Tyrosine kinase inhibitors (TKIs) have been widely accepted as first-line therapy for patients with EGFR-mutant metastatic non-small cell lung cancer (NSCLC), achieving a median progression free survival (PFS) and overall survival (OS) of 8.0 to 13.1 months and 19.3 to 30.9 months respectively. Patterns-of-failure studies suggest that the first progression after first-line TKIs occurs most often at sites of disease known to exist at the baseline. In this retrospective study, we aimed to investigate whether the radical local therapy could offer survival benefit in addition to st NSCLC patients with activating EGFR mutations and treated with TKIs as first-line management after the diagnosis of stage IV disease (either synchronous or metachronous) between 2010 and 2017 at our institution were reviewed. All enrolled patients should receive radical local therapy (either surgery or radiotherapy with curative intent) at least to the main site of disease. We defined the main site as the primary site for synchronous IV-stage patients and at least 1 progression site for metachronous IV-stage patients. OS and PFS were calculated from the first day of IV-stage treatment. Kaplan-Meier method was used for survival estimation and Log-rank test was rendered for survival comparison between groups. A total of 45 patients entered into the final analysis, including 18with synchronous stage IV diseases and 27patients with metachronous diseases. A total of 208 gross tumor sites were identified and 130 of them received local treatment, including 90 sites treated with radical approaches and another 30 sites with palliative therapy. At a median follow-up period of 37.8months, the median OS was 51.4 months, with 1-, 3- and 5-year rate of 97.7%, 58.7% and 25.2%, respectively. The median PFS was 16.4months, with 1-, 2- and 3-year rate of 64.7%, 29.1% and 6.8%, respectively. There was no difference between synchronous and metachronous groups. In 38patients who progressed, 19 (42.2%) involved new metastatic sites only, 12(26.6%) involved initial sites only, and 3 (6.7%) involved both. Metastatic EGFR-mutant NSCLC patients who received TKIs and radical local therapy in our study obviously provided longer OS and PFS compared with historical results using TKIs alone. Prospective randomized evidences are warranted to clarify the clinical efficacy of additional local therapy to first-line TKIs for this highly selective subgroup of patients.