Immune-related adverse events and outcomes during treatment with immune checkpoint inhibitors in non-small cell lung cancer (NSCLC) patients.

e21662 Background: Immunotherapy of cancer has changed the paradigm of treatment of many tumours, specially non-small cell lung cancer (NSCLC). The use of immune-checkpoint inhibitors (ICI) is associated in some patients with the development of new immune-related adverse events (irAEs). Our aim was to study if there is any correlation between the appearence of irAEs and the efficacy of ICI. Methods: We collected data of 104 patients diagnosed of advanced NSCLC and treated with ICI in monotherapy at our institution between December 2015 and December 2019. Several variables as clinical, tumour-related and therapeutical were included and univariate and multivariate Cox regression analysis were performed. Results: Cohort of 84 men and 20 women, median age of 67 years and 86% with Eastern Cooperative Oncology Group (ECOG) Performance Status of 0-1. 89% were active or ex-smokers and 11% had never smoked. 60% of patients had adenocarcinoma histology, 39% scamous and 1% had not otherwise specified (NOS) carcinoma histology. 3% of patients had III-B stage at the moment of start of immunotherapy, 37% M1a, 30% M1b and 30% M1c. 2 patients had driver mutations in EGFR gene. 41% of patients had unknown PDL1 status; 14% had no PDL1 expression, 14% low expression and 31% high expression. 78% of patients had progressed to prior line of treatment, while 22% were treatment-naive. irAEs occured in 65% of patients; 12% developed grade 3 to 4 toxicities. More frequent irAEs were fatigue (54%) and rash (27%). Significant statistical variables in univariate analysis were included in multivariate analysis by Cox regression. The appearence of any grade of iRAE was associated with improved progression-free survival (PFS) (median 17.9 months vs 5.1 months; HR 7.12; p = 0.008). Those patients who experimented any grade of irAE were more likely to achieve stabilization or response than those who suffered progression of disease (HR 13.00; p < 0.001; 95% CI [3.47-48.78]. The use of corticosteroids during treatment with ICI was not related to the benefit of treatment. Conclusions: Appearence of immune-related adverse effects during treatment with ICI was associated with better outcomes in our population. The use of corticosteroids during immunotherapy didn´t have any deleterious effect on the efficacy of treatment.