Best Practice & Research Clinical Obstetrics and Gynaecology 23 (2009) 87–107
Contents lists available at ScienceDirect
Best Practice & Research Clinical Obstetrics and Gynaecology journal homepage: www.elsevier.com/locate/bpobgyn
7
Cancer issues Jo Marsden, BSc, FRCS(Gen Surg), MD, Consultant Surgeon a, David Sturdee, MD, FRCOG, Consultant Gynaecologist b, * a b
King’s Breast Care, King’s College NHS Hospital, Denmark Hill, London SE5 9RS, UK Department of Obstetrics and Gynaecology, Solihull Hospital, Heart of England NHS Foundation Trust, Solihull B91 2JL, UK
Keywords: HRT breast cancer endometrial cancer hormones and cancer
The potential for hormone therapy to cause cancer is the greatest fear for postmenopausal women considering hormone replacement therapy (HRT). Breast cancer is the most common female malignancy, for which HRT is one of many modifiable risk factors, often attracting disproportionate attention. Randomized controlled trials have confirmed that in postmenopausal women aged 50–59 years taking combined oestrogen and progestogen HRT over 5 years, there will be three extra cases of breast cancer per 1000 women. With the use of unopposed conjugated equine oestrogens, there would be four fewer cases over the same time. Women can be advised that the risk of breast cancer is not significantly increased with up to 3 years of combined HRT and up to 5 years of unopposed oestrogen. Unopposed oestrogen increases the risk of endometrial hyperplasia and carcinoma significantly, and this is dose and duration dependent. The addition of progestogen prevents the proliferative effect of oestrogen on the endometrium, and may even reduce the risk of endometrial cancer compared with non-users if given continuously. The use of combined oral contraception in premenopausal women also reduces the risk of endometrial cancer but increases the risk of cervical carcinoma significantly. HRT does not influence the risk of cervical cancer. Epithelial ovarian cancer risk may be slightly increased with longterm use of unopposed oestrogen, is not altered by the addition of progestogen, and is reduced significantly in users of combined oral contraception. The mechanism for these effects is not understood. Colorectal cancer and possibly lung and gastric cancers are reduced by the use of HRT. Apart from a slight increased risk of gallbladder disease and carcinoma with HRT, there are no data linking oestrogen or progestogen with any other malignancies. Ó 2008 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ44 121 424 5390; Fax: þ44 121 424 5389. E-mail address:
[email protected] (D. Sturdee). 1521-6934/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.bpobgyn.2008.10.005
88
J. Marsden, D. Sturdee / Best Practice & Research Clinical Obstetrics and Gynaecology 23 (2009) 87–107
There is a longstanding history that female sex steroids may be carcinogenic, and there is much evidence that oestrogen, whether it be endogenous or of exogenous origin, is implicated in the development of some breast and endometrial cancers in particular. In recent years, large randomized controlled studies such as the Women’s Health Initiative (WHI) from North America have provided valuable data, but unfortunately the manner in which the findings have been publicized, particularly in the mass media, have caused widespread and often inappropriate anxiety and controversy. Many women and their doctors in primary care in the UK are now convinced that hormone replacement therapy (HRT) causes breast cancer, and a large number of doctors will refuse to prescribe HRT because it is too dangerous. The method of presentation of the data can give different perspectives depending on whether the relative increased risk or the absolute number of cases per 10 000 women per year, for example, are used. Relative risk (RR) is a difficult concept to comprehend, and requires a knowledge of the underlying risk of the condition. Receptors for oestrogen and progesterone are present in many of the tissues of the body, particularly those of the female genital tract and breast, so it is not surprising that situations where these hormones are increased will result in stimulation of some tissues and possibly pathological changes. This chapter will consider organs and tissues that may be affected by female sex steroids, and in particular the current data on the impact of HRT on the incidence of cancer. Breast cancer Breast cancer is the most common female malignancy diagnosed in the UK with an overall lifetime risk of one in nine. The incidence is age related, rising rapidly in the fourth and fifth decades and then continuing to increase but not as steeply after the menopause, implicating endogenous sex hormones in its development.1 This is supported by epidemiological evidence and large randomized endocrine breast cancer chemoprevention and treatment trials showing a reduction in the incidence and improvement in the survival of hormone-sensitive (i.e. oestrogen-receptor positive) breast cancer, respectively.2 However, the aetiology of breast cancer is complex due to interactions between reproductive (hormonal), lifestyle and genetic factors whose interplay is poorly understood. HRT is one of many modifiable risk factors identified (others include obesity and moderate/heavy alcohol intake) but it often attracts disproportionate attention. HRT and breast cancer diagnosis Overwhelmingly, clinical evidence evaluating the impact of HRT on breast cancer is from observational case-controlled or cohort studies. Randomized trials have been conducted but, with the exception of the WHI, are either underpowered for reliable conclusions to be drawn on breast cancer incidence or this has not been an a-priori study hypothesis. The findings of these latter studies and their limitations are summarized in a comprehensive review by Collins et al.3 In 1997, the Collaborative Group for Hormonal Factors in Breast Cancer performed a re-analysis of worldwide observational studies of HRT, concluding that breast cancer risk was only increased with long-term exposure (i.e. for more than 5 years). The degree of risk was estimated to be equivalent to that of delaying the onset of the menopause, and a fall in risk following HRT cessation supported a tumour growth promoting effect rather than initiation of malignant transformation.4 Subgroup analysis suggested a greater risk with combined HRT compared with unopposed oestrogen, but as only 12% of women were exposed to the former, this was not considered conclusive. The placebo-controlled WHI confirmed that the addition of an oral continuous progestogen [i.e. 2.5 mg medroxyprogesterone acetate (MPA), daily] to postmenopausal oestrogen replacement [i.e. 0.625 mg conjugated equine estrogen, (CEE)] was responsible for the observed small increase in breast cancer risk.5–9 Risk was restricted to women who had been taking combined HRT prior to randomization, supporting the duration effect shown in earlier observational studies.6–9 In absolute numbers, for women aged 50–59 years using continuous combined HRT for 5 years, three extra breast cancers per 1000 women exposed would be anticipated (Table 1a). With unopposed CEE, four fewer breast cancers would be observed for the same duration of use in women of the same age group
J. Marsden, D. Sturdee / Best Practice & Research Clinical Obstetrics and Gynaecology 23 (2009) 87–107
89
Table 1 (a) Combined hormone replacement therapy (HRT) and invasive breast cancer diagnosis. O, oestrogen in the form of conjugated equine oestrogen; P, progestogen in the form of medroxyprogesterone acetate. Year of study publication
OþP number of events
Placebo number of events
Hazard ratio (95% confidence interval)
2002 Prior use of HRT Never
