Senior Principal Scientist Genentech, United States
Disclosure(s):
Denison Kuruvilla: No relevant disclosure to display
Objectives: To develop a population pharmacokinetic (popPK) model to characterize the pharmacokinetics (PK) of subcutaneous (SC) administration of mosunetuzumab, a CD20xCD3 T-cell engaging bispecific antibody in development for non-Hodgkin lymphoma. Additionally, the study aims to externally evaluate the model’s suitability to describe mosunetuzumab PK in combination with lenalidomide.
Methods: Data were sourced from two clinical trials: Study GO29781 (Phase I/II trial with escalating doses of mosunetuzumab as a single agent in patients with relapsed or refractory B-cell non-Hodgkin lymphoma; NCT02500407) and Study CO41942 (Phase Ib/II trial with a randomized stage evaluating SC versus IV mosunetuzumab in combination with lenalidomide in patients with relapsed or refractory follicular lymphoma; NCT04246086). A previously developed intravenous (IV) popPK model (1) was extended to add SC PK data from Study GO29781 (N = 228, dose range: 1.6 mg to 90 mg). A range of potential models for describing the SC absorption of mosunetuzumab were explored, including simple first-order, transit, and parallel absorption processes. The IV part of the model was fixed, including between-subject variability and previously identified covariate relationships, except for the residual error. Stepwise covariate modeling was performed to explore potential relationships affecting SC absorption. The suitability of the final SC/IV model was evaluated using mosunetuzumab PK data in combination with lenalidomide from Study CO41942 (IV, N=39, dose: 1/2/30 mg and SC, N=78, dose: 5/45/45 mg), using standard goodness-of-fit plots and visual predictive checks.
Results: The SC absorption was described in the updated SC/IV model by two parallel first-order absorption processes and allowing the fraction of dose entering the central circulation to increase over time. The relative bioavailability of the SC formulation was estimated to be 0.896 (95% CI 0.825-0.972). The mean terminal half life at steady-state for SC population was 17.0 days and was similar to that previously estimated for the IV population. No new covariate relationships were identified. The model was successfully validated using SC and IV data at different dose levels from Study CO41942.
Conclusions: The IV popPK model for mosunetuzumab was updated to include SC administration from 228 patients across a broad dose range. No covariate effects of baseline age, baseline weight, sex, race, baseline anti-CD20 drug concentration, drug substance or site of administration were identified on the SC absorption part of the model. The model's validation using external data from Study CO41942 demonstrated its robustness and also indicated that there was no impact of lenalidomide on mosunetuzumab exposure.
