Editorial

A personal initiative for women's health: to challenge the Women's Health Initiative

A. R. Genazzani, Editor-in-Chief, and M. Gambacciani, Menopause Clinic, Pisa, Italy

In the USA, a randomized, placebo-controlled, double-blind trial to evaluate the effects of a combination estrogen-progestogen therapy in postmenopausal women was stopped after 5.2 years by the Women's Health Initiative (WHI) because the health risks were considered higher than the health benefits by a small margin. The results were rapidly published in the Journal of the American Medical Association1. The trial was part of a 40 center study investigating the effects of hormone replacement therapy (HRT) on heart disease and breast cancer, having as secondary end-points the incidence of endometrial and colorectal cancer and fractures in postmenopausal women aged 50 to 79 years. Of the 16 608 women in the study, 8506 were randomized to receive oral conjugated estrogens (0.625 mg die) plus progestogen (medroxyprogesterone acetate (MPA), 2.5 mg die) and 8102 to receive placebo.
Unfortunately, this large trial was conducted on a population with totally different characteristics from those of women typically considered for HRT. In fact, the WHI was not conducted in "healthy postmenopausal women" as repeatedly stated in the paper and the accompanying Editorial1,2. In this 'gold standard' of research (the randomized clinical trial), the population was not ideal to harvest results that ". . . will help inform medical practice and guide millions of women in making critical decisions about their health as they grow older". In fact, a significant proportion of these women were less healthy than the authors appear to suggest: 35% were treated for hypertension, 35% were overweight (BMI 25-29) and 34% obese (BMI > 30), 4% had diabetes, 12.5% had elevated cholesterol levels requiring medication, 6.9% used statins. In the way that they have been reported by the lay press, the WHI study findings of an increased risk of breast cancer, cardiovascular disease, strokes, and deep venous thrombosis are likely to cause disproportionate alarm amongst women. First of all it should be noted that the small increase in the number of patients with breast cancer is in accordance with previous population studies3. The unexpected result was the increase in cardiovascular events, in contrast to all the evidence of experimental, epidemiological and observational studies suggesting that postmenopausal HRT can reduce coronary heart disease4. Generally speaking, the data of a given study describe the risk for the entire population, and the figures reported for the population do not refer to the individual excess risk for a single woman.
In stopping the trial, the WHI investigators noted that women using HRT exhibited a 26% increase in invasive breast cancer (i.e. 38 cases among HRT users versus 30 cases among placebo users per 10 000 person-years), with no significant difference observed for in situ breast cancers, and a 29% increase in CHD events compared with placebo (i.e. 37 cases among HRT users versus 30 among placebo users per 10 000 person-years). In agreement with data already reported from experimental, epidemiological and observational studies4, the HRT-treated group experienced a two-fold rate of venous thromboembolism compared with placebo (34 versus 16 per 10 000 person-years). Endometrial cancer, lung cancer, and total cancer incidences were not affected, as is well known from the literature3. The total death rate, and the death rate for CVD or breast cancer were equally not affected by HRT.
According to the WHI authors, the results of this trial indicate that the average risk in an individual woman is 0.1% per year for breast cancer and for heart attacks, and there is no change in death rates even if the trend in death rates is favourable to HRT1. This WHI publication did not consider, in its overall benefit-risk analysis, a variety of other conditions that may be positively or negatively affected by HRT, including gallbladder disease, diabetes, cognitive function, and quality of life.
In the women who had never been treated with hormones before entering the study, the 5.2 years of HRT did not induce an excess risk in breast cancer. The increased breast cancer risk was seen in women previously exposed to 5-10 years of hormones. Previous exposure to hormones may also have played a role in fracture rates. Women under HRT had a 34% reduction in hip fractures and 24% reduction for total fractures. This is the first solid evidence from a randomized clinical trial that HRT prevents fractures. However, the effect of 5.2 years of HRTwas astonishing in this population at presumably low risk of fractures (on the basis of BMI). Moreover, the fracture rate may have been positively influenced by the fact that a number of women had been previously treated with hormones. On the other hand, the fracture data are in line with the proven efficacy of HRT in preventing bone loss that can lead to osteoporosis, as recently demonstrated in a randomized clinical trial demonstrating that both standard and lower doses of HRT relieve menopausal symptoms and prevent osteoporosis5,6. The WHI demonstrates that HRT can determine a 37% reduction in the risk of colorectal cancer, which is very important due to the relevance of this cancer for women's health and wellbeing3.
The authors stated that the WHI results have broad applicability. In contrast, we believe that the limitations of age, health status and other characteristics remain. The HRT regimen consisted of a daily pill containing 0.625 mg of conjugated equine estrogens and 2.5 mg of MPA.
This combination is one of the most commonly used around the world, particularly in postmenopausal women, for the prevention and treatment of climacteric complaints. Therefore, this treatment is intended for postmenopausal women at least 10-15 years younger than the WHI population. The WHI data and the subsequent recommendations may not apply to other forms of HRT with different estrogens and progestagens, or indeed to other doses of conjugated estrogens and MPA.
HRT is a collective name, but not all types of HRT are the same. Firstly, different age groups have different requirements for estrogen doses, with older women requiring lower doses6,7. Basically in clinical practice, women with a similar age range to the WHI study population would hardly ever be treated with products with a potency similar to the drug used in the study7-9. Although the authors of the WHI paper state that the decision about HRT should be personalized, the clinical management of WHI participants is open to critical questioning. For example, why were 79-year-old women treated with standard doses of HRT? We would never choose that kind of combination for our elderly patients with those clinical characteristics. As gynecological endocrinologists, we have always personalized our selection of therapies long before the WHI publication. In Europe, we have a dozen different progestogens that can make a difference in the outcome, as well as different estrogens for oral, transcutaneous, transdermal, nasal and parenteral routes. In our view, the WHI study raises some serious concerns only for certain categories of postmenopausal women: primarily the obese and the over 65.
The interim results of the WHI trial so far available are by no means conclusive, and we feel it unwise to make ultimate recommendations for clinical management based upon a study performed on a completely different population than that seen in our clinics. HRT is one of the most widely used and extensively studied of all drug therapies and has been used for more than 60 years by over 100 million postmenopausal women. As stated in the NIH news release, for women taking combination HRT for the short-term treatment of symptoms, the benefits are likely to outweigh the risks. We believe that there is currently sufficient evidence overall to support carefully considered HRT use in individuals who need hormones for their symptoms of hot flashes, night sweats, the dryness associated with vaginal atrophy, quality of life and for prevention of osteoporosis.
The biological plausibility of cardiovascular event prevention is not altered by this paper. Most of all, randomized clinical trials do not and cannot eradicate the results of huge amount of experimental, epidemiological and observational data. The possible confounding biases in the observational studies of HRT and coronary heart disease, lead part of the scientific community to reject the already available positive consistent findings on HRT and cardiovascular disease. These scientists rely only on randomized clinical trials assuming that they are the gold standard and that biases are relevant only for other study designs. Large observational studies that have sufficient numbers to correct for potential biases still demonstrate the HRT benefit on blood lipids, vessel wall, and cardiovascular events. The negative or null HRT effect on coronary heart disease reported in the WHI trial, as well as in the secondary prevention HERS trial10,11, may be satisfactorily explained by the characteristics of the studied population (age, health status, concomitant medications). Further randomized clinical trials of HRT for primary or secondary prevention of coronary heart disease are needed. However, we must emphasize that these studies should be performed in women that are appropriate candidates for HRT in real clinical practice. Critically, these must also include different types and doses of HRT.
For the time being, therapeutic recommendations resulting from the WHI study should be restricted to populations similar to that included in this trial. Moreover, in elderly postmenopausal women, low-dose HRT combinations have been proven beneficial in controlling subjective symptoms and preventing osteoporosis and, therefore, there is no need for the doses used in the WHI. This study also included hysterectomized women treated with estrogen alone; the NIH states that there is currently no evidence of increased risk of breast cancer in these women. Therefore, no change is recommended in the management of hysterectomized women undergoing estrogen-only therapy.

