Background: Bridging therapy (BT) is critical for delivering chimeric antigen receptor T (CAR-T) therapy in patients with RRMM as manufacturing times vary from 4-8 weeks and patients require interim disease control. We evaluated the impact of types of BT and response to BT on CAR-T expansion, safety and efficacy outcomes in patients with RRMM receiving ciltacabtagene autoleucel (cilta-cel) or idecabtagene vicleucel (ide-cel).

Methods: In this single-center retrospective analysis, we included patients receiving ide-cel and cilta-cel CAR-T cell therapy for RRMM between May 2021-March 2025. BT was categorized into four groups: (1) No bridging/minimal therapy (steroids or localized radiation only); (2) Conventional myeloma-directed therapy (proteasome inhibitors, immunomodulatory drugs, anti-CD38 monoclonal antibodies, low-dose alkylators); (3) Talquetamab (bispecific antibody); and (4) High-dose alkylator therapy.

CAR-T expansion kinetics were measured via flow cytometry. Mann-Whitney U-test, Fisher's exact, and log-rank statistics were used to analyze continuous, categorical, and time-to-event variables, respectively. Multivariable Cox proportional hazards models were fit for progression-free survival (PFS) and overall survival (OS) outcomes by bridging response.

Results: Amongst 135 patients receiving CAR-T (n=50 ide-cel; n=85 cilta-cel), median follow-up was 12 months (IQR: 7–26). Median prior therapy lines were higher in the ide-cel cohort (6 vs. 4; p<0.001). BT type differed by CAR-T product (p=0.05): talquetamab use was more common with cilta-cel (20% vs. 6%), while high-dose alkylators were more frequent in ide-cel (20% vs. 16%). The overall use of BT (91% and 78%) did not differ significantly between cohorts (p=0.08). BT distribution for cilta-cel cohort was as follows: no BT (n=8; 9%), conventional (n=46; 54%), talquetamab (n=17; 20%), and high-dose alkylators (n=14; 16%); and for ide-cel cohort: no BT (n=11; 22%), conventional (n=26; 52%), talquetamab (n=3; 6%), and high-dose alkylators (n=10; 20%). Partial response (PR) or better with BT was observed in 41% of cilta-cel patients and 24% of ide-cel patients (p=0.04). Overall, talquetamab was associated with higher response rates (92%) compared with patients who received patients who received conventional therapy (23%) and high dose high alkylator (85%) (p<0.001).

In the cilta-cel cohort, all Parkinsonian neurotoxicity cases (n=4) occurred in non-responders to BT vs. no cases in responders (p=0.1). Non-responders also had higher rates of ≥grade 2 cytokine release syndrome (CRS) (31% vs 11% p=0.05) and any delayed neurotoxicity (DNT) (26% vs 11% p=0.1) compared to responders. In the ide-cel cohort, all immune effector cell-associated neurotoxicity syndrome (ICANS) (33%, n=9) events occurred in non-responders to BT (p=0.04). CRS incidence was higher in non-responders (93%) and no BT (91%) compared to responders (67%), approaching significance (p=0.06).

Death within 6 months of CAR-T occurred in 15 patients amongst cilta-cel cohort, with non-relapse mortality (NRM) in 8 (9%). NRM was nominally higher in patients who did not respond to BT at 13% vs 6%. Only 1 patient died in the ide-cel cohort from NRM (ascending paralysis, achieved a CR to BT)

Response to BT did not impact the efficacy of CAR-T therapy. Response rates to CAR-T (≥PR, ≥complete response (CR), PFS and OS) were similar regardless of type of BT or response to BT with both cilta-cel and ide-cel (all p>0.3). Post-CAR-T, 42% of ide-cel and 60% of cilta-cel patients achieved CR, often despite lower grade BT responses. Out of patients who did not respond to BT (stable or progressive disease), 12/27 (44%) ide-cel and 26/39 (67%) cilta-cel patients achieved CR following CAR-T. CAR-T expansion kinetics and area under the curve (AUC) analysis did not differ by BT response or type in both cohorts (all p>0.5).

Conclusions: While response to CAR-T is not dependent on the type of BT or response to BT, BT plays a critical role in safely administering CAR-T therapy. Effective bridging (≥PR) to lower disease burden was associated with an improved safety profile of CAR-T, particularly cilta-cel. Non-responders to BT experienced numerically higher toxicity and NRM, with all Parkinsonism cases with cilta-cel seen in non-responders to BT. Type of BT did not impact outcomes, suggesting that the best BT is one that patient is not refractory to and has the highest likelihood of response, while balancing safety.

This content is only available as a PDF.
2025
Sign in via your Institution