key: cord-0333040-448qf7ol authors: Orlow, I.; Sadeghi, K. D.; Edmiston, S. N.; Kenney, J. M.; Lezcano, C.; Wilmott, J. S.; Cust, A. E.; Scolyer, R. A.; Mann, G. J.; Lee, T. K.; Burke, H.; Jakrot, V.; Shang, P.; Ferguson, P. M.; Boyce, T. W.; Funchain, P.; Ko, J. S.; Ngo, P.; Rees, J. R.; OConnell, K.; Hao, H.; Parrish, E.; Conway, K.; Googe, P. B.; Ollila, D. W.; Moschos, S. J.; Hernando, E.; Hanniford, D.; Argibay, D.; Amos, C. I.; Lee, J. E.; Osman, I.; Luo, L.; Kuan, P.-F.; Aurora, A.; Gould Rothberg, B. E.; Bosenberg, M. W.; Gerstenblith, M. R.; Thompson, C.; Bogner, P. N.; Gorlov, I. P.; Holmen, S. L.; Brunsgaard, E. K. title: InterMEL: An international biorepository and clinical database to uncover predictors of survival in early-stage melanoma date: 2022-05-23 journal: nan DOI: 10.1101/2022.05.21.22275329 sha: 26edb87c193e81d148f3e0290cf55fe1c94558fa doc_id: 333040 cord_uid: 448qf7ol Introduction We are conducting a multicenter study to identify classifiers predictive of disease-specific survival in patients with primary melanomas. Here we delineate the unique aspects, challenges, and best practices for optimizing a study of generally small-sized pigmented tumor samples including primary melanomas of at least 1.05mm from AJTCC TNM stage IIA-IIID patients. This ongoing study will target 1,000 melanomas within the international InterMEL consortium. We also evaluated tissue-derived predictors of extracted nucleic acids quality and success in downstream testing. Methods Following a pre-established protocol, participating centers ship formalin-fixed paraffin embedded (FFPE) tissue sections to Memorial Sloan Kettering Cancer Center for the centralized handling, dermatopathology review and histology-guided coextraction of RNA and DNA. Samples are distributed for evaluation of somatic mutations using next gen sequencing (NGS) with the MSK-IMPACT TM assay, methylation-profiling (array), and miRNA expression (Nanostring nCounter). Results Sufficient material was obtained for screening of miRNA expression in 683/685 (99%) eligible melanomas, methylation in 467 (68%), and somatic mutations in 560 (82%). In 446/685 (65%) cases, aliquots of RNA/DNA were sufficient for testing with all three platforms. Among samples evaluated by the time of this analysis, the mean NGS coverage was 249x, 59 (18.6%) samples had coverage below 100x, and 41/414 (10%) failed methylation QC due to low intensity probes or insufficient Meta-Mixed Interquartile (BMIQ)- and single sample (ss)- Noob normalizations. Six of 683 RNAs (1%) failed Nanostring QC due to the low proportion of probes above the minimum threshold. Age of the FFPE tissue blocks (p<0.001) and time elapsed from sectioning to co-extraction (p=0.002) were associated with methylation screening failures. Melanin reduced the ability to amplify fragments of 200bp or greater (absent/lightly pigmented vs heavily pigmented, p<0.003). Conversely, heavily pigmented tumors rendered greater amounts of RNA (p<0.001), and of RNA above 200 nucleotides (p<0.001). Conclusion Our experience with many archival tissues demonstrates that with careful management of tissue processing and quality control it is possible to conduct multi-omic studies in a complex multi-institutional setting for investigations involving minute quantities of FFPE tumors, as in studies of early-stage melanoma. We are conducting a multicenter study to identify classifiers predictive of disease-specific survival in 159 patients with primary melanomas. Here we delineate the unique aspects, challenges, and best practices 160 for optimizing a study of generally small-sized pigmented tumor samples including primary melanomas of 161 at least 1.05mm from AJTCC TNM stage IIA-IIID patients. This ongoing study will target 1,000 melanomas 162 within the international InterMEL consortium. We also evaluated tissue-derived predictors of extracted 163 nucleic acids' quality and success in downstream testing. Our experience with many archival tissues demonstrates that with careful management of tissue 190 processing and quality control it is possible to conduct multi-omic studies in a complex multi-institutional 191 setting for investigations involving minute quantities of FFPE tumors, as in studies of early-stage 192 melanoma. 