7

Alzheimer’s Disease in Older Men: Are There Gender-Specific Etiological Issues and Treatments?

ROBERT S. TAN and RALPH N. MARTINS

Case History

The classic thinking is that testosterone is an androgenic hormone responsible for normal growth and development of male sex and reproductive organs. It is produced from the testes and also converted from dehydroepiandrosterone (DHEA) in women. There is increasing evidence that androgens may modulate muscle growth, function, and possibly cognitive function as well. We applied this hypothesis to a patient we treated recently in an Alzheimer’s center. This was a patient admitted to our Alzheimer’s center a year ago for terminal care. The patient had a history of advanced Alzheimer’s disease (AD) with associated agitation and delusion behavior. He also had asthma and diabetes mellitus type 2. On admission the patient seemed distant, and he did not answer any questions. On initial physical examination the patient weighed 138 pounds. The initial Mini–Mental State Examination (MMSE) score was 0/30 and the initial total testosterone 160 ng/dL [240–1000 ng/dL]. After informed consent from the family, he was started on an anabolic steroid primarily to increase functionality, but his cognitive parameters were observed.

Fig. 7–1 shows that after implementation of testosterone the MMSE score improved from 0/30 to 5/30, the clock drawing test (CDT) from 0/4 to 1/4, and the Tinetti scale, a fall assessment tool, from 5/2 to 20/28. This case study demonstrates that the use of testosterone in selected AD patients may possibly help with the improvement of cognition and functionality. AD is a terminal disease with little hope of improvement and any treatment modality that would improve the quality of life of a patient is to be investigated further. This patient tolerated the treatment well. It is difficult to conclude if the testosterone treatment alone or the good supportive care resulted in the improvement of the patient.

Do Men Have a Lower Probability of Alzheimer’s Disease and How Can Men Protect Themselves Against It?

To the casual observer, it seems that AD is less common among men. A walk through a nursing home where AD is prevalent would more likely result in an encounter with more female residents than male residents. Men on average live 7 years less than women, and this may account for the seemingly higher prevalence of AD in women, as women tend to live longer and AD is a function of age.¹ However, a recent study conducted in Rochester argues that AD may be as common in men as in women if corrected for the age differences,² though gender-specific risk factors for AD cannot be excluded. For instance, smoking in men has been shown to increase the risk of AD.³ Likewise, HIV-positive men who are carriers of the e4 allele of apolipoprotein E (ApoE) have an increased risk of AD.⁴

The risk of AD in men may be reduced by several factors such as the use of antioxidants, moderate alcohol consumption, and exercise.^5,6 The Honolulu Heart Study examined 3385 men, and those who prophylactically consume vitamins C and E had a 88% reduction in risk for AD.⁵ Another study from the Netherlands suggests that moderate alcohol intake may also decrease the risk for AD in men.⁶ Several epidemiological studies have suggested that lifestyle is important in reducing risk of AD in men. For example, Honolulu Asia Aging Study found that men who walked and socialized regularly exhibited an 80% decreased risk of AD.⁵

Altered neuroendocrine regulation of cognition may also predispose men to AD.⁷ Men, when they age beyond 50 years may suffer from androgen decline, sometimes termed andropause or androgen decline in aging males (ADAM).⁸ This decline in androgen occurs more gradually as compared with women in menopause. The decline is more marked with bioavailable testosterone than with total testosterone, partly because of the changing affinity of sex hormone–binding globulin (SHBG) with aging.⁷ It has been observed that older men who have undergone androgen replacement often report improved memory function or report feeling “brighter.” This suggests that androgen replacement may play an important role in improving memory in men entering into a transitional stage of lowered hormones. Low levels of androgen are not restricted to risk of AD per se but have also been reported for risk of vascular dementia.^9–11 Androgen replacement therapy was associated with improved cognition in vascular dementia patients.^12–14

Another piece of the puzzle has been the observation that gonadotrophins such as the luteinizing hormone (LH) and the follicle-stimulating hormone (FSH) tend to be elevated in AD, and it is unclear if the hypogonadal state results in a rise in these gonadotrophins. Attempts have been made to use antigonadotrophins to treat AD, but this may sound irrational, as there might be further induction of the hypogonadal state, which in itself is suspected to be a risk factor for AD. As mentioned, AD remains a complex illness and has many facets yet to be defined. To date, the use of acetyl cholinesterase inhibitors for AD has not been very successful in terms of a cure. As such, it is prudent to prod other possibilities for therapeutics, as the disease is relentless and devastating.

