Alzheimer's Disease

Table 2

The number of persons over age 85 is now the fastest growing segment of the US population. The number of Americans with AD will double by 2030, and the number of people affected will reach 14 million by 2050. With this increase, there will be a significant increase in the cost of caring for this population beyond the 1990 figure of $100 billion.(17)

The prevalence of AD is greatest in those over age 85 (Figure 3).(17) After the age of 65 or 70, the number of individuals with AD doubles every 5 years, so that by the age of 85 as much as 50% of the population may be affected.(18-20) An additional 20% may have the neuropathologic plaque changes of AD in the absence of clinical dementia.(21) This exponential increase in the apparent increase in AD is seen across racial and ethnic groups and is assumed to be intrinsic to the aging process.

Figure 3

Both in terms of prevalence and incidence, AD is a major health issue for women.(20,22) Women comprise 72% of the population over 85 years of age, and roughly half of this group has Alzheimer's disease. Not only do women constitute a greater proportion of the older population, but AD is expressed earlier in women than in men. This may be related to the estrogen loss that occurs with menopause.

Effect of Estrogen on AD

The best evidence that relates estrogen deficiency to Alzheimer's disease comes from several epidemiological studies. The 5 independent studies in the relative risk (RR) scale in Figure 4 consistently demonstrate a 40% to 60% reduction in the risk of AD in women who have taken ERT.(23-27) This risk reduction supports the role of estrogen in preventing or delaying AD. Relative risk is defined in Table 3.

Figure 4

Table 3

These 5 studies are large epidemiologic studies that provide significant data about estrogen and Alzheimer's disease. In particular, the study that originates with the Leisure World Cohort involved observation of 8877 women over a 14-year period (1981-1995).(24) Using a subset of these individuals, a case-control study of Alzheimer's disease evaluated the impact of estrogen replacement. Of the 3760 women who died in the 14-year period, 248 were women with the diagnosis of AD on a death certificate. These women were the cases. Each case was paired with 5 age-matched controls who died within 1 year of the case, but who did not have diagnosed dementia. The risk of Alzheimer's disease and related dementia was significantly reduced in estrogen users. This reduction was dependent on the dose of conjugated equine estrogen (CEE) used and the duration of exposure to estrogen (Table 4 - odds ratio can be interpreted similarly to relative risk). The longer a woman takes estrogen, and the higher the dose, the greater the protection from AD.

Table 4

In addition, there are 3 small clinical studies that examine the effects of estrogen administration in an elderly population of women. They do not provide the definitive information that epidemiological studies do, and some of them are old.But they are suggestive of trends and serve as a guide for future research. Although these patients were not necessarily diagnosed with AD, the studies evaluate memory and cognition, which often decline in the AD population. One study involved 30 elderly residents of a "home for the aged" whose mean age was 75 years.(28) The women were randomized to treatment with 2 mg of estradiol benzoate IM every week or placebo in an IM vehicle every week for 12 months. The Wechsler memory scale was used as the outcome measure. There was a significant improvement in memory over baseline for the first 12 months with treatment and a decline in memory with placebo. Because baseline performance was impaired, many of these women may have had AD. Although the improvement in memory was modest in magnitude, it is equivalent to about 1.5 years of AD progression. The clinical implication about estrogen's effect in this study was that it might delay the progression of the disease.

Another study involved 50 elderly women in a "home for the aged" who were randomized to treatment with CEE 0.625 mg daily or placebo for 36 months.(29) Many of these women may have had AD. Outcome measures consisted of a standardized interview assessing communication and self-care abilities. The interviewers were blinded in terms of treatment. Treatment with estrogen produced a progressive improvement over the first 12 months followed by a decline (Figure 5). The net effect of treatment was a delay in progression of cognitive decline. This small clinical study again suggests that estrogen therapy may allow the maintenance of independence for several additional years.

Figure 5

A third small clinical study in which the subjects actually had mild to moderate AD involved 30 women, 15 of whom were treated with CEE 0.625 mg daily for 5 months.(30) Psychometric evaluations were performed every month with the Hasegawa Dementia Scale (HDS). Women who were treated with estrogen sustained an improvement in HDS scores. The study suggests that estrogen therapy may improve cognitive function or slow the rate of cognitive decline in patients with established AD.

