Scott, David S.

7546 Background: MRD assessment informs relapse risk in MM but is typically bone marrow-based and invasive. We therefore compared two minimally invasive MRD assays: BM-informed cfDNA whole-genome sequencing (cfWGS) and plasma proteomic MRD (EasyM). Methods: 71 longitudinal plasma samples (22 diagnosis, 49 follow-up) from 22 newly diagnosed MM patients across 8 Canadian sites (TFRIM4) had paired cfWGS and EasyM (Rapid Novor Inc). cfWGS MRD used patient-specific mutation lists from diagnostic BM WGS (Abelman et al., medRxiv 2025). EasyM used mass spectrometry to quantify residual M-protein from baseline. The primary endpoint was progression-free survival (PFS). Results: Among 22 MM patients (8 male, 14 female; median age at diagnosis 62 years), 8 progressed at a median follow-up of 52.6 months (range: 10-79 months). At 1-year maintenance (n = 18 available), cfWGS MRD+ (6/18) predicted inferior PFS (HR 20.12, 95% CI 2.23-181.60, p = 0.007). EasyM any-detect positivity was frequent (16/18, reflecting delayed clearance) and not prognostic (p = 0.99), with poor agreement vs cfWGS (44.4%, κ = 0.118). An exploratory EasyM clearance threshold of 0.93% (hypothesis-generating) classified 5/18 patients as MRD+ and stratified PFS (HR 8.78, 95% CI 1.55-49.82, p = 0.004), improving agreement with cfWGS (17/18, 94.4%, κ = 0.870). Progression clustered in cfWGS+/EasyM-high patients (EasyM >0.93%; 4/5; median 11.9 months post-collection [1.9–22.8]) and occurred once in cfWGS−/EasyM-low patients (1/12, 27.6 months). The only discordant case (cfWGS+/EasyM−) progressed 8.7 months after collection. Higher quantitative MRD burden was associated with inferior PFS (cfWGS: HR 2.92 per 1 SD, 95% CI 1.29-6.59, p = 0.01; EasyM: HR 5.48 per 1 SD, 95% CI 0.79-37.90, p = 0.085). At ~100 days post-autologous stem cell transplantation (ASCT; n = 16 available), EasyM was uniformly MRD+ (16/16) while cfWGS was MRD+ in 7/16 patients. Applying an EasyM clearance threshold (2.13%; 9 cleared vs 7 residual) improved agreement with cfWGS (14/16; 87.5%; κ = 0.746). Of two discordant cases, one cfWGS+/EasyM-cleared patient relapsed 38.6 months post-collection and one cfWGS−/EasyM-residual patient remained progression-free at last follow-up. At post-ASCT, both assays showed consistent but underpowered associations with PFS (cfWGS HR 6.32, 95% CI 0.70–56.76, p = 0.100; EasyM-clearance HR 2.36, 95% CI 0.39–14.18, p = 0.335). Across all post-treatment samples (n = 49), continuous cfWGS burden correlated with EasyM residual M-protein (rho = 0.535, p < 0.001). Conclusions: BM-informed cfWGS MRD strongly stratified PFS, whereas EasyM required a clearance definition due to near-universal early positivity. Concordant plasma burdens support complementary MRD information. Future work will validate these findings in larger cohorts and quantify the incremental value of combined cfWGS and proteomic MRD for early relapse prediction.

Abstract Introduction: Early relapse limits durable remissions in multiple myeloma (MM). Cell-free DNA (cfDNA) fragment patterns capture tumor burden and nucleosome positioning. We tested whether a fragmentomics signature trained at MRD time points stratifies progression-free survival (PFS) when applied to pretreatment plasma. Methods: We performed 30-40× whole-genome sequencing of plasma cfDNA from 43 newly diagnosed MM patients, with 65 additional MRD timepoint samples from 41 of those patients (post-ASCT n=30; 1-year maintenance n=30; other maintenance n=5) across eight Canadian sites (TFRIM4 and IMMAGINE studies) and one U.S. site (Mayo SPORE). MRD testing used multiparameter flow cytometry for all follow-up samples (n=65) and clonoSEQ in a subset (n=31). Logistic and elastic-net models were trained at MRD timepoints on fragmentomics features (short-fragment burden; a combined model integrating fragment-length distributions with inferred nucleosome positioning at MM-specific chromatin accessibility regions [Ordoñez et al., Genome Res. 2020]). A decision threshold was fixed by the Youden index during training and applied unchanged to diagnosis samples. PFS was analyzed by Kaplan-Meier, log-rank, and Cox models. Results: After a median follow-up of 49.1 months, 16 of 43 patients (37%) progressed. Using the prespecified locked threshold, patients classified as MRD+ by the proportion-of-short-fragments model had significantly shorter PFS than MRD- patients (HR = 5.12; 95% CI 1.16-22.57; log-rank p = 0.016), with median PFS of 49.2 months for MRD+ cases versus not reached for the MRD- group. Fixed-horizon PFS favored MRD- samples at 24/36/48 months (100/94/87% vs 85/81/53% for MRD+). A combined model integrating weighted fragment lengths with inferred nucleosome positioning also separated outcomes (log-rank p=0.054); the Cox HR was not estimable because no MRD- patients relapsed. At 24, 36, and 48 months, baseline MRD positivity by the short-fragment model identified 100%, 83%, and 86% of patients who relapsed within those intervals (specificity 44%, 43%, and 52%), while the combined model detected all early relapses across the same horizons (specificity 15-21%), indicating consistently high sensitivity but modest specificity for predicting early progression. Conclusions: A cfDNA fragmentomics signature trained at MRD time points stratified PFS when applied to diagnostic plasma, identifying patients at risk of early relapse. Short-fragment burden was most discriminative, and a combined model with nucleosome positioning recapitulated it with higher sensitivity, albeit lower specificity. These data support cfDNA fragmentomics as a practical liquid-biopsy risk biomarker in MM and suggest higher signals may proxy aggressive disease with greater tumor shedding and chromatin disorganization. Future work will validate findings in larger cohorts and integrate cytogenetic and immune features. Citation Format: Dor David Abelman, Jenna Eagles, Aimee Wong, Saumil Shah, Jeffrey Bruce, Stephanie Pedersen, David S. Scott, Cecilia Bonolo de Campos, Signy Chow, Darrell White, Irwindeed Sandhu, Kevin Song, Esteban Braggio, Shaji Kunnathu Kumar, Alli Murugesan, Tony Reiman, A. Keith Stewart, Suzanne Trudel, Trevor J. Pugh. Baseline cfDNA fragmentomics identifies multiple myeloma patients at risk of early progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2597.

