes

Recent Advances and Challenges in Luminescent Imaging: Bright Outlook for Chemiluminescence of Dioxetanes in Water

Complete Status: Complete First Name: Peter Last Name: Clark Email: pclark@mednet.ucla.edu Organization: UCLA Country: United States References

Acidity generated by the tumor microenvironment drives local invasion

Immunological Effects of Conventional Chemotherapy and Targeted Anticancer Agents

A PET Imaging Strategy to Visualize Activated T Cells in Acute Graft-versus-Host Disease Elicited by Allogenic Hematopoietic Cell Transplant

Imaging of Activated T Cells as an Early Predictor of Immune Response to Anti-PD-1 Therapy

Complete Status: Complete First Name: Jelena Last Name: Levi Email: jlevi@cellsighttech.com Organization: CellSight Technology, Inc. Country: United States References: (1) Gabrilovich DI

Image/Figure: References: 1

First-in-Humans Imaging with (89)Zr-Df-IAB22M2C Anti-CD8 Minibody in Patients with Solid Malignancies: Preliminary Pharmacokinetics, Biodistribution, and Lesion Targeting

Full Name of Abstract's 1st Author : Veronica L. Nagle Complete Status: Complete First Name: Veronica Last Name: Nagle Email: naglev@mskcc.org Organization: Memorial Sloan Kettering Cancer Center Country: United States References: [1] Radiopharmaceutical Chemistry

DiPODS: A Reagent for Site-Specific Bioconjugation via the Irreversible Re-bridging of Disulfide Linkages' submitted manuscript, 2020. References: [1] B.Gleich and

Development and evaluation of a connective tissue phantom model for subsurface visualization of cancers requiring wide local excision

Significance of Phosphorylated Epidermal Growth Factor Receptor and Its Signal Transducers in Human Soft Tissue Sarcoma

Preclinical imaging of epidermal growth factor receptor with ABY-029 in soft-tissue sarcoma for fluorescence-guided surgery and tumor detection

Toxicity and Pharmacokinetic Profile for Single-Dose Injection of ABY-029: a Fluorescent Anti-EGFR Synthetic Affibody Molecule for Human Use

Group A Streptococcus intranasal infection promotes CNS infiltration by streptococcal-specific Th17 cells

GE-180 PET Detects Reduced Microglia Activation After LM11A-31 Therapy in a Mouse Model of Alzheimer's Disease

Initial studies with [11C]vorozole positron emission tomography detect overexpression of intra-tumoral aromatase in breast cancer

Aromatase Imaging with [N-methyl-C-11]vorozole PET in Healthy Men and Women

Resultant percent dense breast (PD) shown on right. C. Scattergram of PD values in the high aromatase (h) and low aromatase (l) groups. D

Early Phase I Study of a (99m)Tc-Labeled Anti-Programmed Death Ligand-1 (PD-L1) Single-Domain Antibody in SPECT/CT Assessment of PD-L1 Expression in Non-Small Cell Lung Cancer

Synthesis and Biologic Evaluation of a Novel (18)F-Labeled Adnectin as a PET Radioligand for Imaging PD-L1 Expression

PET/CT in patients with advanced stage non-small-cell lung cancer

Image/Figure: References: 1

Image/Figure Caption: (A) Chemical structure of

Effect of Dye and Conjugation Chemistry on the Biodistribution Profile of Near-Infrared-Labeled Nanobodies as Tracers for Image-Guided Surgery

Thrombectomy 6 to 24 Hours after Stroke with a Mismatch between Deficit and Infarct

Resolving Thromboinflammation in the Brain After Ischemic Stroke?

Updated Mechanisms of Action and Resistance in Breast Cancer. Frontiers in Oncology

Non-Invasive in Vivo Imaging of near Infrared-Labeled Transferrin in Breast Cancer Cells and Tumors Using Fluorescence Lifetime FRET

Tuning Nuclear Quadrupole Resonance: A Novel Approach for the Design of Frequency-Selective MRI Contrast Agents

Image/Figure Caption: Figure 1. Graphic presentation of the new class of QP-based contrast agents. Complete Status: Complete First Name: Simona Last Name: Baroni Email: simona.baroni@unito.it Organization: University of Turin Country: Italy References

Determination of [11C]PBR28 binding potential in vivo: A first human TSPO blocking study

Kinetic analysis and test-retest variability of the radioligand [11C](R)-PK11195 binding to TSPO in the human brain -a PET study in control subjects

Poster Communication: Preclinical assessment of 18F-LW223; a novel TSPO radiotracer for the detection of inflammation using PET

Magnetic Particle Imaging of Macrophages Associated with Cancer: Filling the Voids Left by Iron-Based Magnetic Resonance Imaging

Image/Figure Caption: MPI and MRI images of nanoparticle uptake in TAMs

Liver Fibrosis, but No Other Histologic Features, Is Associated With Long-term Outcomes of Patients With Nonalcoholic Fatty Liver Disease

Pharmacological inhibition of the vitronectin receptor abrogates PDGF-BB-induced hepatic stellate cell migration and activation in vitro

Radio-opaque ethylcellulose-ethanol is a safe and efficient sclerosing agent for venous malformations

A novel system for submillimeter-resolution tomography with radiolabeled molecules in mice

Performance evaluation of A-SPECT: A high resolution desktop pinhole SPECT system for imaging small animals

Proceedings of Singapore Healthcare

Alzheimer's Dement

Alzheimer's Dis

) Dekkers, I. A.; Roos, R.; Van Der Molen

Image/Figure Caption: Figure 1. MnLMe application at the clinical dose enhances MRI contrast in healthy mice. (A) Representative contrast-enhanced T1-weighted image in the kidneys, liver, heart, and bladder after application of MnLMe at a clinical dose (second panel) in comparison to pre-contrast imaging (first panel). Difference in T1-signal intensity

Zinc as an Imaging Biomarker of Prostate Cancer

Imaging Insulin Secretion from Mouse Pancreas by MRI Is Improved by Use of a Zinc-Responsive MRI Sensor with Lower Affinity for Zn2+ Ions

A manganese(II)-Based Responsive Contrast Agent Detects Glucose-Stimulated Zinc Secretion from the Mouse Pancreas and Prostate by MRI

Image/Figure Caption: Figure 1. Zinc detection mechanism with Mn-based contrast agent

Complete Status: Complete First Name: Andre Last Name: Martins Email: andre.martins@med.uni-tuebingen.de Organization: Werner Siemens Imaging Center Country: Germany References

Imaging cell surface glycosylation in vivo using 'double click' chemistry

Imaging vascular physiology to monitor cancer treatment

Dynamic contrastenhanced imaging techniques: CT and MRI

A lightfluence-independent method for the quantitative analysis of dynamic contrast-enhanced multispectral optoacoustic tomography

Image/Figure Caption: Full Name of Abstract's 1st Author : Shreya Goel Complete Status: Complete First Name: Shreya Last Name: Goel Email: shreya.goel.shreya@gmail.com Organization: The University of Texas MD Anderson Cancer Center Country: United States References

Proceedings of the National Academy of Sciences

Image/Figure Caption: A) Diagram of radiopharmaceutical

Oligonucleotide conjugated antibodies permit highly multiplexed immunofluorescence for future use in clinical histopathology

Intracellular paired agent imaging enables improved evaluation of tyrosine kinase inhibitor target engagement

Proc. Natl. Acad. Sci

Safety and effectiveness of SGM-101, a fluorescent antibody targeting carcinoembryonic antigen, for intraoperative detection of colorectal cancer: a dose-escalation pilot study

Detection of Micrometastases Using SPECT/Fluorescence Dual-Modality Imaging in a CEA-Expressing Tumor Model

Influence of tumour microenvironment heterogeneity on therapeutic response

Oxygen Enhanced Optoacoustic Tomography (OE-OT) Reveals Vascular Dynamics in Murine Models of Prostate Cancer

Mean compartmental pH as a function of the slice (points) compared to real pH (lines) (b). Mean correlation between calculated pH and real pH

1-clinical-protocols-for-the-diagnosis-and-treatment-of-covid-19v7.pdf?sfvrsn=c6cbfba4_2. Released by National Health Commission & State Administration of

Whole-body PET: physiological and artifactual fluorodeoxyglucose accumulations

Image/Figure Caption: Figure 1: CT imaging manifestation during illness (vertical: A1-G1), follow-up CT (vertical: A2-G2) and PET/CT (vertical: A3-G3) among seven

Moreover, no increase of 18F-FDG-uptake was observed (C2-D2the follow-up CT (E2-G2; green arrow), and increased 18F-FDG-uptake was not observed among the mild pneumonia (E3-G3

Identification of novel cancer therapeutic targets using a designed and pooled shRNA library screen

Hu B, Gilkes DM, Chen J: Efficient p53 activation and apoptosis by simultaneous disruption of binding to MDM2 and MDMX

Image/Figure Caption: Figure 1: The polypeptide probe targeting to nucleus DDX24

High Dose Efficiency, Ultra-high Resolution Amorphous Selenium/CMOS Hybrid Digital X-ray Imager

Image/Figure Caption: Propogation based phase contrast X

Tris(2-pyridylmethyl)amine (TPA) as a membrane-permeable chelator for interception of biological mobile zinc(

Image/Figure Caption: Figure 1. DNP-15N MR experiments of 15N-TPA-d6. (a) 15N NMR spectrum of hyperpolarized and thermal equilibrium 15N-TPA

TPA-d6. (i) A diagram of the phantom arrangement

15N spectra at different Zn2+to [15N]TPA-d6 ratios; (iii) 15N images of free

40 ppm); (iv) images of the Zn2+-[15N]TPA-d6 complex (20 ppm). (v) and (vi) show the ratiometric images of each species

Quantitating drug-target engagement in single cells in vitro and in vivo

Determining target engagement in living systems

Simultaneous extracellular and intracellular quantification of EGFR using paired-agent imaging in an in ovo tumor model

Solanki Email: solanki@ohsu.edu Organization: OHSU Country: United States Delegación del Gobierno para el Plan Nacional sobre Drogas

Instituto de Investigación e Innovación en Ciencias Biomédicas de Cádiz-INiBICA

Minocycline benefits negative symptoms in early schizophrenia: a randomised double-blind placebo-controlled clinical trial in patients on standard treatment

Chemical exchange saturation transfer (CEST): what is in a name and what isn't?

A new method for three-dimensional magnetic resonance images denoising

Hypoxia-inducing factors as master regulators of stemness properties and altered metabolism of cancer-and metastasis-initiating cells

Characterizing Vascular Parameters in Hypoxic Regions: A Combined Magnetic Resonance and Optical Imaging Study of a Human Prostate Cancer Model

Image/Figure Caption: Figure: a) Established protocol for FWC surgery: the femur is exposed

Complete First Name: Helen Last Name: Jin Email: helenjingshu.jin@uhnresearch.ca Organization: University of Toronto Country: Canada References: 1

Image/Figure Caption: Figure 1. (A) MicroPET images of

FBZA at 60 min postinjection in B16F10 subcutaneous tumor model (blue arrow) (B) Tumor uptake values and (C) tumor-to-liver ratios of

FBZA at 60 min postinjection in SK-MEL-3 subcutaneous tumor model (green arrow). (E) Tumor uptake values and (F) tumor-to-liver ratios

DMPY2

Stem Cells in Clinical Practice and Tissue Engineering

Image/Figure Caption: Clear border zones at the interection of infarct and normal cardiac apex tissue are shown

Wenzhi Ren et al, Silica Coated Super-paramagnetic Iron Oxide Nanoparticles (SPIONPs): A New Type Contrast Agent of T1 Magnetic Resonance Imaging (MRI)

Complete First Name: M. Zubair Last Name

Image/Figure Caption: Figure 1. Volume variations of tumor mass during esomeprazole treatment for DU145 and PC3 cells (a,c)

Meckel's diverticulitis: a rare entity of Meckel's diverticulum. J Surg Case Reports

Pediatric Radiopharmaceutical Administered Doses: 2010 North American Consensus Guidelines

Taxotere Chemosensitivity Evaluation in Mice Prostate Tumor: Validation and Diagnostic Accuracy of Quantitative Measurement of Tumor Characteristics by MRI, PET, and Histology of Mice Tumor

Image/Figure Caption: The figure represents delineated tumor areas on sodium MRI

Full Name of Abstract's 1st Author : Rakesh Sharma Complete Status: Complete First Name: RAKESH Last Name: SHARMA Email: rksz2009@gmail.com Organization: Florida State University Country: India References: LITE-1 Dependent

Intravenous Delivery of Oncolytic Reovirus to Brain Tumor Patients Immunologically Primes for Subsequent Checkpoint Blockade

Image/Figure Caption: Figure 1: Tracking changes of PD-L1

Full Name of Abstract's 1st Author : Julia Hoebart Complete Status: Complete First Name: Julia Last Name: Hoebart Email: Julia.Hoebart@icr.ac.uk Organization: The Institute of Cancer Research Country

Wireless magnetothermal deep brain stimulation

Noninvasive Deep Brain Stimulation via Temporally Interfering Electric Fields

Optical deconstruction of parkinsonian neural circuitry

Shining light on the head: Photobiomodulation for brain disorders

Noninvasive brain stimulation from physiology to network dynamics and back Nature Neuroscience

Ultrasonic modulation of neural circuit activity

Biogenic gas nanostructures as ultrasonic molecular reporters

References: [1] Upthegrove R, Khandaker GM. Cytokines, Oxidative Stress and Cellular Markers of Inflammation in Schizophrenia

A systematic review of the effect of cannabidiol on cognitive function: Relevance to schizophrenia

Cannabidiol (CBD) as an Adjunctive Therapy in Schizophrenia: A Multicenter Randomized Controlled Trial

The rat brain in stereotaxic coordinates

Table shows mean values ± SEM of brain metabolites in Hipp and PFC. Results are expressed as mean ± SEM. C) Volumetric study: Manual regions of interest (ROI) were segmented as follows (coordinates from Bregma

Full Name of Abstract's 1st Author : Nicolás Lamanna-Rama Complete Status: Complete First Name: Nicolás Last Name: Lamanna-Rama Email: nlamanna@hggm.es Organization: Fundación para la Investigación Biomédica Gregorio Marañón Country: Spain References: 1)

Full Name of Abstract's 1st Author : Irena Pashkunova-Martic Complete Status: Complete First Name: Irena Last Name: Pashkunova-Martic Email: irena.pashkunova-martic@meduniwien.ac.at Organization: Medical University of Vienna Country: Austria References

Angiogenesis as a novel therapeutic strategy for Duchenne muscular dystrophy through decreased ischemia and increased satellite cells

Muscle Biopsy Evaluation in Neuromuscular Disorders Nanette

Murugan NA, Nordberg A, Agren H. Different positron emission tomography tau tracers bind to multiple binding sites on the tau fibril: insight from computational modeling

Full Name of Abstract's 1st Author : Ran Sing Saw Complete Status: Complete First Name: Ran Sing Last Name: Saw Email: ransing.saw@med.uni-tuebingen

References: 1. Bronaugh

Image/Figure Caption: Figure 1 A schematic representation of the in vitro permeation test operated in the PET: (a) The designed PET diffusion cell, (b) A cross-section of the diffusion cell inside the PET, (c) The synthesis of gold nanoparticles, (d) PET imaging of the diffusion cell

Direct Administration of Nerve-Specific Contrast to Improve Nerve Sparing Radical Prostatectomy

The case for early detection

Cancer biomarkers: Can we turn recent failures into success?

Application of positron emission tomography imaging to cancer screening

Gaussia luciferase reporter assay for monitoring of biological processes in culture and in vivo

5. Herschman, H. R. PET reporter genes for noninvasive imaging of gene therapy, cell tracking and transgenic analysis

Image/Figure Caption: Figure 1. (a) HeLa cells treated with tumor-activatable minicircle MCwith MC-pSur and no transfection agent, or a mock treatment (n=3, day 3, p in vivo produce GLuc that is secreted and measured in whole blood (n=6 for "MC-pSur + jetPEI

FHBG retention within PET ROIs at tumor site (n=6, day 9

MC-pSur, minicircle with Survivin promoter

(hydroxymethyl)butyl)guanine

%ID/g, percent injected dose per gram

In vivo confocal microscopy for diagnosis of melanoma and basal cell carcinoma using a two-step method: analysis of 710 consecutive clinically equivocal cases

Validation of the use of a fluorescent PARP1 inhibitor for the detection of oral, oropharyngeal and oesophageal epithelial cancers

Representative examples of H&E and PARP1-IHC stained sections show high and low PARP1 expression in melanoma and nevus, respectively. The dermatopathologist scoring shows higher

Imaging brain activity during seizures in freely behaving rats using a miniature multi-modal imaging system

Chronic widefield imaging of brain hemodynamics in behaving animals

A Compact Head-Mounted Endoscope for In Vivo Calcium Imaging in Freely Behaving Mice

A wireless miniScope for deep brain imaging in freely moving mice

A miniature multi-contrast microscope for functional imaging in freely behaving animals

Optical brain imaging in vivo: techniques and applications from animal to man

Laser Speckle Contrast Imaging: theory, instrumentation and applications

In vivo confocal microscopy for diagnosis of melanoma and basal cell carcinoma using a two-step method: analysis of 710 consecutive clinically equivocal cases

Imaging therapeutic PARP inhibition in vivo through bioorthogonally developed companion imaging agents

Validation of the use of a fluorescent PARP1 inhibitor for the detection of oral, oropharyngeal and oesophageal epithelial cancers

Image/Figure Caption: Investigating PARPi-FL for nuclear labeling and improving diagnosis of basal cell carcinoma

Survival of the fittest: cancer stem cells in therapeutic resistance and angiogenesis

Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer

Survival and self-renewing capacity of breast cancer initiating cells during fractionated radiation treatment

Identification and characterization of cancer stem cells in human head and neck squamous cell carcinoma

Chang, J. C. Cancer Stem Cells Role in Tumor Growth, Recurrence, Metastasis, and Treatment Resistance

The role of CD133 in cancer: a concise review

Image/Figure Caption: Therapeutic efficacy of

Circuitbreakers: optical technologies for probing neural signals and systems

A novel numerical approach to stimulation of a specific brain region using transcranial focused ultrasound

Piezo1 and Piezo2 are essential components of distinct mechanically activated cation channels

The Mechanosensitive Ion Channel Piezo1 Significantly Mediates In Vitro Ultrasonic Stimulation of Neurons. iScience

Electroporation as a vaccine delivery system and a natural adjuvant to intradermal administration of plasmid DNA in macaques

Depth-wise progression of osteoarthritis in human articular cartilage: investigation of composition, structure and biomechanics

Assessment of glycosaminoglycan concentration in vivo by chemical exchange-dependent saturation transfer (gagCEST)

Chemical exchange saturation transfer magnetic resonance imaging of human knee cartilage at 3T and 7T

Cartilage quality assessment by using glycosaminoglycan chemical exchange saturation transfer and (23)Na MR imaging at 7T

High Quality Three-Dimensional gagCEST Imaging of In Vivo Human Knee Cartilage at 7

User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability

Image/Figure Caption: Figure caption: Images from four subjects, the left knee of one OA patient (red box) and three healthy controls, below. Axial (left) and coronal , wherein the medial side of the knee (subjects' left side, L, of right knee) had

Image/Figure Caption: The better linear relationship of 68Ga-NOTA-WL12 than 18F-FDG with PD-L1 TPS on IHC: In a sixty-eight-year-old male with advanced NSCLC, the PD-L1 TPS on IHC was 8%. On 68Ga-NOTA-WL12 imaging, the tumor showed low uptake of tracer (a, b),while on 18F-FDG imaging, the tumor showed intense radioactive concentration (c) . In an eight-year-old female with advanced NSCLC, the PD-L1 TPS on IHC was 80%. On 68Ga-NOTA-WL12 imaging, the tumor showed moderate uptake of tracer (d, e),while on 18F-FDG imaging, the tumor showed intense radioactive concentration (f) . Abstract Body : Introduction: Glioblastomas (GB) are malignant brain tumours with poor prognosis even after surgical resection and aggressive therapy. The participation of the immune system is key for sustained response [1] . We have used MRSI-based nosological images [2, 3] for GB therapy response assessment through tumour responding index (TRI) calculation. An oscillatory TRI pattern (6-7 days) was shown in longitudinal studies. Glioma-associated microglia/macrophages (GAMs) constitute the most abundant non-tumour cell type in the GB microenvironment and can be activated and polarized into an anti-tumour (M1) or pro-tumour (M2) phenotype [4] . The purpose of our present work was to gain further insight into the contribution of immune cell populations to the MRSI spectral pattern changes recorded from Temozolomide (TMZ) -treated preclinical GL261 GB. Methods: C57BL/6 mice bearing GL261 GB tumours (n=39) were treated with TMZ in an immune-enhanced metronomic schedule (IMS) every 6 days, at 60 mg/kg (n=23) [3] , while the remaining (n=16) were administered with vehicle and used as controls. T2w MRI and consecutive 14ms volumetric TE MRSI were acquired every 2 days [1] and nosologic maps calculated [5] . Twenty mice (n=10 controls, n=10 treated) were euthanized at TRI-guided time points for in vitro evaluation, while the remaining ones had longitudinal follow-up until endpoint or curation. Quantitative polymerase chain reaction (qPCR) was performed with genes associated to microglia/macrophage population (F4/80) and its polarization (NOS2 for M1 phenotype and CD206 for M2 phenotype). Results were normalized to two reference genes (HPRT and TPB). Immunostainings for CD3 (lymphocytes) and Iba-1 (microglia/macrophages) were performed in n=6 additional mice [2] . Cured mice (n=8) were followed-up by T2w MRI and in case of non-tumour mass detection within one month of cure, a "re-challenge" experiment with GL261 was carried out. Results-Discussion: IMS-TMZ increased GL261 GB bearing mice survival, from 21±1 days in untreated mice up to 295±279 days, improving previous results [4] . TRI oscillations (6.2±1.5 days, n= 11, Fig. 1A) were in agreement with immune cycle length (Fig 1B) . MRSI spectral changes could reflect immune system action involving lymphocytes and especially microglia/macrophages (GAMs), since they can represent up to 30% of GB mass [6] . Namely, immunohistochemistry shows CD3 and Iba-1 content significantly higher in responding zones. Significant differences were observed for F4/80 (0.71±0.32 vs 0.18±0.08 relative expression; Fig.1C and Nos2/CD206 ratio (0.26±0.16 vs 0.11±0.09 ratio; Fig. 1D ) between responding and control tumours, respectively. Global GAMs population and M1/M2 ratio were higher in responding compared to control tumours. Since M1 and M2 have different metabolic profiles [7] , this population difference could be one of the key causes for the differential MRSI-sampled pattern with time. This agrees with an increase in antitumour immune cells content at TRI-high peak times (Fig 1 A) . Regarding the re-challenged mice, only 1/8 tumour grew after 10 days, which vanished again after being treated with only one additional IMS-TMZ dose. This oscillatory pattern may constitute an immune system activity biomarker for use in future work. Conclusions: Our results indicate that IMS-TMZ treated GL261 GB showed significantly more microglia/macrophages than untreated mice, with predominance of the M1 anti-tumour phenotype, which can be imaged non-invasively by MRSI. This can be of interest in pathologies/therapies in which immune system participation is foreseen. IMS-TMZ induced immune memory in GB cured mice, although the ongoing mechanism needs further clarification. believe this to be the case for the relative values of the dentate gyrus and other immediately nearby structures (see Persuasive Data for B1 maps). 

