Circadian disruption from time zone position and risk of colorectal cancer precursors in women

Background/Aims: Research on the link between circadian disruption, particularly from night shiftwork, and colorectal cancer (CRC) has been limited and inconsistent. Few studies have investigated whether other sources of circadian disruption may be associated with CRC or its precursors. One source is solar jetlag that leads to residents in the western vs. eastern part of a time zone to receive less light exposure in the morning and greater light exposure at night, likely suppressing melatonin release and reducing sleep propensity and sleep duration. The objective of this study was to examine the association between solar jetlag and CRC precursors in the United States. Methods: Our study consisted of Nurses' Health Study (NHS) II participants who received one or more lower endoscopies between 1991-2015. Cases self-reported colorectal polyps in biennial questionnaires and were confirmed by medical record review. As a proxy for solar jetlag, we calculated the distance from the time zone meridian (TZM), based on participant's geocoded residential address histories, which was modeled as a per 5-degree increase in longitude moving east to west within a time zone. Time-varying multivariable-adjusted logistic regression models for clustered data estimated odds ratios (OR) and 95% confidence intervals (CI). Results: Over a 24-year follow-up period, 72,612 NHSII participants received at least one lower endoscopy; 4,450 conventional adenomas and 4,873 serrated polyps were diagnosed. We found no statistically significant association between distance to TZM and conventional adenomas (OR=1.03; 95% CI=0.99,1.07) or serrated polyps (OR=0.98, 95% CI= 0.95,1.02). Results were similarly null in analyses stratified by age at endoscopy (<50 vs. ≥ 50 years), polyp size, anatomical location, malignant potential, or reason for endoscopy. We observed statistically significant effect modification in which positive associations were observed in the Mountain time zone, among those never engaging in rotating night shift work, areas with higher ultraviolet radiation, and lower latitudes. Conclusions: Although we did not observe an association between distance to TZM and CRC precursors, we found significant effect modification in the association by various covariates determined a priori. These findings require further research into the mechanisms of action and confirmation in other cohorts.