Mechanisms of Resistance and Relapse with Talquetamab in Patients with Relapsed/Refractory Multiple Myeloma from the Phase 1/2 MonumenTAL-1 Study

Introduction: Talquetamab (tal), a novel GPRC5D×CD3 bispecific antibody (BsAb), has shown deep and durable responses with overall response rates (ORRs) 71%, including in high-risk populations, and a clinically manageable safety profile in patients (pts) with relapsed/refractory multiple myeloma (MM) in the MonumenTAL-1 study (NCT03399799/NCT04634552). To better understand mechanisms of resistance and relapse in pts treated with tal, baseline and longitudinal GPRC5D expression and immune profiles were assessed in pts treated with the 0.4 mg/kg weekly (QW) recommended phase 2 dose, including a separate cohort of pts who received prior T-cell redirection therapy (prior TCR; chimeric antigen receptor CAR-T cell and BsAb therapies). Methods: MonumenTAL-1 is a first-in-human, phase 1/2, single-arm study of tal monotherapy. Eligible pts had MM and were intolerant to or progressed on established therapies (phase 1) or had ≥3 prior lines of therapy (LOT) including a proteasome inhibitor, immunomodulatory drug, and anti-CD38 antibody (phase 2). Analyses were performed in pts receiving tal 0.4 mg/kg QW who did not receive prior TCR (no prior TCR cohort) and pts receiving tal 0.4 mg/kg QW who received prior TCR (prior TCR cohort). Baseline and on-treatment bone marrow or whole blood were analyzed by flow cytometry for GPRC5D expression and immune cell profiles, respectively. For mechanisms of resistance, samples were analyzed in responders (defined as ≥PR) and nonresponders (defined as stable or progressive disease PD); for mechanisms of relapse, samples were analyzed at baseline and at end of treatment (EOT)/PD. Results: Median prior LOT was 6 and 5 in the prior and no prior TCR cohorts, respectively. T-cell counts at baseline were similar between cohorts, whereas pts in the prior TCR cohort had higher expression of single coinhibitory markers (LAG-3, TIM-3, PD-1), dual coinhibitory markers (PD-1/LAG-3 and PD-1/TIM-3), and activation markers (CD38 and CD25) on CD8+ T cells, along with a higher frequency of immunosuppressive regulatory T cells (Tregs) compared with the no prior TCR cohort. GPRC5D was highly expressed on MM cells at baseline; however, expression was variable among pts in both cohorts and was not associated with clinical response. In both cohorts, correlative analyses with response showed that nonresponders vs responders had a more exhausted peripheral immune phenotype at baseline, indicated by lower T-cell counts and higher frequencies of Tregs and expression of coinhibitory markers (LAG-3, TIM-3) on CD8+ T cells. In the first cycle following tal administration, greater and transient activation of T cells was observed in responders vs nonresponders. In addition, greater recovery of CD3+ T cells was observed and sustained in responders within 2 cycles of treatment, whereas persistence of CD38+ expression on CD8+ T cells, higher presence of Tregs, and sustained expression of LAG-3, TIM-3, and PD-1 on CD8+ T cells was observed in nonresponders, consistent with the baseline correlative analysis in the prior and no prior TCR cohorts (Figure). Finally, analysis of progression samples of pts who initially responded and then progressed with tal showed a higher expression of coinhibitory receptors (LAG-3, TIM-3, CD38, PD-1/LAG-3, PD-1/TIM-3) on peripheral CD4+ or CD8+ T cells in both cohorts compared with baseline, indicative of an exhausted T-cell phenotype at relapse. Overall, these correlative data were comparable between prior TCR and no prior TCR cohorts. Conclusions: Baseline immune profiling in heavily pretreated pts from MonumenTAL-1 suggested that compared to pts with prior TCR, pts with no prior TCR had a favorable immune fitness profile (less T-cell dysfunction and immune suppression). In both cohorts, baseline and longitudinal correlative analyses suggest a mechanism of resistance for nonresponders including lower T-cell counts and higher frequency of Tregs and expression of coinhibitory markers on CD8+ T cells, whereas responders show greater T-cell activation and recovery of CD3+ T cells that was sustained longitudinally. Progression data indicated an exhausted T-cell phenotype for pts who relapsed on tal. Approaches to overcome resistance, including combination trials with anti-CD38 monoclonal antibodies (eg, daratumumab) and PD-1 inhibitors as well as alternative dosing regimens, are currently being studied.