key: cord-0764393-3pxq08qz
authors: Sivapalan, Rohan; Liu, Jinyan; Chakraborty, Krishnendu; Arthofer, Elisa; Choudhry, Modassir; Barie, Philip S; Barouch, Dan H; Henley, Tom
title: A Novel Cell Therapy for COVID-19 and Potential Future Pandemics: Virus Induced Lymphocytes (VIL)
date: 2020-11-27
journal: bioRxiv
DOI: 10.1101/2020.11.26.400390
sha: 442acd653f1c0fa9f819e94ea676673038a9897b
doc_id: 764393
cord_uid: 3pxq08qz

The a priori T cell repertoire and immune response against SARS-CoV-2 viral antigens may explain the varying clinical course and prognosis of patients having a mild COVID-19 infection as opposed to those developing more fulminant multisystem organ failure and associated mortality. Using a novel SARS-Cov-2-specific artificial antigen presenting cell (aAPC), coupled with a rapid expansion protocol (REP) as practiced in tumor infiltrating lymphocytes (TIL) therapy, we generate an immune catalytic quantity of Virus Induced Lymphocytes (VIL). Using T cell receptor (TCR)-specific aAPCs carrying co-stimulatory molecules and major histocompatibility complex (MHC) class-I immunodominant SARS-CoV-2 peptide-pentamer complexes, we expand virus-specific VIL derived from peripheral blood mononuclear cells (PBMC) of convalescent COVID-19 patients up to 1,000-fold. This is achieved in a clinically relevant 7-day vein-to-vein time-course as a potential adoptive cell therapy (ACT) for COVID-19. We also evaluate this approach for other viral pathogens using Cytomegalovirus (CMV)-specific VIL from donors as a control. Rapidly expanded VIL are enriched in virus antigen-specificity and show an activated, polyfunctional cytokine profile and T effector memory phenotype which may contribute to a robust immune response. Virus-specific T cells can also be delivered allogeneically via MHC-typing and patient human leukocyte antigen (HLA)-matching to provide pragmatic treatment in a large-scale therapeutic setting. These data suggest that VIL may represent a novel therapeutic option that warrants further clinical investigation in the armamentarium against COVID-19 and other possible future pandemics.

Like their tumor-resident counterparts, TIL, Virally-Induced Lymphocytes, or VIL, represent 84 those T cells that have been activated in response to TCR-mediated, antigen-specific recognition of 85 protein epitopes from viral particles 9, 10 . Cytotoxic CD8 + T cells play a crucial role in mediating viral 86 clearance in response to many respiratory viral infections including respiratory syncytial virus (RSV), 87 influenza and coronavirus (CoV) 5 . Recent evidence has also demonstrated a critical role for the T cell 88 immune response in the pathogenesis of the recently emerged COVID-19 disease, caused by the novel 89 SARS-CoV-2 coronavirus 11, 12, 13, 14 . In addition to the readily detectible humoral immune response in 90 the context of neutralizing antibodies in convalescent patients who had recovered from COVID-19, 91 these studies have collectively shown strong SARS-CoV-2-specific memory T cells are frequently 92 observed 11, 13 . Furthermore, significantly larger T cell responses appear to correlate with severity of 93 the disease, underscoring the importance of T cells above and beyond the humoral antibody response 94 for combating infection 11 . This is an important consideration for a novel therapeutic, as most 95 prophylactic vaccines currently in development for COVID-19 are designed to focus on eliciting 96 antibody responses to the spike protein of SARS-CoV-2 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 .

In addition, the antibody response in recovered COVID-19 patients has been shown to decline 98 several months after infection, raising concerns that therapeutics or vaccines designed to elicit 99 primarily neutralizing antibody responses may not be sufficient to engender the cellular immunity 100 required for long-term duration of protection or to protect from potential repeat infections 25. Thus, 101 the a priori T cell repertoire, both quantity and quality, may portend COVID-19 disease prognosis and may influence the outcome between mild or severe disease. The treatment of patients with 103 convalescent sera has recently been given US FDA Emergency Use Authorization (EUA) 26 . The transfer 104 of neutralizing antibodies reflects a more passive serological immune engagement versus the more 105 active cellular immune response that SARS-CoV-2-specific T cells would provide.

