Sleep patterns, genetic susceptibility, and gastrointestinal cancer mortality: a large-scale prospective cohort study.

Background: The impact of sleep patterns on gastrointestinal cancer mortality has not been comprehensively explored. Moreover, the interaction between sleep patterns and genetic susceptibility with gastrointestinal cancer mortality remains uncertain.

Methods: The study included a total of 379 845 participants from the UK Biobank. A poor sleep pattern was defined by short sleep (< 7 h/day) or long sleep (> 9 h/day), late chronotype, frequent insomnia, snoring, and frequent daytime dozing. The outcomes were mortality of any gastrointestinal cancers and six site-specific gastrointestinal cancers. Polygenic risk score was generated to characterize genetic risk. Multivariable cox proportional hazards regression models were use to analyze the associations of sleep patterns and genetic susceptibility with gastrointestinal cancer mortality.

Results: A poor sleep pattern was associated with the increased risk of mortality from overall gastrointestinal cancer (HR 1.32, 95% CI 1.16-1.49), esophagus cancer (HR 1.37, 95% CI 1.02-1.85) and liver cancer (HR 2.01, 95% CI 1.48-2.74). Participants with a poor sleep pattern and high genetic risk combination had the highest mortality risks for esophagus, stomach, colorectal, and liver cancer. Significant multiplicative interactions (HR 1.51, 95% CI 1.05-2.18) and additive interactions (RERI 0.54, 95% CI 0.13-0.95; AP 0.34, 95% CI 0.10-0.58) of an intermediate sleep pattern and high genetic risk were observed on esophagus cancer mortality.

Conclusions: A poor sleep pattern is associated with increased risks of gastrointestinal cancer mortality independent of the conventional risk factors, and the association is modified by genetic susceptibility.