*2.2. Study Design, Treatments, and Endpoints*

This single-arm-multicenter study involved daily oral administration of osimertinib (80 mg/day). Osimertinib had to be started at 80 mg/day, and if adverse events (AEs) occurred, dose reduction was performed according to the dose reduction criteria. Administration of osimertinib was continued until the patient met the discontinuation criteria or disease progression. Tumor assessments were performed at baseline, every 6 weeks (± 2 weeks) for 6 months, and then every 9 weeks (± 2 weeks) until disease progression. Baseline brain imaging was performed on a similar schedule. Among patients with T790M mutations, the objective response rate (ORR) was 62% (95% confidence interval [CI]: 54–68) in the AURA extension study (201 patients). In the AURA2 study (210 patients) the ORR was 70% (95% CI: 64–77) and the median PFS was 9.9 months (95% CI: 9.5–12.3) [21–23]. Docetaxel is the standard treatment for elderly patients based on the Japanese guidelines, as it provided an ORR of 22.7% in a study that compared docetaxel to vinorelbine [24]. Another recent study evaluated carboplatin plus pemetrexed for elderly Japanese patients and revealed an ORR of 41.2% [25]. Based on these findings, a required sample size of 31 patients was calculated according to the normal approximation method, with an expected response rate of 60%, a threshold response rate of 35%, two-sided alpha = 0.05, and 1 – beta = 0.8. However, the target sample size was increased to 35 patients to account for potential dropout cases. The primary endpoint for the trial was the overall response rate (ORR), while the secondary endpoints were PFS, OS, disease control rate (DCR), and safety events.
