LAUSR

Although hormone replacement therapy (HRT) use is associated with elevated endometrial cancer(EC) risk, little evidence assesses potential effect-modifiers on HRT-related EC in a long-term follow-up. In this large-scale longitudinal cohort study, we tried to evaluate the association between different HRT types/methods use and risk of EC, and reveal this risk within different body mass index (BMI) groups. In whole cohort, 677 EC occurred during mean 11.6 years follow-up. Cox proportional hazards regression was used to estimate multivariable-adjusted hazards ratios (HRs) and 95% confidence intervals (CIs) with HRT status (never, former, or current) for risk of EC incidence. Current HRT use was not significantly associated with EC risk (HR for current vs. never HRT use: 1.13; 95% CI: 0.92, 1.38) in the whole cohort, but presented a dose-response effect on increased EC risk (HR for >10-year use vs. never HRT use: 1.73; 95% CI: 1.35, 2.21). Moreover, EC risk differed in distinct regimens or subsets (all Pinteraction < 0.05). Estrogen-only use was associated with elevated EC risk (HR for current vs. never HRT use: 1.51; 95% CI: 1.12, 2.04), but women with high BMI (> 30 kg/m2) who currently use estrogen-only harbored decreased EC risk (HR: 0.56; 95% CI: 0.38, 0.82) compared to counterparts without HRT use. Estrogen-only use is associated with increased EC risk, and precise monitoring of EC development for postmenopausal women with long-term HRT use are urgently needed. BMI could serve as an important surrogate to assess this risk.