Background: To date, osimertinib has been commonly used as a first-line treatment for EGFR mutated advanced NSCLC. However, the issue of what constitutes effective treatment after osimertinib failure remains．We evaluated treatment with afatinib plus chemotherapy for EGFR mutated NSCLC resistant to osimertinib. Methods: Patients (pts) with EGFRm (Del19 or L858R) after failure of osimertinib treatment were assigned to a regimen of afatinib 20mg daily combined with carboplatin AUC5 mg/ml/min and pemetrexed 500mg/m2 every 3 weeks followed by maintenance treatment with afatinib plus pemetrexed until disease progression or unacceptable toxicity was noted. The primary endpoint was rate of PFS at 6 months after initiation of treatment (6M-PFSR). The threshold was set at 35% and the expected value at 60% (two-sided P-value of 5% and power of 80%). Thirty-one patients were required, and the target number of patients was set at 35 after accounting for ineligible cases. Secondary endpoints were PFS, OS, ORR, DOR, and Safety. In this study, blood samples were collected before and after study treatment and at the time of PD to examine biomarkers using CAPP-SEQ. This biomarker study is ongoing. Results: Between June 7, 2020 and January 19, 2022, 36 pts were enrolled. One patient was found to meet the exclusion criteria, and the efficacy was analyzed in the other 35 patients. The mean age was 70 years, 60% were women, and 54.3% were nonsmokers. The 6M-PFSR, the primary endpoint, was 57.1%. The confidence interval of 39.3%-71.5% exceeded the threshold of 35%, and the study met its primary endpoint. Notably, 24.1% of pts had a long-term PFS of 1 year or longer. ORR was 48.6%, DCR was 88.6%, median PFS was 8.2M, median DOR was 5.6 M, and median OS was not reached. By mutation type, ORR were similar for Del19 and L858R (46.7% and 50.0%, respectively), but median PFS tended to be longer for Del19 compared to L858R (9.6M and 5.2M, respectively). Pts who responded to previous osimertinib therapy (CR and PR: n = 29) tended to have longer mPFS compared to non-responders (SD, PD, and NE: n = 6) (8.5M and 5.8M, respectively). Adverse events due to TKI and chemotherapy were observed. Diarrhea (52.8%), anorexia (44.4%), fatigue (33.3%), and paronychia (36.1%) were the most frequent adverse events, and these adverse events were predictable and manageable. Interstitial pneumonia developed in three patients (8%), one of whom was the only death in the study. Conclusions: The combination of afatinib and the platinum doublet showed satisfactory efficacy with manageable adverse events in tumors refractory to osimertinib, and met its primary endpoint. OS follow-up and biomarker analyses are still ongoing. This regimen is expected to be a candidate for second-line therapy after osimertinib. Clinical trial information: jRCTs021200005.