In an evolution of the cGMP methods for the expansion of TIL for the treatment of cancer that 107 the researchers employ in a currently active human clinical trial 27 , we sought to develop an adoptive 108 T cell therapy for COVID-19 based on rapid ex vivo expansion of SARS-CoV-2 antigen-specific VIL. Given 109 the crucial importance of a strong virus-specific T cell response for patients with severe disease, 110 especially during the critical days where respiratory distress is common, the adoptive transfer of a 111 quantity of expanded, activated, effector memory T cells capable of mounting a robust virus-specific 112 response may be important to reduce viral load and improve patient outcomes. Thus, we designed a 113 T cell expansion platform comprising of microbead-based artificial antigen-presenting cells (aAPCs), 114 we termed VIPR-particles (VIL-inducing particles Rx), that carry MHC pentamer-peptide or tetramer-115 peptide complexes, specific for immunodominant SARS-CoV-2 epitopes, coupled with costimulatory 116 anti-CD28 antibodies.

Using Peripheral Blood Mononuclear cells (PBMCs) isolated from convalescing COVID-19 118 patients to represent what may be achievable in the clinic for patients actively suffering from the 119 severe disease, we show that these TCR-specific aAPCs can expand virus-specific VIL up to 1,000-fold 120 over a rapid and minimal culturing duration of just 7-days. Furthermore, these expanded VIL are 121 enriched in virus antigen-specificity, show polyfunctional cytokine responses and acquire a T effector 122 memory phenotype, making them highly suited for participating in an active cellular immune response 123 when adoptively transferred back to patients after this minimal ex vivo expansion time. We also 124 demonstrate the broad clinical potential of this platform and its modularity beyond COVID-19 and 125 show in the setting of CMV infection that VIL specific for immunodominant CMV epitopes can also be 126 expanded up to 1,000-fold using CMV-specific VIPR particles over a 7-day culture. To develop a platform for robust and rapid ex vivo expansion of viral antigen-specific T cells from 142 patients exposed to viral pathogens, we generated a micro-aAPC capable of providing an 143 immunogenic viral peptide in the context of MHC Class I or MHC Class II molecules in combination 144 with anti-CD28 stimulation molecules (Fig. 1a) Optimization of antigen-specific micro aAPCs for T cell expansion 171 172 Next we sought to optimize the design of the VIPR particle aAPCs to further enhance the expansion of 173 antigen-specific T cells within the rapid 7-day stimulation culture. To this end, we first investigated 174 the impact that the ratio of T cells to VIPR particles had on the proportion of CMV-specific T cells 175 enriched at day-7. A dose-dependent enrichment was seen with lower doses of particles and higher 176 numbers of T cells, such that an optimal enrichment was observed with a ratio of 20:1 T cells to VIPR 177 particles (Fig. 1d ). In addition, we observed that increasing the ratio of molecules of anti-CD28 178 antibody to peptide-MHC-pentamer also increased the capacity of the VIPR particles for expansion of 179 the antigen-specific VIL population ( 31 . Surprisingly, we found YLQ antigen-specific VIL were 189 barely detectible within the isolated T cell populations from these individuals (Fig. 2a&b ).

However, after 7-day culture with MHC-I VIPR particles, antigen-specific CD8 + T cells could be 191 readily detected by pentamer staining and could be enriched and expanded to frequencies greater 192 than 1% (Fig 2a&b) . While the majority of PBMCs from convalescent individuals with the YLQ matched 193 MHC allele (HLA A*02), included T cells from which antigen specific VIL could be enriched, the overall 194 frequency varied between individuals and did not appear to correlate with either the length of time 195 since they were symptomatic, nor the reported severity of their symptoms (Fig. 2b) 

Rapid VIL expansion results in 1,000-fold enrichment of CMV antigen-specific T cells within 7-198 days 199 200 Having demonstrated that VIPR particles can enrich both SARS-CoV-2 and CMV specific T cells, we 201 evaluated the capacity for the 7-day culture system to expand the overall quantity of virus specific VIL.

