Allogeneic NKT Cells Expressing a CD19-Specific CAR in Patients with Relapsed or Refractory B-Cell Malignancies: An Interim Analysis

Abstract Autologous T cells engineered to express a CD19-specific chimeric antigen receptor (CAR) mediate high rates of complete response (CR) in patients with B-cell malignancies. However, autologous cell therapy products are time- and resource-intensive to manufacture and vary in potency and toxicity. Unlike polymorphic HLA-restricted T cells, monomorphic CD1d-restricted Vα24-invariant natural killer T cells (NKTs) are not alloreactive, and therefore therapeutic NKTs can be generated from allogeneic donors without the risk of graft-versus-host disease (GvHD). Moreover, pre-clinical models suggest that CAR-NKTs have inherent advantages over CAR-T products including an ability to trans-activate NK cells, cross-prime tumor-specific CD8 T cells, and recognize CD1d-positive B-cell lymphoma cells via their endogenous NKT T cell receptor. We report interim results from five patients treated on a phase 1 dose-escalation trial of allogeneic NKTs engineered to co-express a CD19-specific CAR, IL-15, and shRNA targeting beta-2 microglobulin and CD74 for downregulation of HLA class I and class II molecules, respectively (ANCHOR, NCT00840853). Four patients with relapsed/refractory B-cell non-Hodgkin's lymphoma (NHL, cohort A) were enrolled on dose level (DL) 1 (NHL-1, -2, -3) and DL 2 (NHL-4), and 1 patient with relapsed acute B-lymphoblastic leukemia (ALL, cohort B) was enrolled on DL 1 (ALL-1). Primary and secondary objectives of the trial are to assess safety and anti-tumor responses; immune response evaluation is an additional objective. NKTs were isolated from the leukapheresis product of 1 HLA-unmatched healthy individual, transduced with the CAR, expanded ex vivo for 14 days to a total of 2.7×10 9 CAR-positive cells (99.8% NKT purity, 0.04% T cells), and cryopreserved. Patients received 10 7 (DL 1) or 3×10 7 (DL 2) CAR-NKT cells per square meter of body surface area following lymphodepleting conditioning with cyclophosphamide/fludarabine. Adverse events were evaluated per NCI criteria. When accessible, patients underwent core biopsies of an involved site at 2-5 weeks post-infusion. Response to therapy was assessed at 4 weeks per Lugano Criteria (for NHL) or NCCN guidelines (for ALL). The most common adverse effects observed were nausea and grade 3-4 hematologic toxicities related to the lymphodepletion chemotherapy. There were no early adverse events attributable to the cellular product except grade 1 cytokine release syndrome in 1 patient. Of the 4 NHL patients (all with diffuse large B cell lymphoma), 3 had a partial response (NHL-1, -2, and -4, Figure 1A) that evolved into a CR in 1 case (NHL-2). The ALL patient achieved a CR with incomplete hematologic recovery and showed no evidence of leukemia by morphology, flow cytometry, or next-generation sequencing at 4 weeks. We detected in vivo expansion of donor-derived NKT and CAR-NKT cells in the peripheral blood of NHL-4 and ALL-1 that peaked at 1 week post-infusion in both cases as determined by flow cytometry and qPCR. While CAR-NKTs were not detected in the peripheral blood of the first 3 NHL patients beyond 3 hours post-infusion, they were found in tumor tissues collected from the 2 biopsied NHL patients (Figure 1B). In patient NHL-2, we also observed a 2000-fold expansion of recipient NKTs with a skewed TCRβ repertoire; this population peaked at 6 weeks post-treatment and remained elevated through 12 weeks (Figure 1C). An ELISpot assay performed with peripheral blood from NHL-1 and NHL-2 before treatment and at 4 and 6 weeks post-therapy to detect changes in the frequency of T cells reactive to a set of tumor-associated antigens (TAAs) found no evidence for epitope spreading toward the tested TAAs. In conclusion, our data indicate that allogeneic CAR-NKT cells are well-tolerated and can mediate objective responses in relapsed/refractory NHL and ALL patients even at the low doses tested. Our initial results from this first-in-human clinical evaluation of allogeneic CAR-NKTs suggest that NKTs represent a promising platform for "off-the-shelf" cancer immunotherapy. Figure 1 Figure 1. Disclosures Ramos: Tessa Therapeutics: Patents & Royalties, Research Funding; Athenex: Research Funding; Genentech: Consultancy; Novartis: Consultancy. Courtney: Athenex: Patents & Royalties, Research Funding. Metelitsa: Athenex: Patents & Royalties, Research Funding. OffLabel Disclosure: Cyclophosphamide and fludarabine are being used as lymphodepleting agents before immune effector infusion.