First line (1L) durvalumab in patients with PD-L1 positive, advanced non-small cell lung cancer (NSCLC) with a performance status of 2 (PS2). Primary analysis of the multicenter, single-arm phase II trial SAKK 19/17

9088 Background: The safety and efficacy of 1L durvalumab in PS2 patients (pts) with advanced NSCLC is unknown. Important safety data leading to exclusion of pts with relevant respiratory symptoms have been published as an interim report. Here we present the primary analysis of 1L durvalumab in PS2 pts, unsuitable for combination chemotherapy and PD-L1 expression in ≥25% of tumor cells. Methods: In this single-arm, multicenter, phase II trial pts with PD-L1 positive (tumor proportional score, TPS ≥25%), advanced NSCLC with PS2, unsuitable for combination chemotherapy determined by the investigators, in the absence of known contraindications for immunotherapy and sufficient organ function, received a fixed dose of 1500 mg durvalumab every four weeks. The primary endpoint was overall survival (OS) at 6 months. The statistical hypothesis was to improve OS at 6 months from ≤35% to ≥53%. Adverse events (AEs) were assessed according to National Cancer Institute Common Terminology Criteria for AEs (NCI CTCAE) version 5.0. Results: Forty-eight pts were included (29 males, 19 females). Median age was 76 years (range, 37-87). OS at 6 months was 60% (95% CI: 45-74%). OS at 6 months after the exclusion of pts with initially relevant respiratory symptoms was 67% (95% CI: 46-84%, n = 27) compared to the subgroup of pts without this exclusion criteria who were recruited before the amendment (52%, 95% CI: 30-74%, n = 21). Median OS was 8.5 months (95%CI: 4.4-16.7). Objective response rate and median PFS were 17% (95% CI: 8-30%) and 2.5 months (95% CI: 1.8-7.1). Thirty-three deaths (69%) were observed to date. Ten early fatal events considered not treatment-related occurred during the first 5 weeks of treatment. Four out of the first 7 early fatal events (4/7; 57%) were respiratory failures in pts with advanced symptomatic primary lung tumors. Only 3 more early fatal events occurred after the protocol amendment excluding pts with severe respiratory symptoms. Thirty-nine patients (81%) had an AE grade ≥3 (G3). The most frequent AEs ≥G3 were lung infection (19%), dyspnea (15%) and hypertension (10%), respectively. Treatment-related AEs ≥G3 were reported in 9 pts (19%) and included colonic perforation in one patient (grade 5), colitis in 5 pts (10%), hepatitis and increased lipase in 3 pts each (6%). Conclusions: 1L durvalumab in PS2 pts with advanced PD-L1 positive (TPS ≥25%) NSCLC is effective and led to a promising 6-month OS of 60%. Four-weekly durvalumab can be safely offered to pts presenting without severe pulmonary symptoms who are not candidates for chemotherapy. Clinical trial information: NCT03620669 .