Four Is Better Than Three: Will We Finally Agree?

The phase II GRIFFIN study investigated the effects of adding the anti-CD38 monoclonal antibody (mAb) daratumumab in intravenous form to treatment with bortezomib, lenalidomide, and dexamethasone (VRd) in the United States. Improvements were noted in both the primary endpoint of depth of response (stringent complete remission: 67% vs. 48%) and the secondary endpoint of progression-free survival (PFS at four years: 87.2% vs. 70%, hazard ratio [HR]: 0.45 [95% CI 0.21-0.95]) in transplant-eligible patients with newly diagnosed multiple myeloma (NDMM).¹ 

In the confirmatory phase III PERSEUS study, sponsored by the European Myeloma Network in collaboration with Janssen, 709 transplant-eligible patients ages 18 to 70 were randomized to VRd +/– subcutaneous (SQ) daratumumab from 2019 to 2020. Randomization was stratified according to International Staging System (ISS) scores and cytogenetic risk. Patients in the control arm received four 28-day cycles of VRd, melphalan at 200 mg/m² with autologous stem cell rescue, and two 28-day consolidation cycles of VRd, followed by lenalidomide maintenance until progression. Notably, dexamethasone (40 mg) was given on days one to four and days nine to 12, while bortezomib was given at 1.3 mg/m² on days one, four, eight, and 11. Patients in the experimental quadruplet arm also received daratumumab SQ in the induction and consolidation phases, in accordance with the schedule recommended on the prescribing label, and once monthly thereafter. After two years, patients who had achieved a CR and sustained minimal residual disease (MRD) negativity for 12 months discontinued daratumumab and continued single-agent maintenance therapy with lenalidomide, whereas those with lesser responses could continue doublet maintenance with daratumumab and lenalidomide until progression.

After a median follow-up of 47.5 months, the primary endpoint was met, with 84.3% of patients being progression-free in the quadruplet DVRd arm versus 67.7% in the control/triplet VRd arm. Moreover, the HR was nearly identical to that observed in the GRIFFIN trial (HR: 0.42 [95% CI 0.30-0.59], p<0.001). All subgroups — including patients with an ISS score of 3 and those with a high-risk cytogenetic profile — appeared to benefit, except when those age 65 and older (HR: 0.97 [95% CI 0.52-1.81] were compared with those under age 65 (HR: 0.30 [95% CI: 0.20-0.46]). Both the proportion of patients achieving CR or better (87.9% vs. 70.1%, p<0.001) and the rate of MRD negativity (75.2% vs. 47.5%, p<0.001) at the 10^-5 threshold were also higher in the quadruplet arm than in the control arm.

Although follow-up for overall survival (OS) was immature, there were 34 (9.6%) and 44 (12.4%) deaths in the quadruplet and control arms, respectively, including four (1.1%) and one (0.3%) COVID-related deaths. Rates of grade 3+ neutropenia (62.1% vs. 51.0%), thrombocytopenia (29.1% vs. 17.3%), serious adverse events (SAEs) (57.0% vs. 49.3%) — particularly SAEs due to pneumonia (11.4% vs. 6.1%) — were all higher for DVRd than for VRd. However, the rate of treatment discontinuation due to an AE was higher in the VRd arm (22.5%) than in the DVRd arm (9.1%). The median CD34⁺ cell yield during stem cell collection was 5.5 vs. 7.4 × 10⁶/kg for the quadruplet vs. triplet arm.

The delivery of DVRd as outlined in the PERSEUS trial appears feasible in real-world practice. However, the four-day dexamethasone pulses administered in PERSEUS are rarely used today given the increased risks of thromboses, infections, and deaths when compared with those for weekly dexamethasone.²  Although cycle length differed between PERSEUS and GRIFFIN (28 vs. 21 days), both used twice-weekly doses of bortezomib, whereas many patients receive only once-weekly dosing outside of clinical trials.

What remains unclear is the optimal duration of frontline anti-CD38 mAb therapy needed to attain the benefits observed. Unlike the CASSIOPEIA study,³  the present study did not incorporate a second randomization at the time of maintenance. In the CASSIOPEIA trial, patients who received daratumumab during induction did not appear to experience a PFS benefit from the inclusion of daratumumab in the maintenance phase as well. While daratumumab therapy was only continued beyond two years in 36% of patients in PERSEUS, the effects of such protracted anti-CD38 mAb therapy on both health economics and outcomes following the next line of therapy and on OS require longer follow-up. We know, for example, that lenalidomide resistance confers downstream adverse prognosis in relapsed MM⁴ ; however, the use of maintenance lenalidomide until progression yields OS benefits.⁵ 

On the other hand, the PFS curves in PERSEUS and GRIFFIN did not separate during the induction period but diverged later with continued therapy. Furthermore, the rates of MRD negativity increased during maintenance, with the difference in MRD negativity rates between the arms widening over time during the maintenance phase, favoring the addition of daratumumab. Also, in CASSIOPEIA, maintenance daratumumab was given only once every eight weeks, which has been considered suboptimal based on pharmacokinetics.⁶  Ultimately, the optimal duration of anti-CD38 mAb therapy will need to stem from prospective, risk- and/or response-adapted studies such as DRAMMATIC and AURIGA.^7-8  Other questions to investigate within this disease segment include the outcomes of frontline isatuximab and/or carfilzomib as part of quadruplet induction.