No abstract

IMPORTANCETreatment of newly diagnosed multiple myeloma (NDMM) with a quadruplet regimen consisting of a monoclonal antibody, proteasome inhibitor, immunomodulatory imide, and corticosteroid has been associated with improved progression-free survival (PFS) compared with triplet regimens. The optimal quadruplet combination, and whether this obviates the need for frontline autologous stem cell transplant (ASCT), remains unknown. We evaluated elotuzumab and weekly carfilzomib, lenalidomide, and dexamethasone (Elo-KRd) without ASCT in NDMM.OBJECTIVE To investigate the efficacy of Elo-KRd using a measurable residual disease (MRD)-adapted design in NDMM regardless of ASCT eligibility. DESIGN, SETTING, AND PARTICIPANTSThis multicenter, single-arm, phase 2 study enrolled patients between July 2017 and February 2021. Median follow-up was 29 months.INTERVENTIONS Twelve to 24 cycles of Elo-KRd; consecutive MRD-negative results at 10 −6 by next-generation sequencing (NGS) after cycles 8 (C8) and 12 determined the duration of Elo-KRd. This was followed by Elo-Rd (no carfilzomib) maintenance therapy until disease progression. MAIN OUTCOMES AND MEASURESThe primary end point was the rate of stringent complete response (sCR) and/or MRD-negativity (10 −5 ) after C8 Elo-KRd. Secondary end points included safety, rate of response, MRD status, PFS, and overall survival (OS). As an exploratory analysis, MRD was assessed using liquid chromatography mass spectrometry (MS) on peripheral blood samples. RESULTS Forty-six patients were enrolled (median age 62 years, 11 [24%] aged >70 years). Overall, 32 (70%) were White, 6 (13%) were Black, 3 (6%) were more than 1 race, and 5 (11%) were of unknown race. Thirty-three (72%) were men and 13 (28%) were women. High-risk cytogenetic abnormalities were present in 22 (48%) patients. The rate of sCR and/or MRD-negativity after C8 was 26 of 45 (58%), meeting the predefined statistical threshold for efficacy. Responses deepened over time, with the MRD-negativity (10 −5 ) rate increasing to 70% and MS-negativity rate increasing to 65%; concordance between MRD by NGS and MS increased over time. The most common (>10%) grade 3 or 4 adverse events were lung and nonpulmonary infections (13% and 11%, respectively). There was 1 grade 5 myocardial infarction. The estimated 3-year PFS was 72% overall and 92% for patients with MRD-negativity (10 −5 ) at C8. CONCLUSIONS AND RELEVANCEAn MRD-adapted design using elotuzumab and weekly KRd without ASCT showed a high rate of sCR and/or MRD-negativity and durable responses. This approach provides support for further evaluation of MRD-guided deescalation of therapy to decrease treatment exposure while sustaining deep responses.

8513 Background: MRD-negativity in multiple myeloma (MM) assessed by NGS in bone marrow (BM) aspirate is associated with longer progression free survival (PFS) and overall survival. MS can detect monoclonal protein at a heightened sensitivity in peripheral blood (PB). We sought to assess the concordance of MS in PB and NGS in BM, comparing outcomes by MRD status. Methods: MRD was tested on paired PB and BM samples from transplant (ASCT)-eligible pts with newly diagnosed secretory MM who received treatment on a phase II clinical trial (NCT01816971) with KRd for 4 cycles, ASCT, KRd for 14 cycles, and lenalidomide maintenance (LM). Both NGS and MS were evaluable in 36 pts after a total of 18 cycles of KRd (C18) and in 24 pts after 1 year of LM. MS signatures were identified in pretreatment PB samples. C18 and after 1 year of LM PB samples were evaluated using both MALDI-TOF and liquid-chromatography-MS (LCMS) by the Binding Site Group. Paired MRD by NGS was performed by ClonoSEQ. 20/60 samples reached the limit of detection for 10−6 and 40/60 for 10−5. Results: There was substantial concordance between NGS and MALDI-TOF among the 60 samples ( κ= 0.667, 83% agreement) and fair concordance between NGS and LCMS ( κ= 0.348, 63% agreement). However, all 22 discordant samples (8 with NGS depth 10−6, 14 with NGS depth 10−5) were NGS−/LCMS+. 4/16 (25%) of these pts converted to NGS+, and 3/16 (19%) clinically progressed. There was stronger concordance between LCMS and NGS 10−6( κ= 0.615) than with NGS 10−5( κ= 0.375). At a median follow-up of 56 months, C18 LCMS−(n = 9) was associated with superior PFS vs all LCMS+(n = 27; p = 0.03) and independently vs NGS—/LCMS+ (n = 14; p = 0.04). There were 10 events (including 4 deaths) in the C18 LCMS+ group vs 0 in the LCMS− group. Conclusions: MRD assessment by LCMS in PB appears to reach and possibly exceed the sensitivity of MRD by NGS in BM at a depth of 10−5-10−6. LCMS positivity predicted conversion from NGS— to NGS+ in 25% of discordant cases, and LCMS negativity was a better predictor of superior PFS than MRD negativity by NGS. These observations need confirmation in larger prospective studies.

