Kinetics of circulating multiple myeloma cells (CMMCs) in patients undergoing standard-of-care anti-BCMA chimeric antigen receptor T cell therapy for Relapsed/Refractory multiple myeloma

Abstract Chimeric antigen receptor (CAR) T cell therapies against B-cell Maturation Antigen (BCMA) are a highly effective therapy in patients with relapsed/refractory multiple myeloma. However, assessing response in the early period after cell infusion is challenging. Serologic response is often delayed and serial marrow assessment is challenging. Consequently, other biomarkers are of interest. The presence of circulating multiple myeloma cells (CMMCs) has been recognized as an important prognostic marker. Recent advances in cell preservation/isolation have made enumeration and analysis of CMMCs feasible. The presence of CMMCs may serve as a useful biomarker of disease response for CAR T cell therapies. In this report, we describe the kinetics of CMMCs in the post-infusion period (up to day 100) following anti-BCMA CAR T cell therapy, their association with outcomes and the results of downstream CMMC analysis in a subset of patients. This was a prospective cohort study of patients undergoing standard-of-care (SOC) anti-BCMA CAR T cell therapy at Washington University in Saint Louis. Sample collection was performed under our institutional myeloma tissue bank with an associated specific-use protocol for this study (IRB: 202401004). Samples (4.5 mL) were collected prior to lymphodepletion (LDC) and at days 0, 7, 14, 30, 100, 180 and 365 in CellRescue tubes (Menarini Silicon Biosystems (MSB), Huntingdon Valley, PA). Samples were shipped to MSB in real time for CMMC isolation using the CellSearch system. CMMCs were defined as CD38+/CD138+/DAPI+/CD45-/CD19-. In select samples with CMMC > 100, next-generation sequencing (NGS) was performed using a custom MSB panel for myeloma-relevant mutations, copy number alterations, and translocations, which was compared to pre-CAR T SOC fluorescent in situ hybridization (FISH) testing on pre-CAR T bone marrow biopsies. NGS was performed on the Illumina NextSeq 1000 platform. MSB performed all analysis and returned results without access to clinical data. 48 patients were enrolled between April 2024 and July 2025. The median age was 68, 94% were White and 60% were male. The median time from diagnosis was 5.6 years (range: 0.7 – 20.3) and the median line of therapy was 3 (range: 1 – 8). 45 received cilta-cel and 3 ide-cel. The median follow up in the cohort was 148 days (range: 6 – 427). At day +100, the overall response rate was 97% and 62% achieved a complete response or better (≥CR). MRD was assessable in 68% and 60% of patients at day 30 and 100. Of these, 97% and 96% were MRD-negative at <10-6. One patient relapsed during follow up (day 91) with extramedullary disease. Two patients died in remission during follow up from HLH (day 68) and metastatic breast cancer (213). The other 45 patients (94%) remain in remission at last follow up. The median pre-LDC CMMC was 22/4 mL(range: 0 – 13265). The median CMMC at day 0 was 4 (range: 0 – 8861). 76%, 65%, 12%, 10% and 14% of patients had detectable CMMCs at day 0, 7, 14, 30 and 100, respectively. 29% (12/41) of patients had an increase in their CMMC from day 0 to 7. Only 1 patient had detectable CMMCs throughout day 0 – 30 and no patients had detectable CMMCs at all time points. The patient who relapsed on study cleared their CMMCs on day 14 but had detectable CMMCs at progression (8). 7 CMMC samples from 4 patients were analyzed via NGS at pre-pheresis (1), pre-LDC (4) or day +7 (2) time points (median CMMCs: 2790, NGS input: 100 CMMCs). No patients had genomic loss or mutation of BCMA prior to infusion. Day +7 CMMC NGS showed no evidence of early clonal selection. Compared to SOC FISH, CMMC NGS captured 11/12 cytogenetic changes (gain(1q): 4, del(13): 4, del(17p): 2, del(16q): 1) and identified actionable genomic features in 3 patients (GPRC5D loss due to del(12p) (2) and KRAS G12D). Data from this prospective cohort demonstrates that CMMCs are present in most patients at CAR T infusion and CMMC-based NGS is feasible, recapitulating the majority of marrow-based cytogenetic changes and identifiable clinically actionable alterations in 75% of tested patients. CMMCs decreased rapidly during CAR T cell expansion. Due to excellent outcomes in this cohort, further follow up is needed to assess if baseline CMMC levels or CMMC kinetics are risk factors for treatment failure with CAR T therapy. The utility of CMMCs as a method of early relapse detection requires samples from later time points and sample collection is ongoing.