Hormone therapy (HT), previously called hormone replacement therapy, involves administering estrogen or progesterone alone or in combination as a way to treat menopause symptoms ranging from hot flashes to osteoporosis.
Although some studies report positive effects of HT, the Women's Health Initiative (WHI) terminated a long-term study on HT in March 2004 when researchers found that hormone treatment significantly increased women's chances of contracting heart disease, breast cancer, and stroke. Another WHI study recently showed that long-term hormone treatment also increases the risk of cognitive decline and dementia.
"Many menopausal women are at a loss, trying to ease the symptoms of menopause without incurring any additional serious health conditions such as Alzheimer's or heart disease," says Roberta Diaz Brinton, PhD, of the University of Southern California Pharmaceutical Science Center . "We are working toward a greater understanding of the consequences and repercussions of long-term hormonal replacements, to assist these women in making better informed decisions about their health."
"The WHI findings were especially surprising," says Brinton, "given that retrospective and prospective observational analyses have indicated that HT can prevent general cognitive degeneration, as well as the contraction of diseases such as Alzheimer's."
One difficulty in performing HT studies is controlling for all the variables present in the participating women. Some have had hysterectomies, while others still have their reproductive organs. Women also differ in the time lapse between onset of menopause and start of HT, as well as in how long they have been taking hormones. Treatment regimens also differ. Some women take only estrogen, while others take combinations of estrogen and progesterone. Dosages can be cyclical or constant.
Scientists are now examining these variables, in order to shed light on how and why HT may cause deterioration of cognition.
Animal models are proving useful in the study of HT because better controls are possible than with human studies. For example, like women, female mice experience natural cyclical fluctuations in estrogen levels. In addition, during middle age (17 months for mice) hormone cycling becomes irregular and hormone levels begin to decline, similar to women in menopause. Mice are also known to experience memory loss as they get older.
Jodi Gresack, MS, and Karyn Frick, PhD, of Yale University 's Department of Psychology capitalized on the similarities between female mice and women in a study designed to examine the effects of continuous versus cyclical estrogen on cognition.
Researchers removed the ovaries of 32 middle-aged mice before starting them on various courses of HT lasting three months. A continuous group received estrogen injections daily, a cyclical group was administered estrogen every four days, and a control group received daily injections with no estrogen.
After three months, the mice underwent a variety of cognitive tests. In the spatial water-escape radial arm maze test, mice had to swim in a wheel-shaped maze until they were able to locate a submerged platform escape route. Mice were tested every day for two weeks for both spatial reference memory (long-term memory for information that did not change during the test session) and working memory (short-term memory for information that changed in each trial).
Mice on the cyclical regimen made more reference and working memory errors than control mice. The cyclical group also made more reference memory errors than mice receiving continuous estrogen.
Another test focused on object recognition, a type of nonspatial memory. The mice were presented with two identical objects, and one day later they were presented with one of these objects, along with a new object. On the second day, the investigators measured how long the mice spent examining the new object. Because mice have a natural tendency to explore novel objects, mice with good memory for the original objects should spend more time examining the new object. Again, mice in the control and continuous groups outperformed the cyclical HT mice.
"These results suggest that treatment regimen can influence the extent to which estrogen affects memory in aging female mice," Gresack says. "It is important to take into account factors such as the cyclic versus constant nature of hormone therapy, when designing these therapies for menopausal women."
Other researchers studied the effects of continuous versus no administration of HT. Investigators removed the ovaries of mice aged 8 to 12 weeks and either treated them with continuous estrogen for 47 days or did not treat them with estrogen. Researchers then sacrificed the mice at different time periods after estrogen exposure (at 5, 14, 28, and 47 day intervals) and examined them for the production of the proteins associated with neuron repair and the formation of contacts between neurons. The work was done by Mary McAsey, PhD, Robert Struble, PhD, and colleagues at Southern Illinois University School of Medicine; Craig Cady, PhD, of Bradley University ; and Britto Nathan, PhD, of Eastern Illinois University.
After five days on estrogen, the estrogen-treated mice produced more of the proteins important for repair and neuronal function. However, with prolonged, continuous estrogen treatment, this effect diminished, and by day 47 the estrogen-treated mice were similar to the non-estrogen-treated mice in levels of the repair proteins. In addition, at the end of the experiment, mice that did not receive estrogen showed an elevation of a brain protein associated with the negative aspects of brain aging, while estrogen-treated mice did not.
"Still, it is important to consider the cyclical nature of estrogen release, as the hormone is distributed naturally, during reproductive years," McAsey says.
Another possible key to understanding the often conflicting results of HT on menopausal women is to study the combinations of hormones being administered. Researchers at the Center for Aging, Medical University of South Carolina, examined the effects of the progesterone component of many hormone therapies.