As of this writing, there are five drugs that have been approved in the United States by the Food and Drug Administration (FDA) for treating the cognitive symptoms of Alzheimer’s disease: memantine, galanthamine, rivastigmine, donepezil, and tacrine. Tacrine needs close liver function monitoring. These drugs belong to the class of cholinesterase inhibitors. Each acts in a different way to delay the breakdown of acetylcholine. It is believed that acetylcholine in the brain facilitates communication among nerve cells and is thus important for memory. It is not possible to measure patients’ level of acetylcholine and as such treatment by and large is empirical and based on a clinical diagnosis. The reader is referred to other texts for a more detailed discussion of these treatments. Table 7–1 summarizes the drugs used to treat progression of Alzheimer’s disease.

In general, these drugs are most effective when treatment is begun in early stages. They have all been shown to modestly slow the progression of cognitive symptoms and reduce problematic behaviors in some people. Unfortunately, at least 50% of patients do not respond to the acetylcholinesterase inhibitors. Although the effect of these medications is modest, studies show that when they do work, they can make a significant difference in a person’s quality of life and day-to-day functioning (activities of daily living, ADL). The main differences between the drugs are the side effects they produce and the number of times they must be taken daily (Aricept is taken once daily, Exelon and Reminyl are taken twice a day).

One of the new treatments FDA approved for treatment of cognitive problems is memantine. Overstimulation of the N-methyl-D-aspartate (NMDA) receptor by glutamate is implicated in neurodegenerative disorders. In a double-blind, placebo-controlled trial, patients receiving memantine had a better outcome than those receiving placebo.¹⁵ Memantine was not associated with a significant frequency of adverse events. Antiglutamatergic treatment reduced clinical deterioration in moderate-to-severe AD, a phase associated with distress for patients and burden on caregivers, for which other treatments are not available. Many drugs as well as nutraceuticals have been studies including vitamin E and gingko biloba, both of which look promising.

Laboratory Models of Androgens and Brain Function

Estrogens have been thought to be protective against AD in women, but treatment of AD patients with estrogens has by and large not been successful.¹⁶ Animal studies have indicated that testosterone does play an important role in cognition. Many researchers have postulated that the action is through the androgen receptors (ARs). It has been discovered that AR and its associated messenger RNA is abundant in the rat brain. The cerebral cortex and hippocampus are areas that are rich in ARs.¹⁷ In one study the blockade of ARs in mice expressing the major AD genetic risk factor, the ∊4 of apolipoprotein E (ApoE 4) resulted in the development of prominent deficits in spatial learning and memory. Paradoxically, female neuron-specific enolase (NSE)—ApoE4 mice, which are prone to impairments in spatial learning and memory, exhibited memory improvement in response to androgen treatment (Fig. 7–2).¹⁸ In addition, cytosolic AR levels were decreased in NSE-ApoE4 mice and improved memory treatment in the NSE-ApoE4 female mice was associated with increased cytosolic AR levels.

Reproductive hormones influence the development of the central nervous system (CNS). For instance, androgens influence neuritic arborization and the receptive field of individual cells, whereas estrogen induces this communication by forming spines, synapses, and gap junctions. As such, estrogens and androgens act in different but complementary ways to modulate neural development and organization.¹⁹ Castration of animals serves as a model to observe cognitive changes. For example, it has been shown that sexually intact male dogs were significantly less likely than neutered dogs to progress from mild to severe impairment.²⁰ These observations are consistent with the notion that circulating testosterone in aging sexually intact male dogs slows the progression of cognitive impairment. The senescence-accelerated mouse (SAMP) exhibits age-related learning and memory deficits in the ability to perform inferential tasks.²¹ These memory deficits are associated with a corresponding decline in testosterone levels with age. Testosterone replacement improved age-related impairment of learning and memory.²² It has been postulated that cognitive impairment of SAMP8 mice may be related to an interaction of aging and lowered testosterone levels.

