Abstract
Background: Tobacco smoking accounts for 22% of the pancreatic cancer burden in Australia. Smoking may increase therapy-related adverse effects and was associated with shorter survival in the previous analysis of all stages of pancreatic cancer. There is ongoing uncertainty among clinicians regarding the benefits of encouraging smoking cessation in patients with advanced cancer. The objective of this study was to evaluate the association between current smoking and survival outcomes in patients with advanced pancreatic cancer using real-world registry data.
Methods: We studied the impact of ongoing smoking on survival among patients with metastatic and locally advanced pancreatic carcinoma in the Pancreatic Cancer: Understanding Routine Practice & Lifting End Results (PURPLE) Translational Registry. Of 1,454 registry patients, 1,047 had the smoking status of current smokers versus non-smokers recorded and were analysed using inverse probability weighting. Predictor variables were age ≥65 years, locally advanced or metastatic disease, liver metastases, gender, Index of Relative Socio-Economic Advantage and Disadvantage (IRSAD) score, Charlson morbidity index, and presence of >3 metastatic sites. The association between smoking and overall survival (OS), cancer-specific survival (CSS), and progression-free survival (PFS) on first-line systemic treatment was calculated using Cox proportional hazards regression.
Results: A total of 19.4% of analysed patients were current smokers. Smokers were more likely to be male (P=0.048), be diagnosed at a younger age (P<0.001), and have a lower socioeconomic status (P=0.001). The results failed to show an impact of current smoking on OS [hazard ratio (HR) =0.92, 95% confidence interval (CI): 0.84–1.01, P=0.09], CSS (HR =0.93 95% CI: 0.85–1.02, P=0.13) or PFS (HR =0.98, 95% CI: 0.90–1.16, P=0.72) for locally advanced and metastatic disease patients combined.
Conclusions: In this study of survival outcomes, current smoking was not associated with a reduction in OS, CSS, or PFS among patients with incurable pancreatic cancer. This contrasts with the results of previous studies.