ARTICLE AD BOX
ABSTRACT
Objective
To explore the association between smoking, genetic susceptibility and early menopause (EM) and clarify the potential mechanisms underlying this relationship.
Design
An observational and Transcriptome-wide association analysis (TWAS) study.
Setting
UK Biobank and public summary statistics.
Population
139 869 women with full baseline and menopause data, and no gynaecological surgery history.
Methods
Adjusted modified Poisson regression models were developed to determine the smoking and genetic risk effects on EM. TWAS was used to identify gene expression between smoking and EM, with Mendelian randomisation (MR) to infer causality. Enrichment analysis explored regulatory networks of transcription factors, microRNAs and potential therapeutic targets. Small molecule drugs were predicted using drug-gene interaction analysis.
Main Outcome Measures
EM prevalence and common gene expression patterns.
Results
Women with over 30 pack-years of smoking had about 1.5 times higher EM risk, with RRs of 1.39 (95%CI, 1.23–1.56), 1.45 (1.33–1.59) and 1.45 (1.36–1.55) in the low, intermediate and high genetic risk groups. TWAS identified hub genes such as IMMP2L, BMPR2 and HMGN1. MR confirmed daily cigarette consumption as a causal factor in early menopause. Several potential therapeutic targets (e.g., SP600125, INCB18424 and ruxolitinib) were identified.
Conclusions
Smoking reduction significantly lowered the risk of EM. Hub genes and therapeutic targets identified provided new avenues for mitigating harmful effects of smoking.