Endometrial cancer and hormone-replacement therapy in the Million Women Study

Abstract

Background: Postmenopausal women who use hormone-replacement therapy (HRT) containing oestrogen alone are at increased risk of endometrial cancer. To minimise this risk, many HRT users who have not had a hysterectomy use combined oestrogen-progestagen preparations or tibolone. Limited information is available on the incidence of endometrial cancer in users of these therapies.

Methods: 716,738 postmenopausal women in the UK without previous cancer or previous hysterectomy were recruited into the Million Women Study in 1996-2001, provided information about their use of HRT and other personal details, and were followed up for an average of 3.4 years, during which time 1320 incident endometrial cancers were diagnosed.

Findings: 320,953 women (45%) reported at recruitment that they had used HRT, among whom 69,577 (22%) last used continuous combined therapy (progestagen added daily to oestrogen), 145,486 (45%) last used cyclic combined therapy (progestagen added to oestrogen, usually for 10-14 days per month), 28,028 (9%) last used tibolone, and 14,204 (4%) last used oestrogen-only HRT. These HRT types had sharply contrasting effects on the overall risk of endometrial cancer (p<0.0001 for heterogeneity). Compared with never users of HRT, risk was: reduced with last use of continuous combined preparations (relative risk 0.71 [95% CI 0.56-0.90]; p=0.005); increased with last use of tibolone (1.79 [1.43-2.25]; p<0.0001) and oestrogen only (1.45 [1.02-2.06]; p=0.04); and not significantly altered with last use of cyclic combined preparations (1.05 [0.91-1.22]; p=0.5). A woman's body-mass index significantly affected these associations, such that the adverse effects of tibolone and oestrogen-only HRT were greatest in non-obese women, and the beneficial effects of combined HRT were greatest in obese women.

Interpretation: Oestrogens and tibolone increase the risk of endometrial cancer. Progestagens counteract the adverse effect of oestrogens on the endometrium, the effect being greater the more days every month that they are added to oestrogen and the more obese that women are. However, combined oestrogen-progestagen HRT causes a greater increase in breast cancer than the other therapies do. Thus, when endometrial and breast cancers are added together, there is a greater increase in total cancer incidence with use of combined HRT, both continuous and cyclic, than with use of the other therapies.