A longitudinal single-cell atlas to predict outcome and toxicity after BCMA-directed CAR T cell therapy in multiple myeloma

Comprehensive Summary

This study evaluated 61 patients with relapsed/refractory multiple myeloma treated with BCMA-targeted CAR T-cell therapies, comparing ciltacabtagene autoleucel (cilta-cel, n=27) and idecabtagene vicleucel (ide-cel, n=34), with progression-free survival (PFS) as the primary endpoint. Cilta-cel demonstrated markedly superior efficacy, with a complete response (CR) rate of 78% versus 38% for ide-cel (absolute difference +40%) and an overall response rate of 93% versus 67% (+26%). Median PFS was not reached for cilta-cel but was 6 months for ide-cel (p=0.001). Achieving CR was strongly associated with improved outcomes, as patients in CR had not reached median PFS compared to just 3 months in non-CR patients. Notably, cilta-cel was more effective at deepening responses, converting 70% of patients with stable/progressive disease and 80% with partial response into CR, compared to 18% and 46% with ide-cel. Cytokine release syndrome (CRS) occurred in 52% of cilta-cel and 79% of ide-cel patients, but did not correlate with survival, indicating toxicity is not a marker of efficacy. Lower soluble BCMA levels correlated with better responses and greater CAR T expansion (p=0.00013), supporting its role as a real-time biomarker of tumor burden. Mechanistically, successful outcomes were driven by expansion of cytotoxic CD4⁺ CAR T cells, while treatment failure was associated with T-cell dysfunction and exhaustion. Clinically, these findings support prioritizing cilta-cel when available, aiming for deep responses, and using biomarkers like soluble BCMA to guide monitoring, while proactively managing CRS and prolonged immunosuppression at the bedside.

Outcomes and Implications

This research highlights several important implications for clinical care and future investigation in multiple myeloma. Differences in CAR T-cell expansion and persistence between cilta-cel and ide-cel suggest that treatment-specific monitoring strategies are needed, particularly as delayed expansion may affect timing of toxicity and response assessment. The identification of soluble BCMA as a dynamic biomarker supports its use in guiding real-time evaluation of treatment response and disease burden. In addition, the association between CAR T-cell expansion and toxicity indicates that early, proactive management strategies may improve patient safety without compromising efficacy. Baseline immune and tumor characteristics may also help predict outcomes and guide patient selection. More broadly, these findings support continued integration of molecular and cellular profiling into clinical decision making and may inform the development of next-generation CAR T therapies and new therapeutic targets.