Intracellular neurofibrillary tangle formation as a result of tau protein hyperphosphorylation is one of the major neuropathological hallmarks of AD. It has been demonstrated that heat shock–induced hyperphosphorylation of tau protein in the brain of orchiec-tomized male rats can in fact be reduced by testosterone.23 Interestingly, this effect was not seen with estrogens, indicating it was a direct effect of testosterone or its metabolite dihydrotestosterone. It has also been demonstrated that treatment with testosterone decreases the secretion of amyloid β (Aβ) peptide from cerebrocortical neurons of rats.²⁴ Evidence is accumulating to indicate that changes in Aβ precedes tau hyperphosphorylation.²⁵ In addition to reducing Aβ levels, testosterone and dihydrotestosterone can protect cultured hippocampal neurons of rats from cell death by directly blocking Aβ toxicity via an estrogen-independent mechanism.²⁶ Testosterone also exerts its neuroprotective effects through receptor-mediated mechanisms^27,28 by stimulating increased secretion of a metabolite of amyloid precursor protein (sAPP),²⁹ which has been demonstrated to exhibit neurotrophic properties.

Dehydroepiandrosterone sulfate (DHEAS) is synthesized in situ in the brain. Therefore, it has been termed a neurosteroid. Numerous animal studies have demonstrated the neuroprotective and memory-enhancing effect of DHEAS. DHEAS concentrations decline with age in various animals, including the rhesus monkey.³⁰ The data on DHEA and cognition is conflicting. One study demonstrated that cognitively impaired and unimpaired aged rhesus monkeys did not differ in their DHEAS levels.³¹ DHEA replacement enhanced memory in aged mice in several experiments.^27–29 In experiments, DHEA has been shown to significantly reverse pharmacologically induced neurotoxicity with NMDA, dimethylsulfoxide kainic acid, H₂O₂, dizocilpine, ethanol and scopolamine.^4,32–34 Possibly, DHEA may be useful in treating memory impairment based on its antioxidant and neuroprotective effects in the hippocampus. It has been reported that DHEA increased the production of amyloid precursor protein (APP) and promoted the release of its metabolite sAPP from PC12 cells.^35,36 Although the increased production of APP per se may lead to acceleration of mild cognitive disorder presumably as a result of increased Aβ production, the increased APP produced in DHEAS-treated cells is processed through the non-amyloidogenic pathway to release APP, which is known to have neuroprotective properties that may be mediated in part by its ability to promote neurite outgrowth.^36–38 The effects of DHEA on cognition may be mediated through its nongenomic action on several neurotransmitter receptors, including cholinergic neurons.³⁹ Despite the promising data from animals, the effects of androgens on human cognition are less clear and will be discussed in the following sections.

Older men often complain of memory problems. A Texas study of 302 older men found that 36% of patients identified themselves as experiencing andropause on a standardized questionnaire survey and reported memory loss as a symptom.40 Another study in Perth, Western Australia, which employed a similar questionnaire given to 100 men treated for low testosterone, found initial symptoms included memory loss and lapses in concentration (85%), fatigue (96%), mood problems (83%), and loss of libido (77%) (Linda Byart, personal communication). Hormone replacement therapy in this cohort resulted in 86% of men mentioning an improvement in memory and attention span.

Investigators have examined the effect of androgen substitution or withdrawal upon various study populations. By and large, these studies are limited by population size, but are very encouraging. A double-blind study conducted by Janowsky et al⁴¹ demonstrated that transdermal testosterone enhanced spatial cognition of healthy men aged 60 to 75 years when testosterone levels were raised for a 3-month period to those commonly found in young men. Endogenous production of estradiol was decreased in men receiving testosterone supplementation, and estradiol levels correlated inversely to performance on tests of spatial cognitive skills.²⁵ A 6-week randomized, double-blind, placebo-controlled study of healthy older men aged 50 to 80 years found that 100 mg of testosterone enanthate improved both spatial and verbal memory.⁴² Another small randomized, double-blind study of healthy volunteers aged 61 to 75 years demonstrated improved working memory following intramuscular injections of testosterone enanthate 150mg/week. Better performance on tasks involving frontal lobe–mediated working memory was related to a higher testosterone-to-estrogen ratio.²⁸ The subjects in these studies were healthy volunteers who did not have AD.