193 194 195 196 197 198 Introduction 200 Melanoma accounts for the great majority of deaths from skin cancer with an estimated 7,650 deaths in 201 the USA in 2022 (1). The five-year survival rate of melanoma ranges from 98% for localized disease to less 202 than 30% in patients with distant metastases at the time of diagnosis (2, 3) . Factors known to affect 203 progression and survival include primary tumor characteristics (4, 5), presence of nodal or distant 204 metastases at the time of diagnosis (6), anatomic site of the tumor (7), as well as demographic 205 characteristics such as age at diagnosis (8, 9) and sex (9); however, the prediction of individual outcomes 206 based on clinicopathologic factors has limited power. The addition of molecular characteristics has the 207 potential to improve risk classification and help identify individuals more likely to benefit from a 208 differential surveillance schedule and/or therapies. To date, the most comprehensive efforts undertaken to reveal the multifaceted molecular however, was small (n~67) and the thickness of these tumors was larger than the vast majority of primary 215 melanomas, which are usually very thin at diagnosis. Thus, findings may not be generalizable. To be able 216 to translate research findings into the real clinical and diagnostic practice, it is necessary to utilize 217 melanoma tissue that remains available once histopathologic diagnosis is completed as part of the 218 standard care. Therefore, for findings to be generalizable, investigations need to focus on the use of 219 limited archival tissue from small melanomas. In this article we describe our experiences organizing the tissue collection and processing of 221 specimens that are being collected by the multi-institutional InterMEL Consortium (11), a study designed to define molecular subtypes of cutaneous primary melanoma of at least 1. 05 mm in patients diagnosed 223 with AJCC TNM stages IIA through IIID melanoma to improve the prognostic stratification, to guide 224 enhanced surveillance, and/or the use of adjuvant therapies for early intervention, as necessary. The use of targeted and immuno-based therapies for stage III melanoma patients has been shown 226 to be effective in improving recurrence-free survival (12). Also, pembrolizumab has just received FDA 227 approval as adjuvant treatment for stage IIB/C melanoma based on its significant efficacy compared to 228 placebo in preventing both locoregional and distant recurrence in a double-blind clinical trial (13). 229 Therefore, it is pertinent and timely to better identify which patients are at highest risk who would most 230 likely benefit from these treatments and which patients are at lower risk who should be spared of 231 unnecessary, potentially toxic, and costly treatments (14). The InterMEL Consortium is collecting clinical and pathologic data, as well as archived melanoma 233 tissues, to screen for somatic mutations and evaluate miRNA and mRNA expression, methylation, protein 234 expression, and to then correlate these features with survival. A uniquely challenging aspect of the current 235 study is the use of archived primary melanomas, which are typically of limited size. To address this concern 236 our study design involves the systematic co-extraction of DNA and RNA from the same tissue sections and 237 derived cell lysates. This study design allows us to save tissue, and most importantly, it allows us analysis 238 of molecular data from the same portion of tumor, eliminating potential biases due to intratumor 239 heterogeneity. In this report we detail procedures involved in establishing the appropriate infrastructure for our 241 multicenter, integrated study. These include tissues and workflow, sample quality and quantity 242 parameters, and criteria for DNA and RNA distribution for testing, for procuring a balanced proportion of 243 controls (survivors for >5 years with no evidence of disease) and cases (dead within 5-years of follow up) 244 across tests. We also report on the various quality control (QC) parameters used. Based on our experience 245 on already procured data and FFPE tissues from 685 eligible early-stage melanomas from 10 centers, we All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Histopathology-guided co-extraction of nucleic acids from archival 310 tissues 311 For each tumor specimen, systematic marking of qualifying unstained tissue areas is performed. This is 312 followed by careful scraping of tissues and batched co-extraction of nucleic acids using the AllPrep® 313 DNA/RNA FFPE Kit (Qiagen). Details on the preliminary optimization, co-extraction procedures, and 314 aliquoting rules, can be found in the Supporting Information (S1 Appendix, S1 Fig, S2 Appendix). The 315 extraction kits and procedures utilized for procurement of germline DNA (gDNA), and the detailed All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 23, 2022. ; https://doi.org/10.1101/2022.05.21.22275329 doi: medRxiv preprint procedures utilized for the assessment of DNA and RNA quantity and quality, are also found in the 317 Supporting Information (S1 Appendix). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 342 The methodology is described in detail in the Supporting Information, within the S1 Appendix. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 23, 2022. Results from the preliminary assessment of two co-extraction kits done through an independent, pilot set 405 of FFPE tissues include quantity and quality assessment, as well as genetic and epigenetic testing (S2 406 Appendix, S1 Fig, S2 Fig) (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The tumor characteristics significantly associated with the obtained NA quantity and/or quality are There was a positive association between pigmentation and quantities of total RNA 442 obtained, with heavily pigmented tumors rendering higher yields compared to both lightly pigmented and 443 non-pigmented tumors (Fig 2E) . DV200 values were also higher in RNAs extracted from pigmented lesions 444 (Fig 2F) . Ulceration was significantly associated with greater DNA A260/A280 ( Fig 2G) (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Fig 3 depicts the effect of case characteristics (Fig 3A) , time elapsed from sectioning to 451 co-extraction (Fig 3B) , and tumor characteristics (Fig 3C-D (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We evaluated the effect of survival, marginally associated with failures ( Fig 5A) , to assess 492 differences in the attrition between cases and controls. Characteristics associated with methylation 493 screening failure were year of diagnosis (i.e. age of the blocks) (Fig 5B) , time elapsed from sectioning to 494 extraction (Fig 5C) , %dsDNA (Fig 5D) , and amplifiability ( Fig 5D) . Six of 683 (1%) RNA samples failed the 495 Nanostring-based QC. These samples were flagged and then deemed unsuitable for data analysis because 496 of their low proportion of probes above the minimal threshold. Inter-batch replicas showed a very similar 497 miRNA expression profile before (r>0.8) and after correcting for batch effect (r>0.9). There were no 498 variables found in association with Nanostring flags or failures. Of the samples sent for mutations screening with the MSK-IMPACT TM assay, 5 samples were subsequently 503 deemed ineligible, 317/560 (57%) were run through the pipeline, and had data available for the present 504 analysis. On average, the coverage was 249x, with a range of 22x to 762x, and 59 (18.6%) samples had 505 coverage below 100x (Fig 4) . The recommendations to proceed (or hold samples back) with the synthesis 506 of libraries or MSK-IMPACT assay were significantly associated with coverage (p<0.01) (S6 Fig); however, 507 no significant associations were found between the amount of tissue extracted or any of the NA 508 characteristics with samples scored as 'fail' by the genetics core (DNA-QC, Lib-QC). For details on data 509 availability, see S1 Appendix. We have delineated a step-by-step approach for the handling of biospecimens and for data collection in 513 a large multi-center international study, including quality assurance and quality control, in preparation for 514 multi-omics testing from small sample size archival primary melanoma tissue. A unique aspect of this 515 study is the testing of DNA and RNA co-extracted from the same cell lysate obtained from FFPE tissues. This co-extraction is critical for the proper integration of data obtained from multi-omics platforms utilized 517 for the screening of mutations, methylation, and miRNA profiling, with protein expression and mRNA 518 planned in the near future. FFPE tissue blocks of primary melanomas are extremely valuable as they represent the target 520 lesion or 'index tumor' of interest and are accompanied by a vast array of clinicopathologic information. The process of formalin fixation and paraffin embedding helps conserve the cells topography, and blocks All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 23, 2022. ; https://doi.org/10.1101 doi: medRxiv preprint matching germline DNA is not available or the amount and/or quality inadequate, we submit tumor-only 547 batches for analyses against a pool of normal DNA. To this effect, a robust pipeline has been developed 548 to identify germline artifacts, taking