Molecular profiling and kinetics of circulating multiple myeloma cells (CMMCs) predict resistance to bispecific antibodies (BsAbs) in relapsed and/or refractory multiple myeloma (RRMM)

Abstract

Abstract Introduction: Bispecific antibodies (BsAbs) targeting B-cell maturation antigen (BCMA) and G protein coupled-receptor, class C group 5 member D (GPRC5D) elicit deep and rapid responses in relapsed/refractory multiple myeloma (RRMM). Circulating multiple myeloma cells (CMMCs) offer a minimally invasive source for real-time disease monitoring. However, early biomarkers of responses are lacking. We investigated whether CMMC kinetics and molecular features predict treatment response to BsAbs. Methods In this multicenter study, RRMM patients (pts.) receiving standard-of-care BsAbs were enrolled across three institutions. Peripheral blood (PB) samples were obtained at baseline, weekly for 4 weeks, and monthly thereafter until progression. Enumeration of CMMCs was performed using the CELLSEARCH® system Menarini Silicon Biosystems (MSB), Huntington Valley, PA. For molecular profiling, 50-100 CMMCs per patient were flow-sorted and sequenced using a custom targeted NGS panel (Illumina NextSeq 1000 platform) covering recurrent myeloma-related mutations, copy number alterations (CNAs), and IGH translocations. In addition, comprehensive genomic and transcriptomic profiling with targeted DNA/RNA sequencing was performed (in select pts.) using the Menarini Search (MS) assay targeting both cellular and cell-free nucleic acids (cfDNA and RNA) from bone marrow (BM) and matched PB at baseline. CMMC kinetics and molecular profiling were correlated with outcomes. Results A total of 77 pts. were enrolled (median age 70 years, range 49-86), 48% female, 11% black, 26% with International Myeloma Working Group high-risk and 6 median prior lines of therapy (range, 3-16). BsAbs included teclistamab (n= 39), elranatamab (n=10), and talquetamab (n=28). Twenty-three (30%) had prior CAR-T therapy. The best overall response rate was 51% (44% ≥VGPR, 7% PR). The median baseline CMMC count was 93 cells/4 mL (range, 0-36,365). At day +30, 20 of 70 evaluable pts. (29%) achieved complete CMMC clearance, among whom 95% attained ≥VGPR. Among pts. achieving ≥VGPR, the CMMC clearance rate was 45% at day+7, and no pts. achieving ≥VGPR retained detectable CMMCs beyond day +30. Moreover, none of the primary refractory pts. cleared CMMCs at any time point. CMMC resurgence/increase occurred in 80% (8/10) pts. who progressed after initial response. CMMC sequencing was performed on 35 samples from 20 pts. at baseline and serial time points. Common alterations included KRAS (15%), NRAS (5%),TP53 (25%) mutations, 1q gain/1p loss (45%) and t (11;14) (25%), with CNAs and translocations all concordant with available BM FISH. Among pts. with CMMC NGS data, 60% (n=12) harbored high-risk genomics. These pts. showed lower responses (≥VGPR: 33% vs. 100% in standard-risk), higher progression (83%), and increased mortality (55%) at 3 months follow up. Notably, no BCMA or GPRC5D point mutations were detected at any time point. However, monoallelic deletions in 12p (GPRC5D) (N=1) or 16p (BCMA) (N=3) and both (N=2) were detected and associated with heterogeneous outcomes. Two pts. exhibited biallelic 12p and 16p loss at baseline and were refractory to both BCMA- and GPRC5D-directed BsAbs. Seven pts. with matched BM and PB at baseline underwent analysis with MS assay. Clonotypic IgH receptors were characterized through RNA sequencing and showed tight concordance between BM and PB, with recurrent usage of IGHV3-33, IGHV5-51, and IGHV1-46 across pts. and were concordant in all 7/7 cases. Two cases exhibited light chain-only (IGKV1D-33) clonality. cfDNA/RNA analysis demonstrated strong concordance with BM findings, capturing 1q gain,13q deletion, TP53 mutations, and translocations such as t (4;14) with matching NSD2, and MAFB in all 7/7 cases except for t (6;14) which was not detected in PB. Conclusions: CMMC enumeration and molecular profiling offer a minimally invasive, real-time approach to monitor response and resistance to BsAbs in RRMM. Genomic analysis of CMMCs not only mirrored but often outperformed bone marrow cytogenetics, identifying high-risk features—such as biallelic 12p/16p loss—not captured by standard testing. CMMC resurgence preceded clinical progression, supporting its role for serial monitoring for potential early intervention. Given its ability to track clonal evolution and resistance, CMMC profiling may serve as a valuable tool to predict response and early relapse across immune therapies, including BsAbs and CAR-T.