Efficacy and Safety with Extended Follow-up in a Phase 1 Study of BMS-986393, a G Protein-Coupled Receptor Class C Group 5 Member D (GPRC5D)-Targeted CAR T Cell Therapy, in Patients (pts) with Heavily Pretreated Relapsed/Refractory (RR) Multiple Myeloma (MM)

Abstract

Introduction Most pts with MM on current treatments will eventually relapse or fail to respond, including with B-cell maturation antigen (BCMA)-targeting therapies. Pts with RRMM have suboptimal survival rates, which worsen with each new line of treatment; thus, new therapy classes are needed. BMS-986393 is an autologous chimeric antigen receptor (CAR) T cell therapy targeting GPRC5D, a promising target for novel MM therapies. Initial data from a phase 1 study in pts with heavily pretreated RRMM suggested that a single infusion of BMS-986393 was efficacious with a favorable safety profile. At the recommended phase 2 dose (RP2D; 150 × 106 CAR T cells), the overall response rate (ORR) was 91%, and complete response rate (CRR) was 48% (Bal et al, ASH 2023). Here we report data after additional follow-up. Methods Pts had ≥ 3 prior anti-MM regimens including a proteasome inhibitor, an immunomodulatory agent, and anti-CD38 therapy, and autologous stem cell transplant (if eligible); prior BCMA-directed therapies (including CAR T) were allowed. After screening and leukapheresis (bridging therapy optional), pts received lymphodepleting chemotherapy followed by a single infusion of BMS-986393 (25-450 × 106 CAR T cells). The primary objective was safety. Results As of May 20, 2024, 84 pts had received BMS-986393, 26 at the RP2D. Median age was 63 years (range 39-80); 51% were male; 67% were White and 17% Black or African American. 42% had high-risk cytogenetics (del[17p], t[4;14], and/or t[14;16]), 55% had 1q21 gain/amp, and 45% had extramedullary disease. Pts had received a median of 5 prior regimens (range 3-15); 49% had received prior BCMA-targeted therapy, including BCMA-directed CAR T cell therapy in 38% of pts. 76% of pts had triple-refractory disease, and 35% had penta-refractory MM. Across all doses, treatment-related (TR) adverse events (AEs) occurred in 94% of pts; 71% had a grade (G) 3/4 TRAE, and there was 1 TRAE-related death (cytokine release syndrome [CRS]) in pts treated at 450 × 106. Overall, 69 pts (82%) had CRS (66 at G1/2); all but 1 (G5) resolved. Three pts had macrophage activation syndrome/hemophagocytic lymphohistiocytosis (all G3), none of these events occurred at the RP2D. TR neurologic AEs, including immune effector cell-associated neurotoxicity syndrome (ICANS) and other select neurotoxicities (dizziness, ataxia, neurotoxicity, dysarthria and/or nystagmus), were observed. ICANS occurred in 8 (10%) pts (2 at G3) and resolved in 7. Other select neurotoxicity occurred in 10 pts (12%); 5 (6%) at G3 (the rest G1/2); these appeared to be dose-related. On-target/off-tumor nail, skin and oral TRAEs occurred in 19%, 30%, and 31% pts, respectively. Five pts had TR weight loss (all G1/2), and treatment-emergent (TE) infections occurred in 42 (50%; 14 pts at G3/4). After median follow-up of 14.6 months (range, 2.8-25.2 months) with 79 efficacy-evaluable pts, ORR was maintained at 87% (91% for the RP2D). Considering disease features, ORR was 84% for pts with high-risk cytogenetics (26/31), 87% for triple-class refractory disease (52/60), 79% for prior BCMA-targeting therapy (30/38), and 86% for extramedullary disease (31/36). Of pts with minimal residual disease (MRD) data and CR or better, 88% (22/25) were MRD-negative (10−5 depth; at any time point within 3 months prior to achieving CR or better, until the time of progression or death). Of 69 responses, 33 (48%) were ongoing. Median PFS was 14.5 months (95% CI 11.8-20.8). Dose-dependent increases in cellular expansion occurred across dose levels, and all doses consistently reduced soluble BCMA levels. Conclusions After the longest recorded follow-up for a GPRC5D-targeted CAR T therapy, BMS-986393 showed a high rate of durable responses that deepened with time. High response rates occurred irrespective of disease characteristics, including in pts naïve and exposed to prior BCMA-targeted therapy, and with high-risk cytogenetics, triple-class refractory, and extra-medullary disease. Long-term safety data showed a manageable profile of on-target/off-tumor AEs and neurotoxicity, and that the risk of infections may be lower than some BCMA-targeted therapies. These data support BMS-986393 as a potential treatment for heavily pretreated RRMM, which is being investigated in the ongoing phase 2 QUINTESSENTIAL study (NCT06297226). The presentation will report updated data after ~ 18 months' follow-up, including the first overall survival data.