Exposure-response (ER) analyses of efficacy and safety of trastuzumab deruxtecan (T-DXd) to inform dosing recommendation in HER2-mutant non-small cell lung cancer (NSCLC).

Abstract

e20545 Background: T-DXd, a novel human epidermal growth factor receptor 2 (HER2) targeting antibody drug conjugate, was evaluated at doses of 5.4 and 6.4 mg/kg every three weeks in Phase 1 or 2 trials in HER2 mutant (HER2m) NSCLC subjects, including Destiny-Lung01 (DL-01) and DL-02 trials1,2. ER analyses of key efficacy and safety endpoints were conducted to support the dosing recommendation of T-DXd in HER2m NSCLC. Methods: Exposure-efficacy (EE) analysis was performed using confirmed objective response (cORR) data from HER2m NSCLC in Phase 1-2 trials (N = 181). The dataset included primary analysis data cut from DL-011 and interim analysis data cut from DL-02. Exposure-safety (ES) analyses of T-DXd were performed, using pooled data from 11 Phase 1-3 trials (N = 1850) across tumor types, including DL-01 and DL-02, for safety endpoints of adverse events (AEs) associated with dose reduction or discontinuation, Grade3+ (G3+) AEs, serious AEs, G3+Anemia, G3+neutropenia, G3+thrombocytopenia, adjudicated any grade interstitial lung disease (ILD), and G3+ ILD. cORR and all safety endpoints, except ILD were analyzed by logistic regression; any grade ILD was analyzed by parametric time-to-event analysis. The T-DXd or DXd exposure metric and numerous patient-specific covariates were tested in the analysis. Results: T-DXd exposure had a positive but shallow relationship with cORR. The predicted cORR varied from 57% (95% credible interval (CrI): [36, 77]) to 65% (95% CrI: [41, 84]) across 5th and 95th percentile of T-DXd exposure respectively. No clinically meaningful difference in cORR was estimated across 5.4 and 6.4 mg/kg doses with predicted cORR values of 53% (95% CrI: [44, 60]) and 55% (95% CrI: [47, 62]) at T-DXd doses of 5.4 mg/kg and 6.4 mg/kg, respectively. ES relationships were consistent with previous T-DXd ES analyses in breast cancer (BC). Model-predicted AE rates were lower (up to 9%) in 5.4 mg/kg compared to the 6.4 mg/kg dose. Model-predicted safety at 5.4 mg/kg dose in HER2m NSCLC was similar to the approved dose of 5.4 mg/kg in BC. Conclusions: The ER analyses showed no clinically meaningful difference in efficacy between 5.4 and 6.4 mg/kg doses of T-DXd; and safety events were lower in 5.4 mg/kg, thus supporting the 5.4 mg/kg recommended dose for HER2m NSCLC. These findings were later confirmed based on primary analysis in DL-022 and support the utility of ER analyses in dosing recommendation. Reference: Li BT et al. N Engl J Med 2022;386:241-51 Goto K et al. J Clin Oncol 41:4852-4863. Clinical trial information: NCT03505710 , NCT04644237 .