Impact of prior chemoradiotherapy-related variables on outcomes with durvalumab in unresectable Stage III NSCLC (PACIFIC)

Introduction: The PACIFIC trial demonstrated that durvalumab significantly improved progression-free and overall survival (PFS/OS), versus placebo, in patients with Stage III NSCLC and stable or responding disease following concurrent, platinum-based chemoradiotherapy (CRT). A range of CT and RT regimens were permitted, and used, in the trial. We report post-hoc, exploratory analyses of clinical outcomes from PACIFIC according to CRT-related variables. Methods: Patients were randomized 2:1 (1-42 days post-CRT) to up to 12 months durvalumab (10 mg/kg intravenously every 2 weeks) or placebo. Efficacy and safety were analyzed in patient subgroups defined by the following baseline variables: platinum-based CT (cisplatin/carboplatin); vinorelbine, etoposide, or taxane-based CT (all yes/no); total RT dose (<60 Gy/60-66 Gy>66 Gy); time from last RT dose to randomization (<14 days/>= 14 days); and use of pre-CRT induction CT (yes/no). Treatment effects for time-to-event endpoints were estimated by hazard ratios (HRs) from unstratified Cox-proportional-hazards models. Results: Overall, 713 patients were randomized, of whom 709 received treatment in either the durvalumab (n/N = 473/476) or placebo arms (n/N = 236/237). Durvalumab improved PFS, versus placebo, across all subgroups (median follow up, 14.5 months; HR range, 0.34-0.63). Durvalumab improved OS across most subgroups (median follow up, 25.2 months; HR range, 0.35-0.86); however, the 95 % confidence interval (CI) of the estimated treatment effect crossed one for the subgroups of patients who received induction CT (HR, 0.78 [95 % CI, 0.51-1.20]); carboplatin (0.86 [0.60-1.23]); vinorelbine (0.79 [0.49-1.27]); and taxane-based CT (0.73 [0.51-1.04]); and patients who were randomized >= 14 days post-RT (0.81 [0.62-1.06]). Safety was broadly similar across the CRT subgroups. Conclusion: Durvalumab prolonged PFS and OS irrespective of treatment variables related to prior CRT to which patients with Stage III NSCLC had previously stabilized or responded. Limited patient numbers and imbalances in baseline factors in each subgroup preclude robust conclusions.