The landmark Women's Health Initiative Memory Study (WHIMS) program has had an enormous impact on our understanding of how estrogens and estrogen-containing hormone therapy affect cognitive outcomes in postmenopausal women. It is the starting point and touchstone for any discussion on cognition and dementia in women. As reviewed in this chapter by principal WHIMS program investigators, the WHIMS comprised two large randomized placebo-controlled trials of conjugated equine estrogens with and without medroxyprogesterone acetate in women aged 65 years and older. In this study, the two hormone therapy formulations were associated with increased risk for probable dementia (hazard ratio 1.77, 95% confidence interval 1.22 to 2.58). They were also associated with a small adverse mean difference of 0.21 (0.06 to 0.37) points on the 100-point Modified Mini-Mental State examination. Adverse findings were similar for both hormone therapy formulations and for women with and without histories of prior hormone therapy use. The Women's Health Initiative Study of Cognitive Aging and the Women's Health Initiative Magnetic Resonance Imaging Study were conducted on subsets of WHIMS participants. The former found little evidence that conjugated equine estrogens with medroxyprogesterone acetate had a positive effect on cognitive aging. The latter found that the hormone therapy formulations were associated with decreased brain volumes, particularly among women with lower levels of cognitive function at baseline, but mean effects on ischemic lesions were not significant. No subgroups of WHIMS participants have been identified for which initiating hormone therapy appears to convey cognitive benefit.

Casadesus and colleagues make a case that hormonal changes associated with the dysregulation of the hypothalamic-pituitary-gonadal (HPG) axis following menopause/andropause are implicated in the pathogenesis of Alzheimer's disease (AD). Experimental support for this postulate has come from studies demonstrating an increase in amyloid-β (Aβ) deposition following ovariectomy/castration. Because sex steroids and gonadotropins are both part of the HPG feedback loop, decrements in sex steroids result in a proportionate increase in gonadotropins. They provide a review of the basic science relevant to luteinizing hormone (LH) and its receptor as a background for considering LH regulation of cognitive behaviors and AD pathology. Results of their analyses suggest that marked increases in serum LH following menopause/andropause is a physiologically relevant signal that could increase Aβ secretion and deposition in the aging brain. Suppression of the age-related increase in serum gonadotropins using anti-gonadotropin agents, such as leuprolide, is proposed as a novel therapeutic strategy for AD.

This chapter discusses the possibility that men with Alzheimer's disease (AD) exhibit age-accelerated endocrinological change that increases the risk for AD related pathological processes. This age-accelerated endocrinological change is characterized by an earlier decrease in thyroid stimulating hormone, an earlier increase in sex hormone binding globulin, and a subsequent earlier decrease in free or bioavailable testosterone levels. They also show an earlier increase of gonadotropin levels, which may be an appropriate, if inadequate, response to lower free testosterone levels. However, both these processes (the increase in gonadotropin levels as a response to the lowered levels of free testosterone) have the propensity to negatively affect neuronal health, as subsequent chapters in Section 6 will outline in more detail. This acceleration in age-related endocrinological change may be related to stress and morbidity (which are both known to affect thyroid stimulating hormone) and/or possibly, certain genetic polymorphisms that could predispose to an earlier lowering of free testosterone, which, especially when combined with a decrease of androgen receptors in the brain, leads to a lowered protective environment for the aging brain.

This chapter explains why two meta-analyses of treatment studies only found time-limited positive effects of estradiol in both women with and without dementia. The findings in women with dementia do not substantiate the “window of opportunity” theory. Negative effects of longer term treatment (>1 year) have also been reported in both women with and without dementia.

The meta-analyses reported that the most substantial effects of estradiol on cognition were seen in women who had undergone surgical menopause. Another large observational study found an increased risk for dementia when women had undergone surgical menopause. This chapter attempts to explain these findings by speculating that some women who are more at risk for medical indications for surgical menopause have particular genotypes implicated in sex steroid metabolism and synthesis. These genotypes expose these women to high estradiol levels. This may explain why these women show such strong responses to undergoing surgical menopause, as this induces a very sharp decline in their previously high estradiol levels. If these women also have genetic polymorphisms that predispose them to high levels of toxic estrogenic metabolites (such as catecholestrogens) this could lead to DNA damage. In the first instance that could lead to medical indications for surgical menopause, such as ovarian or endometrial cancer, endometriosis, cysts, etc. If these women are given estrogens at a later time in their lives, they might also be more susceptible to dementia, which has previously also been associated with DNA damage. The brain is apparently less able to compensate for this type of damage in later life, which could explain the more substantial negative effects on cognition and dementia risk found in the older women. Future genetic screening for these polymorphisms might allow more targeted treatment for those women who are not genetically at risk.

Estrogen-based therapies often include a progestin to antagonize tumorigenic effects of estrogens in the uterus. While much has been learned about the functional and neuroprotective effects of estrogens in the brain, far less is known about the effects of progestins, particularly specific progestins like progesterone and medroxy-progesterone acetate, used either alone or in combination with estrogenic therapies. In this chapter, Pike and Carroll review many of the effects on cell survival and function, first of estrogens, and then of progestins. While not all progestins are alike, the authors find that prolonged exposure to progestins often will decrease the protective effects of estradiol on cell survival and function. Evidence suggests that a cyclical regimen of estradiol and progesterone may be most efficacious. Ultimately, the development of neural selective estrogen receptor modulators (SERMs) with the potential to circumvent the need for, and hence the negative neural consequences of, progesterone will be an important advance to estrogen-based therapies.

