Introduction: In an earlier report of the ASCEND-8 study (open-label, phase I, three-arm study, treatment-naive patients and pre-treated patients with advanced/metastatic NSCLC), it was shown that ceritinib 450 mg with food had comparable exposure and better gastrointestinal tolerability than 750-mg fasted. Methods: Here, we report efficacy and updated safety data from primary efficacy analysis of the ASCEND-8 study. Key secondary endpoints were overall response rate and duration of response, assessed by blinded independent review committee (BIRC) using Response Evaluation Criteria in Solid Tumors 1.1. Results: In total, 306 patients were randomized to ceritinib 450-mg fed (n = 108) or 600-mg fed (n = 87) or 750-mg fasted (n = 111), of which 304 patients were included in safety analysis and 198 treatment-naive patients (ALK receptor tyrosine kinase [ALK]–positive by immunohistochemistry) were included in the efficacy analysis (450-mg fed [n = 73], 600-mg fed [n = 51], and 750-mg fasted [n = 74]). The BIRC-assessed overall response rate was 78.1% (95% confidence interval [CI]: 66.9–86.9), 72.5% (95% CI: 58.3–84.1), and 75.7% (95% CI: 64.3–84.9), respectively; and the median duration of response (months) by BIRC was not estimable (NE) (95% CI: 11.2–NE), 20.7 (95% CI: 15.8–NE), and 15.4 (95% CI: 8.3–NE), respectively. Based on the safety analysis (n = 304), the 450-mg fed arm showed the highest median relative dose intensity (100% versus 78.5% versus 83.7%), lowest proportion of patients with dose reductions (24.1% versus 65.1% versus 60.9%), and lowest proportion of patients with gastrointestinal toxicities (75.9% versus 82.6% versus 91.8%). Conclusion: Ceritinib at a dose of 450 mg with food compared to 750-mg fasted showed consistent efficacy and less gastrointestinal toxicity.