The T cell cultures were configured to include the same culturing conditions used in neoantigen TIL 203 human clinical trials to enable rapid T cell expansion, thus in addition to the VIPR particles and high IL-2 (6000 IU/ml), IL-7, IL-15 and NAC, T cells were cultured in Gas Permeable Rapid Expansion (G-REX) 205 plates. These culture plates enable gas exchange from the base of the culture well, allowing cells to 206 be cultured with a large ratio of media per surface area and abundant access to nutrients, and have 207 been shown to facilitate a large and rapid expansion of primary human T cells 32 .

After 7-days of culture with VIPR particles, SARS-CoV-2 antigen-specific CD8 + T cells could be 209 robustly expanded in proportion, but most importantly in absolute quantity of T cells, to an average 210 of over 1,000-fold (Fig. 2c) . Thus, we found that cultures seeded with 1x10 6 total CD3 + T cells could 211 reach expanded numbers, on average, between 2.6x10 7 and 4.5x10 7 total cells at day-7. This 2d). After 7-days, antigen-specific CD8 + T cells had increased from approximately 2x10 3 cells to over 217 1.0x10 6 , leading to up to an average >700-fold expansion in cell number.

Collectively these data demonstrate the ability of the VIPR particle expansion protocol to 219 rapidly enrich and expand VIL from low numbers in CMV-positive individuals and near undetectable 220 numbers in COVID-19 convalescent individuals, to significantly large numbers of virus-specific T cells.

The activation and T cell memory phenotype of rapidly-expanded CMV and SARS-CoV-2 223 antigen-specific VIL 224 225 To evaluate the phenotype of SARS-CoV-2 specific T cells and CMV specific T cells that had undergone 226 enrichment and expansion with VIPR particles, T cells were analyzed for expression of cell surface 227 markers indictive of T cell activation. We observed SARS-CoV-2-specific CD8 + T cells expressing co-228 stimulatory and activation markers 4-1BB, OX-40 and CD25, albeit variable between convalescent 229 individuals, and an elevated expression of HLA-DR when compared to non-virus-specific T cells within 230 the culture (Fig. 3a) . The SARS-CoV-2 antigen-specific VIL population also showed a significant 231 expression of the checkpoint markers PD-1, TIGIT, LAG-3, indicating these T cells have acquired a 232 proliferative and activated functional phenotype (Fig. 3b) . The same profile of activation marker and 233 checkpoint gene expression was observed when CMV-specific VIL were stimulated after 7-day rapid 234 expansion with VIPR particles, with a similarly observed variability between different CMV-positive 235 individuals, indicating this culture platform is effective at rapid T cells expansion and activation with 236 multiple viral antigens (Fig. 3c&d) .

We analyzed the memory phenotype of the expanded SARS-CoV-2 and CMV virus-specific T 238 cells by measuring expression of the canonical memory markers CD45RA and CD45RO and categorized 239 the cell populations into either a naïve (CD45RO -, CD45RA + ) or memory phenotype (CD45RO + , CD45RA -240 ). After the 7-day culture in IL-2, IL-7, IL-15 and NAC, the majority of CMV T cells had begun to adopt 241 a memory phenotype, but the virus-specific CD8 + T cells were almost exclusively expressing the 242 highest levels of CD45RO and completely lost CD45RA expression, indicating the antigen-specific 243 population had uniformly transitioned into memory T cells (Fig. 4a ). Further delineation of the T cell 244 memory phenotype by analysis of CD62L expression within the CD45RO + population revealed the 245 virus-specific T cells had robustly differentiated into an effector memory T cell phenotype via 246 downregulation of CD62L (Fig. 4a&b) . The non-virus-specific T cells within these cultures however, 247 consisted of significantly more naïve T cells. The same profile of effector memory T cells was observed 248 when SARS-CoV-2-specific VIL were stimulated after 7-day rapid expansion with VIPR particles, again 249 demonstrating the antigen-specific VIPR particle platform is effective at significantly expanding 250 activated effector memory T cells over a short time-course. (Fig. 4c&d ).

