Population pharmacokinetics and exposure–response relationships of dostarlimab in primary advanced or recurrent endometrial cancer in part 1 of RUBY

Aims: Dostarlimab-gxly is a humanized monoclonal antibody of the IgG4 isotype that binds to the programmed cell death protein-1 (PD-1) receptor and blocks its ligands. RUBY (NCT03981796) is a two-part multicentre study in patients with recurrent or primary advanced endometrial cancer. The overall aims were to characterise the population pharmacokinetics (PopPK) from Part 1 of this study, identify relevant covariates of interest, and assess exposure–efficacy/safety (ER) relationships. Methods: A PopPK model developed using GARNET (NCT02715284) study data for dostarlimab monotherapy was externally validated with RUBY Part 1 study data. Subsequently, the model was updated with data across the two studies. Exposure–safety analyses for adverse events related to dostarlimab alone or in combination with standard of care (SOC) were modelled using logistic regression. Exposure–efficacy analysis included Cox proportional hazards analysis of the primary efficacy endpoint of progression-free survival (PFS). Results: For the model update, 7957 pharmacokinetics observations from 868 patients pooled from both RUBY and GARNET studies were available. The model was consistent with the previous model. Dostarlimab clearance was estimated to be 7.79% lower when dostarlimab was given as SOC combination therapy. However, no significant covariates were clinically relevant. Hepatic or renal impairment did not affect pharmacokinetics. Among the safety endpoints, only rash showed a small yet statistically significant effect (P <.05) in all subjects; however, this was not not deemed clinically relevant. There were no other clinically significant exposure–safety or exposure–PFS relationships. Conclusions: The addition of chemotherapy to dostarlimab had limited effect on dostarlimab PopPK, with no clinically significant covariates or clinically relevant exposure–safety or exposure–PFS relationships.