In the treatment of newly diagnosed multiple myeloma (MM), carfilzomib combined with lenalidomide and dexamethasone (KRd) shows significantly greater efficacy than the long-held cornerstone regimen of bortezomib with the len/dexa combination (VRd), irrespective of patients’ cytogenetic risks or eligibility for upfront autologous hematopoietic stem cell transplantation (ASCT).

“The COBRA study…[achieved] both co-primary endpoints of minimal residual disease (MRD) negativity at 12 months and progression-free survival (PFS),” said first author Dominik Dytfeld, MD, while presenting the study findings at the American Society of Hematology (ASH) 2025 Annual Meeting.

Evolving treatment regimens for newly diagnosed MM have led to highly effective drug combinations and improvements in overall survival, with the VRd regimen long representing a standard of care.

Studies have shown improvement upon VRd with the utilization of the next-generation proteasome inhibitor carfilzomib in place of bortezomib; however, the previous multicenter, phase 3 ENDURANCE trial showed no significant differences between the two regimens in terms of PFS.

With key areas of uncertainty including the frontline KRd treatment of patients who are eligible for ASCT and those with high-risk cytogenetics, Dytfeld and colleagues conducted the phase 3, multicenter COBRA trial, spanning a population of patients with newly diagnosed MM, regardless of their ASCT eligibility or cytogenetic risks.

For the open-label trial, 250 patients were randomized 1:1 to treatment with KRd (n = 126) or standard VRd (n = 124).

Those in the KRd arm specifically received carfilzomib at a dose of 56 mg/m² on days 1, 8, and 15 for 3 consecutive weeks, followed by 1 week off for 12 cycles. For the subsequent 12 cycles (cycles 13-24), the second dose of carfilzomib (day 8) was skipped.

Lenalidomide was given at a dose of 25 mg for 21 days, and dexamethasone was given once a week at a dose of 40 mg in the first 12 cycles and 20 mg in cycles 13 through 24.

After 24 cycles, the treatment was de-escalated to lenalidomide monotherapy until progression or intolerance.

In the VRd arm, patients received bortezomib 1.3 mg/m² twice a week, lenalidomide at a dose of 25 mg for the first 14 days of the 21-day cycle, and dexamethasone 40 mg on days 1, 2, 4, 5, 8, 9, 11, and 12 for eight 21-day cycles.

Patients then received 18 cycles of lenalidomide plus dexamethasone maintenance, followed by lenalidomide 15 mg daily until progression or intolerance.

All patients who were eligible for stem cell transplantation after four cycles had stem cells collected for deferred transplantation.

The patients’ baseline characteristics were well balanced, with a median age of 66 years and 67 years for the KRd and VRd groups, respectively; 26% and 22% had International Staging System stage III; in both groups, 75% had an Eastern Cooperative Oncology Group performance status of 1 or 2, and 23% in each group had high-risk cytogenetic abnormalities, specifically del(17p), t(4;14), or t(14;16).

At a median follow-up of 53 months, treatment with KRd was associated with a 43% greater reduction in the risk for death or disease progression, with a rate of 28% and median PFS not reached, compared with corresponding rates of 41% and a median PFS of 48.8 months with VRd (hazard ratio [HR], 0.57).

The benefits with KRd were also more improved in patients with standard-risk cytogenetics, with median PFS for KRd not reached and median PFS for VRd of 48.8 months (HR, 0.59).

The high-risk population also had greater improvements, with KRd associated with a 48% reduction in the risk for death or progression: While the median PFS for the KRd was not reached, the rate for VRd was 34.9 months (HR, 0.52).

The improvements extended to patients who were eligible for transplantation, with KRd associated with a 60% reduction in the risk for death or progression and a median PFS that was again not reached compared with 40.1 months for VRd (HR, 0.40).

No differences in PFS were observed in the transplant-not-eligible population, with median PFS not reached for KRd and 52.9 months for the VRd group (HR, 1.06).

In terms of MRD negativity at 12 months, the rate in the intent-to-treat population at a 10^-5 threshold was 31% for the KRd group and 17.7% for the VRd group (odds ratio [OR], 2.08; P = .016).

