Emerging data suggest that isatuximab‐KRd and daratumumab‐KRd produce superior depth of response compared with KRd alone, with sustained MRD negativity and encouraging progression‐free survival in NDMM.
Quadruplet regimens incorporating the second‐generation proteasome inhibitor carfilzomib into frontline therapy for newly diagnosed multiple myeloma (NDMM) are producing deep and durable responses, often with high rates of minimal residual disease (MRD) negativity, and may offer an alternative to the long‐standing bortezomib‐based standard. A comprehensive review in HemaSphere brings together current evidence and emerging clinical trial data to outline mounting evidence for the KRd backbone (carfilzomib-lenalidomide-dexamethasone), both as a triplet and in combination with the anti‐CD38 monoclonal antibodies isatuximab (Isa) or daratumumab (D).1
Multiple myeloma accounts for about 2% of all cancers and 10% of hematologic malignancies in the US, with over 36,000 new cases expected in 2025.2 For transplant‐eligible (Te) NDMM, VRd (bortezomib-lenalidomide-dexamethasone) remains the most widely used triplet backbone, often paired with an anti‐CD38 agent. However, the National Comprehensive Cancer Network (NCCN) guidelines also list KRd, Isa‐KRd, and D‐KRd among options for Te patients, despite KRd lacking formal regulatory approval in this setting.3
Whether KRd‐based quadruplets will redefine the frontline standard now hinges on proving they can deliver deeper, longer‐lasting remissions without compromising safety. | Image credit: Dzmitry - stock.adobe.com
Early‐phase studies established KRd’s strong activity in both transplant‐eligible and transplant‐ineligible (Ti) patients. The MMRC Phase 1/2 trial (n = 53) reported near-complete response rates exceeding 38% after 4 induction cycles, with median progression-free survival (mPFS) not being reached after 13 months of follow-up.4 In the NCI/NIH Phase 2 trial (n = 45), 8 KRd cycles followed by lenalidomide maintenance yielded a mPFS of 67.3 months, with ≥ very good partial response (VGPR) in 89% and MRD negativity (MFC, 10⁻⁵) in all patients achieving near‐complete response or better.5 KRd was added to the NCCN guidelines based on category 2A evidence after publication of this trial.
Real‐world data from Memorial Sloan Kettering (n = 389) comparing KRd with VRd showed a 5‐year PFS of 67% for KRd vs 56% for VRd (P = .027) and higher MRD negativity rates (40% vs 27%). KRdalso significantly improved post‐induction complete response or better (≥CR) and ≥VGPR rates compared with VRd.6 Similarly, the IFM‐KRd study (n = f48) reported stringent CR rates of 62% post-consolidation and MRD negativity in 93% by flow cytometry, with a mPFS of 56.4 months.7
In contrast, the larger multicenter head-to-head Phase 3 ENDURANCE trial (n = 1087) enrolled only standard‐risk, non‐transplant‐intended patients and excluded those with high‐risk cytogenetics—factors that have drawn criticism for limiting its applicability. While KRd yielded higher ≥VGPR rates than VRd (74% vs 65%, P = .0015), it did not significantly improve mPFS (34.6 vs 34.4 months).8 These results, coupled with the trial’s restrictive design, have fueled debate and ultimately influenced the decision to exclude KRd from standard first‐line treatment guidelines for NDMM.
