Tuesday, September 28, 2004
Perspective on the WHI: Where Are We Now?
Bruce McEwen, Ph.D.
Head of the Harold and Milliken Hatch Laboratory of Neuroendocrinology, Rockefeller University
Estrogens (E) and progestogens (P) have profound effects on many tissues and organs throughout the body, including the brain. The effects on the brain include nonreproductive processes such as cognition, mood, pain, and motor coordination. The purpose of this meeting is to define and examine the discrepancies between the findings of the Women's Health Initiative on brain and cognitive function after E and P treatment and the basic research and longitudinal/epidemiological studies that have utilized different formulations, doses, duration of treatment, timing, and mode of administration of E and P.
The meeting will emphasize the basic research on E and P actions and what they tell us about the effects of estradiol and different progestogens on neuron survival, neurochemistry, and structural plasticity. The meeting also will examine the effects of estrogens on brain and cognitive function as they relate to aging. We hope that a major outcome of the meeting will be to determine what information needs to be obtained towards development of further hormone "interventions." A secondary outcome will be to assess what lessons we have learned from studies on E and P that will help in designing clinical trials for other classes of drugs.
I. Basic Science: Effects of Estrogen and Progesterone on CNS Cellular Systems and in Animal Models
A. Neurobiology of Estrogen and Progesterone
Estrogen Mechanisms of Action in Neurons: Factors That Determine Outcome In Vitro as Predictors of Efficacy In Vivo
Roberta Brinton, Ph.D.
Department of Molecular Pharmacology and Toxicology, University of Southern California School of Pharmacy
This presentation proposes three factors determine the efficacy of estrogen (ET) or hormone therapy (HT): (1) The time at which estrogen therapy is initiated. The human clinical data and in vitro preclinical data indicate a healthy cell bias of estrogen action. Estrogen or hormone or exposure early in menopause, in a healthy neuron state, reduces the risk of Alzheimer’s disease, whereas estrogen therapy initiated after the disease has developed or decades following menopause is without benefit. In vitro simulations of a prevention or treatment model of ET are consistent with the outcomes of the human trials. (2) Preclinical basic science analyses indicate vastly different dose response relationships of different ET formulations. (3) The type of progestin used determines the outcome of the therapeutic intervention. Different progestogens promote different cellular outcomes; progesterone confers neuroprotection against neurodegenerative insults, whereas medroxyprogesterone acetate does not and may have deleterious consequences for neuron survival. Insights into estrogen and progestin mechanisms of action in the brain provide a framework for understanding the paradox of estrogen benefit to prevent Alzheimer’s disease versus the lack of benefit in treatment trials and in trials when hormone therapy is instituted years following menopause.
Based on estrogen-inducible mechanisms, which have been elucidated in healthy neuron model systems, it is predicted estrogen therapy could be highly efficacious in preventing neurodegenerative disease by promoting neuronal defense and memory mechanisms, while also predicting estrogen therapy is an ineffective strategy for reversing the pathology of Alzheimer’s disease. A healthy cell bias of estrogen action hypothesis that provides a unifying mechanism of estrogen action with implications for both the benefits and risks of estrogen therapy is proposed. Our research is supported by NIA, the National Institute of Mental Health, The Kenneth T. and Eileen L. Norris Foundation, The L.K. Whittier Foundation, the Stanley Family Trust, and the John Douglas French Foundation.
Brinton RD. Impact of estrogen therapy on Alzheimer's disease: a fork in the road? CNS Drugs 2004;18(7):405–422.
Nilsen J, Brinton RD. Divergent action of progesterone and medroxyprogesterone acetate (Provera): impact on estrogen-induced nuclear mitogen-activated protein kinase (MAPK) signaling in hippocampal neurons. Proceedings of the National Academy of Sciences 2003;100:10506–10511.
In Defense of Estradiol and Progesterone
C. Dominique Toran-Allerand, M.D.
Columbia University College of Physicians and Surgeons; New York, NY
The brain of both sexes is a major target of estradiol and a site of estrogen synthesis, not only during development, but throughout adulthood. Estradiol has important influences on neuronal structure and function in the brain. In addition to secreting 17ß-estradiol, the ovary, like the testis, also secretes testosterone. Many of the actions of testosterone in the brain depend on its initial intraneuronal conversion, through aromatization to the estrogenic metabolite, 17ß-estradiol, which represents an additional important source of estrogen for the brain. Estrogens exert many effects on the human brain. In addition to their well established actions on reproductive function through influences on pituitary hormone release, estrogens also influence higher cognitive functions, pain mechanisms, fine motor skills, susceptibility to seizures, mood, temperature regulation, and sleep.
Estrogens also have neuroprotective effects with respect to damage from age-associated, neurodegenerative states such as Alzheimer’s and Parkinson’s diseases and from multiple sclerosis, schizophrenia, and ischemic stroke. After menopause, a declining stimulus from circulating estradiol and testosterone levels (both of ovarian origin), are believed to make estrogen target neurons in the brain less resistant to age- or disease-related processes. Unlike many other molecules that elicit increases in the levels of their own receptors, prolonged estrogen exposure downregulates or decreases estrogen receptor levels. The longer one is on unopposed estrogen without interruption, the greater the extent of receptor downregulation, and the greater the likelihood that, with time, estrogen target neurons will gradually become less sensitive to estrogen, resulting in a diminution of any beneficial effects.
Discontinuing estrogen in a cyclical manner for a few days, as occurs normally during the menstrual cycle, elicits upregulation and a return to normal receptor levels, allowing neurons to become more responsive to estrogen. Exposure of cultured slices of the developing rodent hypothalamus, preoptic area, and neocortex to 17ß-estradiol enhances the growth and differentiation of axons and dendrites, referred to collectively as "neurites." Estrogen stimulation of neurite growth is developmentally regulated and not normally seen in the adult brain.
However, re-expression of this developmental response occurs in the adult, following loss of trophic support, whether induced by estrogen deprivation (ovariectomy) or brain injury, and estradiol can again be shown to influence the growth and differentiation of neurite-derived structures such as axons, dendrites, dendritic spines, and synapses. The negative results of the Women's Health Initiative (WHI) studies must be interpreted within the context of the extensive literature on the biology of the gonadal steroid hormones and their receptors in humans, lower primates, and rodents and should not be used to "damn" all forms of hormone treatment. The therapeutic potential of estradiol and progesterone has not yet received a proper scientific evaluation. It is imperative the science underlying the observed negative results of the WHI studies, namely the failure to consider the biology of estradiol and progesterone and of their receptors be properly and fully addressed in the future.
Toran-Allerand CD. Mini-review: a plethora of estrogen receptors. Where will it end? Endocrinology 2004;145(3):1069–1174.
B. Cognitive Function: Age-Related Changes in Estrogen/Progesterone Levels and Cognitive Function
Hormone Therapy and Cognitive Performance: Reconciling Animal Studies With Clinical Data
Robert Gibbs, Ph.D.
University of Pittsburgh School of Pharmacy
Animal studies provide compelling evidence that, following the loss of ovarian function, hormone therapy consisting of either estradiol or a combination of estradiol and progesterone can enhance cognitive performance and reduce age-related cognitive decline. Several plausible biological mechanisms for these effects have been identified. Among these, recent studies suggest effects on hippocampal function, partly mediated by effects on basal forebrain cholinergic afferents, underlie many of the preclinical effects on cognitive performance that have been described. Notably, the animal studies also suggest hormone therapy must be initiated within a limited time period following the loss of ovarian function to be effective, and the regimen of hormone therapy is extremely important. Most randomized clinical trials, including the WHIMS trial, focus exclusively on effects in women greater than or 65 years of age. We we believe this is past the "window of opportunity" for eliciting a beneficial effect of hormone therapy on cognitive performance.
In addition, studies suggest—in contrast to the beneficial effects of estradiol alone or cyclical administration of estradiol and progesterone—long-term simultaneous combination therapy is not beneficial and may have adverse effects on basal forebrain cholinergic function and cognitive decline. These findings are consistent with many of the recent observational studies conducted in humans and help to account for the negative effect of hormone therapy on dementia observed in the recent WHIMS trial. Therefore, despite these negative findings, we predict postmenopausal hormone therapy can help prevent age-related cognitive decline in women, provided therapy is initiated at menopause at a dose and regimen appropriate for the brain.
