Abstract
The interplay between sleep characteristics, estrogen sulfotransferase (SULT1E1), and lung cancer may involve complex interactions. While prior research has predominantly examined the association of SULT1E1 with breast cancer and the individual effects of sleep characteristics on lung cancer, the potential multifaceted interactions among these 3 factors remain underexplored. This study employed a two-sample univariable Mendelian randomization framework to systematically examine potential causal associations between 8 distinct sleep-related traits and lung cancer, encompassing its various histological subtypes. Mediation Mendelian randomization (MR) analysis was implemented to assess the putative mediating role of SULT1E1 in these associations. Methodological rigor was ensured through the application of multiple complementary MR approaches. Comprehensive evaluation of heterogeneity and horizontal pleiotropy was conducted utilizing the MR-Egger regression intercept test, Cochran Q statistic, and leave-one-out sensitivity analysis. Furthermore, the Mendelian Randomization Pleiotropy RESidual Sum and Outlier was systematically applied to identify and adjust for potential horizontal pleiotropic effects. Moreover, we employed the false discovery rate method to correct for multiple hypothesis testing and control the risk of false-positive results. Mediation MR analysis, revealed a causal association between sleep duration and a reduced risk of squamous cell lung cancer (SCC), with 9.3% of the effect mediated by SULT1E1. Dozing off during the day was causally associated with a reduced risk of lung cancer and non-small cell lung cancer. Lack of sleep was causally associated with a reduced risk of small cell lung cancer. A causal association was observed between reduced sleep duration and an elevated risk of SCC. Mediation analysis demonstrated that SULT1E1 mediated 9.3% of the total effect of reduced sleep duration on SCC development.