into account coverage, allele level, copy number, and tumor purity 549 (Arshi Arora, manuscript in preparation). A related concern is the choice of volume for the elution of 550 nucleic acids during the extraction (normal tissue) and co-extraction (tumors will be naturally more susceptible to degradation, resulting in greater attrition, despite the use of a kit to 566 restore degraded FFPE DNA (Fig 5) . Melanomas with double-stranded DNA quantities below the threshold 567 for methylation and mutation screening (~320 to 350ng double-stranded DNA), are adjudicated in nearly 568 equal numbers for testing with one or the other DNA platform. In addition, in making this choice we 569 considered their case-control status as well as availability (or not) of germline DNA for NGS. We also All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. cellularity. Samples with greater purity and %dsDNA were better scored by the core, but not those with All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 23, 2022. ; https://doi.org/10.1101/2022.05.21.22275329 doi: medRxiv preprint higher RNA integrity. We have not found an explanation for this effect yet, although, no single variable or 595 set of DNA variables was associated alone with the core's prediction for failure, or with the coverage 596 obtained through the MSK-IMPACT TM assay. This differs to some extend to findings from others who 597 reported block age as the most important variable that influenced sample viability for amplicon-based 598 library construction (22) . For methylation, on the other hand, there are clear associations between year 599 of diagnosis (as a proxy of block's age), time elapsed between sectioning and extraction, %dsDNA, and 600 amplifiability, with failures ( Fig 5) . Another important consideration is the jurisdiction of the centers providing tissues and data. For 602 instance, in 2020, a European center expressed interest in joining our efforts and contributing with a 603 substantial number of cases. However, to comply with the General Data Protection Regulation 2016/679 604 (regulation in EU law, on data protection/privacy in the EU), a thorough analysis, numerous audits and 605 reviews are taking place in relation to detailed information on (i) where specimens and data from EU will 606 be located, (ii) who will have access to samples and data, (iii) how data and specimens are protected, and 607 (iv) which systems and applications will be used to process, store, and run data derived for the samples 608 from EU. This major delay may potentially affect the feasibility of including these specimens if samples 609 cannot be procured and tested in time for analyses, especially since tissues cannot be sectioned in 610 advance. International collaborations are important for ensuring worldwide representation of patient 611 samples. Investigators should plan for these international regulatory steps, when planning future 612 consortia. In summary, we find only two variables that, to some extent, could be controlled in future studies 614 of archival tumors and that have a detrimental effect on the quality of the extracted DNA from melanoma 615 tumors: age of the FFPE tissue blocks, and time elapsed between block sectioning and co-extraction. Because of the large individual sample variability, we were not able to define any variable predictive of 617 success on NGS; consequently, we will continue using a combination of quantity and quality parameters All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 23, 2022. ; https://doi.org/10.1101/2022.05.21.22275329 doi: medRxiv preprint when choosing and preparing aliquots for screening DNA samples with the MSK-IMPACT TM , and we will 619 continue aiming to decrease the time elapsed between sectioning, extraction, and testing. The co-620 extracted RNA does not appear to be affected by the age of the block or time elapsed, but surprisingly, 621 presence of melanin might impart some protection against RNA degradation in FFPE tissues. The greater 622 RNA amounts and RIN values obtained in pigmented melanomas, to the best of our knowledge, are not 623 artifactual, and are possibly due to a change in acidity or pH. Although pigment has no impact on the 624 success of the miRNA screening with Nanostring, our observations are intriguing and worthy of further 625 consideration. Of the 685 eligible cases, 65% of the melanomas rendered sufficient material for testing by 626 all three omics platforms. The implication is that in future studies of early-stage melanomas