Efficacy and safety of immune checkpoint inhibitors in a Danish real life non-small cell lung cancer population: a retrospective cohort study.

Background: To investigate effect and toxicity of immune checkpoint inhibition (ICI) in a Danish real-life non-small cell lung cancer (NSCLC) population. By including patients underrepresented in clinical trials, such as those with brain metastasis (BM), higher age, more comorbidity and poorer performance status (ECOG), comparison of unselected patients to clinical trial populations is possible. Material and methods: Real life data were gathered from 118 consecutive NSCLC patients with incurable NSCLC treated with ICI at the Department of Oncology at the University Hospital of Odense, Denmark from September 2015 to April 2018. Immune-related adverse events (irAEs) grades 3-5 were registered prospectively during the same period. Additional patient related data were obtained retrospectively from patients' files. Overall survival (OS) and progression-free survival (PFS) were calculated using the Kaplan-Meier estimates, the log-rank test and cox regression analysis performed for factors affecting survival. Results: Median age for patients was 66 years (IQR 59-71) and 62 years (range: 55-64) for those with BM. Females 63%; adenocarcinoma (AC)/squamous/others 69%/23%/8%; ECOG ≥ 2 10%; bone/brain/liver metastases 36%/18%/15%; PD-L1 (TPS) <1%/ ≥ 1%/ ≤ 49%/ ≥ 50%/NR: 3%/14%/68%/15%; baseline autoimmunity 10%, Charlson's Comorbidity Index Score (CCIS) ≥ 2 39%, treatment line: 1st/2nd/ ≥ 3rd 39%/30%/31%. Median OS for patients receiving ICI in ≥2 line was 11.5 months versus not reached in first line (HR 2.6, [95% CI: 1.3-5.0], p = .005). For patients with BM, the median OS was 8.2 months (HR 1.38, [95% CI: 0.7-2.5], p = .37). Twenty-four percent of patients terminated ICI due to irAE grades 3-5 alone (grade 5, n  =  1), which were not associated with higher age or BM. Conclusions: OS and PFS were comparable to clinical trial reports. Long-lasting remission is also possible in patients with BM. Real-life populations have higher rates of irAE grades 3 and 4 than reported in clinical trials, but it does not seem to impact median OS.