Gibbs RB, Gabor R. Estrogen and cognition: applying preclinical findings to clinical perspectives. Journal of Neuroscience Research 2003;74(5):637–643.
Estrogen Influences on Cognitive Aging in Monkeys
Peter Rapp, Ph.D.
Fishberg Department of Neuroscience, Mount Sinai School of Medicine; New York, NY
Although there is a substantial basis for supposing ovarian decline might influence the course of normal cognitive aging, the effects of menopause and hormone therapy in women have proved difficult to define. Inconsistencies in published evidence appear at least partly attributable to methodological factors, including differences in the specific dose, formulation, and schedule of hormone administration, and variability in the age at which treatment is initiated relative to the onset of menopause. Recent studies in the laboratory addressed some of these issues using a controlled, prospective design in an established nonhuman primate model, testing whether cyclic estrogen therapy following surgical menopause affects normal cognitive aging.
Current results indicate estrogen administration can broadly influence performance in aged ovariectomized monkeys across a battery of neuropsychological assessments, including testing procedures sensitive to the functional integrity of the prefrontal cortex and medial temporal lobe memory system. By comparison, parallel research in young adult monkeys reported more subtle, task-selective effects. These findings suggest sensitivity to estrogen decline and estrogen therapy may increase with advancing age, and they establish a valuable primate model for exploring the neurobiological basis of these effects.
Results from collaborative investigations (to be discussed by Dr. John Morrison) indicate cyclic estrogen treatment can substantially modulate the circuit organization of the prefrontal cortex and hippocampus in monkeys, providing a potential substrate for the observed cognitive influence of treatment. Supported by NIH grants AG10606 and AG 10675.
Roberts JA, Gilardi KVK, Lasley BL, Rapp PR. Reproductive senescence predicts cognitive decline in aged female monkeys. Neuroreport 1997;8:2047–2051.
Estrogen and the Aging Cortical Synapse: Implications for Cognitive Effects in Aged Monkeys
John H. Morrison, Ph.D.
Professor and Chair, Neuroscience; Mount Sinai School of Medicine
The recent controversies surrounding hormone and estrogen therapy (HT and ET) in women have highlighted the need for more extensive use of controlled, prospective studies of nonhuman primates (NHPs) to elucidate the potential for ET-induced cognitive enhancement in aged subjects and the neurobiological underpinnings of such putative behavioral effects. From this perspective, we have been investigating the neurobiological effects of cyclic ET following surgical menopause of young and aged monkeys and whether such treatment affects normal cognitive aging (see presentation by P. Rapp). Our data from short-term treated young and aged female NHPs have demonstrated a robust induction of spines in CA1 with this cyclical ET regimen. In addition, there is a robust E-induced induction of spines in dorsolateral prefrontal cortex following two cycles of cyclical ET.
Recently, we began analyzing hippocampal and prefrontal synaptic indices in aged animals that have undergone long-term cyclical ET, with ongoing assessment of both hippocampal and prefrontal-dependent cognitive tasks (See P. Rapp). These studies revealed an E-induced increase in CA1 axospinous synapses in aged monkeys that stands in contrast to data suggesting aged rats fail to display E-induced synaptic increases in CA1. In addition, quantitative neuronal reconstructions in dorsolateral prefrontal cortex of these aged behaviorally characterized monkeys demonstrate a robust increase in spine number of Layer 3 pyramidal neurons. Furthermore, this increase appears to correlate with performance on the delayed response task, but not on hippocampal-dependent tasks, suggesting E-induced spine increases in this particular class of pyramidal cells may be a substrate for the enhanced cognitive performance induced by this ET regimen. Supported by AG 10675 and AG 06647.
Hao J, Janssen WGM, Tang Y, Roberts JA, McKay H, Lasley B, Allen PB, Greengard P, Rapp PR, Kordower JH, Hof PR, Morrison JH. Estrogen increases the number of spinophilin-immunoreactive spines in the hippocampus of young and aged female rhesus monkeys. The Journal of Comparative Neurology 2003;23:5708–5714.
Tang Y, Janssen WGM, Hao J, Roberts JA, McKay H, Lasley B, Allen PB, Greengard P, Rapp PR, Kordower JH, Hof PR, Morrison JH. Estrogen replacement increases spinophilin-immunoreactive spine number in the prefrontal cortex of female rhesus monkeys. Cerebral Cortex 2004;14:215–223.
C. Neuroprotection
Role of Nonfeminizing Estrogens in Brain Protection From Cerebral Ischemia: An Animal Model of Alzheimer's Disease Neuropathology
James W. Simpkins, Evelyn Perez, ShaoHua Yang, and Yi Wen
University of North Texas Health Science Center; Fort Worth, Texas
Estrogens are potent and efficacious neuroprotectants both in vitro and in vivo in various neurotoxicity models. We determined the structural requirements for neuroprotection in an in vitro assay, using a library of more than 70 novel estratrienes, synthesized to reduce or eliminate estrogen receptor (ER) binding. We observed neuroprotection could be enhanced by as much as 200-fold through modifications that positioned a large bulky group at the C2 or C4 position of the phenolic A ring of the estratriene. Furthermore, substitutions on the B, C, or D rings either reduced or did not markedly change neuroprotection. For this library of compounds, there was a negative correlation between binding to ERs and neuroprotection with the more potent compounds, showing no ER binding. In an in vivo model for neuroprotection, transient cerebral ischemia, efficacious compounds were active in protection of brain tissue from this pro-oxidant insult. Finally, estradiol protected brains from insult-induced Alzheimer’s disease neuropathology, including activation of apoptosis, stimulation of Aß production, hyperphosphorylation of tau, activation of cyclin-dependent kinases, and activation of catastrophic attempts at neuronal mitosis.
Collectively, these results demonstrate nonfeminizing estrogens are neuroprotective and protect brain from the induction of Alzheimer's disease-like neuropathology in an animal model. These features of nonfeminizing estrogens make them attractive compounds for assessment of efficacy in Alzheimer's disease and stroke, as they are not expected to show the side effects of chronic estrogen therapy mediated by ER-medicated actions in the liver, uterus, and breast. Supported by grants AG 10485 and AG 22550.
Liu RSH, Yang E, Perez KD, Yi S, Wu S, Eberst K, Prokai L, Prokai-Tatrai K, Cai ZY, Covey DF, Day AL, Simpkins JW. Neuroprotective effects of a novel non-receptor binding estrogen analogue: in vitro and in vivo analysis. Stroke 2002;33:2485–2491.
Wen Y, Onyewuchi O, Yang S, Liu R, Simpkins JW. Increased ß-secretase activity and expression in rats following transient cerebral ischemia. Brain Research 2004;1009:1–8.
Wen Y, Yang SH, Liu R, Brun-Zinkernagel AM, Koulen P, Simpkins JW. Transient cerebral ischemia induces aberrant neuronal cell cycle reentry and Alzheimer's disease-like tauopathy in female rats. Journal of Biological Chemistry 2004;279:22684–22692.
Yang SH, Wang J, Simpkins JW. Novel strategies for neuroprotection against acute brain damage. In: A Textbook on Neurointensive Care. AJ Layon, A Gabrielle, and WA Friedman (eds). W.B. Saunders Company. 200. pp 849–863.
Ovarian Steroids, Neuroinflammatory Responses, and Aging
Caleb E. Finch, Ph.D.
Andrus Gerontology Center and Department of Biological Sciences, University of Southern California; Los Angeles, CA
The challenges of optimizing hormone therapy for cognitive health include activation of peripheral inflammatory processes and direct effects on glia. Orally administered estrogens may elevate serum C-reactive protein (CRP), which is a risk marker for cardiovascular and cerebrovascular events. CRP is more than a bystander risk indicator by powerful biological effects that include activation of complement and uptake of LDL by macrophages. Elevated CRP, moreover, was associated with risk of dementia, 25 years later (Schmidt et al. 2002).
Estrogens also act directly on brain neurons and glia. The glial targets, although less well described, include many candidates for interventions in dementias. Neuroprotective effects of E2 on glia could result from the stimulation by E2 of increased Aß uptake by microglia (and phagocytosis in general) (Li et al. 2000). A more specific tie to inflammation is that E2 also blocked the activation of NFkB in glial clutures exposed to Aß (Dodel et al.1999). However, the role of extracellular amyloid in the inflammation during Alzheimer's disease or in normal human brain aging is unknown.
