Use of a Foamy-Virus Vector System to Produce an 'Off-the-Shelf' Fcγ-CR-T Cell Product for the Treatment of Hematological and Solid Tumour Malignancies

Introduction Chimeric antigen receptor (CAR) T cells have transformed the field of cancer immunotherapy. Other approaches, such as the use of Fc gamma chimeric receptor (Fcγ-CR)-T cells have further expanded the applicability of such therapies in both solid and liquid tumours with the added benefit of tackling some of the hurdles associated with CAR-T therapies (such as loss or down-regulation of the target antigen). Previous studies have generated Fcγ-CR-T cells from autologous cells, which carry several limitations (prolonged production time, costly, manufacturing failure), whereas the use of Lentiviral vectors (LV) is endowed with extra limitations (packaging limits and mutagenesis risk). Our group has developed an in-house Fcγ-CR-T cell product, using a safer to LV, foamy virus (FV) vector and T cells from healthy donors. Materials We constructed FV vectors expressing the CD16 V158, which has higher Fc binding and is associated with higher tumour killing, the T cell stimulatory molecule CD3ζ and the costimulatory molecule CD28 and an EFP1a promoter. 2nd generation LV vector backbones were purchased from a commercial vendor. Peripheral blood from healthy individuals were used as T cell sources. T cells were activated by anti-CD3/CD28 beads and transduced with CD16-CR, LV or FV vectors. Transduction efficiency was assayed by flow cytometry (FCM) using an anti-CD16 antibody (Ab), on day 3. The human cell lines Raji, Panc01 and DLD-1 were used for functional assays, in the presence of the Abs Rituximab (anti-CD20) and Cetuximab (anti-EGFR), respectively, at a concentration of 0.1 ug/ml. The CR's Ab-binding capacity was assessed by incubating the CD16-CRs with the Abs for 30mins, followed by an anti-human Ig PE antibody and the related fluorescence intensity was assessed. To determine whether Ab binding to CD16-CR can promote aggregation of effector and target cells, CFSE-labelled target cells were mixed with Ab-coated CD16-CRs for 60 mins and the formation of CFSE-PE doublets was assessed by FCM. Their cytotoxic effect was evaluated against the CFSE-labelled Raji or DLD-1 or Panc01 cells at different ratios (5:1, 10:1) for 18 hours, in the presence of Rituximab (0.1ug/ml) or Cetuximab (0.1ug/ml), respectively. The % of live cells was assessed by flow cytometry and calculated as: [1-live targets (sample)/live targets (control)]x100. Results LV and FV vector titers were between 3-5x10^6 TU/ml and 4-5x10^6 TU/ml, respectively. Transduction efficiency ranged from 58.3-69.2% with FV vectors (MOI 3-5) and 85.2-85.9% with LV vectors (MOI 10-20). The Ab-binding capacity of FV-CD16-CRs was determined to be 68.7% (SEM 2.3) and 71.3% (SEM 3.1), (n=3) for Rituximab and Cetuximab, respectively, whereas that of LV-CD16-CRs was 72.1 (SEM 3.3) and 76.5 (SEM 2.8), (n=3), respectively. Cell aggregation of effector and target cells, assessed as doublets, was: (i) 32%, 39% and 36% for FV-CD16-CRs and (ii) 26%, 31% and 29% for LV-CD16-CRs, coated with Rituximab and Cetuximab, respectively, and it was specific for the Raji, DLD-1 and Panc01 cells, respectively. Next, we assessed whether CD16-CR T cells were able to kill target cells in the presence of specific antibodies. Results showed that the FV-CD16-CR-induced % cell lysis was: (i) 26.3% and 51.5% at 5:1 and 10:1 ratio, respectively, in the presence of Rituximab and Raji cells, (ii) 41.5% and 57.7% at 5:1 and 10:1 ratio, respectively, in the presence of Cetuximab and DLD1 cells and (iii) 39.4% and 57.3% at 5:1 and 10:1 ratio, respectively, in the presence of Cetuximab and Panc01 cells. For LV-CD16-CRs the respective % lysis was comparable. More importantly, this lysis was shown to be specific (no lysis noted with untransduced T cells) and significantly lower in the absence of the antibodies. Conclusion Our group has developed for the first time a FV vector for the generation of CD16-CR T cells, with an efficient gene transfer to human T cells and with potent in vitro cytotoxic properties, similar to their LV-derived counterpart. Overall, we provide a proof of concept that allogeneic, in-house Fcγ-CR-T cells derived from a non-pathogenic viral backbone such as the FV, could be a safe, efficient and affordable alternative to LV-derived vectors for immunotherapy. Our future aim is to further modify these Fcγ-CR-T cells, by interfering with immune checkpoint molecules, in order to achieve better tumour penetration and tackle the anergy induced by the tumour microenvironment.