Use of tyrosine kinase inhibitors (TKIs) is the standard therapy for epidermal growth factor receptor (EGFR)-mutated non-small cell lung cancer (NSCLC) with brain metastases. Several studies have shown that adding chemotherapy to EGFR-TKIs could improve progression-free survival (PFS) in patients with EGFR-mutant advanced NSCLC; however, the efficacy of these agents in patients with brain metastases remains unclear. To investigate the efficacy and safety of gefitinib plus chemotherapy (pemetrexed with platinum) compared with gefitinib alone in patients with untreated EGFR-mutant NSCLC brain metastases. This open-label prospective, multicenter, phase 3 randomized clinical trial was conducted in 6 centers in China from January 13, 2016, to August 27, 2021. The median follow-up time was 21.1 months (IQR, 13.5-31.8 months). Patients with untreated confirmed brain metastases and EGFR-sensitive mutated NSCLC were enrolled. The eligible patients were randomly assigned (1:1) to receive gefitinib plus chemotherapy or gefitinib alone. The primary end point was intracranial PFS; secondary end points included PFS, overall survival (OS), intracranial objective response rate, overall objective response rate, and safety. Intention-to-treat analysis was performed. A total of 161 patients (87 [54.0%] women; mean [SD] age, 55 [9.8] years; range, 26-80 years) were enrolled and randomized to receive gefitinib (n = 81) or gefitinib plus chemotherapy (n = 80). The median intracranial PFS was 15.6 months (95% CI, 14.3-16.9 months) in the gefitinib plus chemotherapy group vs 9.1 months (95% CI, 8.0-10.2 months) in the gefitinib group (hazard ratio, 0.36; 95% CI, 0.25-0.53; P < .001). Similarly, the median PFS was significantly longer with gefitinib plus chemotherapy than gefitinib alone (16.3; 95% CI, 14.4-18.2 months vs 9.5; 95% CI, 8.3-10.8 months; P < .001). Gefitinib plus chemotherapy had a better intracranial objective response rate (85.0%; 95% CI, 77.0%-93.0% vs 63.0%; 95% CI, 52.2%-73.7%; P = .002) and overall objective response rate (80.0%; 95% CI, 71.0%-89.0% vs 64.2%; 95% CI, 53.5%-74.9%; P = .03) than gefitinib alone. At data cutoff, the median OS was also significantly longer in the gefitinib plus chemotherapy group vs the gefitinib group (35.0 vs 28.9 months; hazard ratio, 0.65; 95% CI, 0.43-0.99; P = .04). Grade 3 or worse adverse events were more common with gefitinib plus chemotherapy, most of which were manageable. In this randomized clinical trial, gefitinib plus chemotherapy significantly improved intracranial PFS, PFS, and OS compared with gefitinib alone in patients with untreated EGFR-mutant NSCLC brain metastases and could be an optional first-line treatment for these patients. ClinicalTrials.gov Identifier: NCT01951469.