Ariel Buchler, University of Ottawa Heart Institute, abuch052@uottawa.ca

Abstract Body : Introduction: Dysregulation of extracellular matrix remodeling by specific matrix metalloproteinases (MMPs) in vasculature is associated with atherosclerotic plaque vulnerability.1,2 Given that thrombosis underlies most adverse clinical outcomes including myocardial infarction and stroke, MMPs represent compelling biomarkers for imaging by positron emission tomography (PET).1,3 To date, differentiating stable atherosclerotic plaques from those susceptible to rupture using broad-spectrum MMP radiotracers has been unsuccessful potentially due to engagement with non-pathologic MMPs, possessing distinct and opposing functions in disease progression.4,5 Contrarily, MMP-13 has been shown to predominantly exhibit collagenolysis in atheromatous plaques and reduce smooth muscle cell accumulation, contributing directly to plaque destabilization by degradation of the fibrous cap.6-9 As such, a comparative analysis between non-selective [18F] BR-351 and a selective MMP-13 radiotracer, herein entitled [18F] FMBP, was performed to elucidate whether targeting of specific MMP isozymes augments imaging sensitivity, and improves correlation with markers of vulnerable atherosclerotic lesions. Methods: Preliminary screenings of [18F] FMBP and [18F] BR-351 were completed by in vitro autoradiography of en face aortae from atherogenic ApoE-/-mice. Ex vivo biodistribution and autoradiography were analogously performed in C57Bl/6 and ApoE-/-mice under baseline or homologous blocking conditions to assess radiotracer localization, specific binding, and sensitivity. En face aortae were subsequently stained with Oil Red O, sectioned, and subject to immunofluorescence staining with Mac-2 and MMP-13 to correlate radiotracer uptake with histological biomarkers of plaque vulnerability. Results: Automated radiosyntheses afforded [18F] FMBP and [18F] BR-351 in high decay-corrected radiochemical yields, purities, and molar activities. During preliminary assessments, while both radiotracers localized within atherosclerotic lesions and demonstrated measurable specific binding, [18F] FMBP overall lesion uptake was 1.6-fold higher. Ex vivo biodistributions revealed that [18F] FMBP and [18F] BR-351 predominantly undergo renal clearance and hepatobiliary excretion with low accumulation of radioactivity in non-excretory organs (t=30 min). Ex vivo autoradiography and histology of en face aortae revealed that [18F] FMBP, in comparison to [18F] BR-351, exhibited 2.2-fold greater lesion uptake, observable specific binding (63%), improved disease sensitivity (2.7-fold vs 2.5 fold) and superior correlation to extent of lipid accumulation by Oil Red O (R2: 0.80 vs 0.72). As detected by Mac-2 and MMP-13 immunofluorescence staining, extent of inflammation and ECM remodeling were distinctly increased within the core and fibrous cap of atherosclerotic lesions and further correlated to [18F] FMBP autoradiographic uptake (R2 = 0.71 and 0.56, respectively). MMP-13 additionally correlated with Mac-2 (R2 = 0.85). Conclusion: The feasibility of imaging extracellular matrix remodeling in mouse models of atherosclerosis with MMP-targeted PET radiotracers has been established. Selective imaging of MMP-13 with [18F] FMBP showed improved sensitivity, specificity, and correlation to markers of plaque vulnerability relative to pan-selective [18F] BR-351. Altogether, [18F] FMBP has proven useful for the detection of extracellular matrix remodeling in inflamed atherosclerotic lesions, suggesting that selective isozyme imaging represents a promising approach towards the characterization of high-risk atherosclerosis.with high-accuracy of >80% using only BL radiomic features. Additional biomarkers can be used to further enhance prediction of response to therapy.References: 1. Zwanenburg A, Vallières M, Abdalah MA, Aerts HJWL, Andrearczyk V, Apte A, et al. "The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping," Radiology (2020); (5):191145. 2. Robnik-Sikonja, M., and I. Kononenko. (2003) . "Theoretical and empirical analysis of ReliefF and RReliefF, " Machine Learning, 53, Zhi et al. " Evaluation of classification and regression tree (CART) model in weight loss prediction following head and neck cancer radiation therapy," Advances in radiation oncology vol. 3,3 346-355. 7 Dec. 2017 3,3 346-355. 7 Dec. , doi:10.1016 3,3 346-355. 7 Dec. /j.adro.2017 Abstract Body : Exosomes are nano-sized (30-140nm) spherical vesicles, derived from the endosomal system for intercellular communications. Exosomes have clear advantages over synthetic nanoparticles like liposomes or any other metallic or chemically synthesized nanoparticles as a vehicle/probe because of their more biocompatibility, low toxicity and immunogenicity, enhanced permeability (even through blood-brain barrier) and stability in biological fluids, and ability to accumulate in the tumor or lesions with higher specificity. Different imaging modalities and techniques are being utilized to detect specific cells in vivo. Most of the published reports showed the use of peptide or antibody to target specific cells. The use of hydrophobic peptide/s make the imaging agent less attractive due to poor target-tobackground ratio, therefore, investigators have used nanoparticle-based contrast agent to image specific cells. Making of these Nanoparticles (synthetic) or peptide-based imaging agents needs the involvement of extensive chemistry or radiochemistry. Here we propose to utilize cellular and molecular biology (recombinant DNA technology) technology to make biocompatible, biodegradable, and non-immunogenic nanovesicles (exosome) to carry neutrophil or alternately activated macrophage targeting peptide/s to image neutrophils or macrophages in vivo. These exosomes can be tagged with single-photon emission computed tomography (SPECT) compatible radioisotope/s to be used as SPECT imaging probes. Recently our laboratory has achieved a few milestones in exosome technology; (1) we developed a platform (patent pending) to make engineered exosomes using non-tumorous HEK293 cells that carry and express specific cell targeting peptide to detect specific cells when administered intravenously, (2) we used these engineered exosomes as a therapeutic probe to deplete specific cells in the body, (3) we optimized the methods to collect exosomes in shortest possible time, (4) we showed in vivo biodistribution of exosomes collected from different cells by clinically relevant SPECT imaging. Both in vitro and in vivo studies showed the specificity of these engineered exosomes to target specific cells. In vivo SPECT studies showed targeting and delineating the specific cells in the body.References: 1. Rashid MH, Borin TF, Ara R, Alptekin A, Liu Y, Arbab AS. Generation of novel diagnostic and therapeutic exosomes to detect and deplete pro-tumorigenic M2-macrophages. Advanced Therapeutics. 2020 May; doi: 10.1002.adtp.201900209. 2. Rashid MH, Borin TF, Ara R, Angara K, Cai J, Achyut BR, Liu Y, Arbab AS. Differential in vivo biodistribution of 131Ilabeled exosomes from diverse cellular origins and its implication for theranostic application. particularly in combination with existing FDG-PET techniques that highlight increased glucose metabolism. Beyond measuring inflammation, MPI can image other types of immunotherapiesnotably, labeled T cells in an adoptive transfer model6. MPI enables researchers to monitor the efficacy of immunotherapies with the sensitivity, specificity, and linear quantitation.This aids in validating our GATE model of the SPARK. NEMA measurements of commercial clinical SPECT scanners are stated in [1] with an intrinsic resolution of 2.7-4.2 mm whereas commercial preclinical scanners, such as the U-SPECT and A-SPECT, state an intrinsic resolution of 3.2 mm and 2.0 mm, respectively [2, 3] . To our knowledge, the SPARK presents the highest intrinsic spatial resolution measurements reported to date for any gamma camera assessed with the NEMA standard. This is primarily due to the small form factor of SiPM microcells in combination with a 3 mm-thick CsI(Na) crystal. Later work will assess the remainder of intrinsic, system, and tomographic measurements. A thorough understanding of the experimental and theoretical imaging performance will assist in improving data quality by adapting the complex SPECT system matrix based on, for example, photon scatter or septal penetration of collimators. solution containing 10mM of phosphate buffer as a function of pH values, (B1=3µT) acquired at 37°C using a 7T scanner and a 3T scanner (c,d) . GlucoCEST ΔST% map obtained at 7T injecting glucose solutions at 1.5g/Kg (e) and 3g/Kg (f) dose or 3OMG solution at at 1.5g/Kg (g) and 3g/Kg (h) dose via intravenous route. Data are reported as the difference (ΔST %) between the ST effect before and after the intravenous injection. Abstract Body : Approximately 90% of malignancies that occur in the kidney are renal cell carcinoma (RCC).1 Clear cell renal cell carcinoma (ccRCC) is the most common RCC subtype (≈70%). In ccRCC, even after nephrectomy, distant metastases are frequent with lungs being the most common site.1,2 This study aims to image ccRCC-derived lung tumors in mice using 4- [18F] -(2S,4R)-fluoroglutamine ( [18F] FGln), to understand glutamine transport and glutamine utilization in ccRCC. [18F] FGln is a reporter for the alanine-serine-cysteine transporter 2 (ASCT2), a major glutamine transporter, overexpressed in a number of cancers, including RCC. 3 In ccRCC, high ASCT2 expression was associated with lower overall survival and identified as an independent prognostic factor.4 Therefore, understanding the correlation of [18F] FGln and ASCT2 can provide the opportunity to use [18F] FGln and PET imaging as a predictive biomarker.5 [18F] FGln was reliably produced with 8.5±2.9 % (n=7) radioactivity yield.6 Its uptake was investigated in vitro, in ccRCC cell lines SN12C, UMRC3, and its metastatic subline LUNG150, generated from lung metastasis tissue from an orthotopic UMRC3 kidney mouse model. In vivo, subcutaneous and lung-orthotopic models of ccRCC were imaged. UMRC3 and LUNG150 cell lines express luciferase and GFP, facilitating bioluminescence imaging (BLI) of tumor growth. [18F] FGln uptake cell experiments were carried out to observe initial influx of the tracer due to the transporter (2 min) and the net effect of tracer influx and efficiency of [18F] FGln trapping in the intracellular compartment (1 hour). The 2 min influx of [18F] FGln in UMRC3 cells was blocked by saturation of the transporter with 20 mM of glutamine in the media ( Fig 1A) . The same reduction of [18F] FGln was achieved when ASCT2 expression was reduced by siRNA ( Fig 1A) . Expression of ASCT2 in UMCR3 cells and siRNA knockdown was confirmed by Western blots. At 1 hour, [18F] FGln uptake was reduced in a dose-dependent manner, by addition of glutamine ( Fig 1B) . A similar effect was achieved upon treatment with the ASCT2 selective inhibitor V-9302 ( Fig 1B) . In LUNG150 cells, at 2 min, uptake of [18F] FGln is 15-fold higher than in UMRC3 cells (Fig 1C) , despite lower ASCT2 expression. It is possible that the initial influx of glutamine in these cells is mediated by secondary mechanisms that are worth investigation. This is consistent with ASCT2 inhibition with V-9302, which has less effect than shown in UMRC3 cells ( Fig 1D) . Moreover, knockdown of ASCT2 using siRNA in LUNG150 does not lead to a reduction in [18F] FGln uptake ( Fig 1C) . In vivo, lung-orthotopic ccRCC mouse models were generated by injection of UMRC3 or LUNG150 cells into the left lung of nude mice. PET/MRI proved superior to PET/CT to image this orthotopic lung tumor model. Tumor tissue was easily identified using 3D MRI (Fig 1E) . Tracer uptake was determined in the heart, contralateral lung, tumor, and muscle and data reported as the %ID/cc ratio of tumor versus various organs ( Fig 1F) . Tumor-to-muscle was about 1.5, whereas tumor-to-lung and tumor-to-heart were around 1. It is possible, however, that tumor-to-contralateral lung values could be biased as, after necropsy, bioluminescence/tumor growth was observed in both lungs. In summary, [18F] FGln uptake in ccRCC cell lines is tied to the expression of ASCT2. ccRCC orthotopic lung tumors were generated and [18F] FGln PET/MRI experiments performed. UMRC3 lung orthotopic tumors did not show significant contrast over contralateral lung tissue, however, contrast is observed over muscle. PET/MRI imaging of LUNG150 orthotopic tumors and control non-tumor bearing mice are underway. Based on cell studies, the metastatic line LUNG150 shows promise to give statistical contrast over healthy lung tissues.References: 1. Hsieh, J.J., Purdue, M.P., Signoretti, S., Swanton, C.et al. Renal cell carcinoma. Nat Rev Dis Primers 2017, 3, 17009. 2. Bianchi, M., Sun, M., Jeldres, C., Shariat, S.F.et al. Distribution of metastatic sites in renal cell carcinoma: A population-based analysis. Ann Oncol 2012, 23, 973-980. 3 . Schulte, M.L., Fu, A., Zhao, P., Li, J.et al. Pharmacological blockade of ASCT2-dependent glutamine transport leads to antitumor efficacy in preclinical models. Nat Med 2018, 24, 194-202. 4. Liu, Y., Yang, L., An, H., Chang, Y.et al. High expression of solute carrier family 1, member 5 (SLCA1A5) is associated with poor prognosis in clear-cell renal cell carcinoma. Sci Rep 2015, 5, 16954. 5. Hassanein, M., Hight, M.R., Buck, J.R., Tantawy, M.N.et al. Preclinical evaluation of 4-[18F] fluoroglutamine PET to assess ASCT2 expression in lung cancer. Mol Imaging Biol 2016, 18, 18-23. 6. Zhang, X., Basuli, F. Shi, Z-.D., Xu, B., Blackman, B ., Choyke, P.L., Swenson, R.E. Automated synthesis of [18F] (2S,4R)-4-fluoroglutamine on a GE TRACERlab™ FX-N Pro module Applied Rad Isot, 2016, 112, [110] [111] [112] [113] [114] Image/ Figure: Image/Figure Caption: [18F] FGln uptake experiments in UMRC3 cell line at 2 min (A) and 1 hour (B) and in LUNG150 cell line at 2 min (C) and 1 hour (D) . PET/MR image of UMRC3 lung tumors, using 7T Bruker MR scanner equipped with a Cubresa NuPET insert. ROIs: lung tumor (T) in yellow, contralateral lung (CL) in blue, heart (H) in red and muscle (M) in green (E). Ratio of %ID/cc in lung tumor tissue (T) to contralateral lung (CL), muscle (M) , or heart (H); n = 3 mice (F). 

Behlol Khan, University of Saskatchewan, khan.behlol@usask.ca Category: Oncology Abstract Body : Background: About 25-30% of breast cancer (BC) overexpress epidermal growth factor II (HER2) and this is associated with aggressive disease and poor prognosis. Trastuzumab a humanized monoclonal antibody targeting HER2 is approved against HER2positive BC. However, 20 -50% of patients selected for treatment show no response, while up to 70% of initial responders subsequently acquire resistance. The high linear energy transfer and decay characteristics of alpha particles such as actinium-225 (225Ac) make it ideal for treating small lesions. In this study we hypothesize that targeting 225Ac-trastuzumab can enhance the therapeutic efficiency of trastuzumab. Materials and Methods: Trastuzumab was conjugated with an eighteen-membered macrocyclic chelator p-SCN-Bz-macropa and radiolabeled with 225Ac for alpha particle therapy. The radiochemical yield of 225Ac-trastuzumab was >90%. The (radio)immunoconjugate was characterized by flow cytometry, radioligand binding assays, HPLC and internalization rate (live-cell imaging). HER2 positive JIMT-1 cells, a trastuzumab/pertuzumab resistant model with medium to low HER2 expression were used to study the in vitro cytotoxicity. In vivo radioimmunotherapy using 225Ac-trastuzumab was studied using JIMT-1 xenograft following treatment with three doses of 350 nCi/dose administered 10 days apart. In vivo study endpoint was tumor volume > = 1500 mm3. Control treatment groups included PBS, trastuzumab and 225Ac-labeled isotype control IgG. Results: Flow cytometry showed > 90% binding to the cells. In vitro studies in JIMT-1 showed enhanced cytotoxicity (IC50) of 225Ac-trastuzumab (5.5 ±1.14 nM) which was 62 folds less than unlabeled trastuzumab (340 ± 1.77 nM). 225Ac-trastuzumab was effective at inhibiting the growth of JIMT-1 tumors. 6/7 mice treated with 225Ac-trastuzmab had tumor decreased to <= 40 mm3, with one complete remission. Median survival 225Ac-trastuzumab (not yet reached); PBStreated (51 days). Conclusion: 225Ac-trastuzumab showed promising therapeutic efficiency towards HER2 positive BC JIMT-1 with medium-low HER2 expression. Efficiency of this radioimmunoconjugate in 3D spheroids and other HER2 positive xenografts is ongoing. Abstract Body : Background Invasive aspergillosis (IA) is one of the most severe forms of fungal infection in patients with primary or secondary immunodeficiencies. The lungs are involved in the majority of cases of IA and Aspergillus fumigatus (A. fumigatus) is one of the most common causative agents. In view of high mortality and morbidity associated with IA, the development of non-invasive imaging biomarkers becomes necessary to provide a rapid and specific diagnosis. In this study, we aimed to evaluate L-rhamnose, a naturally occurring sugar known to be metabolized by different Aspergillus species, as a potential PET ligand for the diagnosis of pulmonary infection caused by A. fumigatus. Methods In vitro uptake of 3Hrhamnose was tested in live and heat killed A. fumigatus cultures, and was compared to two representative gram negative and gram positive bacteria (Escherichia coli ATCC #25922 and Staphylococcus aureus ATCC #29213) as well as mouse macrophages (Mus musculus J774.1) to assess the specificity of uptake, in comparison to 3H-2-deoxyglucose (3H-2-DG). 2-deoxy-2-18F-fluoro-L-rhamnose (18F-rhamnose) was then synthesized and evaluated for in vitro uptake by A. fumigatus and E. coli. In vivo uptake was then assessed by autoradiography (3Hrhamnose) and PET/CT (18F-rhamnose) in murine models of pulmonary aspergillosis (2 days following post-pharyngeal inoculation). Standardized uptake values (SUVs) of 18F-rhamnose in infected mice were then measured, and compared to animals with sterile lung inflammation (24 hours following post-pharyngeal poly (I:C) administration) and healthy controls. Results In vitro uptake assays showed selective uptake of 3H-rhamnose by live A. fumigatus cultures when compared to heat killed fungi, macrophages, and selected representative of Gram negative and Gram positive bacterial strains. The uptake of 3H-rhamnose was also significantly higher than that of 3H-2-DG in A. fumigatus. Similarly, increased uptake of the tritium labeled ligand was seen in the lungs of the infected mice by autoradiography while there was no appreciable uptake in the healthy mice and minimal uptake in poly (I:C) treated animals (30 minutes uptake period). 18F-rhamnose was specifically internalized by live A.fumigatus cultures when compared to heatkilled cultures and E.coli. In vivo PET/CT imaging with a 60-minute dynamic 18F-rhamnose PET/CT imaging of a pulmonary IA model showed a slight increase of uptake in lung lesions compared to controls and poly (I:C) treated mice ( Figure 1A and 1B). Attempted modification of the molecule with 18F labeling of C6, to improve uptake, led to rapid defluorination. Conclusions Our study showed that A. fumigatus was able to selectively accumulate 18Frhamnose in vitro but to a lesser extent in vivo, likely due to short serum half-life. Even though the uptake of 18F-rhamnose was significantly higher in lungs of infected mice compared to healthy controls and poly (I:C) treated mice, there was only mild delineation of the lung lesions on PET imaging. Additional studies are underway with different labeling approaches to improve the retention of 18F-labeled rhamnose and increase the uptake within infected lung tissues.

Image/Figure Caption: (A) Representative dynamic average PET images, averaged from 520-3520 seconds post injection, of 18F-rhamnose uptake in the lungs of control, sterile lung (poly I:C) and post-pharyngeally infected pulmonary IA (AF IPA) murine models. The first 520 seconds were removed from analysis to reduce potential effects of increased vascularity after injection. (B) Time activity curve of mean 18F-rhamnose uptake in control, Poly (I:C), and AF IPA models from 0-3370 seconds post 18F-rhamnose injection. Abstract Body : Introduction Tumor vascular perfusion and oxygenation dictate outcomes in cancer treatment, including radiation therapy (RT)1. However, precise measurements of these biomarkers are limited by drawbacks of current dynamic contrast enhanced (DCE) imaging methods2. Here, a new analysis method for DCE multispectral optoacoustic tomography (MSOT)3 is presented that can quantitatively measure vascular perfusion with high temporal and spatial resolution, while endogenous MSOT is employed to evaluate changes in O2 saturation (dsO2) as measure of O2 delivery. The method is tested in two xenograft tumor models with different vascular phenotypes and different response to RT. Methods Subcutaneous xenograft models, MIA PaCa-2 (slow growing, less aggressive) and Colo357 (fast growing, highly aggressive) were prepared by injecting female athymic nude mice with 1x106 cells in lower right flank. Tumors were irradiated with two isocentric beams to deliver a single dose of 10 Gy (X-RAD 320, Precision X-ray Inc.). MSOT measurements were acquired 2 days before and 1 and 4 days after RT, using an MSOT inVision 256-TF scanner (iThera Medical). DsO2 was evaluated with endogenous MSOT by switching inhalation gas from 21% O2 to 100% O2. DCE MSOT was performed over 15 min after intravenously injecting 2.5 mmol/kg ICG. Data analysis was implemented in MATLAB. Correlative histopathology confirmed in vivo imaging outcomes. Results/Discussion We used noninvasive MSOT to examine native differences in vascular perfusion and DmsO2 between two molecularly distinct tumor models ( Fig. 1 ) and evaluate early response to RT (Fig. 2) . Baseline DCE MSOT measurements of ICG kinetics indicated > 3-fold lower perfusion in Colo357 tumors compared to MIA PaCa-2 ( Fig. 1c) . Endogenous MSOT of hypovascular Colo357 tumors showed poor response to O2 challenge (DsO2) compared to vascular rich MIA PaCa-2 tumors (Fig.1d ). Endogenous and DCE MSOT could sensitively and precisely detect early changes in the biomarkers in response to RT. A single 10 Gy dose caused a dramatic drop in vascular perfusion in MIA PaCa-2 tumors, 1 and 4 days post-RT (Fig. 2a) . In contrast, resistant Colo357 tumors showed no change in ICG perfusion at day 1 followed by a 2.5-fold decrease at day 4 ( Fig.2b) . DsO2 emerged as a significant biomarker of heterogeneous RT response in the two tumor models (Fig.2c ). Conclusions We demonstrate improved precision and application of endogenous MSOT and DCE MSOT in noninvasive assessment of molecularly different tumor models. Quantitative and synergistic measurements of tumor vascular perfusion and DsO2 enabled early and precise evaluations of response to RT. MSOT derived imaging biomarkers can potentially allow rapid and accurate cancer staging, treatment planning and monitoring, in preclinical and clinical settings.

Michael Luciano, National Cancer Institute, lucianomp@nih.gov Category: New Chemistry, Biology & Bioengineering Abstract Body : In vivo fluorescence imaging has the potential to probe biological processes in their native environments and inform surgical decision making.1 While multiplexed imaging is routine in microscopy, such experiments are challenging in body-wide imaging due to the competitive autofluorescence and scattering encountered with visible (2 Recent progress has demonstrated the benefits of optical imaging using longer wavelengths (1000 to 2000 nm), referred to as the Shortwave-Infrared (SWIR) or NIR-II range. These methods enable high resolution imaging in bulk tissue, with dramatic improvements in both resolution and depth relative to imaging in the NIR region.3 As InGaAs detector-based imaging systems have become more accessible, a critical bottleneck in this field is access to biologically compatible fluorescent probes that operate in this range Indocyanine dyes are broadly used in fluorescence-based experiments. In the visible range, trimethine indocyanines (e.g. Cy3) and pentamethine indocyanines (e.g. Cy5, AF-647), find extensive use in applications spanning routine quantification to advanced microscopy. Derivatives of heptamethine-indocyanines (e.g. Cy7, IRDye800CW and indocyanine green, ICG) are the most broadly applied molecules for in vivo near-infrared (NIR) imaging, including for clinical use.4 The defining chemical feature of these molecules -two-carbon homologation -leads to ~100 nm increases in absorbance maxima, making these scaffolds nearly ideal for simultaneous use in multiplexed experiments. Another feature of the heptamethine variants is their utility in SWIR imaging, due to significant emission extending beyond 1000 nm.5 The use of heptamethine indocyanine "tail imaging" in the SWIR has been investigated in the clinic using the FDA-approved NIR fluorophore, ICG.6 This study extends the biological utility of the indocyanine scaffold to longer wavelengths. Guided by computational design, we report two classes of novel dyes with significant SWIR emission based on the nonamethine indocyanine scaffold. The first is a true per-sulfonated "Cy9" type derivative (FNIR-872, λabs = 872 nm), which has been optimized for monoclonal antibody labeling. We also describe a novel chemical strategy to modify the nonamethine cyanine scaffold with a fused aryl ring, thereby extending the absorbance maximum beyond 1000 nm (FSWIR-1072, λabs = 1072 nm). These dyes exhibit cyanine-like exceptional absorption cross sections, high solubility, and excellent photo/chemical stability. Targeted tumor imaging was successfully performed using FNIR-872-mAb, revealing comparable TBR and tumor signal to clinically used IR-800CW-mAb. In addition, we also describe how using red-shifted (892 nm) excitation wavelengths exhibit reduced endogenous emission (autofluorescence) in vivo from multiple peritoneal organs. Lastly, we demonstrate the utility of mAb and dextran bioconjugates of these novel small molecule dyes in combination with the FDA-approved dye ICG for multicolor tumor imaging experiments in the SWIR region. These efforts demonstrate the high potential of substituted nonamethine cyanines for targeted SWIR imaging, and extend the utility of indocyanine dyes to multiplexed imaging in the SWIR region for the first time. (17), [4465] [4466] [4467] [4468] [4469] [4470] Z.; Fang, C.; Li, B.; Zhang, Z.; Cao, C.; Cai, M.; Su, S.; Sun, X.; Shi, X.; Li, C.; Zhou, T.; Zhang, Y. ; Chi, C.; He, P.; Xia, X.; Chen, Y.; Gambhir, S. S.; Cheng, Z.; Tian, J., First-in-human liver-tumour surgery guided by multispectral fluorescence imaging in the visible and near-infrared-I/II windows. Nat. Biomed. Eng. 2020, 4 (3) , 259-271.

