Hormone Replacement Therapy Update
March 9, 2018
Hormone replacement therapy update
Bottom line: Hormone therapy does not cause cancer or increase mortality. On the
contrary, hormone therapy has many benefits and has been shown to decrease mortality.
The Women’s Health Initiative (WHI) was a randomized controlled primary prevention trial. In the estrogen plus progestin component of the WHI, 16608 postmenopausal women aged 50-79 (mean age 63) with intact uterus, were given conjugated equine estrogens (CEE) 0.625mg/d + medroxyprogesterone acetate (MPA) 2.5mg/d or placebo. There was also an estrogen only component comprised of 10,739 women with hysterectomy that were randomized to receive CEE 0.625mg/d alone or placebo. Parameters evaluated were: CHD (primary outcome), hip and other fractures, invasive breast CA (primary adverse outcome), stroke, PE, endometrial/colorectal/other cancers, death due to other causes. In 7/2002, the CEE+MPA component of the WHI was stopped after 5.2 years (planned duration of 8.5 years) because the risks were thought to exceed the benefits for CHD, stroke, invasive breast CA, venous thromboembolism.1. Nominal 95% confidence intervals (CIs) were used to determine increased risk. They apply to simple trials for single outcomes. They do not account for the
multiple statistical testing issues that occurred in this trial. Adjusted 95% CIs use group sequential methods to correct for multiple analyses over time. When adjusted 95% CIs used, the only parameter that reached statistical significance is deep venous thrombosis (increased risk) and non-hip/vertebral osteoporotic fractures (decreased risk).
Increased risk of DVT was also previously seen when comparing CEE+MPA to placebo in the1998 HERS study. The HERS study also showed no increased risk in MI, CHD death, cancer,stroke or total mortality.
Importantly, differences in risk of VTE are noted depending on the route of estrogen delivery and the type of progestin used. The 2007 French ESTHER study showed increased risk of VTE with oral estrogens and norpregnanes. By contrast, no increased risk was seen with transdermal estrogen, micronized progesterone or pregnane derivatives (including MPA). In 2012, the Million Women Study (MWS) showed increased risk of VTE with medroxyprogesterone acetate.
The 1995 PEPI trial showed unopposed estrogen increased HDL by 5.6 mg/dl and lowered fibrinogen without affecting blood pressure or insulin. Adding micronized progesteronedampened the HDL increase somewhat at 4.1mg/dl with no increased risk of endometrialhyperplasia. MPA had the smallest increase in HDL at 1.6mg/dl.
In a small study of postmenopausal women with coronary artery disease, estrogen withprogesterone increased exercise time to myocardial ischemia, whereas MPA did not.6 Primatestudies show that progesterone had a protective effect on coronary hyperreactivity (persistent severe vasoconstriction), whereas MPA caused prolonged vasoconstriction. Also, MPA reversed the estrogen-mediated protection from coronary vasospasm, whereas progesterone did not.7 Progesterone has been shown to be neuroprotective, while MPA has not.
A 2007 secondary analysis of the WHI evaluated hormone therapy and cardiovascular diseaserisk based on age and years since menopause. Statistical significance was only reached in women who started HT 20 or more years after menopause with a HR=1.28 (95% CI=1.03-1.58, P=.02).
In 2005 E3N-EPIC cohort suggested that synthetic progestins but not micronized progesterone increased breast cancer risk.
In 2011, the Million Women study showed an increased risk of breast cancer if HRT (E2 alone or E2/P) started before or after 5 years after menopause.
In a series of 5 articles between 2011 and 2013, Shapiro et al demonstrated that neither the Collaborative Reanalysis nor the WHI nor the MWS were able to establish whether or not HRT causes breast cancer because they did not adequately satisfy the criteria of causation (time order, bias, confounding, statistical stability and strength of association, dose/duration response, internal consistency, external consistency, biological plausibility).
Million Women study showed risk of endometrial cancer to be lowest in women taking combined continuous HT with a RR of 0.71 (95% CI=0.56-0.90, P=0.005), lower than even non-users who had a RR of 1.0. Confirming other studies, unopposed estrogen users had an increased risk with a RR of 1.45 (95%CI=1.02-2.06, P=0.04).
Progesterone provides more benefits than just endometrial protection and as such, we recommend progesterone even in women who have undergone hysterectomy. Progesterone improves bone density by increasing osteoblast numbers and promoting osteoblast maturation and differentiation.18 Progesterone improves sleep and anxiety. And as mentioned above, progesterone has a neuroprotective effect.
In a 2013 study, the effects of estrogen avoidance in hysterectomized women aged 50-59 years was examined. They concluded that between 2002 to 2011, between 18,601 and 91,610 postmenopausal women (the actual toll being closer to 40,000-48,000) died prematurely because of not using estrogen therapy. Despite the 2004 and 2011 reports from the WHIestrogen alone study showing reduction in mortality of 13 per 10,000 person years, decline in the use of estrogen therapy did not reverse. The primary risk reduction was through decreased coronary heart disease but there was also a reduction in cancer deaths.
In 2017, the data from the WHI was analyzed to look at long-term all-cause and cause-specific mortality. During an 18 year follow-up, no risk of of all-cause, cardiovascular, or cancer mortality was seen with either the CEE+MPA or CEE alone component.