CoV-2 antigen-specific VIL 254 255 To further evaluate the function of the rapidly expanded virus-specific VIL, we performed intracellular 256 cytokine staining and flow cytometry and measured the proportion of the cells that were producing 257 IFN, TNF and IL-2. VIPR particle expanded T cells from convalescent COVID-19 individuals were 258 stimulated after 7-day culture with 20 g/ml of SARS-CoV-2 YLQPRTFLL peptide antigen for 6-hours.

We observed strong expression of all three proinflammatory cytokines within the antigen-specific T 260 cell population (identified by TCR specific pentamer staining), but could not detect expression of either 261 IFN or TNF in the non-antigen-specific T cell population (T cells that do not bind the TCR-specific 262 pentamer) nor within the T cells cultured for 7-days without any VIPR particle expansion (Fig. 5a&b ).

The antigen-specific T cells also showed significantly elevated levels of IL-2 when compared to the 264 non-antigen specific CD8 + population. When analyzed together we see an elevated proportion of cells 265 expressing 1, 2 or all 3 cytokines in combination when compared to the non-SARS-CoV-2-specific T 266 cells within the expanded culture (Fig. 5c&d ).

The same functional response was observed with virus-specific VIL expanded in T cells isolated 268 from CMV-positive individuals and stimulated for 6-hours with pp65 MHC-I epitope peptide antigen.

Intracellular cytokine staining revealed a robust increase in production of all cytokines in the CMV-270 specific CD8 + T cells when compared to the non-specific T cells from the same cultures or non-stimulated controls (Fig. 5e&f ). An elevated frequency of polyfunctional CD8 + T cells expressing 272 multiple proinflammatory cytokines was also seen in the CMV-specific T cell population ( fig. 5g&h ).

Taken together, these analyses demonstrate that elevated numbers of virus-specific VIL can 274 be rapidly expanded in 7-days by VIPR particle culture and form robust activated, polyfunctional 275 effector memory T cells. we calculate an estimated capability to expand and deliver an average of approximately 3.5x10 9 SARS-320 CoV-2 CD8 + and/or CD4 + T cells back to the patient within 7-days (Fig. 6b ).

The 

The VIPR rapid expansion platform is modular and tunable to multiple viral antigens restricted 385 to different MHC alleles. Using validated CMV epitopes, we generated MHC class II VIPR particles 386 against the DRB1*07:01 restricted gB 215-229 antigen and were able to enrich and expand CD4 + T 387 cells (Fig. 1b) . Thus, this technology can be used to expand both T Helper and Cytolytic virus-specific 388 VIL to provide a modality to tune a specific cell therapy product to treat different viral diseases. The 389 enrichment and expansion of CMV-specific T cells demonstrates the flexibility of the VIPR particle 

Harvest PBMCs from HLA-typed blood and isolate T cells

Volume of blood per kg of bodyweight that can be taken from a patient 5 ml / kg Therefore, approx. total volume of blood for T cell extraction in 70kg male 350 ml Number of total T cells per ml of blood 1x10 6 Total number of T cells extracted per patient 3.5x10 8 Estimated % of SARS-CoV-2 VIL during infection 1% Total number of SARS-CoV-2 VIL isolated per patient 3.5x10 6 Average fold-expansion of SARS-CoV-2 VIL ex vivo after 7 days 1,000-fold Total number of expanded SARS-CoV-2 VIL for patient infusions 3.5x10 9 A B + MHC-I VIPR Particle

Infuse back to HLA-typed patients

Antigen-specific Pentamer/Tetramer microbeads 3.5x10 6 VIL 3.5x10 9 VIL Day 7 Figure 6 

Unbiased Screens Show CD8(+) T Cells of COVID-19 Patients Recognize 554

Shared Epitopes in SARS-CoV-2 that Largely Reside outside the Spike Protein

T cell responses in patients with COVID-19

COVID-19 patients display distinct SARS-CoV-2 specific T-cell responses 560 according to disease severity