This chapter explains why two meta-analyses of treatment studies only found time-limited positive effects of estradiol in both women with and without dementia. The findings in women with dementia do not substantiate the “window of opportunity” theory. Negative effects of longer term treatment (>1 year) have also been reported in both women with and without dementia.

The meta-analyses reported that the most substantial effects of estradiol on cognition were seen in women who had undergone surgical menopause. Another large observational study found an increased risk for dementia when women had undergone surgical menopause. This chapter attempts to explain these findings by speculating that some women who are more at risk for medical indications for surgical menopause have particular genotypes implicated in sex steroid metabolism and synthesis. These genotypes expose these women to high estradiol levels. This may explain why these women show such strong responses to undergoing surgical menopause, as this induces a very sharp decline in their previously high estradiol levels. If these women also have genetic polymorphisms that predispose them to high levels of toxic estrogenic metabolites (such as catecholestrogens) this could lead to DNA damage. In the first instance that could lead to medical indications for surgical menopause, such as ovarian or endometrial cancer, endometriosis, cysts, etc. If these women are given estrogens at a later time in their lives, they might also be more susceptible to dementia, which has previously also been associated with DNA damage. The brain is apparently less able to compensate for this type of damage in later life, which could explain the more substantial negative effects on cognition and dementia risk found in the older women. Future genetic screening for these polymorphisms might allow more targeted treatment for those women who are not genetically at risk.

Few comparisons of the efficacy of different estrogenic formulations on brain function have been carried out. While most basic science studies have evaluated effects of estradiol, many of the clinical studies that have evaluated effects of hormone therapies (HT) in healthy older women have used conjugated equine estrogens (CEE). In this chapter, Gleason, Wharton, Carlsson, and Asthana review some of the cellular mechanisms by which estrogens are thought to protect the brain from age-related cognitive decline and Alzheimer's disease (AD), and discuss the pros and cons of oral vs. transdermal therapies and of estradiol vs. CEE. Their analysis identifies numerous advantages of transdermal estradiol vs. oral estrogenic therapies, primarily related to effects on liver enzymes, venus thromboembolic events, and inflammatory cytokines. Collectively, the analysis provides compelling arguments in favor of transdermal estradiol as the therapy of choice. Nevertheless, it is clear that the benefits of estradiol vs. CEE have yet to be proven, as has the efficacy of estrogenic therapies (ET) in the prevention of AD. These issues will be addressed in the recently initiated Kronos Early Estrogen Prevention Study (KEEPS) trial and the ancillary KEEPS study of cognitive aging. Details of the trial including its design and primary goals are discussed.

Hogervorst and colleagues review the complex topic of soy effects on cognition and risk of dementia. In an attempt to address one aspect of this complexity, they conducted a comparative analysis of fermented (e.g., tempe) versus non-fermented (e.g., tofu) forms. Results of their analyses indicate that high intake of tofu was associated with lower cognitive function and an increased risk for dementia, particularly in those participants who were older than 68 years of age. These findings are consistent with previous analyses of hormone therapy and tofu consumption. They also found a complex association with genistein levels. Relatively younger participants (52–68 years of age) appeared to have optimal genistein levels relating to optimal memory function, whereas persons older than 68 years of age with high genistein levels exhibited lower cognitive performance and an increased risk of dementia. These results are reminiscent of the window of opportunity theory (Chapters 4 and 5) or the healthy cell bias theory (Chapter 6). Higher folate levels within tempe (which contains high phytoestrogen levels) may be a mediating factor for its reported protective effects. Further studies to determine the interaction between serum phytoestrogens and folate levels and their relationship to dementia risk are suggested.

The Women's Health Initiative Memory Study (WHIMS) has been much criticized and it has been suggested that subgroups of women exist for whom hormone therapy (HT) might improve or impair cognitive function. Possible modifying variables could be age, smoking, body mass index, and menopausal symptoms. These were included in artificial neural networks (ANN) analyses, which allow testing of complex non-linear higher order interactions of variables to predict outcomes. Artificial neural networks analyses without hidden units could predict responders and non-responders to treatment as well as logistic regression models that included only main effects, which indicated that higher order interactions were not necessary and did not add to the value of the models. There seemed to be no subgroups (e.g., older women who smoke and have a high body mass) for whom HT has a worse or better effect on cognitive function over time. This study also showed that cross-validation is essential in building robust models with many independent variables and should be applied as a standard technique in complex multivariate analyses.

The extent to which testosterone and other androgens might affect cognitive skills in women is not yet well understood. In this chapter, Sherwin reviews changes in endogenous androgens over a woman's lifespan and research findings germane to androgens and cognitive skills in women. For younger women, there is evidence that cyclical changes during the menstrual cycle affect cognitive performance, although it is not possible to tease out effects of testosterone from those of estradiol. In older women, the relation between testosterone levels and cognitive test scores is inconsistent. The ratio of estradiol to testosterone may be important in modulating sex-advantaged cognitive functions in women, with a lower ratio leading to relatively impaired performance on cognitive tasks in which women typically excel.