Background: Osimertinib (OSI), a third-generation EGFR-tyrosine kinase inhibitor (EGFR-TKI), is now a standard treatment for previously untreated EGFR-mutated (EGFRm) advanced non-small cell lung cancer (NSCLC). In the two randomized phase 3 studies, progression-free survival (PFS) and overall survival were statistically significant and clinically longer with gefitinib and platinum-based chemotherapy compared with gefitinib monotherapy. Based on these data, we have planned this phase 2 study to evaluate the safety and efficacy of OSI combined with platinum-based chemotherapy. Patients and Methods: This multicenter phase 2 study enrolled patients (pts) with clinical stage IIIB, IIIC, IVA, IVB or postoperative recurrent, previously untreated EGFRm NSCLC. Pts received oral OSI 80mg once daily (QD), with either cisplatin 75mg/m2 (arm A) or carboplatin [area under the curve (AUC) = 5, arm B], plus pemetrexed (PEM) 500 mg/ m2 every 3 weeks (Q3W) for four cycles. In both arms, maintenance was OSI 80mg QD with PEM 500 mg/ m2 Q3W until disease progression or discontinuation. The co-primary endpoints were the safety and the objective response rate (ORR), and the secondary endpoints included the complete response rate (CRR), disease control rate (DCR), and PFS. Results: From July 2019 to February 2020, 67 pts (34 pts in Arm A; 33 pts in arm B) were enrolled: median (range) age 67 (26-75) years; 43 (64.2%) female; 46 (68.7%) ECOG PS 0; 66 (98.5%) adenocarcinoma; 31 (46.3%) EGFR exon19 deletion, 35 (52.2) L858R, and 1 (1.5%) both. One pt did not comply with the eligibility criteria and was excluded from the efficacy analysis. At data cut off (August 31, 2021), 27 (40.3%) pts [15 (44.1%) in arm A and 12 (36.4%) in arm B] had discontinued the protocol treatment, including 9 (13.4%) pts [5 (14.7%) in arm A and 4 (12.1%) in arm B] due to the adverse event (AE). The rate of grade (G) ≥ 3 AEs were 91.0% (88.2% in arm A and 93.9% in arm B). For the safety, neutropenia, anemia and thromobocytopenia were numerically higher in arm B and the rates of G ≥ 3 were 29.4%/60.6%, 14.7%/27.3% and 0.0%/42.4% in arm A/B, respectively. G ≥ 3 QTc interval prolonged and G ≥ 2 anorexia were observed in 14.7%/21.2% and 26.5%/24.2%, respectively. For the efficacy, the ORR was 90.9% [95% confidence interval (CI); 84.0-97.8%]. The CRR/DCR were 3.0%/97.0% (95% CI; 0.0-7.2%/92.8%-100.0%), respectively. At a median follow-up time of 21.4 months (range, 18.2-25.7), median PFS was not reached in both A and B, with an estimated 12-/24-months PFS rate of 90.4%/70.0%. Conclusions: OSI combined with platinum-based chemotherapy for previously untreated EGFRm advanced NSCLC showed the excellent efficacy with tolerable toxicity. This combination treatment is highly promising and should be validated in the phase 3 study. Clinical trial information: jRCTs031180226.