P-033: External control analysis for KarMMa-3: idecabtagene vicleucel (ide-cel) vs real-world standard of care (RW SoC; Connect MM) for triple-class exposed (TCE) relapsed/refractory multiple myeloma (RRMM)

Introduction: KarMMa-3 (NCT03651128) is a phase III randomized controlled trial that demonstrated ide-cel (a chimeric antigen receptor [CAR] T therapy) extended progression-free survival (PFS) versus So in patients with TCE RRMM (hazard ratio [HR] 0.49 [95% CI: 0.38, 0.65]). Evaluating overall survival (OS) in KarMMa-3 is challenging given that 62% of patients in the SoC arm crossed over to ide-cel upon disease progression (i.e., OS estimates for SoC were confounded by subsequent ide-cel use). An external control arm (ECA) based on RW SoC from the Connect MM disease registry provides OS estimates from the United States, which can supplement the comparative effectiveness of ide-cel versus SoC. The aim of this study was to evaluate the comparative effectiveness of ide-cel versus RW SoC in terms of OS and PFS in patients with TCE RRMM.

Methods: Individual patient data for ide-cel from KarMMa-3 (April 2023 cutoff date; intention-to-treat population) and RW SoC from Connect MM (October 2022 cutoff date) were used. To align with KarMMa-3, Connect MM patients with TCE RRMM (November 2015 to December 2020) were restricted to those who 1) received a non-CAR T index therapy in the 3-5L setting, 2) had no prior CAR T therapy nor allogeneic stem cell transplantation (SCT), and 3) had an ECOG performance score of 0-1. Inverse probability of treatment weighting (IPTW) was used to balance pre-specified patient characteristics. Weights from propensity scores (average treatment effect on the treated) were applied to the RW SoC cohort (KarMMa-3 as the target population). Model covariates included were selected based on published rank-ordering of prognostic factors and data availability: triple-class refractory status, high-risk cytogenetics, disease stage, extramedullary disease, time to progression on last regimen, number of prior lines, time since diagnosis, age, hemoglobin, lactate dehydrogenase, prior autologous SCT, race, and sex. Multiple imputation was performed for covariates with ≤30% missing data. Relative treatment effects from Cox proportional hazards models were summarized as HRs with 95% CIs. A sensitivity analysis was performed using regression.

Results: Model covariates were generally balanced between the KarMMa-3 ide-cel cohort (N=254 patients) and Connect MM RW SoC cohort (N=145 patients) following IPTW adjustment. Median OS was 41.4 months for ide-cel and 11.6 months for RW SoC, while median PFS was 13.8 months for ide-cel and 2.9 months for RW SoC. Ide-cel demonstrated improvements versus RW SoC in OS (adjusted HR 0.37 [95% CI: 0.26, 0.53]), which was more precise than KarMMa-3 crossover-adjusted HR. Ide-cel also extended PFS versus RW SoC (adjusted HR 0.45 [95% CI: 0.32, 0.64]), leading to consistent estimates with KarMMa-3. Sensitivity analyses suggested similar results for OS and PFS.

Conclusions: Ide-cel was associated with improved OS and PFS compared to RW SoC, providing evidence regarding the clinical benefit of ide-cel in patients with TCE RRMM.