Estrogen therapy showed unexpected benefit in a reanalysis of the pivotal tacrine trial that lead to the approval of this drug for the treatment of AD.(31) Tacrine is a cholinesterase inhibitor. The study involved a subgroup of women who were receiving estrogen at the time of randomization to tacrine. Contrary to what would have been expected, tacrine alone had no significant effect on cognitive function. Instead, only women receiving ERT concurrently with tacrine had a beneficial, and significant, response (as measured by the cognitive component of the Alzheimer's Disease Assessment Score test). This study has some important implications for other cholinesterase inhibitors that are being investigated for the treatment of AD. It is possible that such agents may also require the concomitant administration of estrogen for their efficacy.

How these observations on estrogen therapy might translate into the clinical setting was explored in a small double-blind, placebo-controlled study of 20 subjects. Treatment consisted of estrogen (CEE) 0. 625 mg daily for 9 months cycled every third month with progestin (medroxyprogesterone acetate) 5 mg for 13 days.(32) The outcome measure was the Clinician Interview Based Impression (CIBI), in which the patient and the caregiver are interviewed by a clinician blinded to treatment and objective measures of cognitive function. Nine domains of behavior and cognitive function are assessed for change. This measure is designed to detect clinically relevant changes in function as opposed to statistically measurable changes that may not translate into clinically significant changes. Of those subjects on active drug, 80% showed clinical improvement, whereas none of the subjects on placebo showed improvement (Figure 6).

Figure 6

Finally, an additional study is noteworthy because all of its 1124 subjects were free of AD at baseline and were followed for 1 to 5 years.(26) The curves in Figure 7 show the percentage of the study population that was free of disease at certain ages. Women who took estrogen for more than 1 year experienced a dramatic delay of AD onset. But even short-term estrogen therapy produced a delay. This group of women averaged 4 months of estrogen therapy and most likely took the medication to control menopausal symptoms such as hot flushes. This raises the question of how brief exposure to estrogen for treatment of hot flushes could influence AD expression 20 to 30 years later.

Figure 7

One hypothesis to explain such an estrogen effect links the hot flush with an irreversible loss of neurons. Thus, women with a history of untreated hot flushes will have a decrease in neuronal reserve as they enter their 70s and 80s and are therefore more likely to express AD at an earlier age.

There is, in fact, some support for the concept that hot flushes are associated with neuronal loss from a study that involved women who required a hysterectomy and bilateral oophorectomy for benign disease.(33) Their cognitive function was assessed prior to surgery and 2 months later. On tests of verbal memory there was a significant, negative association between the selected domain of cognitive function and the severity of hot flushes. Moreover, these changes in cognitive function could not be explained by changes in depressive symptomatology, anxiety, or insomnia.

It is hypothesized that with the loss of estrogen at the time of menopause, there is a decrease in estrogen-dependent glucose transport into the CNS, thereby resulting in a selective loss of neurons within the hippocampus. This region of the brain is related to memory and is involved in AD changes. Should the relationship between estrogen, glucose transport, and the hot flush prove to be significant, it may be important to examine other perimenopausal therapeutic alternatives for their effect on the vasomotor symptom of hot flushes. For example, raloxifene, the newest selective estrogen receptor modulator (SERM), seems to be associated with an increased incidence of such hot flushes.(34) Roughly 30% of women have experienced hot flushes with raloxifene therapy of 60 mg.

Summary of AD Treatment Options

There are currently few alternatives for treating AD. Cholinesterase inhibitors, tacrine and donepezil, affect the symptoms of AD but not the disease process, and the efficacy of these drugs may require a synergistic interaction with estrogen, which enhances the production of acetylcholine.(31) Epidemiologic evidence consistently indicates that nonsteroidal anti-inflammatory drugs delay the expression and the progression of AD.(35,36) Also, vitamin E, an antioxidant, slows the progression of AD.(37)