Background: CANDOR, a randomized phase 3 study, showed significantly improved progression-free survival (PFS) and overall response rate (ORR) for twice-weekly KdD 56 mg/m²(TW KdD56) vs Kd in RRMM, with a 37% risk reduction in disease progression or death (Usmani, 2019). The benefit of once-weekly carfilzomib (K) at 70mg/m² for the same triplet, KdD (OW KdD70), in RRMM was shown in the nonrandomized MMY1001 study. We performed robust cross-study comparisons to investigate how TW KdD56 compares to OW KdD70. Methods: For alignment of inclusion criteria, the primary analysis set for this cross-study comparison included individual data from CANDOR pts with prior exposure to bortezomib and immunomodulatory drug, and all pts in MMY1001. Propensity score method (Rambaldi, 2019; Kumar, 2019) allowed reliable cross-study comparison, controlling for baseline prespecified covariates such as age, creatinine clearance, performance status, prior treatment exposure/refractoriness, and time from initial diagnosis and relapse. Propensity score adjustment based on inverse probability of treatment weighting (IPTW) was implemented on efficacy endpoints, while safety was evaluated side-by-side. Results: The unadjusted and adjusted propensity score comparison of CANDOR and MMY1001 showed similar efficacy in terms of ORR and PFS in the TW KdD56 and OW KdD70 groups (Table). As propensity score comparisons were for efficacy only, safety comparisons between CANDOR and MMY1001 included differing sample sizes and treatment duration (Table). The safety of KdD56 and KdD70 was consistent with the known safety profiles of individual study treatments. Conclusions: OW KdD70 is an efficacious dosing option with favorable benefit-risk, comparable to TW KdD56. The OW KdD70 dosing option represents a more convenient regimen that might encourage adherence and potentially lead to better outcomes for RRMM pts. Clinical trial information: NCT03158688.

*propensity score adjustment using IPTW; ‡treatment-emergent; NE, not estimable