Genetic Modification of Tumor-Infiltrating Lymphocytes via Retroviral Transduction

Hadas Weinstein-Marom, Gideon Gross, Michal Levi, Hadar Brayer, Jacob Schachter1, Orit Itzhaki and Michal J. Besser1, Hadas Weinstein-Marom, Gideon Gross, Michal Levi1, Hadar Brayer, Ella Lemelbaum Institute for Immuno-Oncology, Sheba Medical Center, Ramat Gan, Israel, 2 Laboratory of Immunology, MIGAL-Galilee Research Institute, Kiryat Shmona, Israel, 3 Department of Biotechnology, Tel-Hai College, Upper Galilee, Israel, 4 Department of Clinical Microbiology and Immunology, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel

Adoptive T cell therapy (ACT) holds great promise for cancer treatment. One approach, which has regained wide interest in recent years, employs antitumor T cells isolated from tumor lesions (“tumor-infiltrating lymphocytes” or TIL). It is now appreciated that a considerable proportion of anti-melanoma TIL recognize new HLA-binding peptides resulting from somatic mutations, which occurred during tumor progression. The clinical efficacy of TIL can potentially be improved via their genetic modification, designed to enhance their survival, homing capacity, resistance to suppression, tumor killing ability and additional properties of clinical relevance. Successful implementation of such gene-based strategies critically depends on efficient and reproducible protocols for gene delivery into clinical TIL preparations. Here we describe an optimized protocol for the retroviral transduction of TIL. As the experimental system we employed anti-melanoma TIL cultures prepared from four patients, recombinant retrovirus encoding an anti-CD19 chimeric antigen receptor (CAR) as a model gene of interest and CD19+ and CD19- human cell lines serving as target cells. Transduction on day 7 of the rapid expansion protocol (REP) resulted in 69 ± 8% CAR positive TIL. Transduced, but not untransduced TIL, from the four patients responded robustly to CD19+, but not CD19- cell lines, as judged by substantial secretion of IFN-g following co-culture. In light of the rekindled interest in antitumor TIL, this protocol can be incorporated into a broad range of genebased approaches for improving the in-vivo survival and functionality of TIL in the clinical setting.

INTRODUCTION

ACT with autologous TIL produced from resected melanoma biopsies combined with nonmyeloablative

lymphodepletion and IL-2 exceeds an overall clinical response rate of 41% [95% confidence interval (CI) 35% to 48%] in patients with highly advanced metastatic melanoma according to meta-analysis (1–5). The TIL protocol consists of three basic steps: (i) patient preconditioning by lymphodepleting regimen; (ii) Isolation of TIL from tumor biopsies and ex-vivo expansion of generated TIL cultures with the anti-CD3 antibody OKT3, irradiated feeder cells and IL-2; (iii) TIL administration i.v. followed by IL- 2 infusion. We have been applying for over a decade an open label phase II clinical study in stage IV melanoma patients (2, 6–11). Our initial objective was to assess the efficacy and safety of ACT with autologous TIL, which was offered as a second line of treatment following IL-2–based therapy until 2011 and, since, as a salvage treatment following standard of care therapy including other immunotherapy modalities (e.g., anti-CTLA-4 and anti-PD-1 antibodies). Following surgical resection and processing of a metastatic lesion, TIL cultures are typically generated in IL-2– containing medium within 2 weeks to 1 month. Established TIL cultures are then further expanded in a rapid expansion protocol (REP) to treatment levels. REP is initiated with an anti CD3 agonist, irradiated feeder cells and IL-2, which drives the expansion of TIL to approximately 1000-fold within 14 days. Clinical results of 57 patients treated at our medical center have previously been published (10). Potential predictors for response were shorter time to TIL culture generation, elevated fold large scale expansion, higher total number of cells, higher total number of CD8 T cells and higher frequency of CD8 T cells in the cell product (10).

In spite of the tremendous progress achieved in TIL therapy, tumor specimens may fail to give rise to viable TIL cultures and a high proportion of melanoma patients treated with either TILACT strategy are non-responders (5, 12, 13). Although many responders exhibit long-lasting tumor regression and complete responders only rarely relapse, new strategies are needed for maximizing response rate and duration and, most importantly, for providing a platform for the treatment of other solid tumors. Undoubtedly, a promising route for improving the clinical efficacy of TIL therapy is the enhancement of the functional properties of TIL cultures via genetic modification. To achieve this goal we have recently created several classes of “genetic adjuvants,” including membrane-attached cytokines, truncated, constitutively active toll-like receptors (caTLRs) and spontaneously homo-oligomerizing, constitutively active tumornecrosis factor receptors (caTNFRs).Using electroporation of antimelanoma TIL with in-vitro transcribed mRNAs encoding membrane IL-2, IL-12, IL-15, caTLR4, or caCD40, alone or in different combinations, we could demonstrate their additive, and often synergistic effects on key parameters concerning TIL function and survival (14–18).

The first clinical assessment of this approach was reported by S. Rosenberg et al. (19) who had successfully introduced the neomycin-resistance gene into anti-melanoma TIL derived from five patients via retroviral transduction. Gene-expression in modified TIL could be detected in the circulation of these patients three weeks to two months post-administration. In following studies, the Rosenberg group demonstrated the retroviral transduction of anti-melanoma TIL with the IL-2 gene (20) or with an inducible IL-12 gene placed under the control of a nuclear factor of activated T cells- (NFAT)- responsive promoter (21). While all these studies employed selected anti-melanoma TIL, Forget et al. recently described a protocol for the retroviral transduction of TIL, evaluating the gene encoding the chemokine receptor CXCR2 (22). Here we describe in detail our optimized protocol for retroviral transduction of anti-melanoma TIL cultures and demonstrate its efficacy in endowing transduced TIL with a new antigenic specificity via an anti-CD19 CAR.