Update on menopause: An expert’s insight on pivotal studies
Confusion about what to do—on the part of both physicians and patients—may be the greatest consequence of recent studies.
The past 2 years have witnessed a flurry of scientific publications on menopause and related therapies, particularly use of the sex steroid hormones. In turn, attitudes about menopause and hormone therapy have changed. Perhaps the greatest consequence of all the attention is the confusion about what to do, on the part of both provider and patient.
Many organizations responded with considered, evidence-based, practical guidelines. The most detailed and practice-oriented of these guidelines is the North American Menopause Society’s (NAMS’s) September 2003 Position Statement on use of estrogen and progestogen in peri- and postmenopausal women (www.menopause.org). Even as this Update on Menopause is being written, the report of the terminated estrogen-only arm of the Women’s Health Initiative (WHI) is in press and may further change clinical practice. NAMS will present an updated report on all these developments at the 2004 scientific meeting in Washington, DC, October 6 to 9, 2004. In the interim, the current recommendations hold, and the following publications are of clinical relevance.
WHIHigher levels of exercise reduce breast cancer risk
McTiernan A, Kooperberg C, White E, et al. Recreational physical activity and the risk of breast cancer in postmenopausal women: the Women’s Health Initiative Cohort Study. JAMA. 2003;290:1331–1336.
- LEVEL II-2 EVIDENCE: COHORT OR CASE-CONTROLLED TRIAL
The risk of breast cancer in postmenopausal women who exercised moderately for only a few hours a week was reduced by 18% compared with inactive women—and risk was reduced more in women who exercised moderately but for considerably more hours per week.
A total of 74,171 postmenopausal women aged 50 to 79, with no history of breast cancer, were enrolled. At a mean follow-up of 4.7 years, an increasing total current physical activity score was associated with a statistically significant reduced risk for breast cancer (P = .03 for trend). The women in whom the 18% (95% confidence interval [CI], 0.68-0.97) reduced risk of breast cancer was observed exercised the equivalent of 1.25 to 2.5 hours per week of brisk walking (5.1-10.0 metabolic hours). Women who exercised the equivalent of 10 or more hours of brisk walking per week had slightly greater reductions.
The greatest benefit was in women with a body mass index (BMI) below 24.1, but benefits were seen in women with BMIs ranging from 24.1 to 28.4. In evaluating the effect of previous strenuous-intensity exercise, a statistically significant decreased risk of breast cancer was seen for women who had engaged in strenuous exercise at age 35 (relative risk [RR], 0.86; 95% CI, 0.78-0.95); no significant associations were found for strenuous exercise at ages 18 or 50.
Modest protection, but encourage exercise anyway
This large, prospective cohort study performed in the mid-1990s strengthens the growing body of evidence that higher levels of physical activity afford modest protection against breast cancer. Recreational physical activity appears to be associated with reduced risk for breast cancer in postmenopausal women; longer exercise durations showed only slightly greater reduction in risk.
The strengths of this study are its large numbers, prospective nature, and detailed reporting of breast cancer outcomes. Limitations include possible confounders such as prior oral contraceptive use, and use of self-administered questionnaires to estimate physical activity.
One very important question is raised by this study: Given the low increase in absolute risk of breast cancer reported by the WHI with estrogen plus progestin1—which barely reached statistical significance (total breast cancer RR, 1.24; 95% CI, 1.02-1.50)—and given the statement in the McTiernan study that “the reduced risk associated with increased levels of total physical activity was seen across all the categories of these variables” (including current or past use, or no previous use of hormone therapy), does the reduction of incidence with physical activity in hormone therapy users lower the level of risk to non-significance or to that of nonexercisers in the placebo group? The answer cannot be determined from this report, but it would be illuminating.
These findings are preliminary, and confirming studies are needed. There is little harm in encouraging women to exercise, however.
1. Chlebowski RT, Hendrix SL, Langer RD, et al. Influence of estrogen plus progestin on breast cancer and mammography in healthy postmenopausal women: the Women’s Health Initiative Randomized Trial. JAMA. 2003;289:3243-3253.