Studies involving testosterone substitution have historically been confounded by concurrent changes in estradiol levels and include patient samples that may not be generalized to the population of elderly male patients experiencing low-testosterone syndrome. A 12-month randomized, controlled trial of 32 hypogonadal men (bioavailable testosterone <60 ng/dL), aged 51 to 75 years, failed to demonstrate any difference in verbal or nonverbal memory with 200 mg testosterone cypionate biweekly; however, visuospatial ability was not specifically examined. The authors hypothesized that their inability to demonstrate the more generalized effects on cognition may have been due to the rise in estradiol levels in the treatment group.⁴³

Sexual reassignments, because of the use of sex steroids, gave an opportunity for researchers to observe the effects of hormones on brain function. For instance, Slabbekoorn et al⁴⁴ demonstrated a profound effect of androgen treatment on spatial ability in female-to-male transsexuals (FMs) over a period of 18 months. After 3 months of cross-sex testosterone treatment, 25 FMs demonstrated a significant improvement on a three-dimensional rotated figures task of spatial ability. Untreated male-to-female transsexuals (MFs) had higher scores on visuospatial tasks than untreated FMs transsexuals.⁴⁴

Recently, in a small 8-week randomized, controlled trial, testosterone was shown to exhibit a curvilinear dose-response effect for different verbal and spatial cognitive functions. Supraphysiological levels of testosterone (200 mg testosterone enanthate/week) reduced visuospatial ability and improved verbal fluency in 30 healthy young men aged 19 to 45 years.⁴⁵ This finding confirms a previous study that showed a curvilinear relationship between spatial performances and circulating testosterone concentrations with optimal effects on spatial performance, associated with intermediate plasma levels of testosterone.³⁸ This is consistent with the hypothesis first proposed by Janowsky et al⁴¹ that high levels of testosterone may be aromatized to estradiol within the brain.

The development of new imaging technology has enabled investigators to explore neuropharmacology in a functional manner. A small study of six men with hypogonadotropic hypogonadism when given testosterone demonstrated enhanced cerebral glucose metabolism as assessed by using F18-deoxyglucose positron emission tomography (PET) with increased visuospatial capacity during a three-dimensional mental rotation task.⁴⁶ Brain localization on PET was consistent with those areas that were observed to be activated by such tasks in a previous study on eugonadal men.²³

The hypothesis that gonadal replacement therapy might prevent or delay AD in postandropausal men has recently been supported by a study of the effect of gonadal hormone withdrawal on the levels of plasma (Aβ. This 4-kd peptide is the major protein component of the extracellular amyloid plaques, which is a neuropathological hallmark of AD47 and is thought to play a key role in its pathogenesis.⁴⁸ Plasma Aβ levels increased in men with prostate cancer in response to flutamide (250 mg, three times daily) and leuprorelin acetate (22.5 mg, weekly for 12 weeks) treatment to lower plasma testosterone. This increase in plasma Aβ paralleled a rapid decrease in both plasma testosterone and 17β-estradiol, and remained elevated over a 6-month period while gonadal hormones remained at low detectable levels (Fig. 7–3).⁴⁹ However, regulation of Aβ production represents only one of several potential mechanisms that need evaluation. Some of these other potential candidate mechanisms are based on testosterone’s ability to block neuronal apoptosis⁵⁰ as well as indirectly to prevent tau phosphorylation,²³ to increase the production of the neurotrophic agent nerve growth factor⁵¹ and insulin growth factor-1,⁵² and to reverse the age-related increases in glial fibrillary acidic protein (GFAP).⁵³ Increased GFAP reflects activation of astrocytes, which have been implicated in playing a more prominent role in the pathogenesis of AD.⁵⁴