using archival 627 tissues, one should aim to accrue sufficient material to allow a 35% attrition, when the goal is to obtain 628 genetic and epigenetic profiling such as ours. Furthermore, when considering the methylation failures, 629 the ideal target would permit 40% attrition. We anticipate that these reported practical guidelines for the 630 procurement, preparation, quality control, and testing of co-extracted DNA and RNA for the screening of 631 actionable somatic mutations, methylation and miRNA using high-throughput platforms, as well as our 632 data and insight will be of value to others with their ongoing and future investigations using valuable 633 archival yet limited tissues. 634 635 636 Weida Gong 645 (Bioinformatician), Amanda Vondras (Bioinformatician), Lan Lin (Database Manager); Study centers 646 include the following: Memorial Sloan Kettering Cancer Center Assistant Research Biostatistician) Gbemisola Elizabeth Ilelaboye 651 (Senior Research Technician); initial optimization assays were performed at MSK by Heta Parmar Anne E. Cust (Epidemiologist and Site PI) Cancer Geneticist and Former Site PI) Peter M. Ferguson (Pathologist), Valerie Jakrot Una Moran (Data Technician), for 659 the Genomics Technology Center: Adriana Heguy (Director), Sitharam Ramaswami Marc Ernstoff (Advisor) Site PI), Shenying Fang Site 664 PI) Co-Investigator) Dermatopathologist 666 and Site PI), Pauline Funchain (Co-Investigator) Former Site PI), Gauri 668 Huntsman Cancer Institute Co-Investigator) Celia Requena (Dermatologist), Victor Traves (Pathologist) Patient Advocate: Michelle Rainka The contribution of our colleagues at the Melanoma Institute Australia and Royal Prince Alfred Hospital is 675 gratefully acknowledged Project Manager), Sharon Bayuga (Associate Director Marketing & Communications: Susan Weil (Design & Creative Services 681 Production Manager) Cancer statistics, 2022 MD: National Cancer Institute Predicting survival 689 outcome of localized melanoma: an electronic prediction tool based on the AJCC Melanoma Database Malignant melanoma in 692 the elderly: different regional disease and poorer prognosis Malignant melanoma: clinical variants and prognostic indicators Melanoma staging: 696 Evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging 697 manual Anatomic location of primary 699 melanoma: Survival differences and sun exposure Sex is an independent 701 prognostic indicator for survival and relapse/progression-free survival in metastasized stage III to IV 702 melanoma: a pooled analysis of five European organisation for research and treatment of cancer 703 randomized controlled trials Age and gender are 705 significant independent predictors of survival in primary cutaneous melanoma The 711 State of Melanoma: Emergent Challenges and Opportunities Pembrolizumab versus placebo as adjuvant therapy in completely resected stage IIB or IIC melanoma 714 (KEYNOTE-716): a randomised, double-blind Retrospective analysis of clinical trial safety data for 716 pembrolizumab reveals the effect of co-occurring infections on immune-related adverse events Memorial Sloan Kettering-719 Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT): A Hybridization Capture-720 Mutational landscape of 723 metastatic cancer revealed from prospective clinical sequencing of 10,000 patients Factors that drive the increasing use of FFPE 727 tissue in basic and translational cancer research Comparison of protocols for 729 removal of melanin from genomic DNA to optimize PCR amplification of DNA purified from highly 730 pigmented lesions A versatile method for the removal of 732 melanin from ribonucleic acids in melanocytic cells Evaluation of Tumor infiltrating 734 lymphocytes in breast carcinoma and their correlation with molecular subtypes, tumor grade and stage Suitability of melanoma FFPE samples for NGS libraries: time and quality thresholds for downstream 738 molecular tests Appendix 1: Methods Appendix 2: Preliminary Results Appendix 3: Supporting Figures Quantity and quality assessment of RNA and DNA co-extracted from FFPE Performance of RNA co-extraction kit and elution on Nanostring TM Effect of FFPE tissue input on the yield of co-extracted nucleic acids S4 Fig. Effect of TILs and pigmentation on DNA purity and amplifiability Effect of the DNA and RNA quality on recommendations to proceed with synthesis of 750 libraries and NGS/MSK-IMPACT TM Comparison of DNA-and library-QC recommendations with coverage obtained in 752 melanomas screened with the MSK-IMPACT TM assay