P893: comparative effectiveness of talquetamab vs physician’s choice therapy in triple-class exposed patients with relapsed/refractory multiple myeloma: a matched cohort analysis

Topic: 14. Myeloma and other monoclonal gammopathies - Clinical Background: Talquetamab is a first-in-class G protein coupled receptor family C group 5 member D × CD3 bispecific antibody and has shown an overall response rate (ORR) of ≥73% in patients (pts) with triple-class exposed (TCE) relapsed/refractory multiple myeloma (RRMM) in the phase 1/2 MonumenTAL-1 (NCT03399799/NCT04636552) trial. Given the absence of a control arm in MonumenTAL-1, indirect adjusted comparisons can be made to determine the comparative effectiveness of talquetamab vs real-world treatments. Aims: To compare the effectiveness of talquetamab vs physician’s choice (PC) therapy in pts with TCE RRMM. Methods: Individual pt-level data from MonumenTAL-1 were included from pts treated with subcutaneous talquetamab at 0.4 mg/kg QW or 0.8 mg/kg Q2W with a data cut-off (DCO) of Sept 2022. An external control arm for MonumenTAL-1 was created from pts who were treated with PC as subsequent therapy after discontinuing study treatment from 4 daratumumab trials (CASTOR [NCT02136134]; DCO: June 2021, POLLUX [NCT02076009]; DCO: Sept 2021, EQUULEUS [NCT01998971]; DCO: 2017–2019, depending on the trial arm, and APOLLO [NCT03180736]; DCO: July 2020) and who met the key eligibility criteria for MonumenTAL-1 (N=449 with 770 eligible subsequent lines of therapy [LOT]). In the base model, baseline characteristics of prognostic variables (refractory status, cytogenetic risk, ISS stage, extramedullary disease, time to disease progression on prior LOT, number of prior LOT, time since diagnosis, age, and hemoglobin) were adjusted using inverse probability of treatment weighting. A full model also adjusted for prior stem cell transplant, ECOG performance status, race, sex, and MM type. Outcomes of interest were ORR, very good partial response or better (≥VGPR), overall survival (OS), progression-free survival (PFS), and time to next treatment (TTNT). For binary outcomes, a Cox regression model with a constant in the time variable was used to derive response ratios (RRs) and corresponding 95% CIs. For time-to-event outcomes, a weighted Cox proportional hazards model was used to derive hazard ratios (HRs) and 95% CIs. Sensitivity analyses assessed the impact of alternative statistical methods and variable adjustment. Results: Of 236 unique regimens observed in the PC cohort, the most common were carfilzomib (K) + dexamethasone (d); pomalidomide (P)d; cyclophosphamide (cyclo)Pd; lenalidomide (R)d; and KRd. After reweighting, baseline variables were comparable across pt cohorts. In the base case analysis, pts treated with either talquetamab 0.4 mg/kg QW (N=143) or talquetamab 0.8 mg/kg Q2W (N=145) showed superior outcomes vs PC (see Table). For comparisons of both talquetamab QW and Q2W schedules vs PC, the fully adjusted models and sensitivity analyses remained in favor of talquetamab. Summary/Conclusion: Both talquetamab QW and Q2W schedules showed superior effectiveness vs PC therapy in long-term follow-up data from 4 clinical trials for all measured outcomes. These results further support the clinical benefit of talquetamab as an effective, novel treatment option in pts with TCE RRMM.Keywords: Multiple myeloma, Bispecific, G-protein-coupled receptors, Real world data