8011 Background: The addition of a monoclonal antibody to triplet induction regimens in patients (pts) with MM with intent for autologous stem cell transplant (ASCT) has resulted in higher overall and deep response rates. In this study we are investigating the impact of the addition of Elo to KRd on complete response (CR) and/or MRD-negative rates in newly diagnosed MM regardless of transplant eligibility. Methods: Pts were enrolled from four MM Research Consortium sites into this phase 2 study. All patients receive 12 cycles of Elo-KRd in 28-day cycles: Elo per standard dosing, K 20/56/70 mg/m2 days 1, 8 and 15, R 25 mg days 1-21, and dexamethasone 40 mg days 1, 8, 15, 22. ASCT eligible candidates can undergo stem cell collection after cycle 4 and then resume treatment; pts who elect to proceed to ASCT are censored for response at that time. Pts MRD(-) (<10-5) by NGS after cycles 8 (C8) and 12 (C12) proceed to Elo-Rd until progression. Patients who convert from MRD(+) to MRD(-) between C8 and C12 receive an additional 6 cycles of Elo-KRd (total 18 cycles) followed by Elo-Rd, and pts MRD(+) after C12 receive an additional 12 cycles of Elo-KRd (total 24) followed by Elo-Rd. The primary endpoint of the study is sCR and/or MRD(-) rate after C8 E-KRd. MRD status was determined by ClonoSEQ next generation sequencing (NGS, <10-5) [Adaptive Biotechnologies]. An improvement in the sCR and/or MRD(-) rate by NGS from a historical 30% to 50% at the end of C8 will be considered promising. Results: 44 pts are enrolled, 39 of whom are evaluable for response (cutoff Jan 10 2021). Median age is 62 years (range 43-81, 23% age >70) and 23 (52%) have high-risk cytogenetic abnormalities (HRCA) including 13 (30%) with >2 high-risk abnormalities (6 pts unknown cytogenetics). 34/39 (87%) have MRD trackable by clonoSEQ. The rate of sCR and/or MRD(-) by NGS at the end of C8 is 19/33 (58%), meeting the statistical threshold for establishing efficacy (2 pts censored for elective ASCT before C8 and 4 pts receiving therapy but have not reached C8). With a median follow-up of 24 months, estimated 2-year progression free survival is 87% (100% for standard risk, 79% for HRCA) and estimated 2-year overall survival is 89% (82% for HRCA). No pt who was MRD(-) by NGS after C8 has progressed, including 6 pts with HRCA. Serious adverse events occurred in 30 pts (68%). 89% experienced treatment emergent AEs, the most common (>10%) of which was pneumonia (14%). One pt had grade 5 myocardial infarction. Conclusions: Elo-KRd demonstrates tolerability consistent with known toxicities of these agents and met the primary endpoint of sCR and/or MRD(-) of >50% after 8 cycles. With longer follow-up, the study results may validate that an MRD-adaptive design for de-escalation of therapy in MM can generate deep responses while reducing treatment exposure. Clinical trial information: NCT02969837.

We performed a phase I study of weekly selinexor, carfilzomib, and dexamethasone (wSKd) in patients with relapsed/refractory multiple myeloma (MM). The primary objective was to identify the maximum tolerated dose (MTD) of wSKd.Secondary endpoints included overall response rate (ORR), progression-free survival (PFS), and overall survival (OS). Prior exposure/refractoriness to carfilzomib was permitted. Thirty patients were enrolled; 26 (87%) had triple-class exposed disease and 6 (20%) received chimeric antigen receptor (CAR) T-cell therapy. Dose level 2 (carfilzomib 70 mg/m 2 Intravenous [IV] on