How does aging alter glial responses to estrogen and inflammatory stimulae? We and others find progressive activation of astrocytes and microglia during normal aging (e.g., twofold increase of CD68-immunopositive microglia in the cortico-striatal bundles and GFAP in astrocytes in hippocampus and cortex). We recently found that aging impairs the E2-dependant astrocytic support of neurite outgrowth (Rozovsky et al., in press). These changes are unexpected, because they arise in the absence of definable neuropathology in rodent models that do not accumulate brain amyloid.
Li R, Shen Y, Yang LB, Lue LF, Finch C, Rogers J. Estrogen enhances uptake of amyloid beta-protein by microglia derived from the human cortex. Journal of Neurochemistry 2000;75:1447–1454.
Rozovsky I, Wei M, Morgan TE, Finch CE. Reversible age impairments in neurite outgrowth by manipulations of astrocytic GFAP. Neurobiology of Aging 2004 (in press).
II. Clinical Studies
Cardiovascular Health and Cognition: Perspectives on Using the Primate as a Model for Human Research
Thomas B. Clarkson, D.V.M.
Comparative Medicine Clinical Research Center, Wake Forest University School of Medicine; Winston-Salem, NC
Among nonhuman primates, the macaques (rhesus and cynomolgus monkeys) are the most widely used for studies of estrogen deficiency and estrogen therapy effects on cardiovascular health and cognition. Studies of cynomolgus monkeys, ovariectomized (OVX) for either short or long periods, have contributed important insights about cardiovascular health both relative to estrogen deficiency and estrogen therapy (Karas and Clarkson, Menopausal Medicine, 2003; Mikkola et al., Annals of Medicine, 2004). Studies with the OVX cynomolgus model have shown:
- Estrogen deficiency results in coronary artery vasoconstriction in response to acetylcholine (vasodilation returns with estrogen therapy)
- Estrogen therapy inhibits the progression of coronary artery atherosclerosis by about 70 percent if initiated soon after OVX—whereas, it has no beneficial effects if initiation of treatment is delayed for a period equivalent to 6 postmenopausal years for women (the basis of the KEEP study)
- The perimenopausal period may be the stage of reproductive life with the greatest acceleration in progression of atherosclerosis and the time of greatest benefit of estrogen treatment
- Psychosocial stress results in premenopausal estrogen deficiency and progression of coronary atherosclerosis
Rhesus and cynomolgus monkeys also have contributed to the estrogen-cognition conundrum. Neither OVX of cynomolgus monkeys for periods up to 6 years in women equivalents nor estrogen treatment of OVX monkeys for equivalent to 3 years had any effect on learning and memory (Voytko, Behavioral Neuroscience, 2000); whereas age-matched rhesus monkeys had cognitive decline associated with the peri- and postmenopause (Roberts et al., Neuroscience Report, 1997).
Because of differences in the reproductive hormone milieu of OVX versus naturally menopausal subjects, there is uncertainty about the relevance of data on OVX monkeys for naturally menopausal women and, consequently, a need for a better monkey model.
Mikkola TS, Clarkson TB. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovascular Research 2002;52:605–619.
Karas RH, Clarkson TB. Considerations in interpreting the cardiovascular effects of hormone replacement therapy observed in the WHI: timing is everything. Menopausal Medicine 2003;10(4):8–12.
Mikkola TS, Clarkson TB, Notelovitz M. Postmenopausal hormone therapy before and after the women's health initiative study: what consequences? Annals of Medicine 2004;36(6):1–12.
Consequences of Surgical Menopause on Cognition and Mood
Barbara B. Sherwin, Ph.D.
McGill University; Montreal, Canada
Premenopausal women who undergo a surgical menopause experience classic menopausal symptoms—such as hot flashes—with greater severity and frequency than those who undergo a spontaneous menopause. Although the exact mechanisms for this phenomenon are unknown, the excess symptomatology in surgically menopausal women is thought to be because of the inability of the HPG axis to adapt to the abrupt withdrawl of ovarian hormones. Evidence that neurotransmitter levels are more severely disrupted orginates from the finding that surgically menopausal women also are at greater risk of dysphoric moods and depression compared to older, spontaneously menopausal women. The relatively few prospective studies on surgically menopausal women indicate they experience decrements in aspects of cognitive functioning—at least in the 6 months following bilateral oophorectomy—which cannot be attributed to aging. Findings from studies of young women undergoing “medical oophorectomy” with LHRH analogs support this conclusion.
Because the ovaries produce approximately one-third of the androgen in women, testosterone (T) levels also decrease significantly during surgical menopause. Evidence from randomized, controlled trials indicate some surgically menopausal women experience decreased libido and energy-level postoperatively that can be reversed—in many cases—by the addition of T to an estrogen therapy regimen.
The ramifications of premature ovarian removal or atrophy affect other organ systems and other populations of women. For example, women deprived of ovarian hormones because they experience premature ovarian failure or those who have been treated with radiation or chemotherapy for nonestrogen-dependent cancers are prematurely vulnerable to diseases of aging (e.g., osteoporosis and coronary heart disease). Little or no information is currently available to determine whether or not they also experience earlier and more severe deficits in cognition with aging.
Finally, dysphoric mood and depression may independently affect cognitive functioning in women, and it is important—both from a clinical and from an experimental perspective—to separate the effects of mood on cognitive functioning from the possible direct effects of hormones on these brain functions.
Sherwin BB. Estrogen and cognitive functioning in women. Endocrine Reviews 2003;24(2):133–151.
CNS Effects of Hypogonadism and Estrogen Therapy in Humans
Schmidt PJ, Roca CA, Rubinow DR
Behavioral Endocrinology Branch, National Institute of Mental Health; Bethesda, MD
In some women, the perimenopause may be a time of increased risk for developing depression. We have employed several strategies to investigate the effects of reproductive aging, hypogonadism, and estrogen therapy on mood. The three groups of women studied were:
- Women with the onset of depression occurring during the natural perimenopause
- Asymptomatic premenopausal women followed prospectively with behavioral and reproductive measures until their final menses (FMP)
- Women with GnRH agonist-induced hypogonadism.
Perimenopausal depressed women did not differ from nondepressed controls in basal measures of reproductive hormones; however, estradiol therapy significantly improved mood in depressed perimenopausal women under double-blind, placebo-controlled conditions. Twenty-nine euthymic, premenopausal women were followed for 5 years (on average) until FMP. Nine of the twelve observed episodes of depression (six of which were first onset) occurred in the 24 months surrounding FMP, suggesting the late perimenopause is a period of increased risk for depression. In younger women, hypogonadism was associated with hot flashes and changes in libido, but not mood or cognitive function. Additionally, in younger women hypogonadism eliminated—and estradiol therapy restored—the characteristic pattern of cognition-activated regional cerebral bloodflow in the prefrontal cortex. Our data suggest hypogonadism does not induce uniform effects on mood, but alterations in estradiol modulate mood in the context of perimenopausal depression and the activity of brain regions implicated in the regulation of mood and cognition.
Schmidt PJ, Nieman L, Danaceau MA, Tobin MB, Roca CA, Murphy JH, Rubinow DR. Estrogen replacement in perimenopause-related depression: a preliminary report. American Journal of Obstetrics and Gynecology 2000;183:414– 420.
Berman KF, Schmidt PJ, Rubinow DR, Danaceau MA, Van Horn JD, Esposito G, Ostrem JL, Weinberger DR. Modulation of cognition-specific cortical activity by gonadal steroids: a positron-emission tomography study in women. Proceedings of the National Academy of Sciences of the United States of America 1997;94:8836–8841.
Schmidt PJ, Haq NA, Rubinow DR. A longitudinal evaluation of the relationship between reproductive status and mood in perimenopausal women. American Journal of Psychiatry (in press).
The Differential Effects of Estrogen on Mood and Cognition in Peri- and Postmenopausal Women
C. Neill Epperson, M.D.
Director, Yale Behavioral Gynecology Program
Background
Changes in neuroendocrine function may predispose menopausal women to psychological disturbances characterized by depressed mood, anxiety, irritability, fatigue, insomnia, forgetfulness, and decline in libido. The acute tryptophan (TRP) depletion paradigm examined serotonergic contribution to mood and cognitive function in menopausal women who were within 4 weeks of recovery from an episode of a major depressive disorder (MDD) and in healthy menopausal women pre- and posttreatment with estrogen.
