Lancet Haematol

Background: Triplet and quadruplet regimens based on bortezomib, melphalan and prednisone (VMP) and lenalidomide and dexamethasone (Rd) with anti-CD38 antibodies are potential treatments for transplant-ineligible patients with newly diagnosed multiple myeloma. However, the high risk of toxic effects in this population requires frailty-based therapy adaptation. We aimed to compare the response of carfilzomib-based triplet and quadruplet regimens with a VMP-Rd regimen in newly diagnosed transplant-ineligible patients with multiple myeloma, considering patient frailty.

Methods: GEM-2017FIT was an open-label, randomised, phase 3 trial at 57 hospitals in Spain. Patients aged 65-80 years were enrolled and assessed for frailty using the Geriatric Assessment in Hematology (GAH) scale. Patients were randomly assigned (1:1:1) to receive 18-cycle induction therapy of VMP 9-Rd 9 (one six-week cycle of melphalan 9 mg/m2 and prednisone 60 mg/m2 on days 1-4; bortezomib 1·3 mg/m2 subcutaneous twice weekly, followed by eight four-week cycles of weekly VMP and nine four-week cycles of lenalidomide 25 mg on days 1-21 and dexamethasone 40 mg weekly), carfilzomib-based triplet (KRd; carfilzomib intravenously 20 mg/m2 [only in the infusion on day 1 in first cycle] or 36 mg/m2 on days 1, 2, 8, 9, 15, and 16 in cycles 1-2, 56 mg/m2 in cycles 3-18, plus Rd) or daratumumab-KRd (D-KRd; daratumumab 16 mg/kg intravenous weekly [cycles 1-2], biweekly [cycles 3-6], and every 4 weeks [cycles 7-18]). All patients who completed induction therapy and consolidation were stratified by measurable residual disease status and both those with undetectable measurable residual disease and detectable measurable residual disease were subsequently randomly assigned (1:1) to maintenance therapy with daratumumab and lenalidomide or no maintenance therapy. The primary endpoint was measurable residual disease negativity after induction, which was assessed in the intention-to-treat population. The trial is registered with ClinicalTrials.gov, NCT03742297.

Findings: Between October 15, 2018 and December 15, 2021, 540 patients were enrolled and assessed for eligibility. 462 were eligible for the study and randomly assigned to VMP 9-Rd 9 (n=154), KRd (n=154) or D-KRd (n=154, with one patient subsequently found to be ineligible). 230 (50%) of 461 patients were male and 231 (50%) were female. Patients were followed up for a median of 33·15 months (IQR 25·82-43·08). The 18-cycle undetectable measurable residual disease rate with a sensitivity level of 10-5 in the intention-to-treat population was higher in the KRd group (83 [54%] of 154 patients; odds ratio [OR] 1·73, 95% CI 1·39-2·16; p<0·0001) and D-KRd group (94 [61%] of 153 patients; 2·03, 1·61-2·57; p<0·0001) than in the VMP 9-Rd 9 group (41 [27%] of 154 patients). The incidence of grade 3-4 neutropenia was lower in the KRd group (37 [24%] of 154 patients) compared with the VMP 9-Rd 9 group (62 [40%] of 154 patients) and D-KRd group (63 [41%] of 153 patients). Grade 3-4 infections occurred in 19 (12%) patients in the VMP 9-Rd 9 group, 23 (15%) patients in the KRd group, and 25 (16%) patients in the D-KRd group. Toxicity-related death occurred in a similar frequency in the VMP 9-Rd 9 (seven [5%] patients) and KRd (five [3%] patients) groups, but was significantly higher in the D-KRd group (13 [8%] patients; OR 0·53, 95% CI 0·22-1·30; p=0·16).

Interpretation: KRd and D-KRd were superior to VMP 9-Rd 9 in achieving measurable residual disease negativity after 18 cycles. This study could contribute to incorporation of quadruplet therapy into clinical practice and supports the need for frailty-based assessment in therapy selection.

Funding: PETHEMA.