M. J. Friedrich

JAMA. 2002;287:29-30.

San Diego-Estrogen is a hormone adept at multitasking. Not only does this ubiquitous molecule influence reproduction, it plays a role in bone and mineral metabolism, cardiac and vascular function, and various aspects of cognition, such as memory. But discovering the ways in which estrogen wields its influence in various organ systems such as the brain has been challenging.

Over the years basic research has suggested mechanisms by which estrogen contributes to brain functioning, but clinical studies have produced conflicting results. For example, a randomized, double-blind, placebo-controlled study showed that estrogen replacement therapy given for 1 year did not slow cognitive decline in women with mild to moderate Alzheimer disease (AD) who have had hysterectomies (JAMA. 2000;283:1007-1015). And a meta-analysis of studies evaluating hormone replacement therapy (HRT) for prevention of cognitive decline showed that HRT may have beneficial cognitive effects in women with menopausal symptoms but not in asymptomatic women (JAMA. 2001;285:1489-1499).

Many questions about estrogen's effects remain to be elucidated, and investigators are seeking answers through ongoing laboratory and clinical studies. As Sanjay Asthana, MD, of University of Wisconsin Medical School, said, "It is my belief that we are just starting to appreciate the complexity of estrogen and related hormones in humans. Therefore it is critical that we continue to systematically address all of the clinical issues concerning estrogen treatment for AD."

Several reports presented at the annual meeting of the Society for Neuroscience provide provocative clues to how estrogen may behave in the central nervous system. Researchers reported on the ability of estrogen and estrogenlike compounds to enhance memory as well as to provide protection against illnesses such as AD and stroke, diseases that increase in women after menopause when estrogen levels plummet.

RALOXIFENE AND COGNITION

Because estrogen replacement therapy can increase the risk of breast and endometrial cancer in postmenopausal women, researchers sought to determine whether the selective estrogen receptor modulator (SERM) raloxifene can provide beneficial cognitive effects, as some studies have shown estrogen provides, without the latter's adverse effects. While at the University of Washington, Seattle-where he worked before relocating recently to Wisconsin-Asthana and his colleagues examined the effects of this designer estrogen on women with AD.

Approved by the US Food and Drug Administration for treatment of osteoporosis, raloxifene has yet to demonstrate memory-enhancing effects. To begin to look into this question, Asthana conducted a small randomized, placebo-controlled, double-blind study with the drug involving 14 women with AD. Preliminary results were reported at the meeting.

Women received placebo or 120 mg of raloxifene daily for 3 months. During that time researchers administered a battery of cognitive and mood tests before the study began; at weeks 4, 8, and 12; and 2 months after treatment was stopped.

The women who received raloxifene showed improvement in delayed verbal recall, which usually declines in persons with AD, said Asthana. Improvement in verbal memory in the raloxifene group, which was evident after the first month of treatment, was sustained throughout treatment and started to diminish following treatment termination. The placebo group showed no significant change in memory during treatment.

Asthana noted that to reduce learning or practice effects in these studies, different versions of the cognitive tests were administered at each time point. Also, each woman in the study underwent a "practice" session prior to starting treatment, a strategy that is believed to reduce practice effects. Asthana's team also compared raloxifene's effect on verbal memory with that of estrogen by examining the verbal memory scores from one of their previous publications (Neurology. 2001;57:605-612), in which women with AD received daily estrogen.

Asthana reported that the performance of women taking raloxifene fell between that of women taking estrogen and those who received placebo. But because treatment with raloxifene is showing a trend toward memory improvement-a trend that will be examined as the study continues and the number of women in it is increased-he said these preliminary results may indicate that raloxifene could serve as an alternative treatment for women who cannot tolerate estrogen or who don't want to take it, he said.

PREVENTING AD

While estrogen's benefits for women with AD remain controversial, evidence is accumulating that bathing the brain in estrogen enhances cognitive function in healthy older women.

Researchers at the University of Washington School of Medicine examined what the benefits might be from a short-term trial of estrogen replacement in cognitively healthy postmenopausal women. Laura Baker, PhD, one of the study authors, noted that it is standard clinical practice to administer progesterone with estrogen to protect against endometrial cancer. And in some cases, she said, testosterone is prescribed to improve libido.