Abstract In this study, we investigated tumor intrinsic mechanisms of resistance to cevostamab, a bispecific T-cell engager targeting Fc receptor-like protein 5 (FCRL5) and CD3 that has shown promising clinical activity and durable responses in advanced MM. Longitudinal bone marrow aspirates from nine MM patients treated with cevostamab were analyzed. All patients had previously been treated with both immunomodulatory drugs (IMiDs) and proteasome inhibitors (PI) and eight patients had received daratumumab prior to cevostamab, which was, on average, the 6th line of treatment (mean treatment duration 220±273 days). Five of nine patients responded with ≥PR. Whole genome and transcriptome sequencing was performed on sorted CD138+ MM cells at baseline, and whole exome and transcriptome sequencing was performed on five and three samples, respectively, following cevostamab discontinuation. Baseline FCRL5 expression levels did not correlate with clinical response. However, in one non-responder patient, FCRL5 expression increased four-fold after 28 days of treatment. Given the location of the FCRL5 gene on chromosome 1q—commonly amplified in advanced MM—we evaluated the presence of Gain(1q) in our study cohort. All patients whose best response was ≤PR (n=6) harbored a Gain(1q), whereas it was absent in those who achieved VGPR and sCR (n=3). Crucially, one of the patients that achieved sCR was initially treated with 60 mg of cevostamab for one year and then escalated to a higher dose cohort (90 mg) in the second year. At progression, a clonal FCRL5E115A missense mutation (allelic frequency = 56%) and a subclonal FCRL5G455Efs*19(allelic frequency = 5%) frameshift deletion mutation were detected, both absent at baseline. The missense mutation was located on the extracellular domain of FCRL5 targeted by cevostamab. To validate the functional impact of FCRL5E115Aand FCRL5G455Efs*19 on the sensitivity to cevostamab in vitro, we generated a FCRL5 knockout (KO) model utilizing CRISPR-Cas9 technology in OPM2 MM cells. In vitro cevostamab cytotoxicity assay with PBMCs and pan T cells from healthy donors at various effector to target (E:T) ratios confirmed a complete loss of cevostamab sensitivity in our FCRL5KO models. cDNA encoding patient-derived FCRL5 alterations (FCRL5E115Aand FCRL5G455Efs*19) as well as FCRL5wtwere cloned into mammalian expression vector and reintroduced into FCRL5KO cells. Cevostamab sensitivity was reinstated by FCRL5wt expression in KO cells. In contrast, FCRL5E115A conferred complete resistance to cevostamab (p&lt;0,001) and FCRL5G455Efs*19 depicted a significant decline in cevostamab sensitivity relative to FCRL5wt expressing models. Additional experiment utilizing direct stochastic optical reconstruction microscopy (dSTORM) are ongoing to monitor the membrane localization of FCRL5 mutants. By integrating genomic, transcriptomic and functional data with clinical outcomes, this work is the first to report on acquired FCRL5 mutations causing resistance to cevostamab. We hypothesize that the prolonged treatment duration may have contributed to the acquisition of the target mutations and note a subsequent durable sCR response to BCMA-targeted immunotherapy in the same patient, suggesting immune-related factors such as T cell exhaustion, were not drivers of cevostamab resistance. Due to the limited sample size, these findings require validation in larger cohorts. Nonetheless, our findings begin to elucidate tumor intrinsic mechanisms associated with cevostamab resistance.