IntroductionThis study validated real-world pharmacokinetic (PK) data using an established population PK (PopPK) model for atezolizumab in Japanese patients with NSCLC and explored the relationship between PK parameters, effectiveness, and adverse events (AEs) for the 1200 mg once every three weeks regimen. MethodsA subgroup of 262 of 1039 patients from J-TAIL consented to this exploratory research for PK evaluation of atezolizumab monotherapy for unresectable advanced/recurrent NSCLC (August 2018 to October 2019; 197 institutions). We evaluated plasma concentrations before the start of the third cycle of atezolizumab infusion classified into quartiles 1 to 4, their association with effectiveness, and the association between atezolizumab maximum plasma concentrations (Cmax) calculated using the existing PopPK model and AEs of special interest (AESIs). ResultsOverall, 175 of 262 patients were included; baseline characteristics were similar to those of patients enrolled in J-TAIL (Eastern Cooperative Oncology Group performance status ≥ 2, 12.0%; age ≥ 75 y, 28.9%; atezolizumab as more than or equal to third-line treatment, 57.5%). Atezolizumab plasma concentrations were similar to previously reported data among Japanese/non-Japanese patients. The overall survival was significantly shorter in patients with lower atezolizumab plasma concentrations in Q1 versus Q2 to Q4, although progression-free survival remained the same. The PK data adequately fit the PopPK model, with the frequency of AESIs increasing as the calculated Cmax at cycle 1 increased. ConclusionsIn real-world Japanese patients with unresectable advanced/recurrent NSCLC, PKs were similar to previous reports. Certain patient populations had shorter overall survival, and atezolizumab plasma concentrations in cycle 3 were lower in this population. Elevated Cmax at cycle 1 may be associated with an increased frequency of AESIs.