Background: Osimertinib, an EGFR tyrosine kinase inhibitor, is mainly metabolized by CYP3A. This study investigated genetic and non-genetic factors influencing plasma osimertinib concentrations in 36 patients with non-small cell lung cancer. Changes in CYP3A activity were evaluated using its endogenous marker, 4β-hydroxycholesterol (4β-OHC).

Method: Pre-dose plasma concentrations of osimertinib, its metabolites (AZ5104 and AZ7550), and 4β-OHC were measured by LC-MS/MS. Genotypes of CYP3A5*3 and CYP oxidoreductase (POR*28), P450 redox partners, were determined in patients. The Glasgow prognostic score (GPS) was adopted to assess cancer-related inflammation using serum albumin (Alb) and C-reactive protein (CRP).

Results: CYP3A5*3 and POR*28 genotypes did not affect the plasma concentration/dose (C/D) ratio of osimertinib and its two metabolites, nor the 4β-OHC concentration. The plasma concentration of 4β-OHC was significantly correlated with the C/D ratio of AZ7550 (ρ = 0.438, P < 0.05), but not with those of osimertinib and AZ5104. The plasma C/D ratio of osimertinib, AZ5104, and AZ7550 significantly correlated with serum Alb (ρ = -0.486, ρ = -0.651 and ρ = -0.400, respectively) and CRP (ρ = 0.638, ρ = 0.546 and ρ = 0.284, respectively). Moreover, those C/D ratios were significantly higher in patients with a GPS of 1-2 compared to those with a score of 0 (P < 0.01, P < 0.01, P < 0.05, respectively).

Conclusion: Cancer cachexia caused the increased plasma concentration of osimertinib and its metabolites, whereas the genotypes of the metabolic enzymes did not. Plasma 4β-OHC was shown to be a limited predictive biomarker for osimertinib pharmacokinetics.