In this issue of Blood, Roussel et al report the findings of the phase 2 trial of frontline carfilzomib, lenalidomide, and dexamethasone (KRd) with autologous stem cell transplantation (ASCT) in transplant-eligible (TE) patients with newly diagnosed multiple myeloma (NDMM).1
The aim of initial therapy for TE patients with NDMM is to maximize the depth and duration of the first response. Multiple studies have shown the impact of depth of response, increasingly measured by minimal residual disease (MRD) negativity, as a surrogate marker for survival.2 Bortezomib plus immunomodulatory drug–based triplet induction followed by ASCT is now considered standard of care for the TE population. The Intergroupe Francophone du Myélome 2009 (IFM-09) trial demonstrated a progression-free survival (PFS) advantage with the bortezomib, lenalidomide, and dexamethasone (VRd)-ASCT arm vs VRd alone (50 vs 36 months), a complete response (CR) rate of 59%, and an MRD− rate of 79% (multiparametric flow cytometry [MFC], 10−4).3 However, high rates of peripheral neuropathy (PN) with bortezomib often lead to treatment discontinuation, limiting induction and consolidation regimens, and resulting in suboptimal responses.
Carfilzomib, a second-generation proteasome inhibitor (PI), has a distinct structure and mechanism of action compared with bortezomib.4 More importantly, carfilzomib therapy is associated with low rates of PN, is able to overcome bortezomib resistance, and has demonstrated efficacy in the relapsed/refractory setting.5 However, the question remains: does KRd produce deeper responses than VRd, thus justifying the cost and toxicity of KRd? To date, there have been no phase 3 randomized studies comparing KRd to VRd incorporating ASCT in NDMM. The phase 3 ENDURANCE trial found that KRd was not superior to VRd. However, this trial was performed in transplant-ineligible (or those without ASCT intention) standard- or intermediate-risk patients with NDMM who were a median age of 65 years.6
For this study, the IFM investigators hypothesized that switching bortezomib to carfilzomib would improve the stringent CR (sCR) rate at the completion of consolidation (primary objective) without increasing toxicity. The trial evaluated the efficacy and safety of 8 cycles of KRd as induction (4 cycles) and consolidation (4 cycles) in TE patients with NDMM, followed by a year of lenalidomide (10 mg) maintenance posttransplant and postconsolidation. All patients proceeded to ASCT after 4 cycles. Of the 46 patients enrolled, 21% had adverse cytogenetics. In 42 evaluable patients after consolidation, the rate and duration of response were impressive: 26 had sCR (61.9%); 27 had at least CR (64%), of whom 92.6% had undetectable MRD by MFC (2.5 × 10−5) and 63.0% had undetectable MRD by next-generation sequencing (NGS) (10−6). With a median follow-up of 60.5 months, the median PFS was 56.4 months with an estimated 5-year overall survival (OS) of 77.8%. The combination was safe and well tolerated with no KRd-related deaths; 4 patients discontinued treatment due to toxicities and 33 patients (72%) had at least 1 dose modification of 1 study drug. The most frequent grade 3/4 adverse events (AEs) were hematological (74%) and infectious (22%). Serious AEs were reported in 31 patients with 8 cardiovascular events (including 2 heart failures, 5 pulmonary embolisms or deep vein thromboses), highlighting the need for close monitoring for these AEs and appropriate antithrombotic prophylaxis for this regimen.
Although it is difficult to make cross-trial comparisons, this study suggests a PFS advantage for KRd of 6 months, and a higher MRD-negativity rate compared with the VRd-ASCT arm in the IFM-09 trial. Several other phase 1/2 trials have also demonstrated the feasibility and safety of KRd-ASCT in the treatment of NDMM. The Multiple Myeloma Research Consortium (MMRC) recently published a similar study utilizing KRd induction, ASCT, consolidation, and 10 cycles of KRd maintenance. Similar sCR (60% postconsolidation) and comparable toxicity rates were observed with 70% of patients achieving MRD negativity using NGS (10−5). Of note, high-risk patients who were MRD− had encouraging 5-year PFS and OS rates of 77% and 81%, respectively.7 Available phase 3 data incorporating KRd and transplant include the UK Myeloma XI trial, which demonstrated the superiority of KRd plus cyclophosphamide (KRdc) induction pre-ASCT (n = 526) compared with lenalidomide, dexamethasone, and cyclophosphamide or thalidomide, dexamethasone, and cyclophosphamide induction (n = 530). Neither control arms included a PI. After a median follow-up of 34.5 months, the median PFS for the KRdc group was significantly longer than the triplet control group (not estimable vs 36.2 months; P < .001) with 75% of patients in the KRdc group achieving MRD negativity (MFC, 4 × 10−5) post-ASCT.8 Findings from the ongoing FORTE trial also demonstrated the efficacy of the KRd-ASCT regimen; this phase 3 randomized study compared 4 KRd cycles followed by ASCT and 4 KRd consolidation cycles (n = 158); 4 carfilzomib, cyclophosphamide, and dexamethasone (KCd) cycles followed by ASCT and 4 KCd consolidation cycles (n = 159); or 12 KRd cycles (n = 157). Patients in the KRd-transplant arm had a 1-year sustained undetectable MRD rate of 68% of patients compared with 54% in the nontransplant arm (P = .02). With a median follow-up of 45 months, the estimated 3-year PFS was 78% in the transplant arm vs 66% in the nontransplant arm (P = .023).9
With the high MRD-negativity rate associated with triplet induction and ASCT, the current question is whether the addition of anti-CD38 monoclonal antibodies (daratumumab [Dara]-KRd or isatuximab-KRd) can further improve depth and duration of response and translate to improved survival, obviating the need for ASCT. Preliminary results with the Dara-KRd quadruplet report an MRD− rate of 83% without ASCT.10 We eagerly await the findings of the ADVANCE trial (NCT04268498), which will compare Dara-KRd with KRd or VRd and the phase 3 COBRA study (NCT03729804) comparing extended KRd (24 cycles) with the established RVd regimen (8 cycles with lenalidomide maintenance) in high-risk patients with deferred ASCT.
The utility and achievement of MRD negativity during maintenance therapy has also yet to be well defined. Can patients with sustained MRD− sCR cease maintenance therapy without the risk of early relapse? A risk and MRD-adaptive approach in the intensification or de-escalation of treatment is still under investigation and will help to guide clinical decisions in the future. Until then, this study demonstrates that KRd induction/consolidation with ASCT achieves deep and durable responses with a manageable safety profile and is a future option for the upfront treatment of TE patients with NDMM.
Conflict-of-interest disclosure: S.J.H. provided consultancy services to, served on an advisory board for, and was an investigator on studies for AbbVie; provided consultancy services to, received honoraria from, served on advisory boards for, received research funding from, and was an investigator on studies for Amgen, Celgene, Janssen Cilag, and Novartis; provided consultancy services to and served on advisory boards for GlaxoSmithKline; provided consultancy services to, received honoraria from, served on advisory boards for, and was an investigator on studies for Roche/Genentech; provided consultancy services to, received honoraria from, and served on advisory boards for Takeda; served on scientific advisory boards for, received research funding from, and was an investigator on studies for Haemalogix; and provided consultancy services to/held an advisory role with Sanofi. J.E. declares no competing financial interests.