Methods
Menopausal women with depression responsive to treatment with estradiol 75 ug/d (N=4); the selective serotonin reuptake inhibitor (SSRI) fluoxetine 20–40 mg/d (N=4); or estradiol 75 ug/d and fluoxetine 20 mg/d (N=3) and peri- and postmenopausal healthy controls (N=18) underwent active and sham TRP depletion sessions. Healthy menopausal women underwent the procedures both before and after estradiol 75 ug/d treatment (ET) for 3 months. Mood and cognition were assessed using standard clinician-administered tests before and 6 hours after administration of the amino acid mixture.
Results
Although active TRP depletion resulted in an 89 percent decline in total and free plasma TRP levels, neither menopausal women who had recently recovered from an episode of MDD or healthy menopausal women pre- or postestrogen administration experienced a worsening of their mood during active TRP depletion. In contrast, TRP depletion was associated with a worsening of performance on the delay paragraph recall subtest of the Wechsler Memory Scale (WMS) in both groups. In addition, the healthy menopausal women experienced a significant worsening in performance on the delayed paired associates subtask of the WMS during TRP depletion prior to estrogen administration. However, after estrogen administration, healthy controls showed no difference in performance on these measures between sham and active TRP depletion sessions.
Conclusions
Results from this pilot study, indicating menopausal women who have recently recovered from an MDD do not experience a worsening of mood with acute TRP depletion is in contradistinction to findings of mood worsening in other previously depressed populations undergoing the same procedures. Although preliminary, the results are suggestive of a distinct—perhaps less serotonin mediated—pathophysiology for depression during menopause. Consistent with previous literature, TRP depletion was associated with worsening of performance on several verbal working memory tasks; in particular, tasks requiring delayed information recall. Importantly, estrogen administration appeared to provide protection against the effects of rapid TRP depletion and serotonin depletion in healthy controls who were tested before and after ET. Although the perimenopausal women in this study appeared to benefit most from estrogen administration, baseline differences in performance may be responsible.
Epperson CN, Naftolin F, Cappiello A, Czarkowski KA, Stiklus S, Anderson GM, Charney DS, Krystal JH. Resistance to depressive relapse in menopausal women undergoing tryptophan depletion: preliminary findings. Psychopharmacology (in review).
Epperson CN, Gueorguieva R, Cavus I, Krystal JH. The impact of tryptophan depletion on cognitive function in peri and postmenopausal women pre and post estrogen replacement. North American Menopause Society, Chicago, IL; October 4, 2002.
Epperson CN, Wisner KI, Yamoto B. Gonadal steroids in the treatment of mood disorders. Psychosomatic Medicine 1999;61:676–697.
Estrogen or No Estrogen: Long-Term Followup
Natalie Rasgon, M.D., Ph.D.
Department of Psychiatry and Behavioral Sciences, Stanford School of Medicine
This presentation will focus on regional cerebral metabolism, as evidenced by position emission tomography (PET) with 18FDG in women at risk for Alzheimer’s disease and effects of estrogen use on the brain metabolism in this population.
Among risk factors for Alzheimer's disease, most notable are mood disorders, hypothyroidism, and familial and genetic risk for Alzheimer's disease. Gender differences in prevalence rates for both mood disorders and Alzheimer’s disease are approximately 2:1. Combined risk for affective disorders and Alzheimer’s disease is present only in women, also suggesting an estrogen effect. Thyroid dysfunction has been associated with reduced brain metabolism, cognitive decline, and depression. A relationship between estrogen and thyroid hormone is suggested by an increased hypothyroidism rate among women, especially aging women. Data on regional cerebral metabolism in hypoestrogenic (postmenopausal) women with major depressive disorder and hypothyroid-depressed women is presented. In addition, estrogen use among older persons at familial and/or genetic risk for Alzheimer’s disease will be protective of regional cerebral metabolism. These data, related data, and a new longitudinal study investigating effects of estrogen discontinuation on cerebral metabolism and cognition in women at risk for Alzheimer's disease will be reviewed.
Dunkin J, Rasgon NL, Fairbanks L, Elman S, Altshuler LL, Rapkin A. Estrogen use in treatment of postmenopausal depression: neurocognitive correlates. Psychoneuroendocrinology (in press).
Rasgon NL, Silverman D, Siddarth P, Miller K, Ercoli LM, Elman S, Lavretsky H, Huang SC, Phelps ME, Small GW. Estrogen use and brain metabolic change in postmenopausal women. Neurobiology of Aging (in press).
Low Dose Estrogen, fMRI, and Cognitive Function
Vincent P. Clark, Ph.D.
The MIND Institute Department of Psychology, University of New Mexico
Although several functional neuroimaging studies examined estrogen therapy’s (ET) effects on brain activity associated with learning and memory, no study has examined such effects on a simple attention task. A pilot study examined the effects of low-dose ET on behavioral performance and brain function during a visual target detection task. Hemodynamic activity evoked by target and nontarget stimuli were examined using event-related functional magnetic resonance imaging (ER-fMRI). Participants included 16 women between ages 73 and 84 who were part of a randomized controlled study to evaluate the effect of ultralow dose micronized estradiol on bone (Prestwood et al. 2003). No significant differences in behavioral performance were found with ET. However, there was evidence ET group participants altered hemodynamic responses to stimuli in various subcortical and cortical brain areas, including occipital, motor, parietal, anterior cingulate, and other prefrontal cortical structures. The findings are interpreted as support for the idea that estrogen may facilitate the efficiency of brain function in postmenopausal elderly women.
Prestwood KM, Kenny AM, Kleppinger A, Kulldorff M. Ultralow-dose micronized 17-beta-estradiol and bone density and bone metabolism in older women: a randomized controlled trial. Journal of the American Medical Association 2003;290(8):1042–1048.
Cognitive and Brain Aging: Using Imaging To Distinguish Potential Risks and Benefits of Estrogen
Susan Resnick, Ph.D.
NIA
Studies of estrogen-containing hormone therapy and cognition in postmenopausal women have yielded inconsistent findings. Several small randomized trials of younger women following surgical menopause and larger observational studies of older postmenopausal women, including the Baltimore Longitudinal Study of Aging (BLSA), suggested benefits of hormone therapy on some cognitive functions, particularly verbal memory. In contrast, a large trial of combination conjugated equine estrogens (CEE) and medroxyprogesterone acetate (MPA) in older women with cardiovascular disease did not support these findings, and data from the Women’s Health Initiative Memory Study (WHIMS) suggested small detrimental effects of combination CEE plus MPA and CEE alone on changes in a global measure of cognitive function. In an ancillary study to WHIMS, the effects of hormone therapy on annual change rates in specific cognitive functions and affect in 2,302 older postmenopausal women without dementia were investigated.
Results of the Women’s Health Initiative Study of Cognitive Aging (WHISCA) CEE and MPA subtrial indicated that the intervention affected only some cognitive functions and that these effects differed across different memory types. Furthermore, effects on specific aspects of memory were evident only after an average of 4–5 years of treatment. These findings indicate hormone therapy differentially affects specific cognitive functions and suggests hormone therapy may have competing risks and benefits on cognition. Thus, it will be important to incorporate imaging assessments into future studies to distinguish potential risks, such as increases in vascular events or inflammation, from potential benefits, such as decreased brain atrophy or amyloid deposition.
Resnick SM, Maki PM. Effects of hormone replacement therapy on cognitive and brain aging. Annals of the New York Academy of Sciences 2001;949:203–214.
Maki PM, Resnick SM. Effects of estrogen on patterns of brain activity at rest and during cognitive activity: a review of neuroimaging studies. Neuroimage 2001;14:789–801.
Vascular Disease in the Brain: An Important Cause of Dementia?
Lewis H. Kuller, M.D., Dr.P.H.
