Minimal residual disease dynamics in post-transplant patients with newly diagnosed multiple myeloma who received daratumumab plus lenalidomide versus lenalidomide alone as maintenance therapy in the auriga study

Abstract

Abstract Introduction:The standard of care following autologous stem cell transplant (ASCT) for patients (pts) with newly diagnosed multiple myeloma (NDMM) is lenalidomide (R) maintenance. In the phase 3 AURIGA study (NCT03901963), adding subcutaneous daratumumab (DARA SC) to R maintenance (D-R) improved MRD-neg conversion (conv) and PFS in anti-CD38–naïve pts who were MRD pos post-ASCT. To better understand the evolution of MRD status during maintenance and its impact on long-term outcomes, MRD dynamics from the AURIGA study are reported herein. Methods:Eligible pts with NDMM were aged 18-79 yrs; had very good partial response or better and were MRD pos (10–5; NGS) following ≥4 cycles of anti-CD38–free induction therapy and ASCT; and enrolled within 6 mo of ASCT. Pts were stratified by cytogenetic risk and randomized 1:1 to receive R maintenance ± DARA SC for up to 36 (28-day) cycles or until progressive disease (PD), unacceptable toxicity, or consent withdrawal. MRD was assessed by NGS at 12, 18, 24, and 36 mo after maintenance start. The primary endpoint was MRD-neg conv rate (10–5) by 12 mo from maintenance start. Results: In total,200 pts were randomized (n=99 for D-R, n=101 for R). Baseline characteristics were well balanced between groups. At a median follow-up of 40.3 mo, overall MRD-neg conv rates at 10−5 were 60.6% for D-R and 28.7% for R (OR, 3.92 [95% CI, 2.16-7.14]; P<0.0001) and at 10−6 were 36.4% and 13.9% (OR, 3.59 [95% CI, 1.78-7.23]; P=0.0003). MRD-neg conv rates (10–5) were higher for D-R vs R at 12 mo (49.5% vs 18.8%) with similar trends at 18 and 24 mo. There was a trend towards decreased risk of PD (per investigator assessment) or death for D-R vs R in pts who achieved MRD-neg conv at 10–5 (n=6/60 vs n=4/30; HR, 0.69 [95% CI, 0.20-2.47]) and who achieved MRD-neg conv at 10–6 (n=1/36 vs n=3/14; HR, 0.23 [95% CI, 0.02-2.27]) but not in pts who achieved MRD-neg conv at 10–5 only (but not at 10–6) or who remained MRD pos (10–5) through the cutoff for this analysis. Sustained ≥12-mo MRD-neg rates at 10−5 were 29.3% for D-R and 7.9% for R (OR, 4.88 [95% CI, 2.09-11.38]; P=0.0001) and at 10−6 were 19.2% and 2.0% (OR, 11.75 [95% CI, 2.65-52.01]; P<0.0001). Among pts who achieved sustained ≥12-mo MRD neg (10–5), 0% (n=0/29) of D-R pts and 25% (n=2/8) of R pts had PFS events. After initial achievement of MRD neg at 10–5 and 10–6, MRD-pos recurrence (prior to PD or subsequent therapy) was observed in 26.7% (n=16/60) and 16.7% (n=6/36) of D-R pts and 30.0% (n=9/30) and 38.5% (n=5/13) of R pts, respectively. In pts who achieved MRD neg at 10–5 only (but not at 10–6), 54.2% (n=13/24) of D-R pts and 41.2% (n=7/17) of R pts experienced MRD-pos recurrence. In pts who experienced MRD-pos recurrence (10–5), PD was observed in 31.3% (n=5/16) of D-R pts and 11.1% (n=1/9) of R pts; of those 5 D-R pts, 3 had cytogenetic high risk at diagnosis per IMS criteria and 4 had MRD recurrence within <7 mo, reflecting the functional high-risk status of these pts. Median time between MRD-pos recurrence and PD/death was 15.9 (range, 5.6-19.2) mo in D-R pts and 13.1 mo for the R pt. One D-R pt and 2 R pts experienced PD (per investigator assessment) or death following MRD-pos recurrence at 10–6. Conclusions:D-R maintenance provided benefit via increased MRD-neg rates, notably at 10–6, as evidenced by a >2.5-fold MRD-neg conv rate and an almost 10-fold sustained MRD-neg rate vs R maintenance alone. Pts achieving MRD-neg conv at 10–6 were less likely to have MRD-pos recurrence than pts who attained MRD-neg conv at 10–5 only (but not at 10–6), and D-R pts were less likely to develop MRD-pos recurrence than R pts. Pts who achieved deeper MRD neg (10–6) or sustained MRD neg (10–5 or 10–6) showed a trend toward improved PFS vs pts with MRD-pos recurrence or who never achieved MRD neg, with few PD events observed in pts with MRD-neg conv. In addition, these data reinforce that sustained MRD neg might be a better predictor of MRD-pos recurrence than MRD-neg results at a single time point, and treatment with D-R may provide better durability compared to R. Finally, the higher MRD-pos recurrence rates observed in AURIGA compared to the PERSEUS study may have been due to the absence of anti-CD38 in induction therapy in AURIGA. Overall, these data highlight the value of D-R maintenance in attaining and deepening MRD neg.