The group--Heather Bimonte-Nelson, PhD, Rachel Singleton, Brice Williams, and Ann-Charlotte Granholm, PhD, DDS ,--had previously demonstrated that ovary removal in aged rats enhanced spatial working memory while simultaneously decreasing elevated progesterone levels. Building on these findings, the researchers set out to determine whether giving the aged ovariectomized rats progesterone reversed these positive cognitive effects.
Thirty rats were used in the study. Ten rats kept their ovaries, and twenty rats had their ovaries removed. The ovariectomized rats were then divided into two groups: those receiving progesterone and a control group that did not receive progesterone.
As in the Yale study, a water maze was used to test working and reference memory. The maze difficulty was changed at increasing rates, forcing the rats to remember greater amounts of information. The rats receiving progesterone exhibited deficiencies in learning and remembering the maze. In addition, rats treated with progesterone also showed problems remembering many items of information, while untreated rats were able to successfully remember the items.
These findings suggest that progesterone is detrimental to learning to navigate through new environments, and that it increases short-term memory deficits in aged rats, Bimonte-Nelson says. "The results support the hypothesis that progesterone negatively affects memory during aging," says Bimonte-Nelson. "The negative outcome of several studies evaluating the combined estrogen/progesterone HT may be due, in part, to unfavorable effects of progesterone. Future research to determine the specific effects that progesterone, alone and in combination with estrogen, has on the brain and on learning and memory are crucial to help us understand the complex effects of female hormone loss and female hormone therapy."
Other recent work examines how two established protective actions of estrogen with relevance to Alzheimer's are affected by the presence of progesterone. Christian Pike, PhD, Martin Ramsden, PhD, and Emily Rosario, MS, of the University of Southern California 's Department of Gerontology, and Andrew Nyborg, PhD, Michael Murphy, PhD, and Todd Golde, PhD, of the Mayo Clinic's Pharmacology Department, performed two studies in rats treated with estrogen, progesterone, or a combination of estrogen and progesterone. One study examined neuron survival and the other looked at beta-amyloid accumulation. Beta-amyloid, a neurotoxic peptide, is a well-established hallmark of Alzheimer's disease.
In both studies, female rats had their ovaries removed. After a two-week recovery period, rats received one of the following treatments: placebo, estrogen alone, progesterone alone, or both estrogen and progesterone.
In the first experiment, levels of beta-amyloid protein were evaluated after a six-week period of hormone treatment. Higher levels of beta-amyloid protein were observed in the hormone-depleted rats compared with control animals. The group receiving estrogen did not experience an increase in levels of beta-amyloid. For the rats receiving the combination of estrogen and progesterone, although progesterone failed to decrease beta-amyloid levels, it did not alter the ability of the estrogen treatment to reduce beta-amyloid levels.
In the second study, rats were treated with a mild dose of neurotoxin after two weeks of hormone treatment. The hormone-depleted rats experienced the greatest amount of neuronal death. In estrogen-treated rats a protective effective against neuronal death was observed, while rats treated with estrogen and progesterone in combination did exhibit neuronal death, suggesting that progesterone inhibited the neuroprotective action of estrogen in this model.
"Together, these data demonstrate that at least some of the neuroprotective actions of estrogen can be inhibited by progesterone," says Pike. "When these results are applied to the human condition, these findings suggest the possibility that the progesterone component of HT may negate at least some of the positive neural actions of estrogen."
In another study, Brinton and Shuhua Chen, MD, of the University of Southern California School of Pharmacy investigated the effect of HT initiation on cognitive degeneration.
The researchers created in vitro (neurons in culture) models simulating three different scenarios of HT timing. In the prevention model, neurons in the hippocampus--a brain region involved in learning and memory--were exposed to estrogen prior to exposure to beta-amyloid. In the treatment model, neurons received estrogen treatment after the beta-amyloid exposure. The control model received only the beta-amyloid treatment.
Neurons in the prevention group had a significantly greater rate of survival than those in the treatment and control groups. Exposure to estrogen did not increase the survival rate of neurons that were first exposed to beta amyloid. The study demonstrates that neurons healthy at the time of estrogen exposure exhibit a beneficial response to estrogen, in neuronal function and survival, says Brinton. In contrast, if the neuron is diseased or dysfunctional at the time of estrogen exposure, there is no benefit.
"This finding is consistent with clinical studies in which women who received estrogen hormone therapy at the time of menopause, before cognitive degeneration becomes apparent, have a lower risk of developing Alzheimer's disease than women who never receive any sort of HT," says Brinton. "These results are also remarkably consistent with clinical data indicating that for women who have already experienced the onset of Alzheimer's disease or who are in their 60s and 70s, hormone therapy may exacerbate the degenerative state."