Image/Figure Caption: Left: Prior biocompatible Indocyanine dyes used in visible/NIR optical imaging and novel NIR and SWIR-emissive nonamethine cyanine dyes reported in this work. Right: 3-Color live imaging of an athymic nude mouse bearing an MDA-MB-468 orthotopic tumor implanted in the mammary fat pad. The mouse was injected with ICG (λexcitation = 785 nm, pseudo-colored green) and FSWIR-1072-Dex-10 kDa (λexcitation = 1064 nm, pseudocolored red) 24 hours after administering FNIR-872-Pan (λexcitation = 892 nm, pseudo-colored cyan) by intraveneous injection and images in all three channels were recorded using long pass filters. Abstract Body : Anaesthesized pigs and other laboratory animals are widely used in positron emission tomography (PET) neuroimaging studies. However, you can hardly anaesthetize the brain without affecting its physiology, and the used anaesthetics may therefore modify tracer binding profiles at receptor targets. We evaluated effects of isoflurane and propofol on the binding of [11C]MDL100,907 on the serotonin 5HT2A receptors in pigs. Two groups of anaesthetised Göttingen minipigs were imaged with [11C]MDL100,907 PET and analysed using regions of interest and statistical non-parametric mapping. The [11C]MDL100,907 binding potentials in striatum under isoflurane anaesthesia significantly exceeded those obtained under propofol anaesthesia, probably due to higher cerebral blood flow in brain during isoflurane anaesthesia. In conclusion, the interactions between the used anaesthetics and brain tracers must be carefully evaluated prior to animal experimentation and in the interpretation of results. Abstract Body : Colorectal cancer (CRC) is the leading cause of death in all of the digestive system cancers and the third in all types of cancer, with a 5-year overall survival rate of only 58%-65%1. Many patients advance to late stages when metastasis has occurred in remote sites and standard treatments are ineffective2. This study aimed to develop a more effective therapeutic approach for CRC based on immunogenic cell death (ICD) and photodynamic therapy (PDT). Here we chose the 18 kDa translocator protein (TSPO), a protein overexpressed in CRC, as the target for Photodynamic therapy (PDT) of CRC. Our TSPOtargeted photosensitizer (IR700DX-6T)3 was found to be located in the mitochondria and specifically bound to TSPO in MC38 CRC cells ( Figure A, B) . PDT with IR700DX-6T caused effective CRC cell death in a concentration-and light dose-dependent manner ( Figure  C ). The mechanisms behind the killing effects were further investigated. The cleaved caspase-3 substrate and morphological changes with typical apoptotic characteristics suggest that our TSPO-PDT treatment induced apoptotic cell death. Remarkably, TSPO-PDT also induced Immunogenic cell death (ICD)4 as evidenced by cell membrane translocation/release of two key DAMPs (damage-associated molecular patterns), including calreticulin (CRT) and heat shock protein 70 (HSP70) after TSPO-PDT treatment (Figure D, E) . Furthermore, our in vivo study using a syngeneic CRC mouse model showed that TSPO-PDT was not only effective on directly treating TSPO+ tumors, but also able to trigger host anti-tumor immune response resembling abscopal effect. Such favorable antitumor immunity was shown to be associated with elevated dendritic cell maturation and CD8+ T cell infiltration, as well as decreased regulatory T (Treg) cells. In conclusion, TSPO-PDT using IR700DX-6T displayed high therapeutic efficacy against CRC cells and tumors. This new therapeutic approach could efficiently treat CRC tumors and cause host anti-tumor immune response, suggesting that TSPO-PDT has great potential in the treatment of CRC. Biomater 28, 160-170, doi:10.1016 Biomater 28, 160-170, doi:10. /j.actbio.2015 Biomater 28, 160-170, doi:10. .09.033 (2015 . 4 Garg, A. D., Dudek-Peric, A. M., Romano, E. & Agostinis, P. Immunogenic cell death. Int J Dev Biol 59, 131-140, doi:10.1387 /ijdb.150061pa (2015 .

Image/Figure Caption: (A, B) IR700DX-6T specifically bound to TSPO. (A) IR700DX-6T was effectively uptaken by TSPO (+) MC38 cells and the fluorescence signal overlaid with mitotracker green (mitochondrial marker). TSPO (-) MCF7 cells showed negligible uptake of IR700DX-6T. (B) DAA1106, a small molecule ligand with a high binding affinity to TSPO, markedly decreased the uptake of IR700DX-6T in MC38 cells. MC38 cells treated with nontargeted IR700DX showed unconspicuous fluorescence signal. (C) Cell viabilities were measured after IR700DX-6T-PDT at various concentrations (0.5-5 μM) and light doses (7.2-27 .0 J/cm2). (D, E) TSPO induced ICD was confirmed by using fluorescence microscope imaging. MC38 cells were treated with PDT using 0.5 μM of IR700DX-6T and 18 J/cm2 of light irradiation at 690 nm. Three hours after the PDT treatment, cells were fixed, permeabilized, and subjected to immunofluorescence imaging (red) of calreticulin (D) and heat shock protein 70 (E). Abstract Body : The innate immune system integrates multiple pro-and anti-inflammatory stimuli, organize key effector functions, and in the proper context, culminate in the respiratory burst, or ROS burst. These systems are deeply conserved across species, although flux through these pathways is most robust in humans. The respiratory burst has long been studied in neutrophils, but is known to occur (with less intensity) in macrophages, microglia, and "immunostimulated" endothelial cells. In neutrophils and macrophages, high energy ROS are produced by NADPH oxidase 2 (NOX2) and myeloperoxidase (MPO). When functioning properly, these cells and ROS bursts provide front line defense to a variety of pathogens and degrade difficult to process foreign bodies, such as carbon nanotubes. When pathogenic, ROS degrade and inhibit the normal function of tissues from cartilage in arthritis to T-cells in tumors.1-5 Contrast MRI, [18F] FDG-PET, or 68/67Ga-Citrate-PET/SPECT lack the sensitivity and/or specificity to detect changes in the innate immunity system in many of these diseases, and novel approaches to detect inflammation are sought.7-11 Herein, we develop and test 4- [18F] fluoronaphthol ([18F]4FN) as a novel radiopharmaceutical to detect ROS burst by PET. A robust and automated synthesis of [18F] 4FN using copper-mediated radiofluorination was developed and validated on a commercial synthesizer (GE TracerLab-FX).12 To date, [18F] 4FN has been successfully synthesized with an average activity yield of 6.8±2.5% (n> = 22), with >99% radiochemical purity, and up to 140 GBq/µmol molar activity. The precursor has now been validated for cGMP production by a third party, ABX. Mechanism-based testing was conducted in vitro. Physiologically-relevant concentrations of MPO and H2O2 could readily oxidize >95% of the reporter probe, significantly more than incubating in H2O2 or vehicle alone (p 95% of cell-associated retention (n> = 3 experiments, n=3-4 replicates per experiment p < 0 .0001), demonstrating selectivity for the reporter vs mitochondrial ROS and PMA-induced macropinocytosis or non-specific binding. A PMA model of mild contact dermatitis (earlobe and base of ear) was utilized to conduct pilot in vivo experiments. PET imaging of [18F] 4FN at 1 hr post injection of the radiotracer in vivo yielded good contrast-to-noise ratios in two independent strains of adult female mice (Balb/c and C57Bl/6N). The reporter was sufficiently robust that both IP and IV injections yielded images with good contrast ratios and large effect sizes (Cohen's coefficient » 0.7). Furthermore, in this model, [18F] 4FN yields superior contrast to [18F]FDG (p=0.004) . In a murine model of whole body toxic shock, there was increased retention in many organs, including the kidneys, and retention was significantly greater than 68Ga-citrate, another small molecule PET imaging agent for inflammation. Finally, in an LPS model of arthritis, [18F] 4FN correlated well with L-012, a validated bioluminescence reporter of ROS and activation of the innate immune system. Focal inflammation, swelling and immune infiltrate was cross-validated by immunohistochemistry. Broadly, [18F] 4FN demonstrated mixed renal and hepatobiliary excretion similar to other clinically translatable PET agents. In summary, [18F] 4FN could be readily synthesized with high molar activity, good yields, was stable in both buffer and mouse plasma and appeared to be a suitable PET agent for monitoring ROS produced by activation of the innate immune system in deep tissues. venous outflow from the kidneys while lactate in the PV predominantly reflects that produced from pyruvate in the small bowel and spleen. Discussion: Distinct vascular contributions can be observed from the liver's dual blood supply, despite the limited lifetime of the HP signal. For optimal HP 13C-pyruvate MRI, the timing of image acquisition is important. These results can help tailor real-time acquisition strategies, such as bolus tracking, to optimize HP 13C imaging of the liver.5 Acknowledgements: This work was supported by NIH grants NIDDK 5R01DK115987 and P41 EB013598. (2) Abstract Body : Liver is the most predominant site for metastases. In most cases, these liver metastases lead to poor prognosis. Therefore, non-invasive detection of liver metastases at early stages is critically important for many types of cancers. However, the sensitivity of imaging detection for liver micro metastases is limited due to lack of molecular biomarker and specific imaging methodology. We first report our finding that chemokine receptor 4 (CXCR4) serves as a biomarker for liver metastases from uveal melanoma (UM). This finding is confirmed by elevated CXCR4 expression in the liver metastases from UM patients. High CXCR4 expression level is further confirmed in liver metastases of UM mouse models, regardless of the expression level in the primary site. We then report the development and optimization of a protein-based MRI contrast agent with significantly improved relaxivities (r1 = 30.9 mM−1 s−1, r2 = 43.2 mM−1 s−1) and strong CXCR4 targeting capability. Taking advantage of molecular MR imaging and of this newly developed CXCR4 targeting contrast agent, we demonstrate that visualization of UM micro-metastases in the liver (of the mouse model), as small as 0.1 mm3, is now possible. The elevated CXCR4 expression in the liver metastases is associated with the specific liver microenvironment. Therefore, molecular MR imaging of CXCR4 is expected to have wide applications in studying liver metastases from different cancer types. We show detection of liver metastases from ovarian adenocarcinoma as an example to prove this concept. The imaging methodology fulfilled by CXCR4 targeted contrast agent is expected to fill the major gap in noninvasive early detection of liver metastases. CXCR4 is associated with uveal melanoma (UM) liver metastases, and detection of micro metastases in the liver by CXCR4 molecular MR imaging. A. High CXCR4 expression was observed in liver metastases in different UM murine models. UM cell lines with high, medium, and low in vitro expression of CXCR4 were used to generate metastatic UM mice models. The CXCR4 expression level in hepatic metastases are high in all three mice models regardless of the expression levels in vitro and in primary site. B. Detection of liver micro metastases by CXCR4 molecular MR imaging. After the administration of CXCR4 targeted contrast agent, the early stage micro metastases in the liver were enhanced and identified in MR image. The metastatic lesions in the H&E staining are well-correlated with the ones that recognized in the MR image. IHC staining confirmed the lesions are UM metastases (HMB45 positive) and with CXCR4 expression (CXCR4 positive). Abstract Body : Background: Stem cell therapy is new strategy for patients with ischemic heart disease. Rat heart micro-level visualization model of cardiac functions, offers the solution in clinical use. Hypothesis: Success of stem cell therapy depends on cell delivery and monitoring the stem cell rehabilitation and functioning in heart by non-invasive ultra-sensitive imaging method. Results: The stem cell repaired cardiovascular territories and reduced infarct related mortality. Non-invasive cardiovascular imaging monitors real-time status of cadiovascular remodeling or differentiated stem cell autografting, possibly, after myocardial dysfunction and recovery. Clear visualization of ventricular muscle fiber orientation, coronary arteries, apex functioning, ejection fraction facilitated evaluation of rejuvenation perspectives. However, the stem cell long-time retention and poor survival rates were major limitations to achieve significant therapeutic effects. Discussion: Right choice of pluripotent cells, stem cell delivery route, real-time monitoring of stem cell trafficking at transplantation sites, are emerging as promising tools. The stem cell molecular event monitoring during delivery, setting and functioning as myocyte cells, may be possible to dog watch the rejuvenation in clinical success. Conclusion: Non-invasive monitoring of stem cell trafficking by ultrahigh 900 MHz MR microscopy provides micro-level details of stem cell therapeutic efficacy in cardiac enlargement and differentiation. Abstract Body : Introduction: Cancer cells are known to remodel other extracellular matrix (ECM) components in the tumor and modify the mechanical properties of the tumor. These mechanical properties provide cues that influence critical cell activities, such as migration, proliferation and stem cell differentiation, which are part of the metastatic cascade in primary tumors [1] [2] [3] [4] . Mechanical properties of tumors studied previously, focused on bulk indicators using bulk tissue compression testing, or cellular mechanics at sub cellular level using atomic force microscopy. However these do not sufficiently address the heterogeneous nature of the tumor. We have therefore, for the first time, developed a micro-indentation method that bridges the gap between the bulk and sub-cellular scale techniques to understand the role of hypoxia in tumor stiffness, by characterizing changes in hypoxia inducible factor (HIF)-1α silenced tumors. Methods: Generation of MDA-MB-231 cells expressing shRNA against HIF-1α (231-HIF-1α) or an empty vector control (231-EV) using lentiviral transduction were performed using previously established methods [5] . 231-EV (n=8) or 231-HIF-1α (n=8) tumors derived from these cells were excised, sectioned and tested using a iNano nanoindenter (Nanomechanics, Inc.) equipped with a 500 μm diameter flat cylindrical probe. A grid based on the size of each section was mapped to cover the maximum area, and the indentation tests were carried out at each of the grid points. Elastic modulus maps of tumor tissue were generated from the test results and the indentation sites were classified as viable, necrotic or interface based on visual observation. Subsequently, tumor sections were fixed in formalin and paraffin embedded. 5 µm-thick slices were obtained and immunostained for major ECM components and crosslinking enzymes such as fibronectin, laminin, COL1A1, lysyl oxidase (LOX) and alpha smooth muscle actin (α-SMA). H&E stained sections were used to annotate viable and necrotic regions. Fractional immunostained areas for each component and enzyme were quantified by computing the fraction of strongly stained pixels identified using Aperio software. Results & Discussion: Microindentation was used to investigate the effects of silencing HIF-1α on tumor tissue stiffness. The elastic modulus in viable tumor regions decreased significantly in 231-HIF-1α tumors compared to 231-EV tumors (Fig 1A-B) suggesting that structural changes occurred in the tumor ECM with HIF-1α silencing. The fractional fibronectin immunostained area decreased significantly in 231-HIF-1α tumor sections compared to 231-EV tumors ( Fig 1C) suggesting that decreased fibronectin density observed with HIF-1α silencing may contribute to the decreased tumor stiffness. The fractional α-SMA immunostained area also decreased in 231-HIF-1α tumor sections compared to 231-EV tumors but not significantly. An increase in the immunostained area for collagen-1, LOX and laminin within 231-HIF-1α tumors indicate a lesser role of these components and enzymes in determining tissue stiffness. Conclusion: HIF-1α silencing is known to inhibit proliferation, impair migration and reduce metastatic abilities of breast cancer cells [6] [7] . Our data suggest that HIF-1α also regulates mechanical cues in the tumor microenvironment.

Characterization of major ECM components and cross-linking enzymes suggest that fibronectin levels may be a major factor in contributing to tissue stiffness. Acknowledgement: Supported by NIH R35CA209960 and R01CA82337. Abstract Body : Adeno-associated viruses (AAVs), single-stranded deoxyribonucleic acid (ssDNA) encapsulated within a 25-nm capsid, are attractive gene delivery vectors. Recently, peptide insertion within key capsid regions has been shown to alter the AAV pharmacokinetics (PK) and result in enhanced brain-specific transduction and reduced off-target accumulation after systemic administration [1, 2] . In the past, AAV PK has been determined via fluorescent, luminescent, or positron emission tomography (PET) reporter gene imaging or classical qPCR. PET imaging of systemically-administered AAV capsids can uniquely offer non-invasive whole-body monitoring and facilitates the estimation of receptor binding kinetics. Here, we set out to develop a viral tracking method using inverse electron demand Diels-Alder reactions (IEDDA). Due to the small volume of AAVs injected (pmol), a multichelator was developed and applied to evaluate the PK and brain accumulation of AAV capsids which display brain tropism [3] . In particular, AAV9-PHP.eB (PHP.eB) and a variant with reduced peripheral accumulation (AAV9-ReB10 (ReB10)) were studied and compared with AAV9. AAV labeling chemistry with IEDDA reaction. Surface lysines were modified with tetrazine-NHS ester (Tz-NHS). The purified Tz-AAVs were reacted with 64Cu pre-labeled multichelators composed of 8 chelator conjugates coupled with transcyclooctene ((NOTA)8-TCO). A 30 min IEDDA reaction afforded 64Cu-AAVs at a yield of 2~10% and radiochemical purity of >98% on instant thinlayer chromatography (ITLC) ( Figure 1A ). K557 and K567 were the predominant locations of the labeled lysines as determined by proteomic analysis. These two residues have not been reported to be directly involved in host receptor binding. The reaction of Tz-AAV with the fluorescent multichelator ((NOTA)8-AF555-TCO)) demonstrated that 3~5 multichelators were coupled with each capsid, and cryo-EM confirmed a similar number of labels on the surface of the capsid. PET/CT imaging and biodistribution of AAVs in C57BL/6 mice. For PK and biodistribution of the three AAVs, PET/CT images of C57BL/6 mice were acquired at 0, 4, and 21 hours after IV administration of 64Cu-AAV9 (421±25 KBq),64Cu-PHP.eB (628±581 KBq), and 64Cu-ReB10 (738±64 KBq) (n=3/group). Brain uptake of 64Cu-PHP.eB (17.0±4.1 %ID/g), and 64Cu-ReB10 (22.1±1.9 %ID/g) at 21 hours was significantly greater than that of 64Cu-AAV9 (0.16±0.02 %ID/g) ( Figure 1B ). While the liver uptake of PHP.eB (35.1±10.4 %ID/g) was significantly increased at 21 hours as compared to AAV9 (18.3±1.2 %ID/g, P=0.0409), liver accumulation of ReB10 (22.2±3.6 %ID/g) was comparable ( Figure 1C ). ROIand Logan analysis of AAVs. Blood circulation half-life of 64Cu-AAV9, 64Cu-PHP.eB, and 64Cu-ReB10 was 5.0, 4.1, and 3.1 hours, respectively. A reversible accumulation model (Logan plot) demonstrated that the 30 min accumulation in the brain after administration was greater for PHP.eB>ReB10>AAV9, with a distribution volume of 0.2, 0.1, and 0.01, respectively ( Figure 1D ). Thus, the affinity of PHP.eB and ReB10 to brain endothelium was estimated to be 20-and 10-fold higher than for AAV9, respectively. In conclusion. PET imaging of AAVs with a multichelator mapped early receptor-mediated binding and longitudinal accumulation over 21 hours. PET identified differences in target and off-target accumulation of peripherally detargeted AAVs. Abstract Body : Personalized cancer medicine strives to prolong treatment efficacy with a variety of molecularly targeted therapies aimed at genetic vulnerabilities in malignant cells. Importantly, deregulation of kinase function in cell signaling pathways has been implicated in numerous cancers leading to the development of tyrosine kinase inhibitors (TKIs) to interact with these kinases for highly specific treatment. Though nearly 50 TKIs have been FDAapproved, TKI monotherapy is seldom curative, largely owing to tumor heterogeneity and acquired resistance. Unraveling the complexities of acquired resistance is challenging because it is a complex interplay of drug distribution and target engagement as well as multifaceted and dynamic interactions between the tumor and its microenvironment. Additionally, intra-tumoral heterogeneity ensures treatment of sensitive subpopulations, while simultaneously promoting the outgrowth of resistant "persister cells." Efficacious therapeutics capable of treating heterogenous tumors requires robust drug target engagement (DTE) in the complex setting of the diseased tissue, where outcomes are dictated by duration, completeness and cellular heterogeneity of drug target engagement. Major contributing factors to therapeutic failure are insufficient DTE, offtarget activity and cell signaling pathway reprogramming as a mechanism of acquired resistance. However, DTE and off-target activity are typically assessed by bulk sampling of plasma or tissue, yielding a heterogenous average, and thus lacking single cell resolution of drug distribution and target binding. There also remains a need for methods of measuring mechanisms of resistance via cell signaling pathway reprogramming. It is necessary to characterize cell signaling pathway reprogramming on the same cell-by-cell basis as DTE for an accurate measure of therapeutic response. Typically cell signaling is measured in cell lysates, which lack spatial context and conventional in situ immunostaining techniques lack the dimensionality to fully characterize therapeutic response. Therefore, no established technology exists to quantify TKI DTE, concomitant with local protein expression, while assessing tumor response heterogeneity. With the ultimate goal of improving TKI therapeutic efficacy, our group has developed a novel fluorescence imaging platform we call TRIPODD (Therapeutic Response Imaging through Proteomics and Optical Drug Distribution and binding). TRIPODD combines innovations made by our group to (1) develop protocols to fluorescently label TKIs that mimic the native drug, (2) advance a novel intracellular paired agent imaging (iPAI) platform to quantify DTE with these fluorescent TKIs, and (3) establish a highly multiplexed immunostaining strategy utilizing DNA barcoded antibodies, enabling in situ oligonucleotide conjugated antibody cyclic immunofluorescence (Ab-oligo cyCIF) imaging. iPAI uses the spectrally distinct, fluorescently labeled targeted and untargeted drug derivatives synthesized by our group to correct for untargeted uptake and facilitate quantitative assessment of DTE. Ab-oligo cyCIF is capable of generating extreme multi-parametric images for quantifying dephosphorylated and phosphorylated protein expression to quantify protein activation, expression, and spatial distribution to unravel mechanisms of therapeutic resistance. Crucially, iPAI and Ab-oligo cyCIF can be performed on the same tissue sample enabling spatially aligned in situ quantification of TKI target engagement and local protein expression. Herein, we demonstrate the complementarity of iPAI and Ab-oligo cyCIF imaging technologies to produce a high dimensional image dataset to quantify spatially resolved tissue DTE, epidermal growth factor receptor (EGFR) cell signaling and cell state, all with single cell resolution. The result is the only fluorescence imaging platform to simultaneously quantify and correlate therapeutic efficacy to the complex interactions of drug binding (iPAI) and proteomics (Ab-oligo cyCIF) at the single cell level. Fully realized personalized therapy will include curated drug selection based on interand intra-tumoral heterogeneity and how these dictate therapeutic response and efficacy for which TRIPODD meets a critically unmet analytical need. Abstract Body : Cyclooxygenase (COX) initiates the production of prostaglandins by catalyzing the rate limiting step through its cyclooxygenase and peroxidase active sites.1 The prostaglandins are implicated in regulation of physiological processes and immune responses in health and diseases. There are two isoforms2 of COX: COX-1 is constitutively present in most cell types whereas COX-2 overexpression is induced by inflammatory and proliferative conditions associated with inflammatory diseases,3 neurodegenerative disorders4 and cancer.5 However, COX-2 levels are indirectly measured by mRNA quantification, western blot analysis, or downstream product quantification (e.g., ELISA assays) these methods are not suitable for determining the effect of local environment on COX-2 activity. COX inhibitors are commonly used in clinics as expansive class of NSAIDs. But it has been difficult to determine the activity of isoform selective COX towards inhibitor development and furthering our understanding of its biological roles. In recent years isoform selective activity-based sensing approaches are emerging as a frontier area in deciphering enzymatic activities at the molecular level.6 This strategy was exploited to develop isoform selective COX-2 probe -CoxFluor to detect its activity in live cells by confocal imaging and flow cytometry.7 It was hypothesized that larger pocket size of cyclooxygenase portion of COX-2 (vs COX-1) can easily accommodate bulkier groups.8,9 So, Arachidonic acid (natural substrate of COX-2) was strategically appended to a reporter dye (Resorufin) through an amide coupling such that the CoxFluor is fluorescently inactive but upon reaction with cyclooxygenase part it forms CoxFluor-PGG2 intermediate which upon subsequent processing by peroxidase portion of the enzyme releases the active dye resorufin (fluorescent). Furthermore, CoxFluor's selectivity was demonstrated in presence of 10fold excess functionally similar enzymes (lipoxygenase, peroxidase, catalase, etc.) and biologically relevant reactive oxygen, reactive nitrogen or other oxidant species. Additionally, by using molecular dynamics calculations it was shown that the presence of large cyclooxygenase binding pocket in COX-2 allows for the ease in CoxFluor's binding compared to COX-1 thus accounting for isoform selectivity. Moreover, CoxFluor enabled interrogating COX-2 activity dependencies on local oxygen concentration. It was hypothesized that since oxygen is a substrate in the reaction its concentration may influence the activity of COX-2 either through direct binding to the natural substrate in the reaction or attaching to the prosthetic heme subunit.10 Thus, on incubating LPS stimulated RAW 264.7 macrophage cell lines under normoxic and hypoxic conditions with CoxFluor it was clearly demonstrated that COX-2 activities changed without any change in enzyme level by flow cytometry and ELISA, respectively. Abstract Body : Introduction: Pancreatic ductal adenocarcinoma (PDAC) is often accompanied by weight loss caused by cachexia. In pancreatic cancer, the syndrome affects nearly 80% of patients and is a major cause of morbidity and mortality 1. A major characteristic of cachexia is accelerated skeletal muscle and fat storage wasting causing nutrient mobilization both directly as lipids and amino acids, and indirectly as glucose derived from the exploitation of liver gluconeogenesis that reaches the tumor through the bloodstream2 causing significant metabolic dysregulation. Early and reliable detection of the onset of cachexia are important to initiate early interventions. Here, we analyzed metabolites in human plasma from PDAC patients with weight loss and with no weight loss using high resolution MR spectroscopy to identify plasma derived metabolic biomarkers of cachexia. These metabolic biomarkers can also play a role in designing and monitoring response to nutritional interventions. Methods: Plasma from 49 patients with PDAC were included in this study under an approved IRB protocol. For 1H MRS analysis, plasma samples were thawed and homogenized using a vortex mixer. Then 350 μL of D2O phosphate buffer saline (NaCl 0.9% in 90% D2O) was added to 250 μL of plasma and mixed. After centrifugation (12000 rpm, 5 min), each sample was transferred to 5 mm NMR tubes and high-resolution 1H MRS was acquired on an Avance III 750 MHz Bruker MR spectrometer using a single pulse and a Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence with water suppression. To investigate the effect of cachexia induced weight loss on plasma metabolic patterns, PDAC patients were divided into four groups (I) PDAC patients with no weight loss history (n=17) (II) with weight loss of 5-15lb (n=13), (III) with weight loss between 16-30lb (n=11), and (IV) with weight loss of more than 31lb (n=8). Results and Discussion: We found that PDAC patients with higher weight loss showed significant plasma metabolic changes compared to PDAC patients with no weight loss. PDAC patients with more 30 lb weight loss showed dramatic metabolic differences when compared with PDAC patients with no weight loss ( Figure 1A ). Total lipids, branch chain amino acids, polyunsaturated fatty acids, alanine, glutamine, glucose, lactate, acetate, pyruvate, beta-hydroxy butyrate were significantly altered in PDAC patients with the most weight loss ( Figure 1A ). The other two groups with a lower weight loss also showed similar trends. We investigated the correlation between mean weight loss of each group with plasma metabolite levels. Lipid (r2 = 0.88), glucose (r2 = 0.7), creatine (r2 = 0.87), lactate (r2 = 0.75) and glutamate (r2 = 0.55) showed a strong negative correlation with weight loss ( Figure 1B ). These results provide the groundwork for plasma metabolite analysis to detect the onset of cachexia and changes with nutritional intervention, and to identify metabolic targets to arrest the progression of PDAC induced cachexia. Acknowledgement: Supported by NIH R01CA193365 and R35 CA209960. Figure 1A : Metabolic heat map showing differences in the metabolic profile of plasma from PDAC patients with no weight loss (n=17), weight loss between 1-15 lb (n=13), 16-30lb (n=11) and 31-70 lb (n=8). The maximum level of a metabolite present in a group was set to 100% to which the corresponding levels in the other groups were normalized. Student t-test was performed and a P-value less than 0.05 was considered significant. (*P-value Figure 1B : Linear regression was performed to investigate the correlation between metabolite levels in plasma of PDAC patients and weight loss. Mean weight loss and mean metabolites concentration were used to determine the correlation. PDAC patients with no weight loss (n=17), weight loss between 1-15 lb (n=13), 16-30 lb (n=11) and 31-70 lb (n=8).

