Durvalumab (durva) after chemoradiotherapy (CRT) in unresectable, stage III, EGFR mutation-positive (EGFRm) NSCLC: A post hoc subgroup analysis from PACIFIC.

Abstract

8541 Background: Standard of care for patients (pts) with unresectable (UR) stage III NSCLC is the ‘PACIFIC regimen’, based on data from the phase 3 placebo (pbo)-controlled trial where consolidation durva following CRT improved overall survival (OS; hazard ratio [HR], 0.68 [95% CI, 0.53, 0.87]) and progression-free survival (PFS; HR, 0.52 [95% CI 0.42, 0.65]), in an all-comer population. However, the benefit of immunotherapy (IO) in pts with EGFRm stage III NSCLC is unclear. We report a post hoc exploratory efficacy and safety analysis from 35 pts with EGFRm NSCLC from the PACIFIC trial (NCT02125461). Methods: Pts with stage III UR-NSCLC, WHO performance status (PS) 0/1 and no progression after ≥2 cycles platinum-based concurrent CRT were randomized 2:1 (1–42 days post CRT) to receive durva (10mg/kg IV q2w for up to 1 year) or pbo, stratified by age, sex, and smoking history; enrollment was not restricted by oncogenic driver gene mutation status or PD-L1 expression. Primary endpoints: PFS (BICR; RECIST v1.1) and OS; key secondary endpoints: objective response rate (ORR) and safety. Treatment effects for the EGFRm subgroup were estimated using an unstratified Cox proportional hazard model; medians were estimated using the Kaplan–Meier method. Statistical analyses were exploratory. Data cut-off (DCO) for the EGFRm subgroup efficacy analysis was 11 January 2021. Results: Of 713 pts randomized, 35 had EGFRm NSCLC based on local testing (durva n = 24, pbo n = 11). In the EGFRm subgroup, more pts in the pbo vs durva arm were male (73% vs 54%), had stage IIIA disease (64% vs 46%), PS 0 (64% vs 54%) and received pre-CRT induction chemotherapy (36% vs 8%). More pts in the durva arm were Asian (63% vs 55%) and had PD-L1 on < 25% tumor cells (67% vs 36%); median age was consistent across arms. At DCO, median duration of follow-up for survival was 42.7 months (range, 3.7–74.3 months) for all randomized pts in the subgroup. Median PFS was 11.2 months (95% CI 7.3, 20.7) with durva vs 10.9 months (95% CI 1.9, not evaluable [NE]) with pbo; HR 0.91 (95% CI 0.39, 2.13). Median OS was 46.8 months (95% CI 29.9, NE) with durva vs 43.0 months (95% CI 14.9, NE) with pbo; HR 1.02 (95% CI 0.39, 2.63). ORR was 26.1% (95% CI, 10.2, 48.4) and 18.2% (95% CI 2.3, 51.8) with durva and pbo, respectively. The safety profile for durva was consistent with the overall population. In the durva and pbo subgroup arms, radiation pneumonitis was reported in 42% vs 36% of pts, and pneumonitis was reported in 17% vs 18% of pts (1 grade 3, pbo arm), respectively. Conclusions: In this post hoc exploratory analysis of 35 pts, PFS and OS outcomes with durva were similar to pbo in the EGFRm population, with wide CIs. The benefit of IO in this population remains unclear. The ongoing LAURA study (NCT03521154) is investigating the efficacy and safety of maintenance osimertinib in pts with locally advanced EGFRm stage III UR-NSCLC with no progression after CRT. Clinical trial information: NCT02125461.