University of Pittsburgh
The results of the Women's Health Initiative cognition study (WHIMS) were unexpected, given the previous reports of benefits of estrogen therapy on cognition. The Cardiovascular Health Study (CHS) of older individuals and the Atherosclerotic Risk in Communities (ARIC) study (in 45–64 year age group) and population studies external to the United States reported the high prevalence of subclinical vascular disease in the brain, particularly white matter grade (WMG) abnormalities and subcortical infarcts. The CHS cognition study reported both high WMG and infarcts are predictors of dementia. Furthermore, in the CHS, both WMG and infarcts were strong, independent predictors of risk of stroke.
The increased risk of stroke in the estrogen and estrogen-progesterone arms of WHI suggests the therapy was associated with: (1) increase in brain vascular disease, (2) many women previously on or currently taking estrogen or estrogen-progesterone may be at high risk, and (3) estrogen and estrogen-progesterone should be provided with great caution—even for younger women—until evaluation of subclinical brain vascular risk is completed.
The vascular disease in the brain is likely preventable by treatment of vascular risk factors, especially in "middle ages." It could be prudent to recommend postmenopausal women (who were on hormone therapy) to maximize adherence to current guidelines for treatment of vascular risk factors (i.e., blood pressure, lipids, diabetes, and antiplatelet therapy).
Women have much less coronary atherosclerosis than men throughout their lifetime. The lower prevalence of atherosclerosis among women is particularly unique to the coronary arteries, as compared to the aorta and to the carotid arteries. These results are consistent in both pathology and imaging studies. Women have lower incidence of myocardial infarction than men throughout their lifetime. They never "catch up" with men. This also is true when comparing diabetic men and women. The extent of coronary atherosclerosis is, however, a primary determinant of the risk of coronary heart disease (CHD) among both men and women, and there is solid evidence the major risk factors for CHD (i.e., blood HDL cholesterol and LDL cholesterol levels, cigarette smoking, hypertension, diabetes) are the major determinants of CHD among both men and women.
We recently have noted that women have less global brain ventricular atrophy as measured by ventricular size on an MRI than do older men. The extent of the ventricular atrophy is an independent predictor of the risk of dementia. High WMG and subcortical infarct—markers of vascular disease in the brain—are associated with the extent of "ventricular atrophy." Markers of peripheral vascular disease, such as increased carotid intimal medial thickness and its risk factors, hypertension, and diabetes, are important predictors of high WMG and subcortical microinfarcts. Women with less ventricular atrophy appear to have lower mortality rates, more than 7–8 years of followup. Older women with less coronary atherosclerosis, age 80 and older, also have lower total mortality.
The likely mechanisms for these interesting findings include differences in sex steroid hormone metabolism, insulin sensitivity, differences in growth factors and body composition, particularly body fat distribution. The key to the "fountain of youth" may not be hormone therapy, but the prevention of weight gain, central obesity, insulin resistance, hypertension, and abnormal lipoprotein patterns among women.
Kuller LH, Lopez OL, Newman A, Beauchamp NJ, Burke G, Dulberg C, Fitzpatrick A, Fried L, Haan MN. Risk factors for dementia the Cardiovascular Health Cognition Study. Neuroepidemiology 2003;22:13–22.
Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MMB. Silent brain infarcts and the risk of dementia and cognitive decline. New England Journal of Medicine 2003;348:1215–1222.
III. Longitudinal and Epidemiological Studies
Levels of Endogenous Sex Steroids and Risk for Brain Aging
Lenore Launer, Ph.D.
NIA
Few epidemiologic studies have examined the association of endogenous sex hormone levels and the risk for multiple brain aging endpoints. Examining associations with endogenous levels will provide the baseline data to consider when planning trials for hormone therapy. Results from two population-based studies will be reviewed. One study is based on a cohort of men and women 55 years and older from Rotterdam, the Netherlands; the second study is based on a cohort of very old men from the Honolulu Asia Aging Study. Results from the Rotterdam Study suggest endogenous estradiol levels are associated with and increased risk for vascular dementia and smaller hippocampal volumes in women rather than men.
In the Honolulu Asia Aging Study, neither testosterone or estradiol were associated with risk for dementia. Testosterone levels were associated with an increased risk for brain atrophy measured on MRI, and estradiol was associated with an increased risk for lacunae. Sex hormone levels were not associated with neuropathologic indicator of brain aging. In population-based studies, there is no indication endogenous sex hormone levels are associated with reduced brain aging.
Geerlings MI, Launer LJ, de Jong FH, Ruitenberg A, Stijnen T, van Swieten JC, Hofman A, Witteman JC, Pols HA, Breteler MM. Endogenous estradiol and risk of dementia in women and men: the Rotterdam Study. Annals of Neurology 2003;53:607–615.
den Heijer T, Geerlings MI, Hofman A, de Jong FH, Launer LJ, Pols HA, Breteler MM. Higher estrogen levels are not associated with larger hippocampi and better memory performance. Archives of Neurology 2003;60(2):213–220.
Geerlings MI, Ruitenberg A, Witteman JC, van Swieten JC, Hofman A, van Duijn CM, Breteler MM, Launer LJ. Reproductive period and risk of dementia in postmenopausal women. Journal of the American Medical Association 2001;285:1475–1481.
Estrogen Therapy and Risk of Alzheimer's Disease in the Baltimore Longitudinal Study of Aging
Dr. Claudia Kawas, M.D.
Institution for Brain Aging and Dementia, University of California, Irvine
We investigated the use of estrogen therapy and risk of developing Alzheimer's disease in the Baltimore Longitudinal Study of Aging (BLSA), a prospective multidisciplinary study of normal aging conducted by NIA. The sample consisted of 472 post- or perimenopausal women followed for up to 16 years in the BLSA with prospective documentation of estrogen use at each BLSA visit. Women were categorized as estrogen users if they had used oral or transdermal estrogens at any time. Cox proportional hazards models with time-dependent covariates were used to estimate the relative risk of developing Alzheimer's disease in women who used estrogen as compared with those who had not used estrogen. Approximately 45 percent of the women in the cohort had used estrogen, primarily around the perimenopausal period.
We diagnosed 34 incident cases of Alzheimer's disease (NINCDS/ADRDA criteria) during followup, including nine estrogen users. After adjusting for education, the relative risk for Alzheimer's disease in estrogen users as compared with nonusers was 0.46 (95 percent CI, 0.209–0.997), indicating a reduced risk of Alzheimer's disease for women who had reported estrogen use. The risk of Alzheimer's disease did not differ in women who used estrogen alone as compared to estrogen and progesterone. Our data did not show an effect for duration of estrogen use. This finding is consistent with several other prospective studies showing a protective influence of estrogen in Alzheimer's disease. To date, however, randomized clinical trials have not confirmed this association. It remains to be determined if this is due to issues such as dosing, duration and timing of exposure, or if unknown confounders are affecting the observational studies.
Kawas C, Resnick S, Morrison A, Brookmeyer R, Corrada M, Zonderman A, Bacal C, Donnell Lingle D, Metter E. A prospective study of estrogen replacement therapy and the risk of developing Alzheimer's disease: the Baltimore Longitudinal Study of Aging. Neurology 1997;48:1517–1521.
A Population-Based Longitudinal Study of Cognitive Functioning in the Menopausal Transition (SWAN)
Peter Meyer, Ph.D.
Department of Preventive Medicine, Rush Medical Center
Background
There is cross-sectional evidence of increasing reported forgetfulness by age and menopausal status as women progress through the menopausal transition. We conducted a longitudinal study to track cognitive performance as women progressed through the menopausal transition and to determine the impact of the transition on cognition. We hypothesized a decline in cognitive functioning occurs as women progress through the menopausal transition, independent of age, educational level, family income, ethnicity, and baseline self-perceived health.
Methods
We began a population-based, longitudinal study in January 1996 with yearly follow-up interviews. After screening for eligibility (age 42 to 52, premenopausal or early-perimenopausal, no exogenous hormone use in the past 3 months and no hysterectomy), 868 agreed to participate. In October 1996, we began assessing working memory (Digit Span Backward) and perceptual speed (Symbol Digit Modalities Test). Including yearly followups through November 2001, we examined cognitive assessment results for 803 women. Women who became pregnant, had a hysterectomy, or began using hormones were censored from that time onward.
Results
Contrary to our hypothesis, we found small—but significant—increases over time during the premenopausal and perimenopausal phases, which are likely due to learning effects. This trend was not accounted for by chronological age, education, family income, ethnicity, or baseline self-perceived health.
