OBJECTIVES : Lenalidomide-based regimens are standard of care for treating newly diagnosed and relapsed multiple myeloma. Challenges remain in defining optimal therapy at time of relapse in the post-lenalidomide setting. For these patients, lenalidomide-free regimens such as Kd and PVd are recommended. Kd was assessed in the CANDOR (NCT03158688) and ENDEAVOR (NCT01568866) trials in patients who received 1-3 prior therapies whereas PVd was assessed in the OPTIMISMM (NCT01734928) trial in patients who received 1-3 prior therapies including lenalidomide. In the absence of a head-to-head trial, a MAIC estimating the comparative efficacy of Kd vs PVd was conducted.

METHODS : Baseline characteristics and progression-free survival (PFS) data from lenalidomide-exposed Kd patients in CANDOR and ENDEAVOR, and PVd patients from OPTIMISMM were used for the analysis. Kd patients from the two carfilzomib trials were pooled and matched to PVd patients using MAIC methodology. Published Kaplan-Meier PFS curves for PVd were digitized to generate patient-level data that reconstructed the curves. Cox regression models were fitted to the matched Kd data and reconstructed virtual patient-level PVd data to estimate the comparative efficacy of Kd versus PVd. The median PFS was compared for the two treatments and the risk of progression and death in terms of hazard ratio (HR) was estimated.

RESULTS : A total of 251 lenalidomide-exposed Kd patients and 281 PVd patients were included in the analysis. After matching, the effective sample size for Kd patients was reduced to 189. In matched patients, the median PFS was 12.1 months and 11.5 months for Kd and PVd respectively. The estimated HR of PFS was 0.96 (95% CI: 0.75, 1.24) for Kd vs PVd.

CONCLUSIONS : After adjusting for cross-trial differences, the MAIC analysis demonstrated similar efficacy (PFS) for Kd and PVd, highlighting that these regimens in the post-lenalidomide space offer a more focused benchmark for such patients.