First-line pembrolizumab for lung cancer in a Spanish real-world study.

Abstract

e23268 Background: KEYNOTE (KN) 024 trial showed that pembrolizumab significantly improves progression free survival (PFS) and overall survival (OS) in patients with previously untreated advanced non-small cell lung cancer (NSCLC) with a PD-L1 expression ≥ 50% without EGFR/ALK alterations. The main aim of this study is to report the outcomes (PFS and OS) of pembrolizumab in a real life setting. Methods: Multicenter retrospective study in 42 Spanish hospitals. Inclusion period: 01/2016 – 07/2020; cut-off date: 31/12/2021. Demographic, clinical and treatment-related data were retrospectively obtained from patients’ medical files. Statistical analysis was carried out using descriptive metrics and survival analysis techniques (Kaplan-Meier, log-rank test, Cox regression). Results: In our cohort of 1005 patients, 74.5% were males, mean age was 66.4 ± 9.7 and mean BMI 25.6 ± 4.6. The majority were current (35.6%) or former smokers (56.6%), with 11 % of patients with ECOG ≥ 2. Patients with a non squamous histology were 79.5% and 95.2% were classified as IV stage (30.3% had hepatic and/or brain metastases, 49.3% had ≥ 2 metastatic sites). Pembrolizumab dose was mainly fixed (83.1% of patients), while the number of treatment cycles was highly variable. 31.8% received concurrent radiotherapy, while previous corticoid (PCE), antibiotic and proton pump inhibitors (PPI) exposure were 18.2%, 7.7% and 40.7% respectively. With a median follow-up of 6.3 months, median OS and PFSwere 19 (95%CI: 16.2 – 21.3) and 8.7 (95%CI: 7.2 – 10.1) months. Several variables were associated with PFS and OS after univariate survival analysis. Age and previous antibiotic exposure were associated to worse OS but not to PFS. Multivariate analysis showed that ECOG ≥2 (HR 2.7 95%CI 2-3.5) and previous PPI exposure (HR 1.3 95%CI 1.1-1.6) were predictors of poor outcome. Conclusions: In our cohort, pembrolizumab showed lower OS and similar PFS than KN 024. Our results confirm that poor PS (ECOG) is a negative predictive factor for effectiveness. [Table: see text]