REFERENCES

1. Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. JAm Med Assoc 2002;288:321-33

2. Fletcher SW, Colditz GA. Failure of estrogen plus progestin therapy for prevention. JAm Med Assoc 2002;288:366-8

3. Genazzani AR, Gadducci A, Gambacciani M. Controversial issues in climacteric medicine II. Hormone replacement therapy and cancer. International Menopause Society Expert Workshop, 9-12 June 2001, Opera del Duomo, Pisa, Italy. Climacteric 2001;4:181-93

4. Genazzani AR, Gambacciani M. Hormone replacement therapy: the perspectives for the 21st century. Cardiovascular disease and hormone replacement therapy. International Menopause Society Expert Workshop, 13-16 October 2000, Royal Society of Medicine, London, UK. Climacteric 2000;3:233-40

5. Lindsay R, Gallagher JC, Kleerekoper M, Pickar JH. Effect of lower doses of conjugated equine estrogens with and without medroxyprogesterone acetate on bone in early postmenopausal women. J Am Med Assoc 2002;287:2668-76

6. Gambacciani M, Monteleone P, Genazzani AR. Low-dose hormone replacement therapy: effects on bone. Climacteric 2002;5:135-9

7. Ettinger B. Personal perspective on low-dosage estrogen therapy for postmenopausal women. Menopause 1999;6:273-6

8. Lobo RA, Whitehead MI. Is low-dose hormone replacement therapy for postmenopausal women efficacious and desirable? Climacteric 2001;4:110-19

9. Gambacciani M, Genazzani AR. Hormone replacement therapy: the benefits in tailoring the regimen and dose. Maturitas 2001;40:195-201

10. Grady D, Herrington D, Bittner V, et al., for the HERS Research Group. Heart and estrogen/ progestin replacement study follow-up (HERS II): Part 1. Cardiovascular outcomes during 6.8 years of hormone therapy. JAm Med Assoc 2002;288:49-57

11. Hulley S, Furberg C, Barrett-Connor E, et al. Noncardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/ progestin Replacement Study follow-up (HERS II). J Am Med Assoc 2002;288:58-66

• Go to the institutional page of Gynecological Endocrinology on the Parthenon Publishing Group website.
  On this website you will also have access to the back issues of the journal.