Background: Despite advances in the treatment of multiple myeloma (MM) almost all patients relapse and high risk features continue to portend a short median survival. The adoptive transfer of B-Cell Maturation Antigen (BCMA) chimeric antigen receptor (CAR) T cells is demonstrating early promise in MM, but the durability of response has not been established. The infusion of genetically modified CD8+ and CD4+ T cells of a defined composition facilitates the evaluation of each subset's function and has contributed to reproducible efficacy and safety in clinical trials with CD19-specific CAR T cells. In this phase I first-in-human study employing a human scFv containing BCMA CAR T cell construct, we report rapid and deep objective responses at a low CAR T cell dose level (5 x 107) suggesting that construct specific features and differences in product formulation may substantially impact efficacy. Methods: Eligible patients had relapsed or treatment refractory MM, ≥10% CD138+ bone marrow (BM) plasma cells (PC), and ≥5% BCMA expression by flow cytometry (FC). Patients were stratified by tumor burden (CD138+ IHC) into two cohorts; 10-30% MM cells [cohort A] or >30% BM involvement [cohort B] to facilitate assessment of impact of disease burden on outcome. Eligible patient's CD8+ and CD4+ T cells were isolated via positive selection, enriched separately by immunomagnetic selection and cryopreserved. The CD8+ and CD4+ T cells were stimulated in independent cultures with anti-CD3/anti-CD28 paramagnetic beads and transduced with a 3rd generation lentiviral vector encoding a fully human BCMA scFv and 4-1BB and CD3 zeta signaling domains. After in vitro expansion, the cell product for infusion was formulated in a 1:1 ratio of CD4+:CD8+ BCMA CAR T cells. A truncated non-functional human epidermal growth factor receptor (EGFRt) was encoded in the transgene cassette to identify transduced T cells. Lymphodepleting chemotherapy preceded infusion of EGFRt+ CAR T cells at a starting dose of 5 x 107 EGFRt+ cells (n=5) for each cohort. Results: Seven patients (median age of 63 years; range 49 to 76) with a median of 8 prior regimens (range 6 to 11) have received treatment. The median %PC in BM (IHC) at enrollment was 58% (range 20% to >80%). In cohort B the dose has been escalated to 15 x107 EGFRt+ cells (n=2). All patients (7/7) had at least one high risk cytogenetic feature (17p- [n=4], t(4;14) [n=2], t(14;16) [n=1]), 71% had ≥ 2 high risk cytogenetic features, 71% had prior autologous stem cell transplant, 43% had prior allogeneic transplant, and one patient (14%) had PCL. The median involved free light chain (FLC) at enrollment was 180 mg/dL (range 40.37 to 502.96 mg/dL; n=5) and the median monoclonal protein was 3.8 g/dL (range 1.6 to 6.5 g/dL; n=5). The overall response rate at 28 days was 100%; at this time all evaluable patients (n=6) had no detectable abnormal BM PC clone by both IHC and high sensitivity FC. Within 28 days of treatment the involved FLC was normal or sub-normal in all patients and the M-protein had decreased by a median of 73% (range 56.25 to 83% reduction). BCMA CAR T cells remained detectable 90 days after infusion, representing up to 41.5 percent of CD3+ lymphocytes. All patients were surviving at a median of 16 wks (range 2 to 26 wks). One patient relapsed (day +60) and a tumor biopsy demonstrated the presence of a BCMAneg PC population, a 70% reduction in the fraction of MM cells expressing BCMA by FC and a fivefold reduction in BCMA antigen binding capacity on MM cells retaining target expression. A cytotoxic T lymphocyte response to the trans-gene product was not identified in this patient. No dose limiting toxicity has been observed during the 28 day monitoring window and treatment has been well tolerated with no cytokine release syndrome (CRS) observed in one patient and grade 2 or lower CRS (Lee Criteria) for all other patients. No neurological toxicity has been observed. Conclusion: BCMA CAR T cells harboring a fully human scFv with a defined composition of CD4+:CD8+ T cells were well tolerated and potent, demonstrating complete objective responses in heavily pretreated high-risk MM at total cell doses as low as 5 x 107. Next generation sequencing and multiparameter high sensitivity flow cytometry studies to evaluate for minimal residual disease are ongoing. Peak expansion levels and persistence of the CAR T cells are being monitored with early findings suggesting an absence of transgene product immunogenicity. Green: Juno Therapeutics: Patents & Royalties, Research Funding. Sather:Juno Therapeutics: Employment. Cowan:Janssen: Research Funding; Abbvie: Research Funding; Juno Therapeutics: Research Funding; Sanofi: Research Funding. Turtle:Caribou Biosciences: Consultancy; Gilead: Consultancy; Bluebird Bio: Consultancy; Precision Biosciences: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Aptevo: Consultancy; Eureka Therapeutics: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Nektar Therapeutics: Consultancy, Research Funding; Juno Therapeutics / Celgene: Consultancy, Patents & Royalties, Research Funding; Adaptive Biotechnologies: Consultancy. Till:Mustang Bio: Patents & Royalties, Research Funding. Becker:GlycoMimetics: Research Funding. Blake:Celgene: Employment, Equity Ownership. Works:Juno Therapeutics: Employment. Maloney:GlaxoSmithKline: Research Funding; Juno Therapeutics: Research Funding; Seattle Genetics: Honoraria; Roche/Genentech: Honoraria; Janssen Scientific Affairs: Honoraria. Riddell:Cell Medica: Membership on an entity's Board of Directors or advisory committees; Juno Therapeutics: Equity Ownership, Patents & Royalties, Research Funding; Adaptive Biotechnologies: Consultancy; NOHLA: Consultancy.
