Anti-BCMA CAR T-Cell Expansion and Its Association with Response and Toxicity in Multiple Myeloma

Background: B-cell maturation antigen (BCMA)-targeted chimeric antigen receptor (CAR) T-cells have revolutionized the management of relapsed/refractory multiple myeloma (RRMM). However, data on CAR-T expansion and persistence and the association with response and toxicities are limited (Fischer et al., Leukemia 2024). We analyzed RRMM patients treated with standard of care CAR-T (idecabtagene vicleucel, ide-cel and ciltacabtagene autoleucel, cilta-cel) to assess the expansion and persistence of BCMA-targeted CAR T-cells and association with response and toxicities. Methods: Circulating CAR T-cells were assessed and quantified by flow cytometry. A PE fluorochrome labeled human BCMA/TNFRSF17 protein was used to detect BCMA-specific CAR T-cells. A total of 56 RRMM patients treated with ide-cel (n=30) or cilta-cel (n=26) were included in the analysis. Response was assessed by IMWG criteria and was categorized into complete response (CR), very good partial and partial response (VGPR/PR), stable disease (SD) and progressive disease (PD). Continuous variables were compared using a Mann-Whitney U-test, while time-to-event variables were compared using a Cox proportional hazard model. Median follow-up was calculated using the reverse Kaplan-Meier method. Results: Median age of our cohort was 67 years, 13% were Black, 25% were Hispanic, 36% had high-risk cytogenetics, 46% had extramedullary disease (EMD). Median prior lines of therapy were 5 and 48% were penta-refractory. Median follow up was 19.7 (95% CI: 12.7, 26.3) months for ide-cel (n=30) and 7.2 (95% CI: 4.9, 19.0) months for cilta-cel (n=26). The median time-to-progression (TTP) was 8.5 (95% CI: 4.2, 21.9) months for ide-cel and 15.3 (95% CI: 4.5, 19.6) months for cilta-cel. The median time to reach the peak CAR-T expansion was 14 (95% CI: 7.3, 15) days for ide-cel, and 14.5 (95% CI: 14, 17.3) days for cilta-cel. Within the ide-cel cohort, peak CAR-T expansion was higher amongst responders (PR or better) compared to patients with SD/PD based on response at 3 months (median 70.4/uL in responders vs 5.3 in non-responders; P<0.001) as well as at 6 months (157.4 vs 6.2/uL; P<0.001). The ide-cel peak levels were not significantly associated with TTP via Cox proportional hazards model (HR 0.99, P=0.107). We next assessed whether peak expansion was associated with toxicities. The peak ide-cel levels were not significantly different between patients who developed cytokine release syndrome (CRS) or not, P=0.933, nor in patients who developed immune effector cell-associated neurotoxicity syndrome (ICANS) or not, P=0.358. Within the cilta-cel cohort, peak expansion was higher in responders compared to non-responders at 3 months (median 148.7/uL vs 65.9/uL) and 6 months (median 128.3/uL vs 89.7/uL), though these did not reach statistical significance (3 months: P=0.273; 6 months: P=0.65). Similar to the ide-cel cohort, peak expansion levels were not significantly associated with TTP (HR 0.99, P=0.59) as well as CRS (P=0.51) in the cilta-cel cohort. However, interestingly, the peak expansion was significantly higher for patients who developed ICANS (median 1269.2/uL) versus those who did not develop ICANS (median 99.7; P<0.001). Conclusions: We demonstrate that CAR kinetics are associated with response and toxicity and that these associations may differ among CAR-T products. In the ide-cel cohort, peak CAR-T expansion was associated with response at 3 and 6 months and no significant difference in toxicities. In the cilta-cel cohort, higher peak CAR-T levels were associated with ICANS development. Neither cohort demonstrated significant association between peak CAR expansion levels and TTP. Future studies are needed to confirm these findings in larger sample sizes.