Subscribe to eTOC


STOCKHOLM, SWEDEN — Estrogen, this year's most maligned hormone, may yet prove to be a winner – but it may win minds, not hearts.

This past summer started out as the worst of times for advocates of hormone replacement therapy (HRT) when they were dealt a crippling blow by the National Institutes of Health (NIH), which announced that it was pulling the plug on the estrogen-progestin (Prempro) arm of the Women's Health Initiative (WHI).

The NIH said that five years of HRT were associated with small but significant increases in the risk for stroke, heart attack, deep vein thrombosis, and breast cancer, while offering protection against osteoporosis and colorectal cancer. The risks, said the NIH, outweighed the benefits.

But two weeks after that announ-cement, research presented at the eighth International Conference on Alzheimer's Disease and Related Disorders (IADRD), held here in Stockholm, suggested that estrogen does appear to be neuroprotective. The findings, which are based on preliminary analysis of data from the Swedish Twin Registry, indicate that this protective effect is not related to length of estrogen therapy. The National Institute on Aging funded the Twins study.


Nancy L. Pederson, PhD, Professor of Genetics and Epidemiology at the University of Southern California in Los Angeles, who is a co-investigator of the Twins study, said the data confirm earlier observations among neuroscientists that “estrogen is clearly protective of the central nervous system.”

Dr. Pederson and colleagues based their findings on data collected by the Swedish Twin Registry; 7,684 female twins aged 65 and older were tested using a 10-item test of memory, concentration, and abstract ability. Women who scored below 13.5 out of a maximum possible 19 points were considered cognitively impaired.

The researchers identified 1,388 women who were cognitively impaired as defined by those test criteria. All but 21 of these women had never used estrogen, while 413 of the 6,926 women who were still cognitively intact had used estrogen; 259 of these women had used it for more than five years.

Higher age and less education were related to cognitive impairment. After controlling for these factors, estrogen use appeared to be linked to better cognitive functioning.


Asked whether she or other researchers will continue estrogen research in light of the increased risks for breast cancer, heart disease, and deep vein thrombosis reported by investigators with the WHI trial, Dr. Pederson noted that there are several ongoing studies of estrogen and its effect on brain function that have “already collected data, and those data are being analyzed.”

She pointed out that since there is “abundant evidence from animal studies suggesting that estrogen is neuroprotective,” she doubted estrogen research would be adversely affected by the decision to halt the WHI study.

Other Alzheimer disease researchers said they agreed with Dr. Pederson's assessment of the current state of estrogen research, and some researchers said they were equally unfazed by the negative press for estrogen.

Zaven Khachaturian, PhD, senior science advisor to the Alzheimer's Association, said in an interview that he considered estrogen “as still promising because it is well known that there are many estrogen receptors in the brain. It may be that we have just not found the right estrogen yet.”

It may also be “that the method of delivery is wrong,” Dr. Khachaturian continued. “We may find more success with an estrogen patch,” he said, adding that he expects an acceleration of estrogen research aided by advances in imaging technology. “As we are able to identify very early stage Alzheimer disease using new imaging techniques, we will also be able to accurately assess the effects of estrogen on brain function,” he added.


Marilyn Albert, PhD, Professor of Psychiatry at Harvard Medical School – who served on the planning committee for the scientific sessions at the IADRD – agreed that there is good evidence that estrogen is beneficial to the brain, but she is not so sure that neuroscientists will escape the negative backwash from WHI.

She said she hoped that the recent estrogen controversy would not preclude further studies of estrogen in Alzheimer disease prevention, but was “still just waiting to see how this all works out.”

Dr. Albert added that her immediate concern is a section of the WHI study that is investigating estrogen's effect on cognition. She noted that the future of that component of the study, for which she is an investigator, is uncertain. Rumors continue to circulate, suggesting that “the data collection for the cognition arm will not be continued,” she said.


If the WHI findings do put a damper on continued estrogen research, selective estrogen receptor modulators, or SERMs, may provide an alternative. The SERM raloxifene has demonstrated estrogen-like efficacy in treating osteoporosis without the breast cancer risks associated with estrogen, said Alzheimer disease researcher Kristine Yaffe, MD, Professor of Psychiatry at the University of California-San Francisco.

Dr. Yaffe reported at the IADRD that early results suggest that high dose raloxifene – 120 mg, which is twice the dose approved for treatment of osteoporosis – can reduce the risk of mild cognitive impairment by 33 percent compared with placebo.

In Dr. Yaffe's study, she and her team studied 5,386 women who participated in the Multiple Outcomes of Raloxifene study for at least three years. The women were randomly assigned to receive either raloxifene at 120 mg, 60 mg, or placebo. Patients who scored in the lowest decile on cognitive tests or who had clinical symptoms completed a standardized evaluation for dementia after a mean of 3.8 years of treatment. An adjudication committee, which was blinded to treatment, used pre-established criteria to classify cognitive status.

Thus far, 742 women have completed this evaluation. “Sixty-eight percent of the women (509) were cognitively normal. But 181 women (24 percent) have developed mild cognitive impairment and 52 (7 percent) women have frank dementia.”

The women randomized to high dose raloxifene were “a third less likely to develop mild cognitive impairment, but we didn't find this protective effect in women taking lower dose raloxifene,” she said. There was also a suggestion of a lower risk of Alzheimer disease and a lower risk for any cognitive impairment.


Dr. Albert said Dr. Yaffe's positive findings are encouraging, but she noted that they are “preliminary.” Moreover, Dr. Yaffe's results are part of a large clinical study funded by Eli Lilly, which manufactures and markets raloxifene (Evista). Nonetheless, Dr. Albert said the findings should not be ignored since Dr. Yaffe is “a well known and highly regarded researcher.”

Dr. Yaffe, too, was cautious when discussing her findings, noting that they are preliminary and are confined to women. Any treatment aimed at preventing Alzheimer disease would have to be effective in both men and women. “We really have to study the drug in men,” Dr. Yaffe said.


Dr. Zaven Khachaturian

Based on the recent “bad news about estrogen,” Dr. Yaffe predicted that more researchers will turn to SERMs as “attractive estrogen alternatives. The results of this study suggest the need for a large trial of raloxifene in both men and women at high risk for developing Alzheimer disease.”


  • ✓ Neuroscientists are concerned about the effect of negative publicity about hormone replacement therapy on future studies of estrogen and cognition.
  • ✓ Despite these concerns, neuroscientists reported at the International Conference on Alzheimer's Disease and Related Disorders on preliminary data suggesting neuroprotective effects of estrogen.
  • ✓ An analysis of data from the Swedish Twin Registry found estrogen to have a neuroprotective effect and that its use appeared to improve cognitive functioning – regardless of the length of therapy.
  • ✓ A research team at the University of California-San Francisco reported that high doses of raloxifene, a selective estrogen receptor modulator (SERM), reduced the risk of mild cognitive impairment by 33 percent compared with placebo.
  • ✓ Neuroscientists predict that more researchers will study SERMs as “attractive estrogen alternatives” for their possible neuroprotective effects.