British Journal of Haematology | 2021
Carfilzomib, dexamethasone and daratumumab in relapsed or refractory multiple myeloma: results of the phase III study CANDOR by prior lines of therapy
Abstract
Bortezomiband lenalidomide-based therapies have become standard front-line treatment for multiple myeloma (MM), resulting in many patients being refractory to bortezomib or lenalidomide at first relapse. This may negatively impact the efficacy of later lines of therapy (LOT) and makes optimal sequencing of MM therapies challenging. Carfilzomib is a second-generation proteasome inhibitor approved in combination with dexamethasone (Kd), and also with lenalidomide (KRd), and daratumumab (KdD) for patients with relapsed/refractory MM (RRMM). Clinically meaningful improvements in progression-free survival (PFS) and overall survival have been observed with KRd versus Rd (ASPIRE) and Kd versus Vd (ENDEAVOR), regardless of the type or number of prior LOT, indicating the superiority of carfilzomib-based regimens relative to previous standards of care in patients with RRMM previously exposed to lenalidomide or bortezomib. Improved clinical outcomes have also been observed in several MM trials combining daratumumab, an anti-CD38 monoclonal antibody, with standards of care, including proteasome inhibitors. In the phase III, randomised, open-label CANDOR trial, KdD improved PFS versus Kd [median PFS not reached vs. 15 8 months; hazard ratio (HR) 0 63; 95% confidence interval (CI) 0 46–0 85; two-sided P = 0 0027] in patients with RRMM (42% of whom were lenalidomide-exposed). Given the challenges of effectively treating patients with RRMM and the common use of front-line bortezomib and lenalidomide, we performed a pre-planned subgroup analysis of CANDOR to evaluate efficacy and safety by number of prior LOT (one vs. two or more), as well as a post hoc analysis of patients with previous exposure or refractory status to bortezomib/ixazomib or lenalidomide. The study design and patient eligibility criteria for the CANDOR trial (NCT03158688) have been previously described and are summarised in the Supplement. A total of 466 patients were randomised 2:1 to either KdD (n = 312) or Kd (n = 154) between 13 June 2017 and 25 June 2018 (data cut-off: 14 July 2019). Across both arms, 43% had one prior LOT, 57% had two or more prior LOT, 42% were lenalidomide-exposed, 33% were lenalidomide-refractory, 91% were bortezomib/ixazomib-exposed, and 33% were bortezomib/ixazomib-refractory. Baseline characteristics were generally balanced between arms across subgroups (Table SI). Results from the subgroup analyses were generally consistent with the overall PFS HR of the study. Among patients with one prior LOT, the PFS HR (95% CI) for KdD versus Kd was 0 68 (0 40–1 14) compared to 0 61 (0 42–0 88) for two or more prior LOT (Fig 1). For patients with one prior LOT, the PFS HR (95% CI) was 0 90 (0 48–1 70) for lenalidomide-naive patients, 0 30 (0 10–0 86) for lenalidomide-exposed patients and 0 11 (0 02–0 52) for lenalidomide-refractory patients. The overall response rate (ORR) was 90 2% versus 76 1% [odds ratio (OR) 2 90, 95% CI 1 30–6 46] among patients with one prior LOT compared to 79 9% versus 73 6% (OR 1 43, 95% CI 0 78–2 60) for two or more prior LOT (Fig 2A). For patients with one prior LOT, minimal residual disease (MRD)-negative complete response (CR) rate was 16 5% versus 1 5% (OR 13 08, 95% CI 1 72– 99 31) compared to 9 5% versus 1 1% (OR 9 03, 95% CI 1 18–68 97) for two or more prior LOT (Fig 2A). The PFS HR (95% CI) was 0 71 (0 45–1 12) for lenalidomide-naive patients, 0 53 (0 34–0 82) for lenalidomide-exposed patients and 0 47 (0 29–0 78) for lenalidomiderefractory patients (Fig 1). The ORR was 87 8% versus 75 0% (OR 2 41, 95% CI 1 23–4 69) for lenalidomide-naive patients, 78 9% versus 74 3% (OR 1 29, 95% CI 0 65–2 54) for lenalidomide-exposed patients and 79 8% versus 72 7% (OR 1 48, 95% CI 0 69–3 20) for lenalidomide-refractory patients (Fig 2B). The MRD-negative CR rate was 13 2% versus 2 5% (OR 5 95, 95% CI 1 37–25 74) for lenalidomidenaive patients, 11 4% versus 0% [OR non-estimable [NE]; 95% CI NE–NE] for lenalidomide-exposed patients and 13 1% versus 0% (OR NE, 95% CI NE–NE) for lenalidomide-refractory patients (Fig 2B). The PFS HR (95% CI) was 0 58 (0 17–2 06) for bortezomib/ixazomib-naive patients, 0 62 (0 45–0 85) for bortezomib/ixazomib-exposed patients and 0 84 (0 52–1 36) for bortezomib/ixazomib-refractory patients (Fig 1). The ORR was 95 7% versus 82 4% (OR 4 71, 95% CI 0 45–49 94) for bortezomib/ixazomib-naive patients, 83 4% versus 73 7% (OR 1 79, 95% CI 1 10–2 92) for bortezomib/ixazomib-exposed patients and 79 0% versus 69 1% (OR 1 68, 95% CI 0 80–3 55) for bortezomib/ixazomib-refractory patients (Fig 2C). The MRD-negative CR rate was 21 7% versus 0% (OR NE, 95% CI NE–NE) for bortezomib/ixazomib-naive patients, 11 8% versus 1 5% (OR 9 00, 95% CI 2 13–38 03) for bortezomib/ixazomib-exposed patients and 7 0% versus Correspondence