Skip to main content
IMS Events Main banner

    MRD and Biomarkers

    (OA-26) Comparative Analysis of ClonoSEQ MRD Testing from Bone Marrow and Circulating Tumor DNA Assessment in Multiple Myeloma

    Saturday, September 20, 2025
    08:00 - 08:10 East Coast USA Time
    Location: Room 718
    Introduction: Multiple myeloma (MM) is the second most common hematologic malignancy in the U.S. Non-secretory MM, a rare subtype, is characterized by the absence of a detectable monoclonal immunoglobulin protein (M-protein) in the serum and/or urine, making disease monitoring particularly challenging. Serial bone marrow assessments are the gold standard for non-secretory MM but are invasive and associated with patient discomfort. clonoSEQ is a next-generation sequencing (NGS) assay that measures minimal residual disease (MRD) by detecting unique rearranged immunoglobulin receptor sequences from DNA in bone marrow (BM MRD); its use in peripheral blood (PB MRD) could offer a more patient-friendly alternative, particularly valuable in non-secretory MM. 

    Methods: We retrospectively analyzed 149 PB plasma samples from 100 patients with secretory MM who had a known dominant clonotype identified in the BM. We evaluated the concordance of PB MRD from ctDNA with BM MRD and other conventional disease biomarkers, including M-protein, free light chain (FLC) ratio, extramedullary disease (EMD), and treatment response (based on the International Myeloma Working Group criteria). Specificity (SP), sensitivity (SN), positive predictive value (PPV), and negative predictive value (NPV) were calculated for PB MRD in various clinical contexts.  

    Results: PB MRD negativity was significantly associated with deep clinical response, including complete response (CR) (p< 0.001) and very good partial response (VGPR) (p< 0.001).  PB MRD positivity was associated with abnormal FLC ratio (74%; 37/50), detectable M-protein (78%; 39/50 and EMD (26%; 14/50). Using BM MRD as the reference, PB MRD+ had a SP of 90.1% and PPV of 94.9%, but lower SN (33.9%) and NPV (21.7%) [TABLE]. Compared to M-protein detection, PB MRD+ had 83.3% SP, 78.0% PPV, 49.4% SN, and 57.9% NPV.  When compared to abnormal FLC ratio, PB MRD+ showed 82.2% SP, 74.0% PPV, 56.9% SN, and 68.2% NPV.  For patients in VGPR or better, PB MRD- had 84.0% SP, 93.6% PPV, 80.6% SN, and 60.0% NPV. Most samples in CR were PB MRD– (83/93; 89.3%), while in relapsed (19/19) and nearly all newly diagnosed samples (8/9) were PB MRD +. Among 9 patients with ≥ 3 sequential samples. all samples at diagnosis or with progressive disease were PB MRD+, and all samples in a CR were PB MRD-. 


    Conclusions: clonoSEQ PB MRD from ctDNA demonstrates high specificity and PPV relative to BM MRD, with strong correlation to markers of tumor burden and response status. PB MRD+ reliably indicates residual disease, while PB MRD– correlates with deep clinical responses. These findings highlight the potential of PB MRD as a non-invasive monitoring tool in MM, and suggest promising clinical utility, particularly for patients with non-secretory disease. BM MRD remains the more sensitive tool for evaluating disease burden, however, further studies are warranted to validate PB MRD as a surrogate for BM-based assessment in clinical practice.