Lessons Learned From the Women's Health Initiative Study
from Southern Medical Journal
A Gynecologist's View of Hormone Replacement
Therapy in Light of the Women's Health Initiative
Physicians form treatment biases based, for the most part, upon the medical
concerns of the population of patients they see. If a gynecologist sees
predominantly early-menopausal women, approximately fifty years of age, who are
otherwise healthy and without heart disease, and who are having hot flashes and
dyspareunia due to vaginal dryness, that physician develops an overall
impression about the utility of hormone replacement therapy (HRT) based upon
experience with that population. On the other hand, if an internist or family
practitioner sees more 65-year-old women who have already been through menopause
and who may already have hypertension and heart disease, that physician will
form a different impression as to the use and need, if any, for HRT. This bias
is known as Bayesian prior probability, and explains why physicians can have
markedly different views about the pros and cons of a specific medical therapy.
The recent early termination of a large, prospective clinical study, testing HRT
versus placebo using primary outcome measurements of coronary artery adverse
events and invasive breast cancer, illustrates this well.
Hormone replacement therapy (estrogen plus progestin) is one such treatment
about which professionals have different views. For many years, there have been
multiple retrospective, case control/cohort, primate, and laboratory studies
indicating that long-term estrogen replacement therapy (ERT) (often mixed with
HRT, but mostly ERT) is associated with a lower incidence of coronary artery
disease in women who did not previously have heart problems. Even from the
initial retrospective report indicating a cardioprotective effect of ERT, this
observation applied to the new occurrence of heart disease, not to women who
already have coronary heart disease (CHD).[1] Gynecologists have never
maintained that, once a woman develops coronary artery disease, estrogen therapy
helps slow its progression. Over the years and with the mass of positive
preventive data, however, nongynecologists began to prescribe HRT with some
frequency to women who already had heart disease in order to see if it could,
indeed, prevent progression. Finally, the Heart and Estrogen/Progestin
Replacement Study (HERS) was conceived to test prospectively whether women who
already had heart disease and whose uterus remained in place (to cause them to
need the addition of progestin to the estrogen) could benefit from HRT. The HERS
demonstrated a rise in CHD adverse events in the first year of the study, and a
reduction in adverse events in years 3 to 5 of the study, with a 0.99 overall
risk for CHD events.[2] Continued monitoring for several years after the study
showed the same thing: no reduced rate of CHD adverse events in women starting
HRT after a diagnosis of CHD.
The recent Women's Health Initiative (WHI) was set up to test both ERT and HRT
in relatively healthy women. One part of the study tested HRT in more than
16,000 women with an in situ uterus, whose average age at entry into the study
was 63 years, and about 7% of whom had previous coronary heart or vascular
problems. This arm of the study was terminated recently due to an aggregate
excess of adverse events.[3] The second arm of the WHI study, which is
prospectively testing ERT alone in women who have had a hysterectomy, is ongoing
as of this writing. We do not yet know the results, other than that monitoring
limits for total adverse events have not been exceeded.
The HRT part of the study was terminated prematurely after a mean of 5.2 years
of follow-up. Looking first at the adverse CHD events from that part of the
study, the risk ratio was 1.29 (95% confidence interval = 1.02-1.63) for women
taking HRT, which means there was a 29% increase, or a difference of 7/10,000
women-years (37 vs 30 per 10,000 person-years) of use. This translates to an
excess of 7 women per 1,000 using HRT for 10 years (ie, less than a 1% excess in
more than a decade of use).
It is my opinion that these increased CHD events are more likely due to the use
of medroxyprogesterone acetate than to estrogen replacement. It is known that
treatment with medroxyprogesterone acetate negates the positive coronary vessel
dilatation that estrogens produce.[4] Additionally, studies of primates have
shown that HRT/ERT needs to be started before the development of cardiovascular
disease and within about 5 to 6 years of menopause in order to have its
cardiovascular protective effect.[5] The same is true for maximum protection
from bone loss. This is the population in which gynecologists believe HRT and
ERT are most useful. To prevent these conditions, the therapy should be started
as soon after menopause as possible, or at least when women are in their 50s,
not for the first time when they are in their 60s.