WHIEstrogen-progestin has no significant effect on gynecologic cancers
Anderson GL, Judd HL, Kaunitz AM, et al, for the Women’s Health Initiative Investigators. Effects of estrogen plus progestin on gynecologic cancers and associated diagnostic procedures: the Women’s Health Initiative randomized trial. JAMA. 2003;290:1739–1748.
- LEVEL I EVIDENCE: RANDOMIZED, CONTROLLED TRIAL
Continuous combined estrogen plus progestin therapy (EPT) does not have a statistically significant effect on either ovarian or endometrial cancer compared with placebo, according to this report.
In this randomized, double-blind, placebo-controlled trial, 16,608 women were assigned to either EPT (0.625 mg/day conjugated equine estrogens plus 2.5 mg/day medroxyprogesterone acetate) or placebo; none of the women had undergone a hysterectomy.
After an average follow-up of 5.6 years:
- There were 20 cases of invasive ovarian cancer in the EPT group (n = 8,506) and 12 cases in the placebo group (n = 8,102). Compared with placebo, the hazard ratio (HR) for invasive ovarian cancer among EPT recipients was a nonsignificant 1.58 (95% CI, 0.77-3.24 [adjusted 95% CI, 0.59-4.23]).
- For endometrial cancer, 27 and 31 cases occurred, respectively, which translated statistically to a nonsignificant hazard ratio for EPT recipients of 0.81 (95% CI, 0.48-1.36).
- For cervical cancers, 8 and 5 cases were reported, respectively, with a nonsignificant HR of 1.44 (95% CI, 0.47-4.42).
Bias against hormone therapy?
The authors concluded that EPT may increase the risk of ovarian cancer but has no significant effect on the risk of endometrial cancer. They commented, however, that, since the EPT arm of the trial was prematurely stopped, the precision of the results is limited and examination of longer-term exposure is precluded.
This paper once again raises the question of whether the writers of the WHI trial have a bias against hormone therapy. In this report, the EPT arm of the WHI trial had an observed annual incidence of 34 ovarian cancer cases per 100,000 person-years—somewhat less than the anticipated population-based rate of 45 per 100,000 person-years. In the authors’ words, the ovarian cancer rate in the EPT group “was elevated (HR 1.58; 95% CI 0.77-3.24 [adjusted 95% CI, 0.59-4.23]) but not statistically significant.” The Kaplan-Meier estimates of cumulative hazards also did not reach statistical significance. Yet in the conclusion of the abstract, the authors state that continuous combined EPT “may increase the risk of ovarian cancer while producing endometrial cancer rates similar to placebo.”
Regarding the conclusion on endometrial cancer risk, the observed incidence for EPT users was 62 per 100,000 person-years, which is also lower than the anticipated population-based rate of 83 per 100,000 person-years. The authors state that this was a “small, nonsignificant reduction” in endometrial cancer risk (HR 0.81; 95% CI, 0.48-1.36). Yet, in the conclusion, while claiming that the nonsignificant difference in ovarian cancer suggests an increased risk, the authors do not state that the nonsignificant reduction in endometrial cancer suggests a decreased risk.
What do the authors expect us to believe, their data or their conclusions? My interpretation of the data in this article is that ovarian and uterine cancers need not be of major concern when determining a woman’s risk-benefit ratio for hormone therapy.
A retraction. Of interest, when this issue was raised in subsequent JAMA correspondence, the WHI authors agreed—representing perhaps the first time that a WHI report publicly retracted a potentially biased conclusion.1,2
Fewer biopsies will be needed with lower dosage. It is not surprising that women taking hormonal therapy containing estrogen had more bleeding and, therefore, more endometrial biopsies than women taking placebo, because a known effect of estrogen is proliferation of the endometrial lining. With the lower-dose hormonal preparations currently available (which result in lower systemic estrogen levels and less endometrial stimulation), uterine bleeding episodes in menopausal hormone therapy users should diminish, along with the number of endometrial biopsies.
