Research

Association Between Prediagnostic Cigarette Smoking and Colorectal Cancer Survival by Molecular Subtypes and Age-onset Status.

Eric Bakwa | 2023

Advisor: Amanda Phipps

Research Area(s): Cancer Epidemiology

Full Text


In the United States, colorectal cancer (CRC) remains the third most commonly diagnosed malignancy in both men and women, and approximately 36% of CRC patients die within the first five years after diagnosis. Several factors are known to affect the survival of CRC patients, including smoking history; however, it is unclear whether previously-observed associations between cigarette smoking history and CRC survival vary across molecular subtypes of CRC or by age at CRC diagnosis.Methods: We conducted a retrospective analysis of survival using data from two complementary study populations: the Colon Cancer Family Registry (CCFR) and the Advanced Colorectal Cancer of Serrated Subtype (ACCESS). This analysis included 4,901 participants from the CCFR and ACCESS studies who completed a baseline questionnaire at the time of enrollment. We used Pearson Chi-square tests of independence to determine if there was a statistically significant difference in the distribution of cigarette smoking status by age at diagnosis as a categorical variable (<50 years and ≥ 50 years) and by different molecular subtypes of CRC. We used univariate and multivariate Cox Proportional Hazards regression analyses to describe associations of smoking with CRC survival overall and within groups defined by age at diagnosis and molecular subtypes. Results: Of the 4,901 participants included in this study, 1,378 participants died due to CRC during study follow-up (average follow-up period = 13.6 years). Ever smoker participants had 27% higher instantaneous risk of dying from CRC than never-smokers (HR=1.27; 95% CI:1.13-1.42). Even after adjusting for different covariates, ever smoker participants continued to have 21% higher risk of dying from CRC compared to never-smoker patients (HR=1.21; 95% CI:1.06-1.38). In analyses stratified by age at diagnosis, the observed association with smoking was limited to individuals with later-onset CRC. In analyses stratified by tumor molecular subtypes, a statistically significant association of smoking with CRC survival was observed only among participants with CRC type 4 (CIMP-low or negative, MSI-low or MSS, BRAF and KRAS wildtype), although similar patterns of association were noted for most other subtypes. Conclusion: Cigarette smoking is associated with poorer survival after CRC diagnosis, particularly for individuals diagnosed with CRC at later ages.