First Name: Santosh K Last Name: Bharti Email: sbharti1@jhmi.edu

Yoko Yamakoshi, ETH Zurich, yamakoshi@org.chem.ethz.ch

Abstract Body : A synthetic lipid derivative with potassium acyltrifluoroborate (KAT) moiety was mixed with commercial lipids to form a lipid nanoparticle with KAT functional group on their surface. This chemically active KAT moiety was used for the reaction with hydroxylamine (HA) derivatives of biomolecules and imaging probes viainstant amide-forming ligation reaction (KAT ligation) [1] worked under physiological conditions with diluted concentration of reactants. In this study, we synthesized a lipid nanoparticle functionalized with an apoB100-mimetic peptide, Gd-chelate, and sulforhodamine B and subjected it to in vivo MRI and ex vivo cryo fluorescent imaging on atherosclerotic apoE-/-mice Lipid nanoparticle (LNP) was initially prepared by self-assembly of lipid materials (phosphatidyl chorine, cholesteryl oleate, triglyceride, and a synthetic oleic acid KAT derivative (OA-KAT)) by sonication and extrusion in buffer [2] . Prepared LNP (LNP-KAT) was stably-dispersed with a diameter of ca. 50 nm, and shown to be a mixture of particles with lipid monolayer and bilayer as observed by cryoTEM ( Figure 1a ).HA derivatives of functional moieties such as (1) LDL receptor-binding peptide motif in apoB100 (apolipoprotein of LDL) (P), (2) Gd(DO3A-monoamide) (Gd), (3) fluorescein (F), and (4) sulforhodamine B (R) were synthetized. By mixing the LNP-KAT with these HA derivatives (1.2 equiv in a concentration of 0.12 mM in total) in pH 5.3 buffer, functionalized LNPs were synthesized in high yield of surface functionalization. No aggregation or degradation of NP was observed. Initial in vitro cell assay of an LNP carrying peptide (P) and fluorescein (F) (PF-LNP) was carried out on monocyte-derived cell line (THP-1, known to express the LDL receptors) in comparison to F-LNP (without peptide) indicating the effect of peptide to enhance the accumulation of LNP into cells presumably viaLDL receptor-mediated mechanism (Figure 1b).The LNP (PGd-LNP) carrying peptide (P) and Gd-chelate (Gd) was tested by in vivo MRI on atherosclerotic apoE-/-mice. All of three tested mice, which were received PGd-LNP viatail vain, clearly shown the enhancement in aortic arch especially in the area of brachiocephalic artery (BA) in 48 hours post-injection and corresponding Gd accumulation by ex vivo ICP-MS analysis. To investigate more detailed biodistribution of LNP, which correlated to the in vivo enhancement, the triply labeled LNP (PGdR-LNP), carrying peptide (P), Gd-chelate (Gd), and sulforhodamine B (R) was synthesized. By MRI in 24 hours post-injection, two atherosclerotic apoE-/-mice, which were injected with PGdR-LNP, showed significant enhancement in BA together with nearby thymus, where macrophage cells are largely existing ( Figure 1c ). Mice were subjected to the ex vivocryo fluorescence imaging analysis. As results (Figure 1d ), clear correlation of MRI enhancement and fluorescence emission was observed in both mice, indicating that the enhancement observed in MRI was directly due to the high distribution of PGdR-LNP efficiently accumulated in atheroplaque in BA and thymus.From these in vivo andex vivo results, it was suggested that PGd-LNP and PGdR-LNP were accumulated into the atheroplaque presumably by the help of LDL receptor overexpressed in atheromas in a similar to our previous results with natural LDL-based MRI-CA on the enhancement of atheroma by MRI in an atherosclerotic mouse model [3] . Furthermore, the enhancement of thymus in current study may be due to the accumulation of LNP through non-LDL receptor-dependent mechanism and presumably by the help of scavenger receptor. Further in vivo studies of LNPs with oxidized peptide tag and without peptide are currently going on to investigate these mechanisms. Abstract Body : Background In recent years, the blockade of immune checkpoint molecules with monoclonal antibodies, like those targeting the PD-1/PD-L1 pathway, has enabled the development of breakthrough therapies in oncology, leading to delayed tumor growth and increased survival. PD-1/PD-L1 expression determined by immunohistochemistry however, correlates moderately with patient survival and response to immunotherapy. PET imaging with radiolabeled anti-PD-1 antibody may allow whole-body detection of PD-1 at high sensitivity and resolution and reveal expression heterogeneity within tumors. Aim Aim of this study was to develop an immunoPET tracer for imaging PD-1 expression using a checkpoint-blocking antibody with proven antitumoral activity. Methods The novel PET tracer was developed by conjugation of anti-PD-1 checkpoint-blocking antibody with the chelator NOTA and labeling with the radioisotope Ga-68. We conjugated the 1,4,7-triazacyclononane-N,N',N''-triacetic acid (NOTA) chelator to Nivolumab (BMS-936558, ONO-4538, or MDX1106, trade name Opdivo; Bristol-Myers Squibb, Princeton, NJ, USA), the first-in-human immunoglobulin G4 (IgG4) PD-1 immune checkpoint inhibitor antibody. The buffer of Nivolumab (10 mg/ml) protein solution was readily exchanged by loading the protein onto a Biorad micro Bio-Spin 6 column, preequlibrated in Sodium Phosphate 0.1M, pH 8.5; p-SCN-Bn-NOTA powder was dissolved in the same buffer at a concentration of 10mg/ml. The mixture of p-SCN-Bn-NOTA and Nivolumab was incubated overnight at 4°C. Then the complex was radiolabelled with 30-50 MBq of Ga-68 in 500 µl of sodium acetate buffer (pH 5.1) at room temperature. The resulting [68Ga]NOTA-Nivolumab was purified using a Biorad micro Bio-Spin 6 column and a sample of the radiopharmaceutical product was tested for quality controls to determine Radiochemical Purity using instant thin layer chromatography (ITLC) and high pressure liquid chromatography (HPLC Abstract Body : INTRODUCTION Due to its overexpression in a variety of tumor types, the chemokine receptor 4 (CXCR4) represents a highly relevant diagnostic and therapeutic target in nuclear oncology. Recently, new imaging probes targeting CXCR4 have been developed for PET imaging of several different hematologic and other neoplasms including leukemia, lymphoma, multiple myeloma, adrenocortical carcinoma or small cell lung cancer and also in other solid tumors and disease conditions, such as splenosis, stroke, atherosclerosis, and myocardial infarction in humans and in animals. Among CXCR4-directed imaging agents, Pentixafor labeled with Gallium-68 (Ga-68) has shown a unique position, because of its elevate affinity and selectivity to CXCR4, its low unspecific binding and adequate distribution profile accompanied by quick renal excretion. Preparation conditions may influence the quality and in vivo behaviour of tracers, especially considering clinical grade compounds, and no standard procedure for the quality controls (QCs) of [68Ga]-DOTA-Pentixafor is available, neither a specific monograph in the Pharmacopoeia does exist. So that, high pressure analytical purity (HPLC) method has become crucial for identification/characterization of the final product, demanding higher resolution than standard thin layer chromatography (TLC) method. Before their use in clinical and research settings, analytical methods must be validated. AIM Aim of this study was to develop a new rapid and simple HPLC method of analysis for the routine QCs of [68Ga]-DOTA-Pentixafor to guarantee the high quality of the finished product before release. MATERIALS AND METHODS At first, we have validated the synthesis process of [68Ga]-DOTA-Pentixafor assessing on three consecutive batches tthe radiochemical purity (HPLC), radiochemical impurities as 68Ga+3 (HPLC and TLC), chemical purity (HPLC and gas chromatography, GC), pH (pH-strips), radionuclidic purity (principal γ-photon), 68Ge-content, Ga-68 half-life (γ-ray spectrometry/counting) and sterility/endotoxin assay (sterility test and LAL test). Validation of analytical HPLC method to determine the chemical and radiochemical purity of [68Ga]-DOTA-Pentixafor was carried out according to ICH Q2 (R1) and EDQM guidelines. HPLC analysis was performed on a Dionex Ultimate 3000 HPLC system (Thermo Fisher Scientific) equipped with a AcclaimTM 120 C18 column 3µm 120Å (3.0mm×150mm), usingwater (0.1% TFA) and acetonitrile (0.1% TFA) as mobile phases. The parameters assessed for the HPLC method validation were: specificity, linearity, precision (repeatability), accuracy and limit of quantification, assessed using a set of standard samples, prepared by serial dilution starting from a ''mother'' solution with the highest concentration of the analyte of interest (Pentixafor) (4, 3, 2.25, 1.25, 0.8, 0.50 µg/mL). RESULTS The mean radiochemical purity of the three consecutive synthesis runs was 99.96% (98.5% measured by ITLC). For the validation of HPLC method, excellent linearity was found between 0.5 and 4 μg/ml, with a correlation coefficient (R2) for calibration curves equal to 0.9995. The average coefficient of variation (CV%) resulted CONCLUSIONS The developed HPLC method to assess the radiochemical and chemical purity of [68Ga]-DOTA-Pentixafor resulted rapid, accurate and reproducible allowing routinely use of this PET tracer as diagnostic tool for imaging CXCR4 expression in vivo, also assuring simple and safe implementation of the produced radiopharmaceutical in the diagnostic activity. Abstract Body : Aim Successful treatment of patients with colorectal peritoneal carcinomatosis highly depends on complete surgical resection of all tumor tissue. Surgical outcomes can be improved by intraoperative imaging using a tumor-targeting antibody conjugated to a fluorophore and a radiotracer, enabling preoperative radionuclide imaging, real-time intraoperative fluorescence guidance and gamma detection. Carcino Embryonic Antigen (CEA) is a suitable biomarker for in vivo targeting of colorectal cancer, and CEA-targeted intraoperative fluorescence imaging has already been investigated in a clinical study in colorectal cancer patients [1, 2] . A CEA-targeted multimodal approach combining fluorescence imaging and radiodetection has been extensively investigated in a preclinical setting, but not in clinical trials [3, 4] . Here we report results of the first clinical study investigating the feasibility, accuracy and safety of CEA-targeted preoperative SPECT/CT and intraoperative fluorescence imaging and radioguidance in patients with colorectal peritoneal carcinomatosis. Methods This phase I single arm intervention study consists of a dose escalation with 3 dose levels (n = 5 per dose level). Patients receive an intravenous injection of the CEA-targeting tracer [111In]In-DOTAlabetuzumab-IRDye800CW. Four days after injection a thoracoabdominal SPECT/CT is acquired to determine intra-abdominal tumor load and to scan for extra-abdominal metastases. One or two days after imaging, surgical resection extended with fluorescence imaging and radioguidance is performed. After standard of care cytoreductive surgery, the peritoneal cavity is examined for residual disease with fluorescence imaging. Blood samples are drawn for safety and pharmacokinetic analysis and vital signs are recorded at every hospital visit. Resected tissue specimens are analyzed microscopically, immunohistochemically (H&E and CEA), and by gamma counting. Results/Discussion No study-related adverse events were seen in the first 10 patients after intravenous injection of 2 or 10 mg of the tracer. Fluorescence imaging was feasible for real-time visualization of peritoneal tumor deposits. Mean fluorescent tumor to background ratios of 1.7 ± 0.6 and 4.1 ± 1.5 were observed in the 2 and 10 mg cohorts, respectively. Preoperative SPEC/CT imaging revealed extra-abdominal lymph node metastasis not seen on conventional imaging modalities in two patients, altering the clinical management strategy. Ex vivo analysis of resected specimens revealed clear correspondence of NIRfluorescence and localization of CEA-expressing tumor cells. These preliminary data show that CEA-targeted multimodal imaging in patients with colorectal peritoneal carcinomatosis is feasible and safe. Abstract Body : Sub-5 nm ultrafine iron oxide nanoparticles (uIONPs) have been shown to exhibit dual T1 (bright) and T2 (dark) contrast for magnetic resonance imaging (MRI) duo to their lower coupled magnetic moment that leads to the increased ratio of longitudinal relaxivity r1 and transverse relaxivity r2. While the composition [1] , size [2] , and aggregation [3] of this class of iron oxide nanoparticles have been generally investigated for improving the dual contrast effect and its applications, there is limited research done in understanding the controlling factors of coating materials on r1 relaxivity. We have previously developed oligosaccharide coated uIONPs by in situ polymerization of glucose on the particle surface [4] , and found that oligosaccharide coating can be modulated for the thickness and hydrophilicity, which in turn may affect the MRI contrast effects. In this work, we report the investigation on further improving the T1 contrast effect by altering oligosaccharide coating applied to uIONPs to adjust the r2/r1 ratio with the objective to better understand the molecular exchange process between the bulky water and hydration of the surface that contributes MRI contrast behaviors of uIONPs. uIONPs with different oligosaccharide coating were prepared with in situ polymerization by altering the ratio of glucose and uIONPs. The results of transmission electron microscopy (TEM), dynamic light scattering (DLS) and zeta potential measurements show that the prepared water-soluble uIONPs with different oligosaccharide coatings have good uniformity and monodispersity ( Fig. 1a) . With the increase of the ratio of glucoses to Fe in the surface polymerization, the hydrodynamic sizes of water-soluble uIONPs increased (Fig. 1b) . The hydrophobicity of the oligosaccharide surface coating was determined by using the hydrophobic molecule pyrene as a reporting probe [5] . After mixing pyrene with different water-soluble uIONPs, it was found that the intensity ratio of the first emission band (at 374 nm) to the third band (at 385 nm) of the pyrene molecules, which is the indication of the hydrophobic properties of coating ligands on the surface, first gradually increased and then decreased as the function of the amount of glucose polymerized on the surface. Thus, the hydrophobicity of the surface gradually decreased and reached the lowest value when the ratio of glucose to Fe increased from 20 to 60. The relaxivities of different uIONPs were measured at the field strength of 3T using a multi-TE spin echo sequence to measure T2 relaxation times and an inversion recovery sequence varying the inversion times to measure T1 relaxation times. The T1 and T2 values at different concentrations of uIONPs were obtained by fitting the signal intensities to the inversion times or echo time using the equation of S1=S0×1-2×e(-Ti/T1)or S1=S0×e-TE/T2 (Fig. 1c) . Then, fitting the obtained T1 and T2 relaxation times with the iron concentration to obtain r1 and r2. The r1 and r2 measurements of the water-soluble uIONPs with different coating showed a dependence of the proton relaxivities on the ratio of glucose to Fe (Fig. 1d ). uIONPs coated with the glucose/Fe ratio of 60 (uIONPs-60) exhibited consistently higher relaxation rates than those coated at a lower glucose/Fe ratio. The increase of the glucose/Fe ratio (from 20 to 60) may play the role in the probability of replacing organic oleic acids on the surface of nanoparticles, and the exchange of water molecules between bulk aqueous phase and magnetic cores. Consequently, the highest r1 relaxivity (4.11 mM-1s-1), comparable to that of Gd-DTPA (Magnevist, 4.22 mM-1s-1), is achieved with uIONPs-60, suggesting the replacement of oleic acids on the surface of the nanoparticles is at the highest level with the reported coating materials and coating method. Whereas, the further increase of glucose/Fe ratio to 200 can only overload oligosaccharide

References: [1] N. Xiao, W. Gu, H. Wang, Y. Deng, X. Shi, L. Ye, T1-T2 dual-modal MRI of brain gliomas using PEGylated Gd-doped iron oxide nanoparticles, Journal of colloid and interface science, 417 (2014) Abstract Body : Background: Vitamin B6 or pyridoxine, in its phosphorylated aldehyde form (pyridoxal 5'-phosphate or PLP) is critical for intermediary metabolism [1] , with pyridoxal being the most predominant of the few aromatic aldehydes normally present in our bodies. An intracellular pyridoxal kinase (PDXK)-pyridoxal phosphatase (PDXP) system controls the intracellular pool of PLP available for metabolism [2] . It was found that low PDXK and high PDXP activity conferred cancer cell resistance to therapy [3] . Importantly, non-small cell lung cancer patients with low PDXK expression (PDXK Methods: The CEST-MRI contrast agent, 2hydrazinonicotinic acid (2-HYNIC), was synthesized from 2-chloronicotinic acid and hydrazine in moderate yield(49%) and sufficient quantity (2 -3 g) [5] . A previously reported CEST-MRI contrast agent preferentially mapping aliphatic aldehydes (MeONA3) was synthesized as previously reported [6] . Hydrazones of 2-HYNIC and MeONA3 were prepared with acetaldehyde, 2-sulfobenzaldehyde, pyridoxal, and pyridoxal phosphate and purified by precipitation. Solution phase Z-spectra were recorded on a 300 MHz Bruker Avance II NMR. CEST-MRI maps were acquired on an MRS 3000 3T MR Solutions system using an EPI protocol, and maps were generated as previously published [7] . PDXK and PDXP expression was confirmed in human small cell lung cancer cell lines (A549 and H460) by Western blot. CEST-MRI was performed on cells treated with 20 mM 2-HYNIC or MeONA3 ± 5 mM pyridoxine. Results: By exchanging the core of our first-generation anthranilic acid-based aldehyde-conditional CEST-MRI contrast agents for nicotinic acid, we have successfully tuned the preference from aliphatic aldehydes to aromatic aldehydes (Fig. 1A ). Proton NMR experiments and molecular modeling calculations suggest that 2-HYNIC favours a planar hydrazone conformation for aromatic aldehydes, while anthranilic acid-derived compound (MeONA3) favours a planar hydrazone conformation for aliphatic aldehydes (Fig. 1B) . Critically, the planar hydrazone conformation is associated with greater CEST-MRI signal [6] , putatively underlying the difference in aldehyde selectivity of these two contrast agents. When applied to pyridoxal vs. PLP sensing, both MeONA3 and 2-HYNIC produced low CEST-MRI contrast (%MTRasym Conclusions:2-HYNIC is a novel, aromatic aldehyde-reactive CEST-MRI contrast agents capable of reporting on the PDXK-PDXP balance that is emerging as a biomarker of lung cancer chemoresistance. Abstract Body : Introduction Gene editing using the CRISPR/Cas9 technology has been shown to be a powerful tool to interrogate the mammalian DNA. However, its application in postmitotic neurons has been challenging due to the size of the CRISPR-associated endonuclease SpCas9 and the low in vivo efficiency. We report a highly efficient knockdown (KD) of the Slc18a2 gene, encoding the vesicular monoamine transporter 2 (VMAT2) in vitro and in vivo using Cas9 from Staphylococcus aureus (SaCas9). Methods The KD efficiency of the sgRNA targeting the Slc18a2 gene was evaluated in vitro in rat primary neurons using surveyor assay and immunofluorescence. Next, AAV vectors for SaCas9 and sgRNA targeting Slc18a2 (n= 14) or LacZ (control, n= 10) were injected into the right substantia nigra of adult rats. Conclusions Our study pioneers the combinatorial use of CRISPR/Cas9 gene editing and multimodal imaging to disclose in vivo the direct link between specific genes and molecular changes observed in disease, and their impact on the whole brain functional connectivity. Using CRISPR/Cas9, we targeted the Slc18a2 gene, engineering the adult rat brain, and reproducing several motor symptoms of Parkinson´s disease. (MSOT) is an emerging noninvasive imaging modality that can detect real-time dynamic information about the tumor microenvironment (TME) in animals and increasingly humans. Particularly, oxygen enhanced (OE)-MSOT can monitor tumor vasculature and oxygenation during disease development or in response to therapeutic treatment. It is a non-ionizing imaging technique exploiting the spectral sensitivity of optical imaging and the high spatial resolution and penetration depth of ultrasound to provide in vivo images with high temporal resolution. In particular, the well characterized endogenous absorption spectra of oxy-and deoxyhemoglobin (HbO2 and Hb, respectively) can be used to evaluate hemodynamics in blood-filled vasculature using a linear regression technique for multispectral unmixing of photoacoustic data. We used OE-MSOT to assess the dynamic changes of oxy-and deoxyhemoglobin following the administration of heme-sequestering peptide 2(HSP2) and cyclopamine tartrate (CycT) using a subcutaneous tumor xenograft model in mice. Methods: Human H1299-luc lung cancer cells were implanted subcutaneously in the flank region of NOD/SCID mice. Mice were randomized into treatment groups that received saline (for control), HSP2 (25 mg/kg every 3 days, in about 50 µl by retro orbital I.V injection), and CycT (by retro orbital I.V 7.5 mg/kg every 3 days) respectively for 3 weeks. For MSOT measurements, depilated anesthetized mice breathed air (21%O2) for 2 minutes followed by 100% oxygen for 5 minutes. In order to examine HbO2 and Hb mouse images were acquired in transaxial sections through the tumor region using seven wavelengths: 680, 715, 730, 768, 800, 850 and 900 nm with an MSOT InVision 256-TF device (iThera Medical, Munich, Germany). Illumination used 7 excitation fiber bundles and detection was based on 270 detectors in toroidal geometry around the mouse. A model-based reconstruction was used prior to multispectral processing. Following the final imaging, mice were sacrificed and tumor tissues were prepared for pimonidazole staining and CD31 staining, separately. Results and Discussion: MSOT reveals the spatial distribution of baseline Hb (blue), HbO2 (red), and total hemoglobin (THb) (Fig.1 A) . OE-MSOT measurements were applied with an oxygen gas breathing challenge to further elucidate the mechanism of action. Oxygen challenge significantly increased the sO2 levels and the amplitude of ΔsO2 response in HSP2-or CycT-treated tumors compared to control tumors ( Fig.1 B&C) . Segmenting the different depth-based concentric regions (Fig.2 B) reveals both THb and kinetic dynamic sO2 levels decrease monotonically with depth in Control tumors (Fig.2  A) .Confirming the OE-MSOT results, the exogenous hypoxia marker pimonidazole was significantly reduced in HSP2-and CycT-treated tumors (Fig.3 A&B) . Meanwhile, the levels of CD31 were significantly increased ( 

Juan Rojas, SonoVol, Inc., jrojas@sonovol.com

Abstract Body : Objective: The ability to see multiple components of cell therapy longitudinally in a preclinical cancer model is key to understanding complex mechanisms of action and therapeutic efficacy. This study illustrates visualization of intratumorally injected cells in live animals via the unique combination of automated, hands-free ultrasound (US) and enhanced dual-reporter bioluminescence imaging (BLI). The objectives were to demonstrate that integrated US and dual-reporter BLI can quantitatively assess a) tumor volume, viability, and growth and b) track cell therapy biodistribution and migration in an anatomical context. As in vivo models are advancing in sophistication, the need for in vivo multiplexing is on the rise. Dual-reporter imaging is attractive because it allows you to derive two signals from one model. However, luciferases other than Firefly luciferase (Fluc) and their respective substrates are less optimal in vivo, largely due to emission wavelength, solubility issues, and fast signal decay. Therefore, in addition to multimodal imaging, orthogonal luciferase/luciferin systems that perform well in vivo are needed to expand the number of channels available. To meet this need, we used NanoLuc luciferase and the newly developed hydrofurimazine (HFz) as our complementary bioluminescent system to Fluc/luciferin. Methods: Five C57Bl6/J mice were subcutaneously inoculated in their flank with 1M Fluc expressing KPC 4662 cells. After 15 days post inoculation, mice were imaged using a novel combined US/BLI instrument (Strata, SonoVol Inc.). First, animals were positioned in the lateral recumbent pose and US volumes were scanned robotically. Subsequently, mice were intraperitoneally (IP) injected with D-luciferin (Promega Inc.) at 150mg/kg and imaged with BLI after a 17 min post-injection waiting period (30 s exposure time). D-luciferin was then allowed to wash out for 2h. Next, 50,000 HEK 293 cells expressing NanoLuc (Promega Inc.) were loaded in a well plate and assayed with hydrofurimazine, HFz (Promega Inc.). Following in vitro confirmation of BLI signal, cells were injected intratumorally in each of the flank tumors of the mice to simulate cell trafficking. Finally, mice were IP injected with 480uL PBS solution of HFz formulated with poloxymer P407. BLI images were captured every 5 min for 70 min using a 1 sec exposure to illustrate the spatiotemporal localization of the injected cells. For quantification, US volumes were segmented with SonoEQTM software (SonoVol Inc.) in 3D to derive tumor volume, and BLI images were rendered with pseudocolor overlay in units of counts/sec. Results: US imaging demonstrated excellent 3D tumor boundary delineation; 15-day tumor sizes ranged from 52 to 248 mm3. BLI showed that both Fluc tumor cells and intratumorally injected Nanoluc cells can be detected with high sensitivity. Overlay visualization confirmed spatial co-localization, and when normalized per second, the signal of the NanoLuc cells was 2.5x stronger than the signal of the Fluc tumor. NanoLuc signal was graphed for 70 min to evaluate the kinetic curve of signal intensity over time. Time to peak was 35 +/-5 min, except for mouse #1394 with the smallest tumor (52 mm3).In this animal, the NanoLuc cells were observed to be migrating away from the tumor region of interest (ROI) towards a lymph node as identified by US (see figure) . This drainage caused a premature decrease in overall NanoLuc signal intensity in the tumor ROI. Conclusion: Integrated US/BLI imaging combined with dual orthogonal luciferases utilizing substrates (luciferins) with good aqueous solubility and bioavailability offer a novel approach for the quantification of cell and immune therapy. Future studies will include additional measurement channels for fluorescence imaging and microbubble-enhanced US.

