Evaluation of Minimal Residual Disease (MRD) Negativity in Patients with Relapsed or Refractory Multiple Myeloma Treated in the Candor Study

Introduction: CANDOR is a multicenter, phase 3, randomized study of adult patients with relapsed or refractory multiple myeloma (RRMM) previously treated with 1-3 prior lines of therapy (NCT03158688). 466 patients received carfilzomib, dexamethasone, and daratumumab (KdD) or carfilzomib and dexamethasone (Kd) in 2:1 randomization (KdD: 312; Kd: 154). Based on the primary endpoint, KdD demonstrated superior progression-free survival (PFS) vs Kd (hazard ratio [HR], 0.63 [95% CI, 0.46–0.85]; P=0.0014). Deep responses and minimal residual disease (MRD) negativity have been associated with improved PFS for patients with RRMM. Herein, we present an analysis of MRD results from CANDOR.

Methods: Details of the dose and schedule were previously reported (Dimopoulos et al., Lancet 2020). The rate of patients with confirmed CR which were MRD negative (MRD[-]CR) in bone marrow aspirate at 12 months (± 4 weeks) measured by next-generation sequencing (NGS; threshold, 1 tumor cell/10^-5 white blood cells) was a prespecified key secondary endpoint. Exploratory analyses included MRD[-]CR at increasing sensitivity (10^-4, 10^‑5, 10^-6) and best overall response MRD[-] status at any time point. All reported responses were by Independent Review Committee and were analyzed for the Intent-to-Treat population. MRD[-] status is at <10^-5 unless otherwise specified.

Results: The best overall MRD[-]CR rate at any time was 13.8% vs 3.2% in the KdD vs Kd arm (Odds ratio [OR], 4.95; P<0.0001) and the MRD[-] rate regardless of overall response status was 22.8% vs 5.8% (OR, 5.15; P<0.0001). At the 12-month landmark, the MRD[-]CR rate was 12.5% vs 1.3% in the KdD vs Kd arm (OR, 11.3; P<0.0001) and the MRD[-] rate was 17.6% vs 3.9% (OR, 5.76; P<0.0001) with the proportion of patients with MRD[-]VGPR being 4.2% vs 2.6%, respectively. The MRD[‑]CR rates at the 12-month landmark for KdD vs Kd were consistent across clinically relevant subgroups (Table).

At the 12-month landmark, KdD treatment resulted in a greater proportion of CR rates (26.9% vs 9.7%) and deeper MRD responses than Kd. Among patients in CR, the depth of response as measured by NGS MRD level at the 12-month landmark was deeper for KdD relative to Kd: cutoff of >10^-4, 36.9% vs 73.3%; 10^-4 to 10^‑5, 16.7% vs 13.3%; 10^-5 to 10^-6, 23.8% vs 13.3%; <10^-6, 22.6% vs 0% for KdD vs Kd, respectively (Figure). Similar to the results at the 12-month landmark, MRD responses independent of the landmark were deeper among patients in the KdD compared to the Kd arm. With median follow-up of 6 months from the 12-month landmark, no patient with MRD[-]CR response progressed or died.

Additional post hoc analyses were conducted within patients randomized to KdD to explore prognostic characteristics for MRD[-]CR. Importantly, prior lenalidomide exposure did not meaningfully impact the MRD[-]CR rate at the 12-month landmark; 13.2% (25/189), 11.4% (14/123), and 13.1% (13/99) for naïve, exposed, and refractory subgroups, respectively. For prior bortezomib, the MRD[-]CR rates were 24% (6/25), 11.5% (33/287), and 6.8% (6/88) for naïve, exposed, and refractory subgroups, respectively. The rates of MRD[-]CR at the 12- month landmark within the KdD arm were consistent across subgroups: patients refractory to the last prior therapy (yes vs no, 10.9% vs 14.3%), number of prior regimens (1-2 vs 3 prior regimens; 13.2% vs 10.1%), prior transplant (yes vs no, 11.8% vs 13.7%), duration of first remission (≤2 vs >2 years, 12.3% vs 13% and ≤1 vs >1 year, 10.7% vs 13.4%), baseline creatinine clearance (≥15 to <50, ≥50 to <80, and ≥80 mL/min, 10.5%, 14.4%, and 11.9%, respectively), age (≤75 vs >75 years, 12.9% vs 8.0%), or dose intensity (< vs ≥ median) for carfilzomib or daratumumab (10.5% vs 14.9% and 9.8% vs 15.6%, respectively). Data on cytogenetics will be included at the time of presentation.

Conclusion: At the primary analysis, patients treated with KdD achieved significantly higher MRD[‑]CR rates vs Kd at the 12-month landmark. Among patients with an MRD[-]CR, the depth of MRD was deeper with KdD vs Kd. With a median of 6 months follow-up, no patient with an MRD[-]CR has progressed; duration of response will be updated at time of presentation. Within the KdD arm, lenalidomide exposure or refractoriness did not diminish the MRD[-]CR rate. These findings support the efficacy of the KdD regimen as an effective treatment for RRMM, including patients who have become lenalidomide refractory.