To determine whether the addition of these hormones would attenuate or negate the potential benefit of estrogen replacement, the researchers randomized 45 women to receive estrogen alone, estrogen plus progesterone, estrogen plus testosterone, or placebo for 3 months. Tests assessing memory and attention were administered before treatment began, at the end of each month, and 2 months after treatment ended.

Preliminary results have shown that when looked at as a group, the women who received estrogen alone or with another hormone performed significantly better on verbal memory tests than those receiving the placebo. When the groups were looked at separately to detect whether testosterone or progesterone detracted from the estrogen benefit, Baker said that women taking estrogen plus progesterone or estrogen alone performed comparably, and their scores exceeded those of the placebo group. However, those receiving testosterone in addition to estrogen did not seem to be performing as well, she said.

A second finding is that the estrogen-treated group as a whole outperformed the placebo group on a test of visuo-spatial memory, said Baker.

Researchers are expanding the study to include 60 women to clarify these results. Baker and colleagues are also identifying variables that may help predict who will respond favorably to estrogen and who will not. "We strongly believe this is the direction to take to help clarify some of the discrepancies in the literature regarding estrogen therapy," she said.

A HEALTHY HIPPOCAMPUS

Another unknown regarding estrogen replacement therapy is whether it has an effect on the size of the hippocampus. By measuring the hippocampal volume size in postmenopausal women, William Jagust, MD, and colleagues at the University of California, Davis, School of Medicine set out to answer this question.

The hippocampus, which is intimately involved in memory, is also severely affected by the pathology of AD, said Jagust. It has been well documented that people with AD have smaller hippocampal volumes than those of the same age who don't have the disease, he said, adding that the volume of the hippocampus in older people seems to some extent to be a risk factor for progression to AD. From this background researchers hypothesized that if estrogen protects against AD, postmenopausal women taking estrogen might have larger hippocampal volumes than those who do not.

The study involved 59 postmenopausal women-46 not taking estrogen, 13 taking estrogen-and 38 men. Study participants, selected from the Sacramento Area Latino Study on Aging, a longitudinal study looking at AD prevalence and incidence in Mexican Americans, were excluded if they already had dementia.

Using magnetic resonance imaging (MRI) to measure hippocampal volumes, the researchers found that the hippocampi in women taking estrogen were larger in volume than in the women who did not take estrogen and also larger than in the men. The difference in size was roughly on the order of 10%, said Jagust. Hippocampal volumes were corrected for head size.

"As far as we know this is the first report of an anatomical basis for this observation," said Jagust. "But it still leaves a lot of questions unanswered, particularly what the mechanism of the change might be."

Because this is an epidemiological study, said Jagust, selection bias of some sort may account for the results. However, he pointed out that the usual variables that account for differences in estrogen use-ie, education and socioeconomic status-did not differ between the groups of women.

Jagust's group will continue to monitor the study subjects to see whether hippocampal volume is predictive of cognitive decline and whether estrogen provides protection against memory loss and the development of AD.

STROKE PROTECTION

Estrogen can prevent damage from stroke, according to new research in rat models, reported Phyllis Wise, PhD, of the University of Kentucky College of Medicine, Lexington.

Wise noted that when women are premenopausal, the incidence of stroke is less than in men; after menopause, however, the incidence of stroke in women rises, suggesting that estrogen plays a protective role.

Using rodents to explore in greater detail how estrogen protects neural tissue, researchers removed the ovaries from 100 rats-to replicate menopause-then administered low doses of estrogen. A week later they inserted a suture in the rats' middle cerebral artery, a procedure that mimics a stroke.

On examining the rodent brains, they found that estrogen does not protect against the initial cell death that occurs within the first few hours after the suture is put in place, but it "dramatically" protects against delayed cell death in the cerebral cortex.

Other work from Wise's laboratory involving mice lacking the estrogen- receptor suggests that estrogen protects the brain from injury by accelerating and amplifying the activity of this receptor (Proc Nat Acad Sci. 2001;98:1952-1957). "It appears to protect the brain against injury by binding to this receptor and activating genes that help the cells survive or suppressing genes that can harm the cells," she stated.

"I believe we have shown that the estrogen- receptor is critically important mechanistically in mediating the effects of estrogen," said Wise, adding that this is a novel action of the estrogen- receptor, which was thought to be involved mainly in reproduction.

"I think the impact of these data are really great," she concluded, "because they will help us strategize how to design estrogens in the future that can have purely protective actions in our bodies and brains without the negative effects"-a sentiment echoed by the other researchers.