Complete Status: Complete Data Augmentation was implemented on each training image to make it more robust to positional variations along with intensity normalization of the image. A Deep-Dense U-Net1 based CNN with skip-connections was implemented to perform automated tumor segmentation. The radiomics parameters2,3 were extracted by integrating the Radiomics Image Processing Toolbox4 in R along with in-house developed Radiomics pipeline written in Python. First order features were directly extracted from the intensity histogram while second order (GLCM) and the higher order (GLRLM, GLSZM) features were extracted using 32 quantization level. The segmentation performance was evaluated based on Dice Similarity Coefficient (DSC) and Jaccard Index. Sensitivity of radiomics parameters to segmentation boundaries was evaluated based on the mean Pearson Correlation Coefficient (r) value of the parameters with respect to delineation by experts. Results: The DSC between the automated algorithm and manual segmentation was 0.941±0.017, 0.942±0.008 and 0.938±0.13 for each of the three observers respectively. The sensitivity of radiomic features to segmentation boundaries was evaluated based on differences in delineation by manual segmentation. Forty-five extracted radiomics features were found statistically significant (i.e., p≤0.05) in correlating to all experts and were used to evaluate the correlation between the features extracted from the automatic segmentation to the manual marking. Out of the 45, 28 features showed significantly high correlation i.e. r ≥ 0.8. The first-order radiomics features correlated significantly to radiomic features derived by experts with r ranging from 0.76-0.99. Higher order features exhibited greater variability in correlation with experts with values of r ranging for 0.314-0.98. Conclusion: Deep learningbased segmentation performed accurate localization and segmentation of tumors in preclinical-MR imaging. However, despite the high agreement in segmentation, there is wide variability in radiomic features suggesting that the sensitivity of the feature to segmentation in a consideration in utility. The automated model mitigates inter-observer variability in radiomic parameters and make their analysis more robust and reproducible. Abstract Body : The identification of lymph node status is recognized as a critical prognostic factor in many solid cancer treatment and is essential to guide future adjuvant treatment [1] . Currently, surgical removal of lymph node followed by pathology examination is commonly performed as a standard-of-care to detect node metastases. However, the conventional pathology protocols are time-consuming and examine less than 1% of lymph node volumes by Hematoxylin and Eosin (H&E) staining, 5-μm-thick slices of lymph node taken at 2-mm intervals by pathologists. Using these routines, it is estimated that 20-60% of micrometastases (0.2-2mm diameter) [2] were undetected. In response, we developed a paired-agent imaging approach by employing a control imaging agent to allow rapid, quantitative mapping of microscopic populations of tumor cells in lymph nodes to guide pathology sectioning to reduce the false negative rate. By employing paired-agent imaging strategy, a recent study demonstrated the ability to detect fewer than 200 cancer-cells using a wide-field non-invasive imaging of breast cancer spread to axillary lymph nodes in a human breast cancer xenograft model [3] . To test the suitability of the staining/rinsing protocol for the pair-agent imaging approach, a targeted and control imaging combination was used to evaluate the potential for the agents to diffuse into and out of swine lymph nodes. IRDye800-Aby029, an anti-EGFR affibody labeled with IRDye800-CW (LICOR) as targeted imaging agent and IRDye700-DX (LICOR) was hydrolyzed as control agent. A green fluorescence protein transfected human cancer cells, MDA-MB-231-GFP, known to overexpress epidermal growth factor receptor (EGFR), were cultured as tumor spheroids with 250 µm in diameter and implanted in swine lymph nodes to mimic the clinical condition of cancer spread to tumor draining lymph nodes. Micrometastases and tumor-free control nodes were compared to evaluate the diffusion of both agents into and out of intact nodes. Lymph nodes were immersed in saline solution to remain moisture and infused via two needles by co-administrating both targeted and control imaging agent at a rate of 0.3mL/min controlled by a syringe pump for 1 min, followed by 5 min rinsing with PBS at the same infusion rate. Subsequently, lymph nodes were frozen-sectioned as 200 um thick slice and imaged under an 80-um resolution fluorescence imaging system (Pearl, LICOR) to evaluate the spatial distribution of both agents in the entire lymph nodes. Fig.1 demonstrates cross-sectional fluorescence signal of both control and targeted imaging agents in the pig lymph node. Neither the targeted nor the control agent fluorescent maps could be used to identify the absence or presence of tumor. However, ratioing the targeted/control images to estimate binding potential [4] (unitless measure of receptor concentration) allowed tumor-bearing from control nodes to be separating perfectly: measured binding potential of targeted agent between micrometastases and control nodes were 1.33 ± 0.47 and 0.05 ± 0.64 respectively. These finding demonstrates the promise of using paired-agent approach by two needle infusion staining and rinsing protocol to significantly improve the sensitivity of cancer detection for breast cancer sentinel lymph node biopsy. Abstract Body : OBJECTIVE: The primary objective of this prospective study was to determine the sensitivity of 68Ga-PSMA-11 PET/CT in detection of prostate cancer in a biochemical recurrence (BCR) setting on a per-patient basis confirmed by histopathology/biopsy, conventional imaging, and/or follow up. METHODS : A total of 91 subjects with confirmed prostatic adenocarcinoma based on histopathology and biochemical recurrence following treatment were enrolled in this prospective study. The Gleason score ranged from 6(3+3) to 10(5+5). 7 subjects had a Gleason score 3+3 (7.7%), 20 subjects had a Gleason score 3+4 (22%), 22 subjects had a Gleason score 4+3 (24.2%), 17 subjects had a Gleason score 4+4 (18.7%), 16 subjects had a Gleason score 4+5 (17.6%), 6 subjects had a Gleason score 5+4 (6.6%), 2 subjects had a Gleason score 5+5 (2.2%) and 1 subject with an undefined Gleason score (1%). The median age of all enrolled subjects was 66 years (range: 52 to 88 years). All subjects received 3 to 7 mCi of 68Ga-PSMA-11 (average 4.7 mCi). PET/ CT imaging was performed 60 ± 15 minutes after injection. For subjects with a PSA, less than 2 ng/ml an additional late scan at 180 minutes post-injection was also done. PET/CT images were read by an experienced nuclear medicine physician aware of clinical data. Evaluation of the results of the 68Ga-PSMA-11 PET/CT scan was compared to other available conventional anatomical and functional imaging modalities. This study received FDA authorization and approval from Biomedical Research Alliance of New York Institutional Review Board (BRANY IRB). RESULTS : All the images were of high quality. For subjects with a PSA less than 0.2 ng/ml the detection rate was 43%, with a PSA between 0.2 ng/ml to 0.5 ng/ml was 53.3%, with a PSA between 0.5 ng/ml to 1.0 ng/ml was 60.8%, with a PSA between 1.0 ng/ml to 2.0 ng/ml was 72%, with a PSA between 2.0 to 5.0 ng/ml was 100% and with a PSA of more than 5.0 ng/ml, the detection rate was also 100%. The average detection rate of 68Ga-PSMA-11 PET/ CT scan was 78. Figures  1a)-b) . This filar-type architecture is based on the use of copper microwires (diameter of 150 µm). In Figure 1a ), the copper wire is bent to form an elliptical-shape loop with typical dimensions of 3 mm for the long axis and 200 micrometers for the short axis. In Figure 1b) , the twist microcoil consists of two strands twisted over a length of 3 mm. A constant pitch of 500 micrometers was used to twist the two strands resulting in 6 identical windings. The microcoils were matched and tuned to the proton resonance frequency using extra-corporeal variable capacitors. NMR experiments were performed at 7 and 17.2 T. NMR spectra and MR images were acquired using external coils for RF excitation. For in vivo experiments, cannulae were stereotaxically positioned the day before the insertion of the NMR microcoils in the cortex of male wistar rats. Results/Discussion The quality factor of the loaded coils was larger than 100. The detection volume of the microcoils, derived from RF simulations and from the MR images were measured to be 500 nL for the twist coil and 250 nL for the loop coil. For in vitro results, the linewidth of the water peak (full width at half maximum) was 6 Hz at 7 T. Figure 1c) shows a typical 1H MRS spectrum (PRESS sequence) obtained in vivo from the rat brain using a loop microcoil. In this particular example of water-suppressed acquisition (512 averages, 17-minutes acquisition) at 7 Tesla, main resonances of brain metabolites (NAA, Glu, Gln, Cr, Cho, etc) can be easily identified and quantified. Compared to an acquisition with a conventional external RF coil, the gain in sensitivity was found to be on order of 1500. Conclusions The MRS/MRI results obtained in vitro and in vivo illustrate the relevance of the microcoil design with respect to spectral resolution, detection sensitivity, spatial selectivity and limited invasiveness. Foreseen applications include the investigation of metabolism in submicroliter volumes in physiological and pathological conditions with metabolic dysfunctions (tumoral environment, neurodegenerative pathologies, etc). Acknowledgment The study received financial support from the « Laboratory of Excellence » TRAIL ANR-10-LABX-57 (research program Insight) and from France Life Imaging (FLI).

Image/ Figure Caption: Figure 1 . a) implantable loop microcoil. b) implantable twist microcoil. c) NMR spectrum acquired at 7 Tesla in the rat brain using a PRESS sequence with water suppression. Total acquisition time : 17 minutes. Acquisition volume : 250 nL. Abstract Body : Introduction: Chemical Exchange Saturation Transfer (CEST)-MRI allows to investigates seeral aspects of tumor microenvironemt, including metabolite, proteins and acidosis. CEST-based tumor pH imaging is hampered by long acquisition times (due to long saturation pulses and multiple offsets sampling) that reduce both the volume coverage and spatial resolution capability, hence the information on tumor pH heterogeneity that can be obtained. At the preclinical level only few multislice CEST sequences have been implemented to solve this issue, but not yet optimal in terms of SNR [1] , geometrical distortions [2] and image artifacts [3, 4] . This work aims to improve both volume coverage and spatial resolution by implementing an uneven irradiation scheme, as in [1] , with a centric-reordered RARE readout, for a whole tumor pH imaging for assessing tumor acidosis heterogeneity. Methods: The proposed sequence is composed of a first long saturation period that induces the steady state CEST contrast and a second short saturation pulse repeated after each image acquisition to compensate signal losses.Since the slices are acquired just after the second short saturation pulse rather than following each long first saturation period, the total scan time for the whole volume is dramatically reduced. Different acquisition parameters (k-space acceleration factor, matrix size, slice scan order, saturation times and power, voxel size, slice thickness) were simulated and optimized. CEST images were acquired on a 7 T scanner at 310 K. Other MRI parameters were TR/TE:3.49/~12000 ms, Matrix size:64×64 and axial slice with a thickness of 1.5 mm.Multislice sequence accuracy and homogeneity to measure pH along the 3rd dimension was tested in vitro on a pH-varying iopamidol phantom. In vivo validation was performed by acquiring multislice pH images covering the whole tumors following iopamidol i.v. injection. Results: Figure 1 shows The observed calculated pH values were close to the experimental (pH-meter) ones along all the slices, suggesting that the pH measurement is not affected by the slice position. Moreover, mean calculated pH values were highly correlated to the experimental ones with low standard deviations, confirming the accuracy along the third dimension. The optimized fast multislice sequence was applied in vivo to obtain multislice 2D pH images (Figure 1d ). The 3D data set was exploited for 3D pH rendering of the whole tumors and to evaluate the pH gradients in the tumoral region (Figure 1e ). Conclusions: The proposed fast multislice CEST-MRI protocol was tested and validated in vivo allowing high volume coverage and pH sensitivity. The optimized multislice sequence can cover a whole tissue in less than 10 minutes, providing a new tool for 3D pH heterogeneity evaluation of whole tumors or organs both in physiological and pathological conditions. Acknowledgments: Funding: This work was supported by the European Union's Horizon 2020 research and innovation programme [1, 2] To precisely study the function of DDX24, a specific probe of DDX24 is needed. We provide a polypeptide probe with a highly specific binding to DDX24 and demonstrate its potential value in DDX24 related disease precision medicine. Procedures: We designed a polypeptide, LT(p)FEHYWAQLTS(p)-RRRRRRRR, a modified polypeptide targeting to the binding complex of DDX24/MDM2/p53 [3] [4] [5] . CHO cells transfected with the DDX24 or vector control were incubated with the fluorescent probe along with the scrambling control peptide. Tumor subcutaneous xenograft models of DDX24 expressing CHO or vector control cells were generated in balb/c nude mice. In vivo fluorescence imaging was carried out to assess the binding specificity for fluorescent probe. For radiolabeling, the NOTA-modified peptide probe was dissolved in NaOAc buffer and incubated with [64Cu]CuCl2. A C-18 Sep-Pak Plus cartridge was used to trap the product. The collected product was evaporated under gentle nitrogen flow in ice bath and dissolved in normal saline for other experiments. The radiolabeled product was tested on Radio-TLC. MicroPET/CT were performed in balbc-nu mice engrafted subcutaneously DDX24 (+) and DDX24 (-) CHO tumors xenografts. Results: When the designed FITC-probe was incubated with CHO cells transfected with DDX24-mcherry showed substantial fluorescence compared to control CHO cell. The peptide probe was located in the nucleolus area, while the control probe was scattered in CHO cells transfected with DDX24. Fluorescence imaging showed intense uptake of Cy5-peptide probe in xenografts at 8-12 hours post-injection and DDX24-mcherry transfected CHO cell xenografts showed substantial fluorescence compared to control CHO xenografts. For radiolabeling, the Rf value of the product after purification is 0.0. The peak of the 64Cu2+ was not found, the RCP (radiochemical purity) of product was more than 95%. The RCY (radiochemical yield) was 62%. DDX24+ tumor xenografts were clearly imaged by microPET/CT at 6 or 12 h p.i. of 64Cu-NOTA-LT(p)FEHYWAQLTS(p)-RRRRRRRR. Abstract Body : Sensitive detection of two biological events in vivo has long been a goal in bioluminescence imaging. Antares, a fusion of the luciferase NanoLuc to the orange fluorescent protein CyOFP, has emerged as a bright bioluminescent reporter with orthogonal substrate specificity to firefly luciferase (FLuc) and its derivatives such as AkaLuc. However, the brightness of Antares in mice is limited by the poor solubility and bioavailability of the NanoLuc substrate furimazine. Here, we report a new substrate, fluorofurimazine (FFz), with even higher brightness and better pharmacokinetics than furimazine in vivo. After establishing the superior performance of FFz, we demonstrated the potential of multiplexing of FFz with orthogonal ATP dependent beetle luciferase systems. We show that we can use Antares with fluorofurimazine to track tumor size and AkaLuc with AkaLumine to visualize CAR-T cells within the same mice, demonstrating the ability to perform two-population imaging with these two luciferase systems. has a favorable nuclear spin of ½ and a wide chemical shift range (900 ppm), yet 15N NMR is much less widely used as a tool in biological systems compared to 13C and 1H because of its poor sensitivity.3 However, 15N NMR in combination with hyperpolarization techniques such as dissolution dynamic nuclear polarization (d-DNP) or parahydrogen induced polarization (PHIP) is beginning to find its way into the biological realm. Hyperpolarization refers to technologies that enhance the NMR signal-to-noise ratio by amplifying nuclear spin polarization.4 Here we report a HP-15N based probe, 15N-enriched, d6-deuterated tris(2pyridylmethyl)amine (TPA) (Fig. 1) for the detection and quantification of total freely available Zn2+ in tissues. TPA is a tripodal ligand that has excellent selectivity for Zn2+ over other common biological cations. Most optical and Gd-based Zn2+ responsive agents contain a sensing moiety structurally derived from TPA.5-7 This compound is known to distribute across cell membranes6 so in principle should detect freely available tissue Zn2+ in all compartments and not just Zn2+ ions released from cells in response to a biological stimulus.7 It was hyperpolarized by microwave driven dynamic nuclear polarization (DNP)4 and demonstrated that the polarized ligand had favorable T1 and linewidth characteristics for 15N MRI (T1=70.9 s, at 9.4T). Chelation experiments with hyperpolarized 15N-TPA-d6 and Zn2+ ions revealed that the 15N NMR signal of 15N-TPA-d6 shifted 20 ppm upfield upon complexation. The 15N spectra also indicated that Zn2+ exchange between free ligand and 15N-TPA-d6-Zn2+ complex is slow on the experimental NMR timescale (Figure 1c ). The Zn2+ detection limit of hyperpolarized 15N-TPA-d6 was found to be around 5 µM Zn2+. Ligand protonation and Ca2+ binding did not interfere with Zn2+ sensing. 15N chemical shift images of phantoms containing HP 15N-TPA-d6 in the absence and presence of Zn2+ were collected to illustrate the feasibility of in vivo imaging of Zn-distribution (Figure 1d ). We also demonstrate how HP-TPA can be used to quantify freely available Zn2+ in human benign prostatic hyperplasia (BPH) tissue.

Cells were then administered therapy and validation studies were repeated to verify that the iPAI reagents were indeed able to predict cellular response, determined via simultaneous viability analysis. These in vitro studies provide a conduit to iPAI implementation in future in vivo work, where non-small cell lung cancer (NSCLC) models will be utilized to study both monotherapy and combination therapy that could provide a durable, long-term treatment response. Interactions between AML and the supporting BM microenvironment play important protective and promotive functions for cancer development1. A hallmark of the BM is its low oxygen partial pressure, which is required for long-term maintenance of HSCs2. While in solid tumours hypoxia has been associated with poor prognosis3, its role in leukemias is not yet fully understood due to the inaccessibility of the BM space for in vivo study. Recently, an intravital imaging mouse model was developed that makes it possible to view and monitor leukemia cells orthotopically within the femoral BM for a few hours a session, repeatedly, over the course of two months4. Here we applied the novel femur window chamber (FWC) technique to study the role of hypoxia on AML proliferation, maintenance, and therapy resistance, and how these effects may be mediated by the chemokine receptor CXCR4. Methods: To monitor cellular hypoxia, the murine myeloid cell line 32DKit was transfected with a fluorescent hypoxia reporter (GFP driven by five hypoxia response elements5). 32DKit cells were also modified to express varying levels of CXCR4 to study the relationship between the chemokine, hypoxia, and AML development. 1x106cells were intravenously injected into each group of 8 week old C3H/HeJ mice, which were then transplanted with the FWC two weeks post-inoculation. FWC installation involved shaving the femoral corticalis to reveal the BM cavity, then securing a cover glass on top4 (Fig. A) . Fluorescent cellular activity within the BM was monitored two times a week for four weeks via confocal microscopy (Zeiss LSM710). Leukemic burden was quantified as the percent total fluorescent signal within the BM space while levels of hypoxia were quantified as the ratio of GFP to mCherry signal above threshold intensities. Nonfluorescent leukemic and healthy mice were also imaged as controls. AML response to hypoxia-targeted treatment was evaluated using mouse groups administered with 10μg/kg of the antibiotic echinomycin6, intraperitoneally injected for five consecutive days one week post-inoculation. Results: Preliminary results validate the utility of our AML hypoxia reporter cell line in detecting varying hypoxic levels both in vitro and in vivo (Fig. B-C) . Interestingly, we observed that AML cells tended to engraft as (and then proliferate within) distinct cell clusters surrounding sinusoids within the vascular niche of the BM, with cells localized at the edges of the clusters being more hypoxic than those towards the center. Longitudinal monitoring of the BM further revealed leukemia-associated changes in the overall proportion of hypoxic cells within the microenvironment. Treatment with echinomycin induced a noticeable transformation in AML cellular conformation, altering the large cell clusters seen prior to treatment to a more individualized and diffuse cellular arrangement three weeks post-treatment. We further examined the influence of CXCR4 on leukemia progression given its role in normal hematopoietic cell homing, growth, and proliferation, which may in part be mediated by BM hypoxia. AML cells induced to overexpress CXCR4 were observed to result in higher leukemic burdens when compared to unmodified cells (Fig. D-F ). Discussion: Current efforts are now underway to clarify the role of hypoxia in activating AML CXCR4 by examining the direct spatiotemporal relationship between the two in vivo. Future studies will focus on targeting these factors using novel therapeutic agents, characterizing drug effect in real-time using the FWC, and examining the potential of anti-CXCR4 and hypoxia combinatorial strategies. Integration of molecular imaging into therapy development can provide invaluable biological readouts regarding treatment action in vivo. Our FWC model provides an innovative means by which to quantitatively study AML cell response to new therapies within the living BM for preclinical trials. Human angiotensin-converting enzyme 2 (hACE2) or the A-peptide of hACE2 from the binding interface, have shown specifically binding to S1 subunit in spike (S) protein of SARS-CoV-2 [1] . After binding of receptor-binding domain (RBD) in S1 subunit of S protein, the heptad repeat 1 (HR1) and 2 (HR2) domains in S2 subunit of S protein interact with each other to form a sixhelix bundle (6-HB) fusion core, bringing viral and cellular membrane into close proximity for fusion and infection [2] . Herein, we reported our effort on developing potent recombinant ACE2, A-peptide, or HR peptide as PET radiotracers for SARS-CoV-2. Method: The recombined protein ACE2-PD (binding to S1), A-peptide (binding to S1) and HR-peptide (binding to S2) were modified by biotin, fluorescent dye Cy5 and NOTA to facilitate in vitro study, fluorescence cell sorting and in vivo PET imaging respectively. The biotinylated peptides were carried out in vitro assays to detect if they can specifically bind to S protein.