The adrenal steroid hormone DHEA is known to decrease with aging and is commonly used as an anti-aging nutritional supplement. Clinical studies of the effects of DHEA and its sulfate DHEAS on cognition have produced conflicting results. Fifty-two community-dwelling AD patients (mean age 76 years) were compared with a control group of age- and gender-matched healthy elderly men and women on a test of everyday memory. No differences were seen between the AD patients and controls in DHEAS levels. However, AD patients with higher levels of DHEAS scored better than those with lower levels.⁵⁵ A longitudinal study of 883 community-dwelling men (mean age at entry 53 years, range 22–91 years) was conducted for as long as 31 years (mean 11.5 years), with biennial assessments of multiple cognitive domains. Neither the rates of decline in mean DHEAS nor the mean DHEAS concentrations within individuals were related to cognitive status or cognitive decline.⁵⁶ It should be noted, though, that most of the previous studies are relatively small, and a definitive outcome must await properly controlled large-scale studies where dose effects are also evaluated, as recently reported by Racchi et al.⁵⁷

FIGURE 7-3. Plasma concentration of (A) β amyloid, (B) testosterone, and (C) estradiol before and after flutamide and leuprorelin acetate treatment. (Adapted from Gandy S, Almeida OP, Fonte J, et al. Chemical andropause and amyloid-B peptide. JAMA 2001;285:2195–2196, with permission.)

The finding that gonadotrophins may be raised in AD has led to the possibility of using antigonadotropins such as leuprolide in the treatment of AD.58 In individuals with AD, there is a twofold elevation in the serum concentrations of the gonadotropins, LH, and FSH compared with age-matched controls. Whether elevated gonadotropins actually cause AD or merely represent a feedback response to low testosterone levels remains to be determined. However, we have recently found that LH does promote increased Aβ secretion in cultured neurons (G. Verdile, C. Atwood, and R. Martins, unpublished results). LH has been localized in the cytoplasm of pyramidal neurons where it is found to be increased together with neurofibrillary tangles in AD brain compared with age-matched control brain. Although the functional consequences of increased neuronal LH are unknown, it is notable that LH is primarily localized to those neurons that are known to be vulnerable to AD-related neurodegeneration. Elevated serum and cortical neuron levels of LH, coupled with the decline in sex steroid production, could play important roles in the pathogenesis of AD. However, if a role for LH in AD is established, leuprolide treatment per se, while reducing LH, would also further deplete the already low testosterone levels, which are important for both reducing Aβ levels and simultaneously increasing the levels of the neurotrophic factor, sAPP. Thus testosterone replacement would be needed to complement the protective effect of leuprolide.

Discussion of Case History

Currently, there is no recommendation of the use of androgens for AD, and it is not FDA approved. The treatment of this patient with advanced AD was based on his hypogonadal state. The aim was to increase functionality with an anabolic hormone, and offer him a better quality of life. The observed improvement in cognition was a bonus, and he did become more communicative. He had improvements in his MMSE and CDT (which is a measure of visual-spatial skills). His functionality as measured by the functional instrument of measure (FIM) scale and the ADL scale also improved. Paradoxically, it has been reported in the literature that sexually disruptive male AD patients can benefit from estrogen treatment. This patient did not turn hypersexual, but that has been noted to occur in some AD patients. Carefully monitoring is needed.

Conclusion and Key Points

Many men become hypogonadal as a result of aging, although some men exhibit changes as early as their 40s. It has been established that androgen replacement may help restore energy and libido. An added major benefit may be the restoration of cognitive function. Many patients with advanced AD also suffer from the wasting syndrome, which can lead to falls. Testosterone being an anabolic steroid may help restore functionality in these patients.

• AD is seemingly less common in men because of the shorter life span of men, and AD is a function of aging.

• Men can reduce their risk of AD by consuming vitamin C and E.

• Moderate alcohol, exercise, and socializing may also decrease the risk for AD.

• In in vitro studies, androgens increase neuritic arborization and the receptive field of cells, whereas estrogens induce communication between cells.

38. Sisodia SS, Koo EH, Beyreuther K, Unterbeck A, Price DL. Evidence that β-amyloid protein in Alzheimer’s disease is not altered by normal processing. Science 1990;248:492–495