Tumor clearance, T-cell fitness, and minimal residual disease (MRD) outcomes in patients with relapsed/refractory multiple myeloma (RRMM) treated with cevostamab plus pomalidomide and dexamethasone: Biomarker analyses from CAMMA 1 Arm B

Abstract Background Fc receptor-homolog 5 (FcRH5) is a type I membrane protein that is ubiquitously expressed on multiple myeloma cells. Cevostamab, a FcRH5xCD3 T-cell-engaging bispecific antibody, has shown encouraging activity and manageable safety as monotherapy in late-line RRMM (Richter et al. ASH 2024). Combining cevostamab with anti-myeloma agents that augment T-cell activity could enhance efficacy. Pomalidomide (Pom) is an immunomodulatory drug (IMiD) that exhibits T-cell co-stimulatory effects and is clinically active when given alone or with dexamethasone (dex) in RRMM (San Miguel et al. 2013). In the Arm B randomized dose-expansion stage of the Phase Ib CAMMA 1 study (NCT03275103), cevostamab plus Pom-dex induced deep and durable responses and had manageable safety in patients (pts) with RRMM (Mian et al. IMS 2025). We report exploratory biomarker analyses from Arm B that aimed to characterize tumor kinetics, MRD, and immune correlates of response. Methods In the Arm B randomized dose-expansion cohorts, eligible pts had RRMM and had received ≥1 prior IMiD and ≥1 prior proteasome inhibitor as part of ≥1 prior line. Pts were randomized (1:1) to receive 70mg (low-dose [LD]) or 105mg (high-dose [HD]) cevostamab given Q2W in Cycles (C) 1–6 (28-day cycle) and Q4W in C7+, after a pre-phase with double or triple step-up dosing. Pom (2mg Day [D] 1–21) and dex (20mg D1, 8, 15, and 22) were given from C1+. Treatment continued until disease progression or unacceptable toxicity. Soluble B-cell maturation antigen (sBCMA), a surrogate biomarker for tumor burden, and cytokines were measured in plasma using LC/MS and multiplex ELISA assays on ELLA platform. MRD was assessed using Adaptive clonoSEQ® (sensitivity: 10−4 to 10−6). Immunophenotyping was carried out using multiparameter conventional and spectral flow. Given the negative impact of prior exposure to BCMA-targeted agents on outcomes (Richter et al. ASH 2024), analyses were conducted in BCMA-naïve pts only. Results As of June 11, 2025, 64 pts were randomized and treated in the Arm B dose-expansion cohorts (LD, n=32; HD, n=32). Of these, 29 pts in the LD group and 25 pts in the HD group were BCMA naïve (median age: 65 years in both groups [range: LD, 40–76; HD, 48–80]; male: 48.3% and 56.0%, respectively). Among these pts, median number of prior lines of therapy was 2 [range: 1–6] in both groups and the majority were refractory to their last line of therapy (LD, 72.4%; HD, 68.0%). At data cut-off, median time on study in BCMA-naïve pts was 10.4 months (range: 6.3–15.5) in the LD group and 8.3 months (range: 3.3–15.7) in the HD group. The objective response rate (ORR) and very good partial response (VGPR) or better rate were 86.2% and 72.4% in the LD group and 88.0% and 76.0% in the HD group. At cut-off, 84.0% of BCMA-naïve responders in the LD group and 86.4% in the HD group remained in response, with deepening responses observed over time. Baseline sBCMA levels and effector-to-target ratios in BCMA-naïve pts were comparable in the LD and HD groups. CD8⁺ T-cell subset distribution and activation markers also showed no difference between groups, while heatmap clustering confirmed immunological comparability for T-cell fitness. In BCMA-naïve pts, cevostamab plus pom-dex induced rapid, deep, and durable declines in sBCMA levels. Median sBCMA decreases were 54.9% by C1D1 pre-infusion (after the first target dose in the cevostamab pre-phase), 83.0% by C1D15 and 90.9% by C2D1 pre-infusion. Notably, no significant difference in sBCMA kinetics or the depth of sBCMA depletion was observed between the LD and HD groups. In the BCMA-naïve group, 25 pts achieved a complete response or better (CR+) and 20 pts were evaluable for MRD assessment. Of these, 14 pts achieved best MRD negativity at the 10−6 threshold level, 3 pts at the 10−5 threshold and 1 patient at the 10−4 threshold. Two pts had calibration failure at screening, and 5 pt samples were not reconciled at cut-off. There was no significant difference for CR+MRD-negativity rate between the LD (9/13, 69.2%) and HD (8/12, 66.7%) groups at the 10−5 threshold in CR+ pts.Conclusions Cevostamab plus Pom-dex induces high ORR and VGPR or better rates, and durable remissions, in BCMA-naïve pts with RRMM. Combination treatment is also associated with deep and durable declines in sBCMA levels and high rates of CR+MRD-negativity. Updated biomarker data will be presented, including baseline immune fitness and correlation with MRD negativity.