Assessment of pharmacodynamic efficacy biomarkers from A phase 1, first-in-human study of arlocabtagene autoleucel (arlo-cel) in relapsed and refractory multiple myeloma (RRMM)

Abstract

Abstract Introduction: The chimeric antigen receptor (CAR) T-cell therapy arlocabtagene autoleucel (arlo-cel; BMS-986393) targets G protein-coupled receptor class C group 5 member D (GPRC5D), a validated target in RRMM. Arlo-cel has demonstrated deep, durable responses in patients with RRMM in a first-in-human, phase 1, multicohort, dose escalation and expansion study (NCT04674813). The objective of this pharmacodynamic biomarker analysis is to further explore arlo-cel efficacy and evaluate soluble B-cell maturation antigen (sBCMA) and minimal residual disease (MRD) as exploratory biomarkers of patient responses. Methods: Patients with 1-3 or ≥3 prior anti-MM regimens were eligible to enroll; prior BCMA-directed therapies, including CAR T-cells, were allowed. The primary objective was safety; secondary objectives included clinical activity per IMWG Uniform Response Criteria and pharmacokinetics. Exploratory objectives include assessment of sBCMA and MRD. As a biomarker of tumor burden, sBCMA concentration was measured from serum at baseline (BL) and longitudinally until progression. sBCMA nadir was defined as the lowest concentration measured for each patient following arlo-cel treatment. The threshold for deep sBCMA clearance was defined in relation to the levels cleared by patients in complete response or better (≥CR). MRD was assessed via next-generation sequencing of bone marrow aspirate at pre-specified timepoints; MRD-negative status was defined at a depth of 10-5. Analyses (4Apr2025 datacut) were conducted in patients with available data and reported here for those receiving ≥3 prior lines of treatment (≥3pLoT). Analyses of the patient cohort with 1-3 prior anti-MM regimens are ongoing. Cox proportional hazards models and Kruskal-Wallis tests were used to assess statistical significance, as appropriate. sBCMA nadir and MRD progression-free survival (PFS) models used the landmark approach, which only considers survival time after the relevant biomarker assessment. Results: In the ≥3pLoT efficacy-evaluable cohort (n=79), after a median 23.3 mo follow-up (range, 3.8-39.7 mo), median PFS (95% CI) was 18.3 mo (11.8, 21.9). Overall response rate was 87.3% and ≥CR was achieved by 53.2% of patients. Median overall survival in all treated patients (N=84) was not yet reached. Baseline sBCMA concentrations were not associated with best overall response (BOR); a similar distribution of BL sBCMA levels was seen in patients achieving no response, partial response (PR), very good partial response (VGPR), and ≥CR. Baseline sBCMA was not associated with PFS (P=0.16). In the ≥3pLoT cohort, greater sBCMA clearance was associated with depth of clinical response following arlo-cel treatment, with the lowest median sBCMA nadir levels observed in patients achieving VGPR and ≥CR. The median time to sBCMA nadir was 2 mo after arlo-cel infusion and 64.9% (50/77) of patients had sBCMA nadirs below the deep sBCMA clearance threshold. Further, median PFS (95% CI) after sBCMA nadir landmark was significantly longer in patients with nadirs below vs above the deep clearance threshold for sBCMA (19.0 mo [17.1, NR] vs 3.9 mo [2.8, 12], P<0.0001). In ≥3pLoT patients with available MRD data, MRD-negative rates at Months 3, 6, and 12 ranged from 76-81%. Patients who were MRD-negative had significantly longer median PFS after their landmark MRD assessments at Month 3 (P<0.001) and Month 6 (P<0.0001) than those who were MRD-positive, while those who were MRD-negative at Month 12 had yet to reach median PFS after their MRD assessment. Higher arlo-cel expansion was generally associated with lower sBCMA nadir levels. Conclusions: Efficacy and pharmacodynamic data from this first-in-human phase 1 study demonstrate that arlo-cel led to deep responses in RRMM, even in patients with high BL tumor burden. Increased clearance of sBCMA after arlo-cel treatment was associated with deeper clinical response and sBCMA nadir levels were prognostic of longer landmark PFS. The relationship between robust CAR T-cell expansion and reduced sBCMA nadir levels reinforces that arlo-cel drove the observed deep therapeutic responses. MRD negativity was also associated with more favorable landmark PFS. These phase 1 findings based on early time points demonstrate the potential use of sBCMA and MRD as pharmacodynamic biomarkers of arlo-cel treatment efficacy and will continue to be monitored in the ongoing phase 2 (NCT06297226) and phase 3 (NCT06615479) studies of arlo-cel.