Population pharmacokinetic and pharmacodynamic analysis of osimertinib.

Abstract

e20536 Background: Osimertinib is an oral, potent, irreversible, CNS active EGFR-TKI, selective for sensitizing (EGFRm) and T790M resistance mutations, indicated for the treatment of patients with T790M positive advanced non-small cell lung cancer who have progressed on or after EGFR-TKI therapy. Osimertinib pharmacokinetics (PK) were evaluated using a population approach and pharmacodynamic (PD) relationships using appropriate modeling approaches. Methods: To understand the impact of covariates on osimertinib PK, a population PK analysis was performed using data from patients who received osimertinib (20–240 mg) during the AURA studies. Exposure metrics were derived from a PK model and used to assess the exposure-response (safety/efficacy) relationship. Efficacy analysis included patients who were T790M positive (n = 710) and safety analysis included all dosed patients (n = 1088). The impact of covariates on exposure-response was assessed. Models accounting for rare safety events were applied to quantify the association between events and exposure. Results: Population PK analyses supported dose- and time-independent PK of osimertinib with no clinically meaningful covariates identified. Patients in the highest exposure quartile (Q4) had a numerically shorter median progression-free survival (8.3 months [95% CI 6.9, 10.5]) compared with patients in Q1, Q2 and Q3 (all 11.2 months [95% CIs 9.7, 12.7; 8.5, 15.6 and 8.7, 13.7, respectively]). A model-based analysis indicated that this effect is likely due to a larger number of patients in Q4 with poor prognostic features, i.e. worse performance status (WHO 1 or 2) and lower baseline serum albumin compared with Q1, Q2 and Q3, rather than to osimertinib exposure. Model-predicted probability of a relationship between osimertinib exposure and LVEF changes was not evident. Model-based analysis predicted that, compared with the median probability (0.03), the probability of a patient experiencing interstitial lung disease may increase with increasing osimertinib exposure (Q1 probability 0.01 [steady-state AUC 6361 nM*h] vs Q4 0.06 [24460 nM*h]) at the 80 mg dose. Conclusions: Population PK and PK-PD analysis is supportive of 80 mg as an appropriate dose for osimertinib. Clinical trial information: NCT01802632; NCT01802632; NCT02094261; NCT02151981.