If you read the WHI HRT study carefully, you will see that the main reason it
was stopped is that the incidence of breast cancer, one of the study's primary
outcome measurements, exceeded the limits of monitoring. It is not clear why
colon cancer and endometrial cancer were not considered primary outcomes, since
the evidence of risk reduction for those cancers with ERT and HRT is well
established. If the protective effect of HRT on those cancers had been included,
it would have changed the decision to stop the study. The study found a risk
ratio of 1.26, or a 26% increase (38 vs 30 per 10,000 person-years) of breast
cancer and a reduced risk ratio (0.63) for colorectal cancer rates. Colon
cancers were reduced by 37% (10 vs 16 per 10,000 person-years). Lung cancer,
endometrial cancer, and, notably, total cancer rates, were not different in the
HRT group versus those given placebo. In light of the result of no net increase
in overall cancer rate, I do not believe the outcome of this study is reason to
stop prescribing HRT to all women.
Nevertheless, there are other data to support the slight increase in the
incidence of breast cancer with the use of HRT/ERT (risk ratio, about 1.3), and
that the risk may increase when used for more than 5 to 10 years. I think that
gynecologists who have not been sure until now as to whether there is a real
increase in breast cancer from HRT need to change how they counsel patients. It
is wise from both a patient-care standpoint, as well as a medical/legal
standpoint, to inform a woman of what appears to be increased breast cancer
risks along with the other risks and benefits of taking HRT. Annual mammograms
for patients treated with HRT are mandatory.
At present, studies indicate that long-term ERT produces a small increase in
well-differentiated breast cancer lesions, but it also appears to reduce colon
cancer almost equally. Tamoxifen use is suspected of increasing the risk of
colorectal cancer,[6] and so far there is no evidence that the use of raloxifene
hydrochloride reduces colorectal cancer risk to the extent that estrogen use
does. Many nongynecologists are unaware of the colon cancer relationship, but
they should counsel patients about it if these alternatives are used.
Based on the Journal of the American Medical Association (JAMA) article and the
report of the discontinuance of the HRT arm of the WHI study, the standard of
care would be to inform menopausal women that HRT may produce a small absolute
increase in adverse cardiac or vascular events in the first year or two it is
used, and that it is associated with a small absolute increase in grade 1 breast
cancers. We have already been informing these patients of the small but real
increase in the risk of thrombophlebitis and pulmonary emboli. Informed consent,
however, should also include telling patients that there appear to be long-term
benefits, with reduced risk of cardiovascular disease, Alzheimer's disease,
osteoporosis, colon cancer and acute macular degeneration of the eye causing
blindness. If patients are symptomatic, short-term benefits of HRT include
improvement in the symptoms of hot flashes, genital atrophy and dryness, and
mood irritability. Therefore, the risks and benefits of HRT must be carefully
weighed for each individual woman, just as taking low-dose aspirin to prevent
coronary artery events must be weighed against adverse stroke events. The
physician should discuss these risks and benefits thoroughly with each woman.
I have been switching my patients who were taking HRT to combination therapy
that does not include medroxyprogesterone acetate as the progestin. That is a
personal preference, however, and the data are not clear. We must reevaluate the
situation when the results of clinical trials using other HRT combinations are
reported.
Frederick R. Jelovsek, MD
References
Hammond CB, Jelovsek FR, Lee KL, et al: Effects of long-term estrogen
replacement therapy. I. Metabolic effects. Am J Obstet Gynecol 1979; 133:525-536
Grady D, Herrington D, Bittner V, et al: Cardiovascular disease outcomes during
6.8 years of hormone therapy: Heart and Estrogen/Progestin Replacement Study
follow-up (HERS II). JAMA 2002; 288:49-57
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. JAMA
2002; 288:321-333
Adams MR, Register TC, Golden DL, et al: Medroxyprogesterone acetate antagonizes
inhibitory effects of conjugated equine estrogens on coronary artery
atherosclerosis. Arterioscler Thromb Vasc Biol 1997; 17:217-221
Clarkson TB: The new conundrum: do estrogens have any cardiovascular benefits?
Int J Fertil Womens Med 2002; 47:61-68
Rutqvist LE, Johansson H, Signomklao T, et al: Adjuvant tamoxifen therapy for
early stage breast cancer and second primary malignancies. Stockholm Breast
Cancer Study Group. J Natl Cancer Inst 1995; 87:645-651