Conclusions
Transition through the menopause is not accompanied by a decline in working memory and perceptual speed.
Meyer PM, Powell LH, Wilson RS, Everson-Rose SA, Kravitz HM, Luborsky JL, Madden T, Pandey D, Evans DA. A population-based longitudinal study of cognitive functioning in the menopausal transition. Neurology 2003;61:801–806.
Estrogen Exposures in Midlife, Memory, and Dementia
Victor W. Henderson, M.D., M.S.
Departments of Health Research and Policy and Neurology and Neurological Sciences, Stanford University
Important questions concern the relation between memory or dementia and the loss of ovarian estrogen production characteristic of menopause. Because memory complaints are common during this time of life, one question is whether menopause is linked to impaired memory. A second is whether estrogen loss increases a woman’s risk of Alzheimer’s disease. Much of our information comes from studies of hormone therapy (HT) in older women, considerably beyond the age of menopause. Some animal studies imply that hormone effects might depend on the age at which therapy is initiated or the timing of initiation in relation to menopause.
Because objective measures of episodic memory are seldom examined in relation to natural menopause, women in the Melbourne Women’s Midlife Health Project (age 52 to 63 years) were assessed with a word-list learning task (Henderson et al., Neurology 2003). Estrogen exposures for 326 women who had not undergone surgical menopause were inferred from menopausal status, time from the final menstrual period, HT use, and serum estradiol concentration. Cross-sectional analyses revealed no relation between episodic memory performance and these estrogen exposures. For natural menopause, these findings imply episodic memory is not substantially affected during the menopausal transition or early postmenopause.
With respect to Alzheimer's disease, experimental evidence from the Women’s Health Initiative Memory Study in postmenopausal women older than age 64 shows that HT initiation increases dementia risk. In apparent contrast, observational studies, based partly on hormone exposures in younger women, link HT to Alzheimer's disease risk reductions. The Multi-Institutional Research in Alzheimer Genetic Epidemiology study examined the relation between HT used for more than 6 months and Alzheimer's disease risk in 426 women with Alzheimer's disease and 545 nondemented controls (Henderson et al., Journal of Neurology, Neurosurgery, and Psychiatry, in press). There was a significant interaction between age and HT use, and in stratified analyses, a significant protective association occurred only in the youngest age tertile (50–63 years; OR = 0.35, 95 percent CI = 0.19–0.66). Among other possibilities, this finding is consistent with the view that HT may protect younger women from AD or that HT used during the menopausal transition or early postmenopause may reduce AD risk. (Alzheimer’s Association IIRG-01-2684; NIH AG09029)
Henderson VW, Guthrie JR, Dudley EC, Burger HG, Dennerstein L. Estrogen exposures and memory at midlife: a population-based study of women. Neurology 2003;60:1369–1371.
Henderson VW. Hormone therapy and Alzheimer's disease: benefit or harm? Expert Opinion on Pharmacotherapy 2004;5:389–406.
Henderson VW, Benke KS, Green RC, Cupples LA, Farrer LA. Postmenopausal hormone therapy and Alzheimer's disease risk: interaction with age. Journal of Neurology, Neurosurgery, and Psychiatry (in press).
Are Epidemiologic Studies Worthless as an Indication of Trial Results, or Does “Nixon's Law” (Timing Is Everything) Prevail?
John C.S. Breitner, M.D., M.P.H.
Director, GRECC, VA Puget Sound Health Care System; and University of Washington; Seattle, WA
Epidemiological studies that suggest “protection” against dementia have provided impetus for several candidate strategies of prevention against Alzheimer’s disease. Prominent among these strategies is treatment with nonsteroidal anti-inflammatory drugs (NSAIDs) and use of hormone therapy (HT) at menopause and beyond. However, at least two treatment trials with NSAIDs (one with two treatment arms) have failed, and several randomized trials have tested HT as a treatment for Alzheimer's disease, with dismal results. Does this mean the epidemiological data are worthless, or is there a critical distinction between treatment and prevention? It is now clear that “secondary prevention” with NSAIDs (i.e., prevention of Alzheimer's disease onset in those with Mild Cognitive Impairment [MCI]) does not work, at least not with selective COX-2 inhibitors. The WHIMS results show HT exposure exaggerates Alzheimer's disease risk within the first few years of treatment. What is going on?
Epidemiological data may provide an explanation. The Rotterdam Study showed a fivefold reduction in Alzheimer's disease incidence with 2 or more years of NSAID prescriptions, but there was no effect of NSAID exposures within 2 years of dementia onset. The 2 years before onset is when cases of incipient Alzheimer's disease will mostly have MCI. The Cache County Study similarly showed no benefit of current NSAID exposure at baseline, but a 60 percent reduction in Alzheimer's disease incidence with past exposures. The HT data are more dramatic. Two early cohort studies showed strong protection against Alzheimer's disease with lifetime exposure to HT, but made no distinction in timing of exposure relative to onset of Alzheimer's disease. Presumably, many of the HT users in these studies had exposures from menopause onwards.
The Cache County Study showed similar “protection” with lifetime HT exposure, but no effect of exposures within the prior 10 years. In fact, such recent exposures were associated with a doubling of Alzheimer's disease risk (i.e., similar to what was seen with recent exposures in the WHIMS trial). We suggest—for both NSAIDs and HT—the epidemiological data are not only consistent with available trials data; they explain the disparities among different studies. Near-term prevention with HT seems impossible, but prevention with NSAIDs may be demonstrable, provided exposures predate disease onset by a few years.
Breitner JC, Zandi PP. Effects of estrogen plus progestin on risk of dementia. Journal of the American Medical Association 2003;290(13):1706–1707.
Zandi PP, Carlson MC, Plassman BL, Welsh-Bohmer KA, Mayer LS, Steffens DC, Breitner JC. Cache County Memory Study Investigators. Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County Study. Journal of the American Medical Association 2002;288(17):2123–2129.
Shumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, Hendrix SL, Jones III BN, Assaf AR, Jackson RD, Kotchen JM, Wassertheil-Smoller S, Wactawski-Wende J, WHIMS Investigators. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the 1 Women's Health Initiative Memory Study: a randomized controlled trial. Journal of the American Medical Association 2003;289(20):2651–2662.
Methodologic Challenges in the Nonexperimental Study of Hormones and Dementia: Findings From the Kaiser Permanente Women's Memory Study
Diana B. Petitti, M.D., M.P.H.; Valerie Crooks, D.S.W.
Department of Research and Evaluation, Kaiser Permanente Southern California
Epidemiologic studies of the hormone use and dementia show mixed results, although several suggest a lower risk of dementia in hormone users. These studies, particularly those that use the case-control design and rely on self-reported prior use of hormones, are subject to several biases.
We have conducted a study that involved baseline cross-sectional assessment of dementia and cognitive impairment among 3,924 women 75+ years of age who were members of Kaiser Permanente Southern California Medical Care Plan in 1998 and who had been continuously enrolled in the plan since 1992. Hormone therapy (HT) was determined using computer-stored prescription data for the period 1992–1998. Women classified as HRT users at baseline filled prescriptions for an oral estrogen in each of the 7 years prior to entry to the study. The mean duration of HRT use at baseline was 27 years (+ SD 11), and the mean age at initiation of HRT use was 50 years (+ SD 11). Cognitive function and dementia were assessed using the Telephone Interview of Cognitive Status supplemented by the Telephone Dementia Questionnaire and medical record review.
At baseline, women with cognitive impairment or dementia who were classified as HRT users by prescription were less likely than women without impairment to report they had used HRT.
| |
Prescription Classification |
Prescription Classification |
Prescription Classification |
| |
HRT User
N=1,657 |
Nonuser
N=1,621 |
HRT User
N=134 |
Nonuser
N=130 |
HRT User
N=68 |
Nonuser
N=71 |
| HRT by Self-Report |
|
|
|
|
|
|
| Current HRT User |
91.2% |
4.8% |
82.1% |
3.9% |
72.1% |
2.8% |
| Not Current User |
8.8% |
94.1% |
17.2% |
95.4% |
25.0% |
91.5% |
| Unknown/Not Sure |
0.0% |
1.2% |
0.8% |
0.8% |
2.9% |
5.6% |
The adjusted OR for dementia in HRT users by prescription compared with nonusers of HRT by prescription was 0.79 (95 percent C.I. 0.62 to 1.01). The adjusted OR for dementia based on self-reported current HRT use was lower than when based on prescription data: 0.68 (95 percent confidence interval 0.47, 0.97). Use of self-reported information on HRT use would have led to an erroneous conclusion about the association of HRT use with dementia.