The Cy5 labeled peptides were tested by FACS and cell immunofluorescence with S protein positive HEK-293 cells. S protein positive and negative HEK-293 cells were seeded in athymic nude mice to test the binding specificity of NOTA-modified peptides for S-protein of SARS-CoV-2 in vivo. The 64Cu-Labeled PET radiotracers including [64Cu]-NOTA-hACE2, [64Cu]-NOTA-A-peptide and [64Cu]-NOTA-HR-peptide were performed in vivo experiments. Results: The saturation binding revealed that biotin-ACE2 recombinant protein, biotin A-peptide and biotin HR-peptide specifically bound to the S protein with high affinity (Kd = 0.046 nM, 3.88 nM, 3.56 nM, Bmax = 2.0, 2.6, 3.0 (A450), respectively) ( Figure 1 ). The competition assay determined binding potency for S1-RBD recombined protein (ACE2-PD, A-peptide) or S protein full-length recombined HR-peptide, IC50 value of 3.09 nM, 10.3 nM, 15.7 nM respectively ( Figure 1 ). FACS revealed that these peptides specifically bind to the cells expressing S protein.Immunofluorescence showed that the A-peptide was stained around the cellular membrane of S protein positive HEK-293 cells. The PET imaging revealed that the 64Cu-A-peptide-NOTA can specifically bind to S-protein positive tumor in live animals ( Figure 1 ). Conclusion: Our data demonstrated that newly designed 64Cu-labled NOTA-peptides radiotracer specifically binds to S protein in vitro and in vivo. These peptide radiotracers represent a promising technique for imaging and quantification COVID-19 in clinic. Fig. 1C ). This uptake pattern was reproduced in an SK-MEL-3 human melanoma xenograft Foxn1nu mouse model. The tumor-to-liver ratio of [18F] DMPY2 was more than 4.5-and 7.6-fold higher than that of [18F] This stem cell option also repairs the myocardial infarction area in heart or vascular territories and ultimately reduces the infarct-related mortality. Materials and Methods: Non-invasive cardiovascular imaging monitors the real-time status of cardiovascular remodeling or differentiated stem cell autografting. Cardiac magnetic resonance imaging (MRI) and bioluminescence are robust non-invasive monitoring techniques to visualize cardiovascular structure changes due to myocardial dysfunction or restorative myocardial recovery. Results: The present study highlights the sources, types, delivery methods of stem cells in cardiovascular treatment, advantages and current limitations of stem cell monitoring, scopes of ultra-high field cardiac 900 MHz Bruker Biospin MRI and bioluminescence methods applied in stem cell transplantation, to translate stem cell molecular events into clinical success and evaluation of rejuvenation rate with future perspectives. At 900 MHz magnetic field, microscopy offers the images up to 15 micron spatial resolution of cardiac tributaries and myocyte behavior. The imaging method is capable of recording physiological functionality of beating heart of animal placed in magnet. Possibly, pluripotent cells are MRI visible. Discussion: The in vivo imaging techniques are useful in dynamic monitoring of cardiac stem cell therapy following myocardial infarction. Choice of stem cells and mode of delivery are very crucial in getting successful stem cell therapy positive outcome. Cardiovascular remodeling evaluation by MRI has merits because it is safe, sensitive, lacks radiation, provides good resolution, generates a real-time events' blueprint or first-hand information of myocardial viability with functional information of cardiac territories and their physiochemical changes in cardiac functions during stem cell rejuvenating process and after myocardial repair. Present time, ultra-high magnetic field CMR possibly has preclinical prospects as in vivo noninvasive molecular imaging or restorative monitoring reporter of rejuvenating stem cell genes to evaluate success of transplantation and cardiac repair. On the other side of coin, researchers are continuously developing new real-time physiological cum functional MRSI options to explore new stem cell molecular probes and smart MRS imaging sequences with improved MRI sensitive specific stem cell differentiation and rejuvenating detection by targeting energy metabolites, myocardial viability, and vital physiochemical molecules. Noninvasive monitoring is necessary and bioluminescence or other radionuclide methods may be alarming because the potential biological damage caused by radionuclide exposed reporter genes and bioluminescence induced immune responses is concern in differentiating stem cells. Seriously, all these issues need research to minimize artifacts within safe limits. With the help of stem cell imaging and monitoring, transplantation of stem cells sooner or later will be optimized for the effective long-lasting therapy of myocardial infarction and heart failure on some day. Conclusion: The right choice of stem cells, pluripotent stem cell delivery, transplantation and real-time monitoring of stem cell trafficking enhances the stem cell therapeutic efficacy in cardiac engraftment and differentiation. [2] This pitfall can be addressed by combining ICIs with oncolytic virotherapy (OV), an approach to immunologically pre-condition tumours for the onset of ICI treatment. [3, 4] However, the combination of ICIs and OV is still in the early stages of development and further research is required to better understand and fully exploit its potential. Hence, we investigate systemic expression of the immune checkpoint programmed death receptor ligand 1 (PD-L1) post-OV using immunoPET. An IgG2b-based anti-PD-L1 positron emission tomography (PET) radioconjugate (89Zr-DFO-αPD-L110F.9G2) was used to study spatiotemporal dynamics of PD-L1 expression after OV in a syngeneic mouse model of oral carcinoma. For OV, a single dose of a genetically engineered herpes simplex type 1 virus (RP1) was injected intratumourally. 89Zr-DFO-αPD-L110F.9G2 (2 MBq) was administered intravenously (~10 μg of radiolabelled co-injected with ~100 μg of unlabelled anti-PD-L1 antibody) and static PET/CT scans with concomitant biodistribution studies were performed on days 3 and 7 post-OV. PD-L1 immunohistochemistry (IHC) was used to confirm antigen expression levels in tumour and spleen samples. Multiparametric flow cytometry (FC) was used to assess expression of PD-L1 and closely related immune checkpoint receptors on CD45+ cells residing in tumours and spleens. Using PET, we found that RP1 injection significantly increased 89Zr-DFO-αPD-L110F.9G2 uptake in spleens and tumour draining lymph nodes 3 days after treatment. By day 7, 89Zr-DFO-αPD-L110F.9G2 uptake in these organs had returned to pretreatment levels. In contrast, tumours of treated mice showed no significant change in radioconjugate uptake. Levels of intratumoural and intrasplenic 89Zr-DFO-αPD-L110F.9G2 uptake were in line with PD-L1 IHC performed on representative sections of both organs. FC analysis showed a trend toward an increase in PD-L1+CD45+ tumour-infiltrating cells on day 3 but not day 7 after treatment, while PD-L2+CD45+ cells followed an inverse trend. Interestingly, PD-1 expressing CD45+ cells were elevated on day 7 after OV in tumours and spleens likewise. In summary, intratumoural RP1 injection leads to early, systemic but not intratumoural PD-L1 upregulation, as detected via anti-PD-L1 immunoPET. Forthcoming studies will assess the potential of 89Zr-DFO-αPD-L110F.9G2 to guide the onset of anti-PD-(L)1 ICI treatment post-OV. (1). This has led the researchers to promote the use of anti-inflammatory/antioxidant drugs as therapeutic strategies in the treatment of SCZ. In this sense, N-acetylcysteine (NAC) has gained special attention, since it increases the biosynthesis of the antioxidant glutathione by regulating the cysteine amino acid availability (2) . To date, NAC has been tested in patients with schizophrenia or first-episode psychosis as adjunctive drug to antipsychotics, but the results so far diverge between studies. Aim. To evaluate whether NAC administration during pregnancy may prevent neuroanatomic, biochemical and behavioral deficits in the maternal immune stimulation ( Conclusions. Our study demonstrates that the administration of NAC during pregnancy prevented the deficits in anxiety in the MIS model. Moreover, it could prevent the atrophy in thalamus and prefrontal cortex reported in schizophrenia (4, 5) , and also slightly prevent some metabolic brain changes, such as the deficit of N-acetyl-aspartate in hippocampus. Together, these results suggest that NAC could be useful as therapeutic strategy to prevent or halt the progression into a schizophrenia-like condition after infection-mediated neurodevelopmental disturbances. 

Nicolás Lamanna-Rama, Fundación para la Investigación Biomédica Gregorio Marañón, nlamanna@hggm.es

Abstract Body : Introduction: Oxidative stress (OS) and inflammation might play an important role on the pathophysiology of mental diseases, which suggest a potential benefit of antiinflammatory/antioxidant compounds in these pathologies [1] . Cannabidiol (CBD), a major nonpsychotropic constituent of cannabis, shows antiinflammatory/antioxidant properties, generating certain interest as a therapeutic strategy for mental disorders [2] . In this sense, CBD has been used as adjunctive therapy in schizophrenia, reducing positive psychotic symptoms and improving better cognitive performance [3] . However, its efficacy in earlier stages of the disease in patients at high-risk of psychosis has not been evaluated yet. Objective: To evaluate the effects of CBD treatment during adolescence on schizophrenia-related deficits in the maternal immune stimulation (MIS) rat model of schizophrenia at adulthood. Methods: Pregnant Wistar rats were administered with Poly I:C (4 mg/kg) or saline in gestational day 15. Male offspring were treated with CBD (10 mg/kg) or its vehicle (VH, ethanol: cremophor: saline; 1:1:18) i.p. from post-natal day (PND) 35 to PND50. Four groups were evaluated attending to the study factors: MIS condition (Saline, MIS) and treatment (VH, CBD) (N=12 per group). Prepulse inhibition test (PPI) was performed on PND70 to evaluate sensorimotor gating deficits. Single-voxel magnetic resonance spectroscopy (MRS) of hippocampus (Hipp) and prefrontal cortex (PFC), and T2weigthed brain images were acquired on PND100. MRS results were normalized to total creatine, and analysed by means of LC Model software. Volumetric brain changes were determined by conventional regions of interest (ROI) and voxel-based morphometry (VBM) analysis with SPM12. Data were assessed by means of 2-ways ANOVAs followed by Bonferroni post-hoc tests (p Results: MIS-offspring showed PPI deficits compared with saline-offspring and CBD showed a trend in the prevention of these behavioral deficits ( Fig.1.A) . ANOVA analysis of MRS data showed a treatment effect in glutamine + glutamate (Gln + Glu) and Nacetylaspartate + N-Acetylaspartylglutamate (NAA + NAAG) in the PFC; and in glycerophosphocholine + phosphocholine (GPC + PCh) and NAA + NAAG in the Hipp. A MIS effect was produced on macromolecule 20 + lipid 20 (MM20 + Lip20). Bonferroni post-hoc test showed that CBD increased Gln + Glu in the PFC of MIS-offspring, as well as NAA + NAAG and MM20 + Lip20 in the PFC of saline-offspring ( Fig.1.B ). In addition, we found a reduction in volume in the Hipp and PFC, together with a ventricle enlargement in MIS-offspring compared with saline-offspring ( Fig.1.C) . CBD effects on volumetric changes are still in progress, but we hypothesize that CBD will prevent the volumetric abnormalities in our MIS model. Conclusions: Our study demonstrates that the administration of CBD during adolescence almost prevented the sensorimotor gating deficits in the MIS model. Moreover, CBD prevented some brain metabolic abnormalities suggesting an improvement of the neuronal integrity and the glutamatergic status, mainly in the PFC. Together, these results suggest that CBD could be a potential therapeutic preventive strategy for the onset of schizophrenia.

Irena Pashkunova-Martic, Medical University of Vienna, irena.pashkunova-martic@meduniwien.ac.at

Abstract Body : Recently toxicity concerns associated with the long-term use of low molecular weight acyclic gadolinium based contrast agents (GBCA) have resulted in the restriction of their administration by the European Medicines Agency (EMA) and triggered risk warnings from the U.S. Food and Drug Administration (FDA). Therefore, the research on novel chelators for paramagnetic centers as part of magnetic resonance imaging (MRI) contrast agents with better stability and safer profile are subject of intense research during the past few years. Furthermore, the development of new molecularly targeted contrast agents or agents that can sense pathological changes in the tumor microenvironment is among the key goals in modern oncological research. Main purpose of this work was the assessment of novel Salinomycin (SAL) based theranostic probes for magnetic resonance imaging (MRI). SAL is a natural antibiotic produced by Streptomyces albus. SAL was found to selectively eradicate breast cancer stem cells (CSCs) and to be at least 100-fold more effective than paclitaxel (a commonly used breast-cancer drug). Selective towards tumors, theranostic agents can revolutionize early diagnosis and therapy of cancer. In this study, first paramagnetic complex of SAL with Gd(III) was synthesized and characterized by various spectroscopic methods such as IR, EPR and ESI-MS. The potential of the complex as a theranostic agent for MRI was evaluated by in vitro relaxivity measurements on a 9.4 T MR scanner. The elemental analysis demonstrated that the complex is of a composition [Gd(C42H69O11)3(H2O)3]. The ESI-MS spectrum of the compound corresponded to elemental analysis data. In the IR spectrum of the Gd(III) complex two characteristic bands at 1540 cm-1 and 1400 cm-1 corroborated monodentate coordination mode of carboxylate anion to Gd(III). The shift of the band for the carbonyl group from 1700 cm-1 to 1690 cm-1 compared to the IR spectrum of salinomycinic acid proved participation of the carbonyl group of the antibiotic in weak hydrogen bonds. The strong broad band at 3400 cm-1 confirmed the presence of hydrogenbonded hydroxyl groups. The EPR spectrum of the Gd(III) complex of SAL consisted of several overlapping signals as three of which with effective g-factors 6, 2.8 and 2 dominated. The recorded EPR spectrum described Gd3+ ions in a crystal field with low symmetry and zero field splitting parameter D>0.3cm-1. Overall the spectral data revealed tricapped trigonal prismatic molecular geometry of the Gd(III) complex of SAL. Three salinomycinate monoanions are coordinated to the metal center via deprotonated carboxyl group and terminal hydroxyl group. Three water molecules, located in the inner coordination sphere of Gd(III) stabilize the structure of the complex by hydrogen bonds with the organic ligand. ultrasound. This experiment confirmed the ability of ultrasound generated by our setup to penetrate regions of the brain deeper, making it suitable for in vivo use in areas of the brain both relatively superficial and deep. The non-invasiveness and penetrative reach of this approach makes it promise for eventual clinical translation, while the cell-type selectivity could facilitate the elucidation and management of neural circuits involved in specific behaviors or disorders. Abstract Body : Introduction: Duchenne muscular dystrophy (DMD) is the most commonly inherited pediatric musculoskeletal disorder. It is a progressive neuromuscular degenerative disorder caused by either the absence or dysfunction of dystrophin [1] . This results in a loss of cell membrane integrity in skeletal and cardiac muscle, and the brain of those affected. Within DMD muscle, necrotic fibers are often observed in groups, and it has been previously hypothesized that this may be due to a reduction in regional blood supply. Furthermore, repetitive regeneration and degeneration cycles deplete the muscle stem cell pool leading to adipose tissue accumulation and the onset of ischemia [2] . Current methods of diagnosis are invasive and localized to certain areas of the body despite DMD being a systemic disorder [3] .

Here we aim to develop an imaging protocol to non-invasively model ischemia and tissue perfusion parameters in DMD patients. We hypothesize that CT will help model DMD disease progression by quantifying changes in hemodynamic and ischemic factors in the heart, brain, and skeletal muscle. Methods: Two cohorts of DMD mice (n=6) were scanned using CT to collect both longitudinal and acute data at differing time points 4-5 weeks, 8-10 weeks, and 15-20 weeks. These critical periods are associated with disease progression and correspond to prefibrotic, fibrotic, and post fibrotic condition respectively. An age-matched wild type control strain (C57bl/10) was used for comparison. Mice in the acute cohort were euthanized at their respective time points and underwent wire myography to study vascular reactivity and validate CT findings. Immunohistochemistry (n=3) was performed for both validation and quantification of the ischemic outcome. Hypoxia-inducible factor 1-alpha (HIF-1a) and myeloperoxidase, were targeted to assess transcriptional regulation within the hypoxic-ischemic environment. In addition, H&E staining (n=1) was used to observe cell morphology and characteristics pertaining to pathology, while Mason's trichrome staining (n=3) confirmed collagen deposition. Results: DMD mice have qualitatively shown increased levels of collagen deposition within the brain, cardiac, and skeletal tissue in comparison to the wild type strain. In addition, preliminary data has suggested increased levels of collagen deposition correlate with increased HIF-1a and myeloperoxidase expression. Furthermore, wire myography has shown that DMD mice are less reactive to dilatory responses when compared to contractile responses at the 15-20 week time point. These findings correlate well with past CT data collected by our lab. Discussion: Currently, there is little knowledge of functional tissue perfusion parameters in DMD patients. This research will be essential in developing therapeutics to restore tissue integrity and function when the natural ability of the tissues for repair is exhausted. Further, it will serve as a noninvasive diagnostic measure to assess disease progression prior to the onset of fatal complications. Abstract Body : Introduction New approaches for the early monitoring of treatment response are needed for the selection of the optimal therapeutic strategy to improve the outcome of cancer patients. Sequential 18F-FDG-PET/CT acquisitions are the gold standard at clinical level for assessing early response as a decrease of tumor metabolic activity in several tumors1. However, this nuclear methodology has several drawbacks mainly associated to radiation exposure and high costs that, in turn, limit its wide applicability. GlucoCEST MRI relies on the use of unlabelled D-glucose and it may be applied on ordinary MRI scanners available at research and clinical settings2. It has been proposed as an alternative to the radiolabelled FDG, but few studies have evaluated the glucoCEST capability to monitor the response to therapy and compared to the FDG-PET technique. This work aims at validating the potential of MRI-GlucoCEST methodology in monitoring the response to two different treatment regimens based on doxorubicin (DOXO) and dichloroacetate (DCA) in breast tumor (4T1) bearing mice.Methods Balb/C female mice (n= 12 for each group) were inoculated with 4x104 4T1 murine breast cancer cells at both mice flanks. The two therapeutic treatments were based on doxorubicin (dose 5mg/kg, four intraperitoneal administrations in one week), a potent anthracycline antibiotic and DNA intercalator that inhibits cancer cell growth3 and on Dichloroacetate (intraperitoneal daily administration 200 mg/kg/day supplemented with oral administration with power dissolved in water at 100 mg/kg ad libitum), whose mechanism of action consists of the inhibition of pyruvate dehydrogenase kinases (PDKs), which reverses the Warburg effect reducing lactate production and proton accumulation in the extracellular tumor region4. GlucoCEST MR images of control and treated groups were acquired before and after one week of treatment and compared with the analogous groups undergone to the 18F-FDG-PET/CT imaging. CEST protocol was performed on a Bruker 7T MRI scanner. Z-spectra before and after glucose (3g/kg) intravenous injection were acquired and CEST contrast was calculated between POST and PRE images. For PET, animals were fasted overnight before the intravenous 18F-FDG injection (about 5.48 MBq/mouse) PET static acquisition was performed for 30 minutes starting 45 minutes after 18F-FDG administration. Results Doxorubicin treated mice showed a marked decrease in tumor growth, being statistically significant from day 21 (after the last doxorubicin treatment), compared to the control group ( Figure 1A) . The FDG-PET technique showed no difference in mean %ID/cc values between the control and the treated groups ( Figure  1C) . Conversely, the calculated average GlucoCEST values between pre-and post-treatment highlighted a significant decrease of the metabolic activity ( Figure 1B and D) . When mice were treated with DCA any difference was observed in the tumor growth rates between the two groups. Both glucoCEST MRI and 18F-FDG PET did not observe any difference in glucose uptake between the treated and control groups following DCA treatment Conclusions In conclusion, for assessing the response to doxorubicin treated mice, GlucoCEST MRI appears more sensitive than 18F-FDG-PET in reporting the therapy response as it parallels the observations relative to the tumour size reduction. DCA treatment did not result in a therapeutic effect, with similar observed glucose uptake between the two techniques. These findings support the view that glucoCEST may be a sensitive technique in monitoring therapeutic effects, but more work appears necessary to identify cancer types and specific treatments where the methodology can be effective. Funding: This work was supported by the European Union's Horizon 2020 research and innovation programme (Grant Agreement No. 667510) (-7.8, -8.1, -8.2, -7.5 Kcal/mol) at the four sites were comparable to MK-6240 To further understand the roles of exosomes and MVs in cancer, here we developed a new EV reporter system that enables comparative analysis of distinct EV classes in vivo. Methods: We developed a new EV imaging probe using a highly sensitive red-shifted bioluminescence resonance energy transfer (BRET) protein called red enhanced Nano-lantern (ReNL) with a palmitoylation signal peptide (Palm). Since palmitoylated reporter proteins have been presumed to label all EV subpopulations, we assessed their labeling efficiency in the fractions of exosomes and MVs derived from murine mammary carcinoma 4T1 cells stably expressing PalmReNL. Cellular uptake of the reporter exosomes and MVs by various cell types was analyzed by fluorescence microscopy and measuring luminescence signals. Furthermore, we used in vivo bioluminescence imaging to determine the biodistributions of the reporter exosomes and MVs in healthy and mammary tumor-bearing mice. Results: (1) Exosomes and MVs carrying PalmReNL exhibited red-shifted emission spectra upon the addition of a luminescent substrate. (2) PalmReNL was more efficiently expressed in exosomes than MVs. However, individual MVs expressed a significantly higher number of reporters due to their larger surface areas compared to exosomes.(3) Culturing the reporter exosomes or MVs with various cell types for 2 h showed the highest cellular uptake efficiency in the recipient cells and the signals decreased over longer periods in culture, followed by the increased proliferation of the recipient cells. (4) Both reporter exosomes and MVs administered intravenously showed similar biodistributions. Both EV classes preferentially accumulated in the lung, followed by liver, spleen, heart, bone, lymph nodes, and brain. (5) Finally, circulating reporter MVs significantly promoted metastatic tumor cell growth in the lung tissues. Summary/Conclusion: Using PalmReNL reporter proteins, we demonstrated the successful tracking of distinct classes of EVs both in vitro and in vivo. Both reporter exosomes and MVs derived from 4T1 cells rapidly transferred the biological signals after their cellular uptake. Moreover, despite their different sizes, both EV classes showed similar organotropisms in healthy mice. Our data suggest that cancer cell-derived MVs, in addition to exosomes, are a critical regulator of systemic promotion of metastasis.

First Name: Masamitsu 

Lauren Brodsky, University of Pennsylvania, lbrod@sas.upenn.edu

Abstract Body : Introduction:Current standard-of-care tests for atherosclerosis can detect calcification of arteries only at later stages of the disease. Commuted Tomography (CT) calcium score, for example, is only effective once appreciable amounts of plaque have formed. 18F-Sodium Fluoride (NaF) is a radiopharmaceutical with affinity of fluoride to hydroxyapatite. For this reason, it is appealing for bone imaging. Recent studies have shown that NaF could also be used to detect calcified micro-deposits within coronary plaque. The goal of this study was to test if the baseline subject characteristics are associated with the short term changes in coronary microcalcification assessed by NaF PET/CT. Methods: This retrospective study included healthy female (N = 8, age 52 ± 10 years, BMI 24 ± 1.7 kg/m2) and male (N = 15, age 50 ± 10 years, BMI 27 ± 2.9 kg/m2) participants who had NaF PET/CT scans taken two years apart. Imaging was performed 90 minutes after intravenous injection of 2.2 MBq of NaF per kilogram of body weight. The analysis regions were selected on CT images by drawing volumes of interest around the entire heart using a semi-automatic segmentation method (PMOD Technologies LLC, Switzerland).SUVmean and SUVmax were calculated in the same regions of the registered PET images. Percent change in SUV between the two time points were correlated against baseline age, BMI, cardiovascular risk factors, and blood chemistry. Results:In males, SUVmean was positively correlated with baseline BMI (r = 0.89, pDiscussion:High BMI is a known risk factor for atherosclerosis. Our data showed that rate of increase in coronary microcalcification over time measured by NaF PET/CT is associated with baseline BMI and some clinical risk factors in males. Lack of such associations in females could be due to low sample size (N = 8). Further prospective studies are needed to determine if baseline BMI and clinical factors could be used to predict rate of increase in coronary microcalcification which could provide the basis for managing the progression of atherosclerosis in patient-specific manner. Keywords:Coronary, Calcification, atherosclerosis,18F-Sodium Fluoride, NaF, PET/CT, BMI, blood pressure, pulse rate Figure 

Ran Sing Saw, Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, Eberhard Karls University of Tuebingen, ransing.saw@med.unituebingen.de

Abstract Body : Introduction: Positron emission tomography (PET) imaging of alpha-synuclein (α-syn) aggregates in Parkinson's disease (PD) would be an invaluable tool for non-invasive diagnosis of the disease as well as facilitating the development of novel treatments. However, no validated α-syn radiotracer is currently available. Here we evaluate MODAG-005, a derivative of anle138b, which modulates α-syn aggregation1, by using in vitro binding assays and an in vivo fibril inoculation model. Methods: In vitro binding affinity and selectivity of [3H] MODAG-005 were evaluated in saturation binding assays using human recombinant α-syn, amyloid-beta (Aβ1-42) and hTau46 fibrils. To evaluate the pharmacokinetic profile in vivo, [11C]MODAG-005 was intravenously injected into healthy rats, as well as rats which were intrastriatally inoculated with 4 µL of α-syn fibrils (30 µM), with a contralateral buffer injection as internal control. Using the volume of the whole striatum based on the rat brain atlas provided in PMOD (44 mm3), the molarity of inoculated fibrils in the right striatum was calculated to be ≈ 2.8 µM. 60-minute dynamic PET scans were performed on healthy rats (n = 3) and inoculated rats at 1, 4 and 7 days post-inoculation (dpi) (n = 6 per time point). Images were analyzed by drawing volumes of interest (VOIs) both on the whole striata based on the rat brain atlas and on the hot spot based on isocontour automatic detection (IAD) set to 70 % hot threshold. IAD was applied in a 2.5 mm sphere drawn into the hot spot of the fibril-inoculated striatum, which was subsequently reflected contralaterally. Binding potential (BPND) in the right, fibril-inoculated striatum was calculated from individual time-activity curves from 30 to 60 minutes, using the left, buffer-injected striatum as the reference region (DVR-1). The presence of inoculated fibrils was validated ex vivo using thioflavin S staining. Results: [3H] MODAG-005 in vitro binding experiments revealed Kd values of 0.2 nM towards α-syn fibrils, 7 nM towards hTau46 and > 100 nM towards Aβ1-42 fibrils.[11C]MODAG-005 showed an excellent penetration of the blood-brain barrier and a fast clearance from the brain, with peak standardized uptake values of more than 1.5 in all rats, indicating a good brain uptake independent from intracranial surgery. For the inoculation model, atlas-based whole striata VOIs revealed BPND of 0.3 ± 0.1, 0.2 ± 0.1 and 0.1 ± 0.1, whereas the IAD analysis yielded BPND of 0.7 ± 0.3, 1.4 ± 0.6 and 0.7 ± 0.4, for 1, 4 and 7 dpi, respectively. Comparing inoculated rats at each time point to healthy rats, the BPND values based on the whole striata VOIs of inoculated rats were statistically significantly higher than healthy rats at 1 dpi (p Conclusion: MODAG-005 exhibited an excellent brain availability, good washout kinetics, as well as in vivo binding to localized α-syn aggregation with sufficient signal-to-noise ratio in rats. The potential of the ligand as α-syn radiotracer will be further evaluated at lower fibril concentrations to determine the limit of detection and in human pathological brain tissues.

