Smoking status and immunotherapy outcomes in smoking-associated cancers

e15097 Background: Improved immunotherapy (IO) outcomes have been observed among non-small cell lung cancer patients with a current or former smoking history. This is thought to be a consequence of increased immunogenic mutation burden among smoking-related cancers. We set out to explore the association between smoking status and immunotherapy outcomes in lung and other smoking-associated cancers. Methods: This was a retrospective analysis of 200 consecutive patients with advanced, smoking-associated solid tumor types, treated with single-agent anti-PD1/PDL1 therapy at a single center between July 2014 and February 2018. The primary outcome was overall survival from date of IO initiation. The secondary outcome was overall response, defined as radiographic complete response or partial response, by RECIST 1.1 criteria. The primary predictor was smoking status (former/current smoker vs. never smoker). The primary and secondary outcomes were analyzed using multivariable Cox proportional hazards models and multivariable logistic regression models, respectively. Models were adjusted for age and sex, and stratified by cancer type. Results: The majority of patients were male (64%) with a history of smoking (72%); the average age was 67.1 ± 11.4 years. Cancer types represented were: non-small cell lung cancer (NSCLC, N = 81), hepatocellular carcinoma (HCC, N = 41), urothelial carcinoma (BLCA, N = 39), head and neck squamous cell carcinoma (HNSC, N = 21), and renal cell carcinoma (RCC, N = 18). Over a median follow-up of 11.3 months (range 0.5-53.2), there were 96 deaths and 27% of evaluable patients achieved radiographic response. Response was not evaluable in 27 patients. In multivariable regression analysis, smoking status was not significantly associated with overall survival nor overall response in any cancer type examined (Table). Conclusions: Smoking status was not associated with outcomes in our cohort of IO-treated patients with smoking-associated cancers, though sample size was limited. [Table: see text]