P-191: Comparison of efficacy outcomes for Carfilzomib plus Dexamethasone and Daratumumab (KdD) versus Pomalidomide plus Bortezomib and Dexamethasone (PVd) and D-Pd in relapsed or refractory Multiple Myeloma

Background With increasing utilization of immunomodulatory agents (particularly lenalidomide) for the management of newly diagnosed multiple myeloma, there is an emerging need for efficacious lenalidomide-free options at relapse. In the CANDOR trial, carfilzomib plus dexamethasone and daratumumab (KdD) demonstrated improved efficacy over Kd for the management of relapsed or refractory multiple myeloma (RRMM) and has a tolerable safety profile. Two pomalidomide-containing triplet regimens with bortezomib (PVd) or daratumumab (D-Pd) plus dexamethasone have demonstrated improved efficacy over Vd or Pd alone in randomized clinical trials (OPTIMISSM and APOLLO, respectively). Herein, naive comparisons and matching-adjusted indirect comparisons (MAIC) of progression-free survival (PFS) were performed for KdD versus PVd and KdD versus D-Pd in lenalidomide-exposed patients. Methods These analyses used baseline characteristics and PFS data from the CANDOR, OPTIMISSM, and APOLLO trials. For each analysis, a naive comparison (without adjustments) and MAIC were performed. For MAIC, KdD patients in CANDOR who were previously exposed to lenalidomide (based on propensity scores) were matched to published summary baseline characteristics of PVd patients in OPTIMMISM and D-Pd patients in APOLLO. Weighted PFS outcomes were calculated. Variables included in the matching analysis were age, disease stage (International Staging System), lenalidomide refractoriness, and number of prior lines of treatment. For the comparison of KdD versus D-PD, bortezomib refractoriness was also included as a matching variable. Scenario and subgroup analyses were conducted to assess the robustness of results. Results Overall, 123 lenalidomide-exposed KdD patients in CANDOR, 281 PVd patients in OPTIMISSM, and 151 D-Pd patients in APOLLO received treatment. There were no notable differences in baseline characteristics considered for the matching procedure. Naive comparisons for PFS significantly favored KdD versus PVd (median PFS, 25.9 months vs 11.5 months; hazard ratio [HR]: 0.545 [0.397–0.747]) and KdD versus D-Pd (median PFS, 25.9 months vs 12.8 months; HR: 0.629 [0.445–0.890]). After matching, PFS also significantly favored KdD versus PVd (median PFS, 25.0 months vs 11.5 months) and KdD versus D-Pd (median PFS, 25.0 months vs 12.8 months). The HR for PFS in the MAIC was 0.539 (0.395–0.736) for KdD versus PVd and 0.677 (0.474–0.966) for KdD versus D-Pd. Conclusion This analysis showed that KdD extended PFS compared with PVd and D-Pd in patients with RRMM and previous lenalidomide exposure. A comparison of overall survival was not considered due to immature data in the studies considered for this analysis. These results suggest KdD offers clinically meaningful improvements over pomalidomide-based triplet regimens for patients with RRMM.