Anupama Datta, Institute of Nuclear Medicine and Allied Sciences/DRDO, anupama@inmas.drdo.in

Abstract Body : Aim/Introduction: Earliest pathological changes to occur after the incidence of Traumatic Brain Injury (TBI) are considered to be the formation of Aβ42-containing diffuse amyloid plaques which are one of the most prominent hallmarks of Alzheimer's Disease and is a risk factor for developing associated neurological disorders. Early detection of these plaques consequently remains vital for better regulation and management of these disorders. Considering this, we developed a cost-effective, chalcone scaffold-based SPECT radiotracer which will specifically bind detrimental Aβ1-42 plaques. The bifunctional chelating agent, pentapa-en-NH2 was explored for the development of SPECT agent. We hypothesise that the molecule will display enhanced binding affinity and blood brain barrier penetration. Methods: The bifunctional chelator was synthesised through a series of steps in high purity and further conjugated to the chalcone derivative. The molecule was conjugated to 99mTc and further used for pre-clinical studies. A repetitive mild TBI mouse model was developed, and uptake of tracer was studied. In vitro binding assay with Aβ1-42 aggregates displayed high binding affinity of Ch-(pa)2 with an inhibition constant of 3.98 ± 0.62 nM. The fluorescent data displayed peaks of absorption/emission at 410/540 nm and exhibited a blue shift with a 10-fold increase in emission intensity on binding with Aβ aggregates. Blood kinetics of the radiocomplex performed on normal rabbit displayed fast clearance (t1/2(F) = 32 ± 0.04 min; t1/2(S) = 3h 55min ± 0.03 min). A repetitive mild TBI mice model was established and ex vivo staining of TBI mice model brain sections with Ch-(pa)2 showed specific binding of complex to amyloid plaques. In vivo scintigraphy displayed more than the two-fold increase of radioligand concentration in TBI model as compared to sham mice owing to higher amyloid burden in TBI model. Ex vivo biodistribution analysis demonstrated high blood-brain barrier penetration in TBI model mice with brain uptake of 5.24 ± 0.31% ID g−1 at 2 min p.i. and 3-fold higher than brain uptake in sham mice. These preliminary studies reveal the enhanced amyloid binding affinity by bivalent approach and offer a new perspective in 99mTc probes for β-amyloid imaging. 

Rishikesh Pandey, CytoVeris Inc., rpandey@cytoveris.com

Abstract Body : There are over 300,000 new cases of breast cancer per year in the United States and breast conservation surgery (BCS) is a standard of care for the treatment of early-stage breast cancer. Even with advancements in the standard of care, repeat surgery is necessary in 20% to as high as 40% of BCS cases due to the histopathological determination of positive margins days or even weeks after surgery. Cytological analysis such as touch preparation and frozen section histopathologic assessment are practiced, but these approaches are inadequate owing to sensitivity and specificity concerns, sample preparation artifacts, and sampling errors. Moreover, these methods are time-consuming and complex which further inhibit their broader applicability. Consequently, there is an urgent need for an intraoperative tool that can quickly, accurately, and non-invasively evaluate margins of the resected breast tissue to allow complete removal of all cancer tissue during primary surgery, reducing and ultimately eliminating the need for secondary procedures. Several spectroscopy-based approaches have been proposed for intraoperative margin detection. Intraoperative tools for margin assessment ideally should detect cancer at the surface as well as at a specified depth of tissue due to the current standard requirement of no residual within 2mm of the surface of the excised tissue for ductal carcinoma in situ. Raman spectroscopy, in particular, has emerged as a potential tool owing to its ability to provide real-time depth-resolved biomolecular information in a label-free manner. While spatially offset Raman spectroscopy (SORS) has shown excellent promise in providing depthselective Raman spectral information, it requires suffers from poor signal to noise ratio from deep layers due to diffuse scattering of light from tissue. Here we present a novel technique that we call Angular Depth Resolved Raman Spectroscopy (ADRRS). We use a non-diffracting, selfhealing beam to interrogate the tissue sample at an oblique incident angle. As such, the Raman signals from different depth layers can be detected at different lateral x positions as illustrated in Fig. 1 . The use of Bessel beam that propagates in scattering media with less beam divergence enables sensitive Raman measurements from sub-surface tissue layers. This talk will present our recent imaging results in determining its applicability in the surgical setting. Abstract Body : Inflammation is associated with a range of diseases including autoimmune and cardiovascular diseases and cancer. The ability to image active inflammatory processes greatly enhances our ability to diagnose and treat these diseases at an early stage. An in vivo multimodal PET/19F MRI imaging probe was designed based on a perfluorocarbon nanoemulsion (NE), which labels monocytes and macrophages in situ and can be used to visualize inflammatory macrophages with high sensitivity, specificity and versatility. The NE has 89Zr encapsulated within its fluorous oil by a novel fluorous chelator, producing a highly stable PET/MRI tracer for in vivo inflammation imaging. Pharmacokinetic analysis of blood samples using both dosimetry and 19F NMR measurements confirmed a high degree of agent stability in vivo. Following a single injection of 89Zr NE, probe utility was demonstrated in murine diseases models of acute inflammation, inflammatory bowel disease (IBD) and solid tumor using PET/CT and 19F/1H MRI in the same animal. Good correlation was observed between hotspot signals from PET and 19F MRI in each of the models tested, with notable differences suggesting that these techniques are highly complementary. In all mouse models studied, over 10-fold greater agent uptake was observed at the lesion site compared to controls. Histology studies confirmed the presence of NE in macrophages in inflammatory lesions. The high sensitivity of PET enables unambiguous whole-body imaging of macrophage burden at clinically relevant dosage, while 19F MRI yields higher resolution views of the underlying lesion localization. Overall, the 89Zr NE probe is valuable as a multimodal preclinical biomarker, particularly for imaging tissues where 18F-FDG normally has a high intrinsic uptake. In this regard, we introduced a pseudo quantitative PA (QPA) imaging approach for depicting different sized optical absorbing molecules. In this work, we performed proof-of-concept simulations along theoretical considerations and validated QPA imaging using phantom materials. Methods: QPA images were simulated using an acoustics toolbox for MATLAB (k-Wave; Mathworks, Natick, MA). An advanced time-domain model of ultrasound (US) wave propagation was implemented given a center frequency of 10 MHz, sampling frequency 100 MHz, and transducer aperture size of 14.8 mm. The overall dimension of the computational grid was set to 14.8 x 14.8 mm with a grid size (spacing) of 11 µm. PA data was reconstructed using time reversal reconstruction. After, the PA data fields were processed using a novel processing approach that links the local US signal to the size of optical absorbing molecules [1] . More specifically, the spectrum of the PA signal was decomposed using a pair of nth-order Gaussianweighted Hermite polynomial (GWn) functions before taking the signal envelope from a Hilbert transform. The envelope of the original unfiltered data is assigned to the green channel, low frequency information (GH2) to the red channel, and the high-frequency information (GH8) to the blue channel, which completes the RGB colormap and QPA image display. A series of phantom materials were constructed from gelatin embedded with a small volume of optical absorbing molecules with known size. Raw radiofrequency (RF) PA data was collected after laser excitation using the Vevo 3100 + LAZR-X system (FUIJIFILM VisualSonics Inc, Toronto, Canada) equipped with a linear array transducer. PA signal analysis was the same as detailed for the simulation study. All results obtained using QPA processing were compared to traditional PA images. Results and Discussion: Simulated images obtained from the two different phantom types suggest that the QPA images can accurately depict the relative size of different locally spaced optical absorbing molecules. QPA images exhibit a change in hue from bluer to red when the scatterer size increases, which is consistent with theoretical considerations. Phantom materials embedded with different-sized optical absorbers revealed the same trend. A comparison of QPA to traditional PA imaging results revealed a considerably narrower point spread function (PSF) for the former, which suggests QPA improves localization of optical absorbing molecules and yields images of higher spatial resolution. Collectively, these preliminary findings help validate this novel QPA method and more results are forthcoming.

References: [1] Basavarajappa L, Hoyt K. High-frequency quantitative photoacoustic imaging and pixel-level tissue classification. IEEE International Symposium on Biomedical Imaging, 308-311, 2020. This newly reported protein-based hybridization (host-guest assembly) is believed to represent a simple and general strategy for researchers to use to overcome the issues of low solubility and poor cell permeability of newly designed fluorescent probes. Despite the success shown using fluorescent probes, biological systems are complex with more than one species being involved in a given pathway. As a result, current fluorescent imaging platforms can provide a limited picture to a particular biological study. Exploiting various photophysical probes of specific fluorophore scaffolds, so called dual/multi-responsive fluorescent probes or AND-logic gates have been developed (Scheme 1b).8 A novel AND logicbased fluorescence probe for the simultaneous detection of ONOO− and GSH was developed. Cellular imaging experiments using this probe demonstrated the ability to monitor the coexistence of metabolically produced ONOO− and GSH and evaluate a therapeutic effect.9 This was recently extended to the detection of ONOO-, however, only when the fluorescent probe was bound to amyloid beta peptide, a key biomarker found in Alzheimer's disease.10 Overall, a new strategy has been developed for the detection of a biologically important species through applying newly developed synthetic methodology. Host-guest chemistry has been employed to overcome solubility issues of fluorescent probes and multi-responsive fluorescent probes have been developed that are suitable for complex biological systems. . FTIR revealed the presence of DFO at the surface of GNPs (hydroxamate peaks: 1629.0cm-1, 1569.0cm-1; amine peak: 3312.0cm-1). XPS revealed the O=C-N C1s peak of DFO at 287.49eV, and upon Zr(IV) chelation, chelated Zr3d5/2 peak at 182.4eV with no ionic Zr3d5/2 (183.38eV). TGA showed a two-step degradation at 189 and 320°C corresponding to DFO and PEG, respectively. Overall, the physicochemical characterisation techniques confirmed the synthesis of ultrasmall GNP functionalized with DFO capable of chelating 89Zr(IV) for PET imaging. The permeation of GNPs through biological (porcine skin) and polymer (latex and nitrile) membranes was studied with PET ( Figure 1d-f ). These membranes were placed in the designed diffusion cell, and 5 mL of a phosphate-buffered aqueous solution containing 89Zr -GNP (~5 MBq; 0.15pmol/mL Zr; and 0.22 or 0.1mM of Au for skin and polymeric membranes, respectively) was added in the DC. DC and AC were imaged continuously for a duration of up to 42 hours. Results: Imaging the permeation of GNP revealed kinetic profiles with unprecedented time resolution. Particles kinetic profiles showed three permeation phases. Radiolabeled GNPs were detected at a concentration as low as 6.8x10-6nM of 89Zr-DFO and 1nM of Au, which is in general 1000 time more sensitive compared with conventional spectroscopic methods.4

Permeation parameters were extracted from these curves: lag time, influx, and diffusion coefficients. Real-time monitoring revealed GNP penetrated in the skin and out with an influx of 7.8 and 4.9µM mm-2 hr-1, respectively. Alternatively, the GNP permeation profiles through polymeric membranes showed an influx of 12.8 and 22.5nM mm-2 hr-1 for latex and nitrile membranes, respectively. Conclusion: This study confirms the strong potential of PET as a highly-sensitive, real-time imaging instrument to monitor the permeation process of nanoparticles (and other substances) across membranes. The study demonstrated quantitatively the permeation profile of GNPs through the skin and polymeric membranes. In the future, this technique will allow the performance of diffusion tests at an unrivalled sensitivity, which is needed for measuring the diffusion of highly potent or toxic molecules across membranes. While some procedures can be supplemented with electromyographical monitoring, this nevertheless amounts to little more than one's hands and eyes, leaving ample room for new methodologies that may enhance detection and visualization of nerves, providing improved surgical outcomes for both patients and surgeons. Fluorescence-guided surgery (FGS) has the potential to revolutionize surgical outcomes by selectively enhancing tissue visibility. Fluorescent molecules with the greatest potential for clinical translation emit in the near-infrared (NIR) region between 650-900 nm, where tissue absorbance, light-scattering and autofluorescence are all at their local minima. [2] However, a limited number of fluorescent molecules have been developed to image nerve tissue, and the vast majority of these are plagued with high non-specific uptake in surrounding tissues and only emit at visible wavelengths. Development of a NIR nerve-specific fluorophore is challenged by the fact that nerve-specific fluorophores must be of sufficiently low molecular weight to cross the blood-nerve-barrier (BNB), however achieving the degree of conjugation (e.g., number of unsaturated hydrocarbons) requisite of long-wave fluorescent emission unavoidably increases molecular weight. [3] Herein, we report our efforts to synthesize libraries of systematically-modified far-red to NIR fluorophores. Nerve-specificity was verified using both murine and swine models following direct and systemic administrations. [4] Modification of lead candidates from our 700 nm library of nerve dyes, facilitated the design of our first-in-class NIR nerve-binding fluorophores, fully compatible with the 800 nm channels already present in clinical imaging systems. Comprehensive molecular engineering of these lead compounds have also resulted in new, stable compounds that are water soluble and demonstrate improved photophysical properties compared to parent compounds, without the need to compromise inherent brightness or nerve-specificity Finally, through fluorescence-guided laparoscopic surgery we have successfully identified the first NIR nervebinding fluorophore candidates for clinical translation. This exciting technology is poised for integration into the day-to-day clinical workflow and will enable direct, high-contrast nerve visualization in real-time.

First Name: Antonio (10mM, v/v = 5:95) solution at flow rate of 4 mL/min. For the autoradiography experiment, mouse spinal cords sections are fixed in formalin for 10 min and washed with PBS. Sections are incubated with 10 μCi of 18F-labeled gabapentin with pregabalin (0 -5,000 nM) in PBS for 30 min and then washed with ice cold PBS (3 x 0.5 min). After washing, the tissues are dried, exposed to phosphor screen and imaged 24 hours later. 18F-labeled gabapentin was synthesized in ~20% decay corrected radiochemical yield based on radio-HPLC. The identity and purity was confirmed by analytical HPLC. Autoradiography in mouse spinal cord sections shows binding of 18F-labeled gabapentin in dorsal horn region consistent with the reported expression of α2δ-1 receptors.5 Over 85% of the binding could be displaced by 5 μM of the selective α2δ-1 drug pregabalin demonstrating high specificity.

First Name: Yu-Peng Abstract Body : Background: Melanoma is the most aggressive skin cancer with the highest associated mortality [1] . The 5-year survival rates range from 98.3% (stage I) to 16% (stage IV) emphasizing the importance of diagnosis at an early stage [2] . Non-invasive morphological imaging approaches such as dermoscopy and high-resolution, morphological reflectance confocal microscopy demonstrate high sensitivity but moderate and variable specificity [3] . Noninvasive gene assays based on PRAME (PReferentially expressed Antigen in MElanoma) expression show similarly high sensitivity but moderate specificity [4] ; neither of these approaches have shown potential in stratifying lesions based on malignant potential. Noninvasive molecular imaging using targeted fluorophores such as PARPi-FL (targets PARP1, which is overexpressed in cancers [5] ), can enable functional evaluation of PARP1 expression to stratify lesions which are malignant and/or with higher malignant potential [6] , thereby improving diagnostic specificity for melanoma detection and prognosis. As a first step towards PARPi-FL imaging in vivo, we investigated the differences in PARP1 expression in the spectrum of melanocytic lesions including melanoma and benign nevi, using histology and PRAME expression as gold standards. Methods: We investigated PARP1 and PRAME expression using immunohistochemistry (IHC) in 25 lesions (14-melanoma, 11-benign nevi) in 5 mm-thin formalin-fixed paraffin-embedded serial sections. IHC was optimized on an automated Leica-Bond stainer platform using optimized protocols. IHC and hematoxylin-and-eosin (H&E) tissue sections were digitized using a MIRAX slide scanner. Manual scoring (by a board-certified dermatopathologist) and IHC quantification (software-based) were performed to assess the differences in PARP1 expression in 14 specimens. Since PARP1 can localize to the nucleolus in addition to the nuclear volume, differential PARP1 staining patterns can be observed in nucleus and nucleolus. Thus, the nuclear and nucleolar positivity were separately evaluated for manual scoring. Manual dermatopathologist scoring was based on number of positive nuclei and nucleoli in the lesion area over the range 0-4, with 0 (absent), 1 ( Results: PARP1 expression was found in all benign and melanoma tissues at different levels. Prominent nucleolar staining was observed in all melanoma tissues, an important feature for detection of malignancy and aggressiveness in pathology [7] . Based on dermatopathologist evaluation, melanoma samples had higher scores for both nucleus, nucleolus area positivity and overall score (range 3-4) as compared to benign nevi (range [1] [2] [3] [4] . Benign nevi showed high nuclear positivity in some nevus (3 out of 8 specimens were scored 4), however they showed low positivity for nucleolus and overall score. In quantification, melanoma showed significantly higher percentage area and intensity for PARP1 expression as compared to benign nevi (p Conclusions: Melanoma shows higher PARP1 expression in terms of intensity and area positivity in direct comparison with benign melanocytic lesions, especially in nucleolus relative to nucleus. These findings will be confirmed on a larger data set to build statistical classifier for benign versus melanoma using both nuclear and nucleolar positivity. Given the IND status of PARPi-FL for use in humans, the results will establish the basis for subsequent in vivo clinical imaging studies for molecular melanoma diagnosis in patients. 

A miniature 'plug-and-play' microscope for lifetime imaging in preclinical models of brain disease Janaka Senarathna, Johns Hopkins University School of Medicine, dmmj.senarathna@gmail.com

Abstract Body : Neuroimaging in preclinical models of brain disease is a critical tool enabling both a rigorous assessment of disease pathology and the efficacious development of new drugs. However, brain diseases are complex: i.e. they present densely convoluted neuronal, microcirculatory and immunological dysfunctionalities. Moreover, brain diseases often shift their disease phenotype with progression, therapy, and the extent of therapeutic resistance. State-ofthe-art neuroimaging systems today (e.g. MRI, PET, Multiphoton or confocal microscopy) are often bulky, available only through imaging centers with limited access, expensive, timeconstrained, and typically require the animal to be anesthetized. Consequentially, neuroimaging can only be performed for short durations at discrete time points in the disease lifecycle, critically under-sampling the disease progression. Moreover, independent of the effects of the disease model being investigated, anesthesia alters brain function and may result in erroneous conclusions about disease progression and response to therapy. While miniaturized microscopes that enable imaging in unanesthetized mice have been designed to circumvent this problem1-6, they often lack the capacity to interrogate multiple neurobiological variables, confining the assessment of neurophysiological function. To address these limitations, we developed an affordable, miniature multi-contrast microscope7. It permits interrogating multiple neurophysiological variables in freely behaving mice via multichannel imaging with fluorescence (FL)1, intrinsic optical signals (IOS)8 and laser speckle contrast (LSC)9, incorporated within a footprint comparable to that of a US quarter (Fig. 1a,b) . One can use the FL channel to image neuronal activity with a voltage sensitive dye or a genetically encoded calcium indicator, as well as monitor fluorescently tagged cells. The IOS channel can be used to image microvascular structures, vasodilation and vasoconstriction, while the LSC channel permits imaging blood flow in the brain. A 3D printed disposable plastic head-mount is implanted atop a surgically prepared cranial window that provides optical access to the brain (Fig. 1c) . Weighing 3 g after support from a strain relief, the microscope can be affixed to the head-mount to conduct neuroimaging at a resolution of 5 µm over a 3x3 mm2 field of view (FoV) at 10 frames per second. We also created an easy-to-use graphical user interface (Fig. 1d) to control all parameters related to image acquisition and illumination, thereby making it a 'plugand-play' device with minimal technical knowhow needed for operation. Moreover, one can pack the microscope onto a briefcase for convenient transportation. After validation against a non-miniaturized imaging system, we demonstrate the utility of our microscope by conducting lifetime imaging of vascular changes induced by a brain tumor. We imaged a 9L-GFP tumorbearing mouse daily with our multichannel microscope (Fig. 1e-k) . We monitored tumor progression in vivo with the FL channel (Fig. 1e) wherein by day 14 (D14) the entire FoV was occupied by tumor. We successfully tracked the stages of angiogenesis in vivo with the IOS channel (Fig. 1f , and red arrows): microvascular sprouting (D3-D6, Fig. 1j,k) , anastomoses and flow establishment (D7-D9), and development of sinusoidal vessels (D7-D15). Fig. 1g shows concomitant, but delayed increase in microvascular perfusion as tumor vessels anastomose with brain microvessels (D7-D9), with mature vessels eventually becoming well perfused (D9-D15). Microvascular connectivity (MC) maps (Fig. 1h) illustrate early stage coherent blood volume fluctuations (i.e. vasodilations and vasoconstrictions) in areas with sprouting tumor vessels (D3-D6), which become more correlated (D7-D9), and finally show a decrease of MC due to formation of abnormal blood vessels (D10-D15). Trends of these metrics are plotted in Fig. 1i . We believe that our miniature microscope will herald a new era in preclinical brain disease research by enabling novel insights into the dysfunction of multiple neurobiological variables over the disease lifetime.

