Related Experiment Videos
Tibolone and endometrial cancer: a cohort and nested case-control study in the UK
Corinne S de Vries1, Susan E Bromley, Hilary Thomas
1Department of Pharmacoepidemiology, Postgraduate Medical School, University of Surrey, Guildford, UK. c.de-vries@surrey.ac.uk
Objective:
Case series and spontaneous reports of endometrial cancer have raised the question as to whether the use of tibolone (introduced into the UK in 1991) is associated with an increased risk of endometrial cancer. This study set out to evaluate whether tibolone use is associated with an increased risk of endometrial cancer.
Methods:
Age-adjusted incidence rate ratios (IRRs) of endometrial cancer were calculated for tibolone use compared with the use of other hormone replacement therapy (HRT). Separate sets of controls, matched for age and general practice, were compared with cases, all nested within a cohort of HRT users identified from the UK General Practice Research Database (GPRD). Conditional logistic regression analysis, adjusted for potential confounders, was used to study the association between tibolone use and the risk of endometrial cancer.
Results:
4995 women used tibolone as their first HRT product; 10 783 (4.3%) of the users of combined HRT had changed to tibolone at some time during the study period. Amongst women whose HRT began with tibolone, the age-adjusted IRR relative to those who started with combined sequential HRT was 1.83 (95% CI 1.19, 2.82). The nested case-control study comprised 162 cases, each matched to two sets of 972 controls. There were 43 tibolone-exposed subjects, 28 of whom had used other HRT before or after tibolone. The adjusted odds ratio of the risk of endometrial cancer in women who had ever used tibolone, compared with users of combined sequential HRT, was 1.54 (95% CI 1.03, 2.32) in the age-matched set and 1.58 (95% CI 1.01, 2.47) in the practice-matched set. Sensitivity analyses did not decrease the risk estimates found.
Discussion:
Tibolone may be associated with an increased risk of endometrial cancer compared with conventional forms of HRT, but our data are fragile. Residual bias and uncontrolled confounding cannot be excluded, and follow-up time is insufficient to draw any firm conclusions with respect to the endometrial safety of tibolone.
