Background and importance Pembrolizumab is an anti-PD1 antibody used to treat metastatic non-small cell lung cancer (m-NSCLC).

Aim and objectives To analyse the efficacy and safety of first (1L) or successive lines (≥2L) of treatment of m-NSCLC with pembrolizumab monotherapy.

Material and methods A retrospective observational study was conducted in 82 patients with m-NSCLC treated with pembrolizumab monotherapy between January 2018 and April-2020. Collected clinical data were analysed with SPSS V.23.0. Progression free survival (PFS), overall survival (OS), calculated with the Kaplan–Meier method, and objective response rate (ORR), using iRECIST criteria, were used to measure efficacy. To analyse safety, adverse effects (AEs) were evaluated using NCI-CTCAE-V.5.0 criteria. Results were compared with the clinical trials KEYNOTE-024 and KEYNOTE-010.

Results 45 (34 men) patients received 1L pembrolizumab (200 mg/3weeks). Median (Me) age=66 years (range (R)=32–83). Histology: 35 non-squamous, 6 squamous. PD-L1 expression: Me=80% (R=55–100%). None ALK, EGFR or ROS mutation. 35 patients presented ECOG=0–1 and 10 ECOG ≥2. Duration: Me=5 cycles (R=1–37). PFS was Me=5.1 months (95% CI=0.5 to not reached). Median OS was not reached, although the OS rate was 66% at 6 months and 60% at 12 months. ORR=40%. 15% of patients died within a month of treatment. 36 patients (80%) had AEs, the majority grade 1–2. 6 patients (13%) had grade 3–4 AEs.

37 (30 men) patients received ≥2L pembrolizumab (2 mg/kg/3 weeks). Age Me=67 years (R=46–90). Histology: 18 non-squamous, 14 squamous. PD-L1 expression: Me=10% (R=1–90%). Two EGFR mutations, no ALK or ROS mutations. Previous treatment always included a platinum doublet regimen. Three patients had received two or more previous lines, 34 only one. 30 patients presented ECOG=0–1 and 7 ECOG ≥2. Duration: Me=4 cycles (R=1–24). PFS was Me=3.4 months (95% CI=1.9 to 4.7) and OS was Me=9.1 months (95% CI=5.1 to 13.2). ORR=16%. 24 patients (65%) had AEs, the majority grade 1–2. 3 patients (8%) had grade 3–4 AEs.

Conclusion and relevance Pembrolizumab in 1L for m-NSCLC in our patients presented lower PFS and OS than those recorded in the KEYNOTE-024 study with similar ORR. This can be partially explained by the greater deterioration in our patients at the beginning of treatment (22% ECOG ≥2) compared with the KEYNOTE-024 study. Pembrolizumab in ≥2L had slightly lower PFS, SG and ORR than those recorded in KEYNOTE-010. AEs were mostly grade 1–2, and less frequent than in the clinical trials.

Conflict of interest No conflict of interest