First Name: Janaka Abstract Body : Background: Clinical examination followed by dermoscopy and Reflectance confocal microscopy (RCM) provide high sensitivity (80-95%) and moderate specificity (60-70%) for diagnosis of basal cell carcinoma (BCC)-the most common cancer in the world [1, 2] . RCM is a high-resolution label-free quasi-histopathological imaging modality based on the detection of reflected (singly-backscattered) light from cellular organelles and tissue [3] . RCM imaging has improved the clinical BCC diagnostic specificity by 2x, beyond which the grayscale contrast has reached a fundamental limit. This is mainly attributed to the weaker back-scatter from chromatin in nucleus: ~100 photons/pixel relative to the surrounding cytoplasm and dermis (~104 to ~106 photons/pixel) because of which structures with high nuclear:cytoplasmic ratio such as BCC tumors, lower epidermis layers and hair follicles are hard to visualize and differentiate, limiting BCC diagnosis [4] . The diagnostic accuracy may be dramatically enhanced with the use of nuclear-targeted fluorescent probes that demonstrate transepidermal delivery following topical application. PARPi-FL is one such small-molecule fluorescent reporter that targets nuclear poly-(ADP-ribose) polymerase [PARP] enzymes [5] . PARPi-FL has shown specific nuclear labeling, high contrast (tumor:normal) imaging [6, 7] , and has an investigational new drug (IND) status for in vivo studies [8] . To investigate the potential clinical utility of PARPi-FL, three important questions were investigated: i) BCC PARP1 expression, ii) PARPi-FL BCC labeling and diagnostic accuracy, and iii) modeling of transepidermal PARPi-FL delivery by passive diffusion following topical application. Methods: PARP1 expression was evaluated on thin FFPE tissue sections of BCCs (n=95). PARPi-FL labeling, and diagnostic accuracy on discarded thick surgical specimens of BCCs (n=85) and normal skin (n=20); while permeability was on normal ex vivo tissue from mastectomy procedures (n=5). In vivo permeability and skin toxicity were evaluated in live anesthetized Yorkshire pigs (n=3). PARP1 expression and PARPi-FL was quantified as area positivity and intensity, respectively, in tumor and normal structures. Diagnostic accuracy was evaluated with blinded reading by two expert confocal readers to assess sensitivity, specificity, positive and negative predictive values (PPV, NPV). Positive nuclear staining in epidermis and underlying dermis indicated successful permeability. Results: Higher PARP1 expression was found in BCCs as compared to normal structures such as epidermis, hair follicles and sebaceous glands (p Conclusion: Higher PARP1 expression and PARPi-FL staining was observed in BCC tumors. PARPi-FL staining was found to be nuclear-specific and highlighted BCCs in the FCM images, thereby improving the sensitivity (by 10-20%) and specificity (by 4%) of RCM+FCM over RCM alone. The permeability and passive diffusion of PARPi-FL through intact stratum corneum following topical application was also confirmed in both ex vivo human skin and in vivo in a pig model in a clinically feasible time range (10-30 minutes). Thus, our innovative findings demonstrate the utility and establish the foundation for prospective in vivo PARPi-FL studies for BCC diagnosis in patients. (Singh 2003 , Eyler 2008 . These refractory tumor cells also drive metastasis and on surviving treatment lead to cancer recurrence (Hermann 2007 , Lagadec 2010 . Consequently, the CSC population represents an attractive therapeutic target for hard-to-treat tumors. Selectively eliminating CSCs by targeting CSC biomarkers could have broad clinical impact as CSCs have been isolated in virtually every type of cancer (Singh 2003 , Al-Hajj 2003 , Matsui 2004 , Taylor 2005 , Han 2014 , Suetsugu 2006 ). In addition to therapeutics, companion diagnostic imaging probes for CSC markers offer to not only stratify patients amenable to CSC targeted therapy, but to also provide valuable information regarding tumor staging and prognosis ( Daywood 2014 , Chang 2016 . Biomarkers suited towards targeting the CSC population for treatment and diagnostic imaging are ideally surface-expressed, sparingly found on regular tissues and lost upon differentiation. Cancer Stem Cell marker CD133 well-embodies these criteria and is exceptionally useful for studying CSCs (Glumac 2018) . Previously developed murine and humanized antibodies targeting CD133 for therapy and diagnostic imaging are rare or underperforming. Cancer stem cell resilience to chemotherapy has stifled some CD133 targeted treatment efforts (Dean 2005) . Other preclinical therapy studies have shown promising antitumor efficacy especially when multiple doses of the therapy were administered (Swaminithan 2013) . Unfortunately these preclinical therapy studies which utilize humanized or murine antibodies would be unsuited for clinical translation due to the human anti-mouse antibody response (HAMA) (Sgro 1995). The long term goal of this project is to leverage highly localized, potent radiation from emerging therapeutic radionuclides with novel, fully human monoclonal antibodies to selectively eliminate cancer stem cells in human patients. Here, RW03, a monoclonal antibody with specificity for CD133 is evaluated as the basis of a versatile targeting platform against CSCs. Robust bioconjugation strategies were used to stably radiolabel RW03 with lutetium-177 achieving >50% radiochemical yield and >99% radiochemical purity (RCP) after a 1 h reaction in slightly acidic (pH 5.5) buffer. Similar strategies were used to produce CD133 targeted companion diagnostic PET/SPECT imaging probes. Radiolabelled conjugates were routinely stable (>95% RCP) in saline for over 1 week, and for several days in mouse serum. Seminal studies evaluating the biodistribution of RW03 directly labelled with lutetium-177 demonstrated tumor uptake of 65% ID/g and tumor/ blood ratio Abstract Body : Manipulating a specific neuron or neuron circuits in the brain is a typical way understanding the brain function and treating brain disfunctions [1, 2] . Ultrasound, especially for low intensity low frequency ultrasound, is an emerging technique that can noninvasively modulate neuron activity in targeted brain regions with fine special temporal resolution and deep brain penetration [3, 4, 5] . Ultrasound is therefore a promising tool for both probing brain function and treating brain diseases with the advantage of non-invasiveness. However, the minimum focal spot of an ultrasound beam is much larger than a single neuron or a specific small set of neurons. These make it difficult for the ultrasound to probe or target the complex neural circuits entangled with interdependent different neurons. The underlying mechanism of ultrasound neuron modulation is thus the key issue for the best usage of ultrasound neuron modulation in both the basic and clinic neurosciences. Due to the nature of mechanical force of ultrasound, the mechano-sensitive ion channel displayed a promising candidate. Piezo family is so far the most sensitive mechanical ion channel, that could respond quickly to forces as low as 10 pN which is consistent with the scale of ultrasound induced forces [6, 7] . Base on this, piezo family is likely to be a factor modulates the ultrasonic neural modulation. To test this hypnosis, several experiments studying the role of exogenous and endogenous piezo1 in the ultrasound neuron modulation were performed. Firstly, 293T was used to test whether piezo1 mediates the ultrasound effect on cells. The results showed that overexpressing the piezo1 in the 293T induce the calcium influx whereas the unexpressed one has little response to ultrasound (n = 9, ***p < 0.0001). This indicates the piezo1 mediated the ultrasound effect on cells. To confirm whether the ultrasound neuron modulation is also mediated by piezo1, primary neurons were adopted. Firstly, the endogenous expression of piezo1 in the primary neurons were tested by immunofluorescent staining. And the function of piezo1 was assessed by calcium imaging with a piezo1 agonist Yoda1 and a piezo1 antagonist GsMTx-4 ( Fig A) . After the confirmation of piezo1 function in primary neurons (n = 15, ***p < 0.001, Fig A) , the ultrasound bio-effect was tested with or without the GsMTx-4. The data showed that the ultrasound alone could activate piezo1, initiating calcium influx with a dose depend of ultrasound intensity. And these effects can be blocked by GsMTx-4 (n = 9, * P < 0.05, ** P < 0.01, *** P < 0.001, Fig B) . Similarly, the ultrasound increased nuclear c-Fos expressions in primary neurons with a dose depend of ultrasound intensity. And these effects can be reduced if the neurons were pre-treated with GsMTx-4 (n = 3, * P < 0.05, ** P < 0.01, Fig C) . These findings demonstrated that the piezo1 mediates the ultrasound neuron modulation in primary neurons. To further explore the signalling implications of Piezo1-mediated ultrasound neuron modulation, a mouse neuronal cell line, CLU199, was applied. Using western blot, the protein expression of phospho-CaMKII, phospho-CREB and c-Fos were tested. The results showed that the ultrasound increased the expression of phospho-CaMKII, phospho-CREB and c-Fos in a dose dependent manner of ultrasound intensity. But Piezo1 knockdown significantly reduced this effect (n= 3, * P < 0.05). These findings indicate the piezo1 mediated ultrasound neuron modulation in vitro and may also provide a possible mechanism of ultrasound brain modulation in vivo. By controlling the expression of piezo1 we may able to target specific neuronal pathways or nuclei in both the basic and clinical neurosciences ( Fig D) [8] . Abstract Body : Image based systems biology is emerging as a powerful new paradigm to better understand the role of the tumor microenvironment (TME) in cancer progression, metastasis and response to therapy. In spite of the recent advances in vascular imaging approaches, imagebased vascular systems biology has remained challenging because currently there are no methods to label blood vessels simultaneously in all three primary imaging modalities, i.e. magnetic resonance imaging (MRI), micro-CT (µCT) and optical microscopy. Consequently, this poses a hurdle to the integration of 3D vascular imaging data with complementary image contrasts acquired at multiple spatial scales, e.g. white matter fiber distributions from MRI (for brain tumors), bone contrast from µCT (to study bone metastasis), protein expression from multiphoton (MPM) or light-sheet microscopy (LSM), all of which can provide invaluable inputs to cancer systems biology models. Moreover, since the spatial resolution of these data may span several orders of magnitudes and having common landmarks visible across spatial scales is often not possible, data integration via image co-registration has also remained challenging. Therefore, we developed a "multicontrast" vascular contrast agent combination that makes blood vessels visible across MRI, CT and optical microscopy. This enabled the development of an easy-to-use method called VascuViz that provides a multimodal, multiscale 3D vascular imaging capability in intact, unsectioned tumors. Here, we showcase the utility of VascuViz for multicontrast and multiscale characterization of the TME in an orthotopic MDA-MB-231 breast cancer xenograft using MRI (40 µm), µCT (9 µm) and MPM ( VascuViz enabled concurrent imaging and visualization of the tumor vasculature with complementary image contrasts such as T1w-and Dw-MRI (Fig. 1a, b) , euclidian distance map from µCT (Fig. 1c, d) , collagen (col) data from second harmonic generation (SHG) imaging and green fluorescent protein expression (GFP) data from cancer cells using multiphoton microscopy (MPM) (Fig. 1e-g) . VascuViz facilitated data integration via a "vascular fiducials" based image co-registration method1. This enabled the generation of co-registered 3D maps of the TME at the macroscopic (40 µm), mesoscopic (9 µm) and microscopic ( 200 µm) in the tumor center (Fig. 1c) . Co-registration between EDM and ADC maps revealed that regions with high mean ADC values (> 0.001 mm²/s) matched with those where inter-vessel distances were in the range of 151 -300 µm (Fig. 1f-j) . These results complemented 3D MPM data ( In conclusion, we developed a multimodality, multiscale vascular imaging approach, VascuViz, using MRI, µCT and LSM and showcased its utility in an orthotopic breast cancer xenograft. In addition to enabling high-resolution vascular mapping in intact unsectioned tumors, this method provides for the first time, the ability to integrate vascular data across multiple image contrast mechanisms from the spatial scale of individual endothelial cell to the whole-tumor. This method has direct applications in image based systems biology of cancer2 and other vascular disease models (e.g. stroke). Bhargava (c) the microvascular network data obtained from µCT at 9 µm spatial resolution. Here, the vascular contrast is due to the presence of radio-opaque BVu. Blood vessels are scaled by their mean diameters (9 to 160 µm). The yellow dotted line represents the tumor boundary overlaid from T1w-MRI (a). Co-registration between T1w-MRI and µCT enabled the generation of the (d) tumor euclidian distance map (EDM) at 9 µm spatial resolution that revealed the presence of inter-vessel distances as large as 350 µm in the tumor center. These regions matched with those that showed high ADC values (i.e. >0.001 mm²/s) in (b). MPM image of a 20 µm thick section through the tumor center revealed (e) capillaries (red), green fluorescent protein (GFP) expression from cancer cells (green) and SHG based collagen (Col) data (blue). Here, the blood vessel contrast is due to the presence of GalRh. Representative regions of the tumor rim (red box) and tumor center (blue box) from (e) are magnified 10X in (f, g), respectively. Abstract Body : Accidental nerve transection or injury is major morbidity associated with many surgical interventions, resulting in persistent postsurgical numbness, chronic pain, and/or paralysis. Nerve-sparing can be a difficult task due to patient-to-patient variability and the difficulty of nerve visualization in the operating room. Fluorescence-guided surgery (FGS) to aid in the precise visualization of vital nerve structures in real-time during surgery could greatly improve patient outcomes. Currently, almost all clinical FGS systems have an "800 nm" channel designed to image the U.S. Food and Drug Administration (FDA) approved indocyanine green (ICG), which is mainly used to assess blood flow or detect sentinel lymph nodes during surgery as a non-specific fluorescent contrast agent. After nearly two decades, only a limited number of fluorescent molecules have been developed to image nerve tissue, where the vast majority of are plagued with high non-specific uptake in surrounding tissues and mostly fluoresce at visible wavelengths. Developing a nerve binding agent that has near infrared (NIR) fluorescent properties has been a complex and challenging task. While nerve-specific fluorophores must be of sufficiently low molecular weight to cross the blood-nerve-barrier (BNB), the degree of conjugation (e.g., number of unsaturated hydrocarbons) requisite for long-wave fluorescence emission necessarily increases molecular weight. To overcome these opposing constraints, we have taken a medicinal chemistry approach that facilitated the design of our first-in-class NIR nerve-binding fluorophore library, which is derived from 700 nm analogs through purposeful repositioning of the polar groups and effective elongation of the electronic push-pull system. However, such structures cannot be readily attained via traditional oxazine syntheses. Adapting modern organic synthesis secured the production of these ICG spectral analogs while maintaining nerve-specific contrast that are fully compatible with the 800 nm channels already present in clinical imaging systems. Utilization of these novel NIR nerve binding fluorophores during FGS will significantly improve nerve visualization, providing an immediate benefit to both patients and surgeons. Additionally, our study design and results demonstrate the feasibility of our molecular engineering approach which can be readily adapted to create other tissuespecific fluorophores. Abstract Body : Objectives: Combinatorial library approaches such as the one-bead onecompound (OBOC) and phage display offer a useful tool for rapid identification of novel receptor-targeted peptides, however subsequent lead optimization towards successful imaging agents remains an ongoing challenge.1,2 This work aims to optimize the binding affinity and selectivity of short peptides identified from various combinatorial libraries that target the integrin avb6, a cell surface receptor that is overexpressed in numerous aggressive cancers.3 C-terminal extension of a short αvβ6-binding peptide with amino acids taken from the well-known αvβ6selective A20FMDV2 peptide derived from the foot-and-mouth disease virus (NAVPNLRGDLQVLAQKVART) has previously shown to significantly improve the binding affinity and selectivity for integrin αvβ6 as well as serum stability.4 Here, we further explore the effect of C-terminal extension of peptides on the binding profiles, serum stability and secondary structure of two peptides from our OBOC library (1 and 2),4 an RTD peptide originally discovered via phage display (3) where the αvβ6-binding DLXXL motif was first identified,5 and two RGD octamers discovered from a sunflower trypsin inhibitor1-based phage display known as SFLAP3 (4)6 and SFITGv6 (5)7 ,in in vitro assays. The lead candidate was PEGylated with the goal to further improve serum stability prior to in vivo evaluation. Methods: All peptides were synthesized using standard Fmoc chemistry on solid-phase. Binding affinity and selectivity of peptides were evaluated by competitive binding ELISA with purified integrins (avb6, avb3 and avb8). Secondary structure of the peptides was determined by circular dichroism (CD) spectroscopy. Radiolabeling of peptides at the N-terminus was performed on solid-phase with [18F] FBA, followed by HPLC purification. The 18F-peptides were evaluated in vitro (binding and internalization) in DX3purob6 (αvβ6 +) and DX3puro (αvβ6 -) cells, and their stability evaluated in mouse serum at 1 h post-incubation (p.i) at 37°C. Results: All modified peptides (6-10) exhibited significantly higher affinity for avb6-integrin in both ELISA (0.4-12 nM). They demonstrated a 1.5-5-fold increase in binding to DX3purob6 cells (3.4-38% binding) with increased selectivity (up to 12:1) relative to the unmodified parent peptides. CD data revealed that all peptides (1-10) exhibited random coil conformation in CD buffer, but in 30% TFE a-helices were observed only for the modified peptides (6-10). [18F] 7 showed 20±0.4% binding to DX3purob6 cells with highest selectivity ratio of 11.8:1, but was only 16% intact in mouse serum at 1 h p.i. This peptide was PEGylated and both PEGylated peptide variants [18F] 11 and [18F] 12 showed ~40% binding to DX3purob6 cells, >99% stability in mouse serum at 1 h p.i. and [18F] 12 showed 2-fold higher selectivity for DX3purob6 cells than [18F] 11 (Table  1) . Conclusions: Five peptides were synthesized and modified by C-terminal extension. All extended peptides showed improved affinity and selectivity for the integrin αvβ6. CD data suggested that the formation of an α-helix is important for selectivity. [18F] 12 is currently under further evaluation in vivo. Abstract Body : Patients with advanced cancers die because their cancer develops resistance to all available therapeutic strategies. "Personalized cancer medicine" strives to use biomarkermatched molecularly targeted therapies to overcome known resistance mechanisms. However, there is currently no methodology to measure the highly interdependent levels of available molecular drug targets, amount of drug binding and the resulting cellular response in a single dynamic, living system. To overcome these limitations, we have developed a dynamic, fluorescence-based, three-compartment model termed intracellular paired-agent imaging (iPAI). iPAI is a fluorescence-based approach that utilizes fluorophore labeled small molecule therapeutics as imaging agents to measure drug target engagement. Herein, we report our designed zwitterionic, water-soluble, and cell-permeable fluorophore pairs, based on the classic tetramethylrhodamine (TMR) and Silicon-substituted TMR (SiTMR). Termed Sulfo-Rh and Sulfo-SiRh, both dyes showed substantially improved optical stability against solvent polarity changes compared to the base fluorophores, TMR and SiTMR. In addition to matched structural features, these newly synthesized fluorophores also exhibited matched photophysical properties, including net charge, molecular size, Stokes' shift, full width at half maximum (FWHM), and brightness. Together, these spectrally-distinct fluorophores offer a unique opportunity for further development into fluorescent labels for quantifying intracellular targets in live cells. In proof of concept studies, we have developed targeted and untargeted versions of the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor Erlotinib using our novel water-soluble and cellpermeant fluorophore pair. The similarities of the iPAI targeted and untargeted agents to the parent drug were characterized using competitive binding and cytotoxicity assays in cancer cell lines with varied epidermal EGFR expression. Further validation studies were completed demonstrating the ability of iPAI to image drug target engagement, providing a chemical tool kit for quantitative small molecule imaging in the living system. Abstract Body : In the infectious disease domain, exploring hots-pathogen interaction using in vivo imaging is a rapidly evolving area. Although non-human primates (NHP) provide a good model for infectious diseases, their exploration for viral transmission and dissemination by in vivo imaging has not been used extensively. This can be explained by the limited access to adequate structures for imaging these large animals with a high resolution while having specific regulations according to the level of confinement for the human pathogens. Our main objectives are to develop minimally invasive technologies for the longitudinal monitoring of infections, host response and treatments in NHP. NHP models for immune related disorders and human infectious diseases (such as CoVid-19, SIV, flu, yellow fever or whooping cough for instance) for studying pathophysiology and for human vaccines, immunotherapies and anti-microbial treatments are developed. In vivo imaging (near infrared fluorescence, probe based confocal endomicroscopy, echography, two-photon microscopy and PET-CT) is performed in BSL2 and BSL3 conditions. The two-photon microscope and the PET-CT camera suites are separated into two sides: a biologically "hot" side (pathogens present) and a biologically "cold" side (pathogens not present). For the PET-CT camera, a transparent biocontainment polycarbonate tube extends the hot side of the imaging suite into the bores of the imaging system machines on the cold side ( Figure 1 ). The animal is placed on a patient table in the hot side of the barrier wall, which then moves it into the biocontainment tube to be imaged. The clear tubes also allow scientists to observe the animals being imaged from the cold side. The imaging machines themselves are located outside of the containment on the cold side of the barrier wall and are thus accessible for adjustments and maintenance without requiring technicians to enter the hot side and also to allow the decontamination of the imaging equipments easily. The behavior of skin antigen presenting cells following intradermal immunization with different vaccine vectors was characterized using probe based confocal endomicroscopy in order to better understand the mechanisms leading to the induction of cellular and humoral immune responses after vaccination1,2. Furthermore, noninvasive in vivo imaging procedures were developed to track bacterial localization and cellular interactions with host cells in the lower respiratory tract of challenged and naturally infected animals in a model of whooping cough in baboons3. Ongoing studies demonstrates the susceptibility of NHP to SARS-CoV-2 infection and the value of CT and PET-CT for characterizing the lesions of respiratory tract and score the disease severity and treatment impact. Development of PET-CT approaches using immuno-PET or PET reporter gene imaging to track pathogens and immune cells at the whole body scale is also ongoing. This was the retrospective study which was performed to evaluate the efficacy of 68Ga-DOTATOC PET/CT in patients with diagnosis of or suspected NET with metastases.A subgroup of patients with metastases and unknown primary after initial work-up was analyzed. The 68GaGa-DOTATOC Whole body PET/CT was done .This was evaluated by the SUV max uptake values. The study was considered true positive if the positive primary site has significant uptake and which was confirmed by histology or follow-up imaging. The scan was considered false positive if no primary lesion was found corresponding to 68aGa-DOTATOC positive site. All negative scans for primary tumor were considered false negative. A scan was classified unconfirmed if 68Ga-DOTATOC PET/CT suggested a primary, however, no histology was obtained and imaging follow-up was not confirmatory. RESULTS:Forty patients with known metastatic NET and unknown primary underwent 68Ga-DOTATOC PET/CT.After evaluation and quantification the study was True positive, false positive, false negative and unconfirmed rates for unknown primary tumor were 38%, 7%, 50% and 5% respectively CONCLUSION:The efficacy of 68Ga-DOTATOC PET/CT is an effective modality in localization of unknown primary in patients with metastatic NET as compared to other modalities including F FDG PET-CT . The amount of vascular uptake was graded using a 4-point scale 0=no uptake 1=less than liver 2=similar to liver 3=higher than liver Grade 0-1 was negative, 2 was moderately positive and 3 was markedly positive. This PET/CT was correlated with clinical indices of ITAS (Indian Takayasu Activity Score) and Kerr/National Institute of Health (Kerr/NIH) Serum acute-phase reactants ESR ,C-reactive protein [CRP]) levels ,Interleukin-6 (IL-6) and Soluble IL-6 receptor (sIL-6R RESULTS: 43% of 54 PET-CT were negative, 31% were moderately positive, and 26% were markedly positive. A significant correlation between the SUV uptake and both ESR and CRP levels was found and correlated. Significantly higher ESR values were observed in pts with markedly positive PET/CT (49.4 + 36.5 mm/1st h) compared with moderately positive (27+ 21 mm/1st h, p = 0.0001) and inactive scans ( 22.7 + 15.9 mm/1st h, p=0.0001). CRP levels were 0.8+1.0 mg/dL in pts with inactive scans, 1.3+ 2.2 mg/dL in pts with moderately positive (p=0.001) and 3.0 + 3.6 in patients with markedly positive scans (p = 0.0001). Higher levels of IL-6 resulted in patients with markedly positive scans (10.0 + 8.9 pg/ml) compared to those with inactive scans (8.1+18.5 pg/ml, p=0.013). We found no association between sIL-6R levels and vascular FDG uptake. There was a significant association between vascular FDG uptake and both ITAS and Kerr/NIH scores CONCLUSION:The above findings of PET/CT is a very useful tool for evaluating disease activity in patients with LVV.

Full Name of Abstract's 1st Author : Sikandar Shaikh Abstract Body : Introduction: Glucosaminoglycans (GAG) are a family of polysaccharides which have an extensive role in the structure of the human body. Osteoarthritis (OA), one of the world's most common musculoskeletal disorders and age-dependent conditions, involves breakdown of joint cartilage, whose molecular structure is largely composed of extracellular glucosaminoglycans [1, 2] . GAG-weighted chemical exchange saturation transfer (gagCEST) MRI has been demonstrated a number of times in the literature [3] [4] [5] [6] . In this study, a previously described gagCEST protocol is applied for imaging knee cartilage in older adults, a subset of whom have been diagnosed with OA. Methods: This protocol was approved by the Institutional Review Board of the Veterans Affairs Medical Center of Philadelphia. Eighteen subjects participated in the first stage of this study; all were male veterans of the United States armed forces over the age of 55, including both patients diagnosed with osteoarthritis of the knee (n = 8) and healthy controls. Patients underwent scanning of the knee in which OA was present; in controls with no knee pain, the scan was always performed on the right knee. All MR imaging studies were performed in a Siemens 7T whole-body MRI scanner (Siemens Medical Systems, Erlangen, Germany) using a circularly polarized (CP) transmit /28-channel receive array knee coil of inner diameter 15.4cm to 18cm ( Quality Electrodynamics, Mayfield Village, OH). The imaging protocol consisted of a localizer, a T2 weighted structural image with the full field of view of the knee coil, and a series of measurements specific to the CEST experiment with a field of view (slab) restricted to a smaller volume. These scans were all performed with custom sequences which are described in detail in [6] . Post-processing was done in Matlab. Results: Even amongst healthy controls, gagCEST "hot spots" frequently occurred in the medial articular cartilage; other subjects had relatively homogeneous gagCEST maps. The figure shows images from four subjects, the left knee of one OA patient (red box) and three healthy controls. Axial (left) and coronal (right) views are shown of the same location in the knee joint, with T2weighted structural images shown by themselves alongside the gagCEST overlay. No ROI is drawn on the gagCEST maps; the jet colorscale shown is 0-20% normalized asymmetry, rendering pixels with no signal navy blue. (It can be seen from this representation that gagCEST is largely able to distinguish cartilage from other tissues, with some occasional spurious signal.) In the figure, the OA subject (red box) clearly has low gagCEST signal throughout the joint, which was typical for all patients. Healthy controls whose CEST maps showed notable features frequently presented the pattern visible in the coronal images here, wherein the medial side of the knee (subjects' left side, L, of right knee) had patches of elevated signal. Conclusions: While this data has so far been subjected to only visual qualitative analysis, it was observed that subjects with heterogenous gagCEST tend to have elevated signal in the medial part of the knee cartilage, which is known to be under high amounts of biomechanical stress in many individuals and is frequently the site of joint pain. Patients with OA generally had very sparse gagCEST signal overall, limiting detection of similar patterns without more quantitative analysis; ongoing work includes the integration of our gagCEST maps with automatic segmentation and thickness calculations performed by PyKneeR [7] The possibility that increased GAG is a physiological response to joint stress suggests that longitudinal studies monitoring the strength of the gagCEST signal in aging joints could shed light on the metabolic etiology of osteoarthritis. In addition, gagCEST can be used to monitor the effecticacy of common interventions such as injection of hydroluoronic acid.