The influence of hormone therapies on type I and II endometrial cancer: A nationwide cohort study
Menée au Danemark à partir de données portant sur 914 595 femmes âgées de 50 à 79 ans (durée de suivi : 14 ans), cette étude évalue l'effet de traitements hormonaux substitutifs sur le risque de cancer de l'endomètre par type histologique
Résumé en anglais
The influence of hormone therapy (HT) on risk for endometrial cancer is still casting which type of HT the clinicians recommend. It is unrevealed if HT has a differential influence on Type I versus Type II endometrial tumors, and little is known about the influence of, e.g., different routes of administration and about the influence of tibolone. We followed all Danish women aged 50–79 years without previous cancer or hysterectomy (n = 914,595) during 1995–2009. From the National Prescription Register, we computed HT exposures as time-dependent covariates. Incident endometrial cancers (n = 6,202) were identified from the National Cancer Registry: 4,972 Type I tumors and 500 Type II tumors. Incidence rate ratios (RRs) and 95% confidence intervals (Cls) were estimated by Poisson regression. Compared with women never on HT, the RR of endometrial cancer was increased with conjugated estrogen: 4.27 (1.92–9.52), nonconjugated estrogen: 2.00 (1.87–2.13), long cycle combined therapy: 2.89 (2.27–3.67), cyclic combined therapy: 2.06 (1.88–2.27), tibolone 3.56 (2.94–4.32), transdermal estrogen: 2.77 (2.12–3.62) and vaginal estrogen: 1.96 (1.77–2.17), but not with continuous combined therapy: 1.02 (0.87–1.20). In contrast, the risk of Type II tumors appeared decreased with continuous combined therapy: 0.45 (0.20–1.01), and estrogen therapy implied a nonsignificantly altered risk of 1.43 (0.85–2.41). Our findings support that continuous combined therapy is risk free for Type I tumors, while all other hormone therapies increase risk. In contrast, Type II endometrial cancer was less convincingly associated with hormone use, and continuous combined therapy appeared to decrease the risk.
