Biomarker Correlates of 3 Year Response in Cartitude-1: A Phase 1b/2 Open-Label Study of Ciltacabtagene Autoleucel in Patients with Relapsed or Refractory Multiple Myeloma

Background Ciltacabtagene autoleucel (cilta-cel) is a chimeric antigen receptor (CAR)-T cell therapy FDA approved for relapsed/refractory multiple myeloma (RRMM) after ≥4 lines of therapy (LOT). This report summarizes clinical response data from the study closeout of the Phase 1b/2 CARTITUDE-1 trial and respective correlative biomarker analyses. Methods Eligible patients (pts) received ≥3 prior LOT or were double refractory to proteasome inhibitor (PI) and immunomodulatory (IMiD) drugs; all received prior PI, IMiD, and anti-CD38 antibody therapy. Primary endpoint was overall response rate (ORR) and safety; secondary endpoints included progression free survival (PFS), overall survival (OS), duration of response (DOR), and minimal residual disease (MRD) negativity at 10−5. Pt and CAR-T drug product (DP) characteristics were evaluated to identify biomarkers of cilta-cel response. Results Of 97 pts, median follow-up was 33.4 mo (range, 1.5-45.2). Response rate was 97.9%; median duration of response (DOR) was 33.9 mo (95% confidence interval [CI], 25.5-not estimable [NE]). Median PFS was 34.9 mo (95% CI, 25.2-NE), with an estimated PFS rate of 47.5% at 36 mo. Median OS was not reached (NR), with an estimated OS rate of 62.9% at 36 mo. Of 49 pts evaluated for MRD, 26 had MRD negativity sustained for ≥12 mo; of these, 20 pts had sustained MRD-negative complete response or greater; median PFS was NR in these subgroups. No new safety signals were reported. CAR-T DP had a mixture of transduced (median [range], 16% [5%-32%]) and nontransduced T cells and a balanced distribution of CD4+ (median [range] frequency, 12% [2%-28%]) and CD8+ CAR+T cells (6% [2%-20%]). CAR+ T cell median peak expansion was at 730 cells/µL between 12 and 14 days post infusion, persisting a median (range) of 100 (20-912) days. Deep and durable responses (eg, best response, DOR) were achieved despite variable CAR-T cell expansion and lack of detectable CAR-T cell persistence over time. Some DP characteristics (eg, high CAR+CD8+ stem-like phenotype, low CAR+CD4+ Treg-like phenotype) were associated with longer DOR. Pt characteristics associated with baseline inflammatory markers correlated with shorter DOR. Pts with high-risk cytogenetics, tumor burden ≥60% bone marrow plasma cells, or baseline plasmacytomas demonstrated high ORR but numerically shorter PFS. Conclusions A single infusion of cilta-cel was associated with a longer PFS than that previously reported for therapies approved for heavily pretreated RRMM. Correlative analyses could support identification of DP, pt, and tumor characteristics to help predict DOR to cilta-cel.