When considering the results at the 10^-6 threshold, the improvements with KRd at 12 months were even more pronounced, with rates of 19% vs 7.3% with VRd (OR, 3.01; P = .008).

No differences were observed in the overall response rate; however, the rate of complete response was significantly higher in the KRd group at 70.6% compared with 53.2% in the VRd group (OR, 2.11; P = .005).

Furthermore, no differences were observed in overall survival, with follow-up on the survival outcomes ongoing.

Rates of toxicity were as expected, with similar rates of adverse events (AEs) seen in both arms.

Grade 3 or higher AEs occurred in 72% of patients in the KRd arm vs 62% in the VRd arm, with the most common grade 3 or higher hematologic AE being neutropenia, occurring in 21% of patients in the KRd arm and 11% in the VRd arm.

For non-hematologic grade 3 or higher AEs, lower respiratory tract infections occurred in 10% of patients in both arms.

Other AEs included peripheral neuropathy of any grade, occurring in 17% and 56% of patients in the KRd and VRd arms, respectively, and cardiac events of any grade, occurring in 18% and 10% of patients.

Treatment-related deaths occurred in two patients in the KRd arm, attributable to pneumonia and COVID.

Overall, “KRd demonstrated anticipated toxicity, with higher rates of neutropenia and cardiac events, but less neuropathy compared to the VRd,” said Dytfeld of Poznan University of Medical Sciences, Poznan, Poland.

He added, “KRd produced deeper responses in terms of the higher rates of this complete response and MRD negativity. Thus, we believe the COBRA results support further investigation of the KRd-based frontline regimen in patients with newly diagnosed multiple myeloma.”

Commenting on the study, Edward A. Faber, Jr, DO, associate professor of medicine, director of the Adult BMT & Cellular Therapy Program, and Hematology Disease Leader at the University of Cincinnati Cancer Center, in Cincinnati, who was an author on the ENDURANCE trial, said that the new findings build upon previous research.

“What we learned from ENDURANCE is that complete response and very good partial response favored KRd versus VRd; however, the toxicity profile abrogated this benefit, thus PFS and OS were similar,” he told Medscape Medical News.

“The current study supports similar conclusions — grade 3 adverse effects more for KRd than VRd and higher cardiac side effects,” he added. “Of course, neuropathy was much higher, but that is well known for VRd and consistent.”

Faber noted the issue of duration of dosing that was also a topic of interest in the talk’s Q and A — specifically, that patients in the KRd arm received 12 months of therapeutic doses, continuing with a lighter version for a total of 24 months, whereas patients in the VRd arm received only eight 3-week cycles, representing about 8 months, before moving on to maintenance with lenalidomide plus dexamethasone.

“Assessments were done at 12 months, with one arm still receiving ‘full dose’ and the other arm already 4 months into maintenance,” he pointed out.

“It may have been a better comparison if patients had received 12 months of the ‘full-dose’ therapy in both arms.”

Finally, while the once-weekly dosing with KRd appeared to have over-performed previous studies, including the A.R.R.O.W. trial, on which Faber was also a co-author, and VRd also performed better, albeit not to the same degree, “in the end, these results seem to be consistent with existing data,” he noted.

Faber noted the additional caveat of the field having progressed from triplet to quadruplet treatment regimens as the standard of care, which Dytfeld addressed in the Q and A.

Further commenting, Murali Janakiram, MD, of the Division of Myeloma, Department of Hematology, City of Hope National Medical Center, Duarte, California, agreed that the findings should be interpreted in the context of the notably different induction periods in the study.

“If induction is planned for a longer time without a CD38 antibody, then KRd for 24 cycles is superior to VRd for six cycles and maintenance,” he told Medscape Medical News.

Dytfeld reported consulting or other relationships with Takeda, Amgen, Johnson & Johnson, Bristol Myers Squibb (BMS), Pfizer, BeiGene, and GSK. Faber reported having no disclosures. Janakiram reported having relationships with BMS and Johnson & Johnson.