Our study demonstrates empirically that women who are demented or cognitively impaired do not accurately recall their prior hormone use. Studies that rely on recall of hormone use as the basis for classification of exposure to HRT are subject to bias. The bias would overestimate the protective effect of hormone use for dementia. Epidemiologic studies must validate recalled hormone use or base the classification of HRT exposure on an independent information source to avoid “dementia recall bias.”
Buckwalter JG, Petitti DB, Crooks VC. Estrogen replacement and risk of Alzheimer disease. Journal of the American Medical Association 2003;289:1101–1102 (letter).
Petitti DB, Buckwalter JG, Crooks VC, Chiu V. Prevalence of dementia in users of hormone replacement therapy as defined by prescription data. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 2002;57:M532–M553.
Wednesday, September 29, 2004
IV. Clinical Trials
A. Prevention and Treatment Trials
Hormones and CHD: Discrepancies Between Laboratory Studies, Observational Studies, and Clinical Trials
Dr. Jacques Rossouw, M.D.
National Heart, Lung, and Blood Institute
Both primary and secondary prevention trials of hormones have found no benefit (or harm) for CHD. Laboratory studies elucidated several mechanisms by which estrogen may improve coronary arterial physiology. They also found mechanisms by which estrogen might increase coagulation, inflammation, and might trigger coronary events in advanced lesions. Animal studies suggest hormones may retard early atherosclerosis, while both animal studies and human angiographic trials indicate hormones do not influence the progression of raised lesions. Hormone use in the primary prevention cohort studies is associated with lower risk of CHD. The hormone use in these studies would mostly have started at the age of menopause—in women whose arteries would be closer to normal on average than women in the clinical trials.
One postulate worthy of further study is that estrogen may have a beneficial effect in normal or near-normal arteries, but the opposite effect in the presence of established atherosclerosis. However, at the average age of menopause, a substantial proportion of women had raised lesions, and a smaller proportion had advanced lesions. Also, the apparent benefit of hormone use was found in secondary prevention observational studies (i.e., in women with compromised arteries). It is likely that uncorrected biases in the observational studies lead to an overestimation of any benefit of hormone use. Endogenous estradiol may be responsible for the later onset of coronary disease in women compared to men; if so, then the appropriate test of the estrogen hypothesis would employ transdermal estradiol in a young population of postmenopausal women. Some parallels can be drawn between the data on hormones and CHD and data on hormones and dementia.
Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM, Ockene J, Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's Health Initiative randomized controlled trial. Journal of the American Medical Association 2002;288(3):321–333.
Anderson GL, Limacher M, Assaf AR, Bassford T, Beresford SA, Black H, Bonds D, Brunner R, Brzyski R, Caan B, Chlebowski R, Curb D, Gass M, Hays J, Heiss G, Hendrix S, Howard BV, Hsia J, Hubbell A, Jackson R, Johnson KC, Judd H, Kotchen JM, Kuller L, LaCroix AZ, Lane D, Langer RD, Lasser N, Lewis CE, Manson J, Margolis K, Ockene J, O'Sullivan MJ, Phillips L, Prentice RL, Ritenbaugh C, Robbins J, Rossouw JE, Sarto G, Stefanick ML, Van Horn L, Wactawski-Wende J, Wallace R, Wassertheil-Smoller S, Women's Health Initiative Steering Committee. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. Journal of the American Medical Association 2004;291(14):1701–1712.
Manson JE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL, Trevisan M, Black HR, Heckbert SR, Detrano R, Strickland OL, Wong ND, Crouse JR, Stein E, Cushman M, Women's Health Initiative Investigators. Estrogen plus progestin and the risk of coronary heart disease. New England Journal of Medicine 2003;349:523–534.
Grodstein F, Clarkson TB, Manson JE. Understanding the divergent data on postmenopausal hormone therapy. New England Journal of Medicine 2003;348:645–650.
Design and Analysis of the Women's Health Initiative Memory Study
Steve Rapp, Ph.D.
Professor of Psychiatry, Public Health Sciences, Dermatology; Wake Forest University School of Medicine
The Women’s Health Initiative Memory Study (WHIMS) is the largest RCT of the effects of estrogen-alone (E-alone) and estrogen plus progestin (E+P) on the incidence of dementia, mild cognitive impairment, and changes in global cognitive functioning. Women (7,480) between the ages of 65 and 79 years at enrollment were studied in two parallel trials. Women were evaluated annually and followed for an average of 4 years (E-alone) to 5 years (E+P). The results from both trials were contrary to the prediction of a protective effect, suggested by prior studies. These unexpected findings have raised questions and highlighted the need for additional analyses and studies. This presentation will review the methods used in WHIMS and the key findings. Planned follow-up studies and analyses also will be described.
Implications of the Women's Health Initiative Memory Study
Dr. Sally Shumaker, M.D.
Public Health Sciences, Wake Forest University of Health Sciences
The Women's Health Initiative Memory Study (WHIMS) results have challenged researchers to pause and rethink evidence, suggesting a protective effect of hormone therapy on cognition in postmenopausal women. Reconciling disparate results is difficult and requires consideration of several methodological issues. This presentation will discuss how the design of WHIMS influenced the results. Also, we will review how the unique WHIMS dataset allows additional questions about such issues as the duration of effect, effect in “younger” women participating in Women's Health Initiative, and possible underlying mechanisms. Limitations of WHIMS will be presented, along with a discussion of the methodological and ethical constraints and ideal characteristics of future clinical trials on HT and dementia.
Shumaker SA, Legault C, Kuller L, Rapp SR, Thal L, Lane DS, Fillit H, Stefanick ML, Hendrix SL, Lewis CE, Masaki K, Coker LH, Women's Health Initiative Memory Study. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. Journal of the American Medical Association 2004;291(24):2947–2958.
Espeland MA, Rapp SR, Shumaker SA, Brunner R, Manson JE, Sherwin BB, Hsia J, Margolis KL, Hogan PE, Wallace R, Dailey M, Freeman R, Hays J, Women's Health Initiative Memory Study. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women's Health Initiative Memory Study. Journal of the American Medical Association 2004;291(24):2959–2968.
PREPARE: Preventing Postmenopausal Alzheimer's Disease With Replacement Estrogen
Mary Sano, Ph.D.
Professor, Department of Psychiatry; Director, Research and Development; Mount Sinai School of Medicine, Bronx VAMC
The PREPARE Study was designed as a double-blind, placebo-controlled multicenter trial of estrogens for the prevention of Alzheimer's disease supported through the NIA-funded grant # RO 1 AG15922. This controlled trial was based on a wide range of observational and laboratory data that suggested estrogen had a beneficial cognitive and neuronal effect. Like the Women’s Health Initiative (WHI), it selected conjugated equine estrogens—the most commonly used form of estrogen at the time of study initiation—as the agent of interest.
The PREPARE Study recruited women aged 65 and older as participants. Like the Women's Health Initiative Memory Study (WHIMS), the primary outcome was incident dementia and used a two-tiered process for assessment of dementia with the first depending on results of cognitive testing and the second being a comprehensive evaluation of those who did not satisfactorily pass the first tier. There were several differences. For example, the PREPARE Study selected only those who were cognitively healthy based on a screening assessment and attempted to exclude those with cardiac risk factors. Most importantly, the study only recruited those who had a family history of dementia in a first-degree relative based on the assumption this would accelerate cognitive change, thereby shortening the trial. Additionally, the neuropsychological battery played a major role in this study, and a composite memory score was generated and acted as an alternate primary outcome with efficacy based on significance in either measure. Initially, the trial was based on 5 years of observation rather than a specific number of outcomes, theoretically allowing us to see the effects of continuous estrogens on cognitive and functional outcomes occurring after conversion to dementia.
During the course of this study, the results of the WHI were monitored and informed changes in our protocol—consent process and information given to our patients. Cardiovascular and other health outcomes for which WHI was originally formulated resulted in discontinuation of this study. However, the PREPARE Study was not affected until results of cognitive outcomes were revealed. Currently, consistent with the WHIMS report of worse cognition and dementia outcomes in the estrogen-treated arms, we have discontinued study medication, but continued blinded followup. This decision is being evaluated and will be reviewed by the DSMB.