1. Utian WH. Hormone therapy and risk of gynecologic cancers [letter]. JAMA. 2004;291:42.-
2. Anderson GL, Judd HL, Kaunitz AM, et al. Hormone therapy and risk of gynecologic cancers—Reply. JAMA. 2004;291:43.-
MILLION WOMEN STUDYBreast cancer risks increased by estrogen plus progestogen
Beral V; Million Women Study Collaborators. Breast cancer and hormone-replacement therapy in the Million Women Study. Lancet. 2003;362:419–427.
- LEVEL II-2 EVIDENCE: COHORT OR CASE-CONTROLLED
Current use but not past use of postmenopausal hormone therapy is associated with an increased risk of incident and fatal breast cancer, especially for estrogen-progestogen therapy (EPT), according to this large observational study from Britain. Risks increased among current users as total duration of use increased.
A total of 1,084,110 women aged 50 to 64 were enrolled between May 1996 and March 2001 and followed to the study finish (end of 2002). Mean follow-up was 2.6 years for breast cancer incidence and 4.1 years for mortality. Nearly half of the women had used postmenopausal hormone therapy, either estrogen therapy alone (ET) or EPT. Primary endpoints were diagnosis of breast cancer and death from breast cancer.
Current ET or EPT use (compared with nonuse) was associated with a statistically significant increased risk of both breast cancer incidence (RR, 1.66; 95% CI, 1.58-1.75) and breast cancer mortality (RR, 1.22; 95% CI, 1.00-1.48). Past use did not increase the risk of incident (RR, 1.01; 95% CI, 0.94-1.09) or fatal disease (RR, 1.05; 95% CI, 0.82-1.34), and the risk decreased with time since last use.
The risks associated with ET and with EPT differed significantly. Current ET users had a 30% increased risk for breast cancer (95% CI, 1.21-1.40) while current EPT users had a 100% increased risk (95% CI, 1.88-2.12). However, vaginal or other local EPT formulations did not increase the risk (RR 0.67, 95% CI, 0.30-1.49). No significant differences in risk were found between specific types or doses of EPT or between continuous combined and continuous cyclic regimens.
Limitations of observational studies
This extremely large observational study found levels of breast cancer risk associated with ET and EPT similar to those reported by the WHI and as predicted in 1997 by the Collaborative Group on Hormonal Factors in Breast Cancer study.1 The Million Women Study implicates an expanded number of ET and EPT products and routes of administration.
Acting as devil’s advocate, I will point out that this is an observational study with large potential for error. The major weakness is that it is a snapshot of hormone therapy use taken at the time of the women’s entry into the study, which was at the time of their 3-year mammogram. No further information was gleaned from the women regarding subsequent changes in hormone therapy use, such as whether they terminated use or changed the dose or route.
Also, the patient-provided data at entry showed a 96% agreement with the actual prescription written by the physician. The 4% variance, although it seems small, is of some concern given the narrow difference in relative risks and the large number of study participants. This is a weakness of any observational study; even if the prescription is filled, evidence that it was actually taken is inadequate.
Finally, the authors report that current use of hormone therapy at baseline increased the risk of breast cancer, although the relative risk was not as large as for disease incidence. They were not able to come up with reliable estimates of mortality attributable to breast cancer.
“Million Women” can be accepted only as an observational confirmation of a small increase in absolute risk of breast cancer.
In conclusion, the Million Women Study can be accepted only as an observational study providing confirmation of a small increase in the absolute risk for breast cancer in women on hormone therapy.
Further implications. The suggestion that these results apply to products beyond those tested in the WHI is in agreement with the NAMS Advisory Panel’s 2003 statement on Postmenopausal Hormone Therapy, which supports the view that although it is not possible to make general conclusions about all members of the estrogen and progestogen families, an improved benefit-risk profile of other EPT agents cannot be assumed.
1. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet. 1997;350:1047-1059.