The FORTE trial (n = 474) further supported KRd’s potency. Among patients undergoing ASCT, KRd achieved ≥VGPR in 70% post‐induction vs 53% with KCd (carfilzomib-cyclophosphamide-dexamethasone), and 1‐year sustained MRD negativity rates were 47% with KRd plus ASCT vs 35% with KRd alone. In the trial’s maintenance phase, KR significantly improved 3‐year PFS over lenalidomide alone (75% vs 65%; HR, 0.64).9
The randomized Phase 3 IsKia/EMN24 trial (n = 302) is the first to directly compare Isa‐KRd to KRd in patients with Te NDMM. After induction, ASCT, and consolidation, MRD negativity at 10⁻⁵ was achieved in 77% vs 67% (P = .049), with MRD negativity at 10⁻⁶ in 67% vs 48% (P < .001). Benefit was consistent across cytogenetic risk groups, and stem cell mobilization was unaffected by the quadruplet.10
The GMMG‐CONCEPT trial exclusively enrolled high-risk patients with ISS stage II/III plus del(17p), t(4;14), t(14;16), or ≥3 copies of 1q21. Among Te patients (n = 99), Isa‐KRd achieved MRD negativity in 68% post‐consolidation, with 80% maintaining MRD negativity for ≥1 year. Median PFS was not reached at 6 years, showing durability in a traditionally poor‐prognosis cohort.11
The Phase 2 MASTER trial (n = 123) used MRD‐guided consolidation after 4 D‐KRd induction cycles and ASCT. MRD negativity at 10⁻⁵ was achieved in 81%, with 71% of patients stopping therapy under MRD surveillance. Two‐year progression rates post‐therapy were 9% for patients with 0 to 1 high‐risk chromosomal abnormality, but 47% for those with ≥2.12
Elderly fit patients with Ti NDMM were specifically evaluated in the Phase 3 GEM2017FIT trial (n ≈ 460), where D‐KRd achieved MRD in 84% of evaluable patients vs 75% for KRd. However, frailer patientswith a Geriatric Assessment in Hematology score ≥20 had higher toxicity‐related discontinuations, suggesting careful selection is warranted in older adults.13
KRd and KRd‐based quadruplets share a hematologic toxicity profile, with Grade ≥3 neutropenia in 20% to 47% of patients and infections in up to 28%. Carfilzomib carries a higher cardiovascular risk than bortezomib, with Grade 3 to 5 cardiac, pulmonary, or renal events reported in 16% of KRd recipients in ENDURANCE vs 5% for VRd. Conversely, peripheral neuropathy rates were lower, <1% for KRd vs 8% for VRd.
Dosing strategies vary, with most quadruplet trials using weekly carfilzomib at 56 mg/m² during induction, with some transitioning to every‐other‐week schedules during maintenance to improve tolerability and reduce treatment burden. The choice between weekly and twice‐weekly dosing can be individualized based on patient fitness, logistical factors, and response depth.
Multiple ongoing Phase 3 trials, including MIDAS (Isa‐KRd) and ADVANCE (D‐KRd), will clarify whether KRd‐based quadruplets can match or surpass VRd‐based regimens as the frontline standard.14,15 Highlighting the potential clinical impact, the authors note, “The recommendations may lead to more frequent use of these combinations as frontline therapy for patients with NDMM, and their efficacy can be maximized by incorporating strategies to mitigate safety concerns, including the establishment of optimal dosing and administration schedules.” Whether KRd‐based quadruplets will redefine the frontline standard now hinges on proving they can deliver deeper, longer‐lasting remissions without compromising safety.1
References
Personalized Care Key as Tirzepatide Use Expands Rapidly
April 15th 2025Using commercial insurance claims data and the US launch of tirzepatide as their dividing point, John Ostrominski, MD, Harvard Medical School, and his team studied trends in the use of both glucose-lowering and weight-lowering medications, comparing outcomes between adults with and without type 2 diabetes.
Listen
For Better Cancer Outcomes, It’s All About Access
August 21st 2025New drugs aren't the only advances in oncology. Innovation includes collaboration between to remove barriers to remove barriers to care, according to experts who gathered for a session of the Institute for Value-Based Medicine in Arlington, Virginia.
Read More
Nomogram Aids Early Detection of Myasthenic Crisis Risk
August 19th 2025Despite nurses often being the first health care workers many patients encounter, they frequently lack adequate tools to optimally evaluate their patients’ conditions, as in the case with myasthenia gravis and being able to predict patients at risk of myasthenic crisis.
Read More