Despite drug discontinuation, we can learn several things from this study. First, we may be able to examine differences in participation, retention, and treatment effect among those with a family history of dementia. A common observation of positive estrogen effect in epidemiological data is that it reflects a delayed outcome from a treatment taken earlier in life. Thus, this study may permit us to assess a closer parallel to the epidemiological phenomenon. A third equally important opportunity will be to assess the use of family history of dementia as an enrichment strategy. In addition, this cohort—with a relatively wide range of ages—will allow us to examine the interaction between family history and age on cognitive outcomes. Finally, this presentation will address the need to create a process to integrate results from other ongoing studies in the clinical trial plan, a concept important to any trial with known agents.
Martin BK, Meinert CL, Breitner JC, ADAPT Research Group. Double placebo design in a prevention trial for Alzheimer's disease. Controlled Clinical Trials 2002;23:93–99.
Sano M, Bell K, Jacobs D. Cognitive effects of estrogens in women with cardiac disease: what we do not know. The American Journal of Medicine 2002;113:612–613.
Silverman JM, Smith CJ, Marin DB, Mohs RC, Propper CB. Familial patterns of risk in very late-onset Alzheimer disease. Archives of General Psychiatry 2003;60:190–197.
Results of the ADCS Estrogen Treatment Trial for AD
Ruth A. Mulnard, R.N., D.N.Sc.
Department of Neurology, Institution for Brain Aging and Dementia; University of California, Irvine
Inspired by reports from several small uncontrolled clinical studies that suggested improvement in cognition for women with Alzheimer’s disease who were treated with estrogen therapy, a randomized, double blind, placebo-controlled trial was conducted to determine the affect of estrogen (unopposed 12-month exposure to Premarin administered at 0.625 md/day versus 1.25 mg/day versus placebo) on global, cognitive, and functional decline in postmenopausal women with Alzheimer's disease. Women (120) with mild to moderate Alzheimer's disease (MMSE scores between 12 and 28) who had had hysterectomies were randomized, with 97 completing the trial.
Based on intent-to-treat analyses comparing combined estrogen groups with placebo and comparing the different dosages of estrogen with placebo, there were no significant differences between treatment groups for the CGIC and the secondary outcome measures (including other measures of global functioning, mood, memory, attention, language, motor function, and activities of daily living) with the exception of the CDR, which suggested worsening among patients on estrogen.
Overall, the results of this study suggest Premarin is not indicated for the treatment of established Alzheimer's disease in women with hysterectomies. Despite our negative results, many observational and natural history studies suggest estrogens may exert a protective effect against the development of Alzheimer's disease. The results of this study do not answer other pertinent questions concerning the effective use of estrogen therapy that differs in formulation, route, regimen, or timing and duration of administration. Only rigorous prospective clinical trials will answer these questions.
Mulnard RA, Cotman CW, Kawas C, van Dyck CH, Sano M, Doody R, Koss E, Pfeiffer E, Jin S, Gamst A, Grundman M, Thomas R, Thal LJ. Estrogen replacement therapy for treatment of mild to moderate Alzheimer's disease: a 1-year randomized controlled trial. Journal of the American Medical Association 2000;283(8):1007–1015.
Thal LJ, Thomas RG, Mulnard RA, Sano M, Grundman M, Schneider L. Estrogen levels do not correlate with improvement in cognition. Archives of Neurology 2003;60(2):209–212.
Alzheimer's Disease: Efficacy of Transdermal 17 Beta-Estradiol
Sanjay Asthana, M.D., F.R.C.P.(C)
University of Wisconsin-Madison Medical School
The neurobiology of estrogen provides convincing evidence the hormone could favorably alter the pathobiology of Alzheimer’s disease. However, recent findings from the Women’s Health Initiative (WHI) and Women’s Health Initiative Memory Study (WHIMS) have raised serious concerns about the safety and feasibility of extended therapy with both unopposed and opposed oral conjugated equine estrogen (CEE) in postmenopausal women. Additionally, the findings of both WHI and WHIMS have identified critical hormone therapy-related issues that need to be systematically evaluated before drawing definite conclusions concerning the safety and efficacy of estrogen.
Among others, some of these issues include the potential differential efficacy and safety of various forms of estrogen (i.e., CEE versus estradiol), effects of different routes of administration (i.e., oral versus transdermal), timing of hormone therapy (perimenopausal versus postmenopausal period), cyclic versus continuous therapy, and choice of the progestational agent (medroxyprogesterone versus natural progesterone). To date, besides two preliminary studies by the author, none of the larger studies involving patients with Alzheimer's disease have systematically evaluated any of the above-identified issues.
Dr. Asthana's presentation will include a brief discussion of prior studies justifying systematic evaluation of hormone therapy-related issues raised by the WHI and WHIMS. Additionally, the presentation will include discussion of data from two randomized studies involving administration of unopposed transdermal 17ß-estradiol to postmenopausal women with Alzheimer's disease. Finally, the presentation will identify pivotal methodological issues (including selection of sensitive neuropsychological tests) that should ideally be addressed in designing studies of hormone therapy for postmenopausal women both with and without dementia.
Asthana S, Baker LD, Craft S, Stanczyk FZ, Veith RC, Raskind MA, Plymate SR. High-dose transdermal estradiol improves cognition for women with AD: results of a randomized study. Neurology 2001;57:605–612.
Asthana S, Craft S, Baker LD, Raskind MA, Birnbaum RS, Lofgreen CP, Veith RC, Plymate SR. Cognitive and neuroendocrine response to transdermal estrogen in postmenopausal women with AD: results of a placebo-controlled, double-blind, pilot study. Psychoneuroendocrinology 1999;24:657–677.
Baker LD, Sambamurti K, Craft S, Cherrier M, Raskind MA, Stanczyk FZ, Plymate SR, Asthana S. 17beta- estradiol reduces plasma Abeta40 for HRT-naive postmenopausal women with Alzheimer's disease: a preliminary study. The American Journal of Geriatric Psychiatry 2003 Mar-Apr;11(2):239–244.
B. SERMS and Cognition
The Effect of SERMs on Cognitive Function and Dementia
Dr. Kris Yaffe
The basic biology of SERMs will be reviewed with particular focus on mechanisms within the central nervous system. Clinical data on SERMs with cognitive outcomes will be reviewed. Results also will be presented in light of findings from the Women's Health Initiative Memory Study. Finally, ongoing trials of SERMs will be discussed.
Yaffe K, Krueger K, Sarkar S, Grady D, Barrett-Connor E, Cox D, Nickelsen T. Cognitive function in postmenopausal women receiving Raloxifene therapy for three years: results from the Multiple Outcomes of Raloxifene Trial. New England Journal of Medicine 2001;344:1207–1213.
Yaffe K, Kreuger K, Cummings S, Blackwell T, Henderson V, Sarkar S, Ensrud K, Grady D. Effect of Raloxifene on prevention of dementia and cognitive impairment in older women: the Multiple Outcomes of Raloxifene Evaluation (MORE) Randomized Trial. The American Journal of Psychology (in press).
C. Design of Clinical Trials
Ethical Framework for the Design and Conduct of Randomized Controlled Trials
Franklin G. Miller, Ph.D.
Department of Clinical Bioethics, NIH
Appreciation of the ethically significant differences between clinical trials and medical care is central to the ethics of randomized controlled trials (RCTs). An ethical framework governing the design and conduct of RCTs will be presented. It consists of seven basic requirements: (1) potential scientific and clinical value of the research question, (2) scientific validity of the study design, (3) fair selection of research subjects, (4) favorable risk-benefit ratio, (5) informed consent, (6) prospective independent review, and (7) respect for enrolled subjects. Ethical issues relating to the enrollment of subjects with diminished capacity to give informed consent will be discussed.
Emanuel EJ, Wendler D, Grady C. What makes clinical research ethical? Journal of the American Medical Association 2000;283:2701–2711.
Chen DT, Miller FG, Rosenstein DL. Enrolling decisionally impaired adults in clinical research. Medical Care 2002;40(9 Suppl):V20-9.