The race for coronavirus vaccines: a graphical guide

Single-shot Ad26 vaccine protects against SARS-CoV-2 in rhesus 566 macaques

Safety and immunogenicity of the ChAdOx1 nCoV-19 vaccine against

A Study of Metastatic Gastrointestinal Cancers Treated With Tumor Infiltrating Lymphocytes 603 in Which the Gene Encoding the Intracellular Immune Checkpoint CISH Is Inhibited Using CRISPR 604 Genetic Engineering

An investigation of the effects of the 607 antioxidants, ebselen or N-acetyl cysteine on human peripheral blood mononuclear cells and T cells

Cytomegalovirus (CMV) phosphoprotein 65 makes a large contribution to 611 shaping the T cell repertoire in CMV-exposed individuals

Cell Responses According to Human Leukocyte Antigen Class I Allotypes and Intraindividual 615 Dominance

SARS-CoV-2 epitopes are recognized by a public and diverse repertoire of 618 human T cell receptors

Accelerated production of antigen-specific T cells for preclinical and clinical 621 applications using gas-permeable rapid expansion cultureware (G-Rex)

COVID-19 length of hospital stay: a systematic review and data synthesis

Prompt Predicting of Early Clinical Deterioration of Moderate-to-Severe 628

COVID-19 Patients: Usefulness of a Combined Score Using IL-6 in a Preliminary Study

Simplified method of the growth of human tumor infiltrating lymphocytes in gas-632 permeable flasks to numbers needed for patient treatment

Generation of tumor-635 infiltrating lymphocyte cultures for use in adoptive transfer therapy for melanoma patients

Internal checkpoint regulates T cell 645 neoantigen reactivity and susceptibility to PD1 blockade

Neoantigen-Specific Adoptive Cell Therapies for Cancer

Safety and efficacy of cryopreserved autologous tumor infiltrating lymphocyte therapy

lifileucel) in advanced metastatic melanoma patients who progressed on multiple prior therapies 655 including anti-PD-1

Reduction and Functional Exhaustion of T Cells in Patients With Coronavirus 658 Disease 2019 (COVID-19)

Efficacy of Adoptive T-cell Therapy Is Improved by Treatment with the Antioxidant N-Acetyl Cysteine

Which Limits Activation-Induced T-cell Death

Memory T cells need CD28 costimulation to remember

Differential T cell costimulatory requirements in CD28-deficient mice

Role of CD28-B7 interactions in generation and maintenance of CD8 T cell 671 memory

Duration of TCR stimulation determines costimulatory requirement of T 674 cells

T cell activation determined by T cell receptor number and 677 tunable thresholds

Manufacturing Dendritic Cells for Immunotherapy: Monocyte 680

Virus-specific T-cell therapy to treat BK polyomavirus infection in bone 683 marrow and solid organ transplant recipients

COVID-19 Outbreak

Effect of Dexamethasone on Days Alive and Ventilator-Free in Patients 689 With Moderate or Severe Acute Respiratory Distress Syndrome and COVID-19: The CoDEX 690

Representative proportion of SARS-CoV-2 CD8 + T 735 cells expressing 1, 2 or 3 cytokines. d, Extended analysis of SARS-CoV-2 VIL polycytokine function as 736 SPICE representation. e, Analysis as in a, for CMV-specific CD8 + T cells. f, Histograms analyzed as in b 737 summarizing the data obtained with CMV antigen specific CD8+ T cells (n=3). g, Representative 738 proportion of CMV CD8 + T cells expressing 1, 2 or 3 cytokines. h, Extended analysis of CMV-specific VIL 739 polycytokine function as SPICE representation

Allogeneic VIL Therapy Platform: An adoptive cell therapy for the treatment of individuals 742 suffering from severe symptoms of COVID-19. a

These 746 antigen-specific VIL expand at an average of 1,000-fold prior to adoptive transfer back to HLA-matched 747 patients to mediate a T cell immune response to support the eradication of the SARS-CoV-2 virus and 748 to engender protective immunity against repeat infection. b, Estimations of viral-specific T cell 749 numbers generated ex vivo for patient