Therapeutics for Cognitive Aging: Preserving Mental Vitality across the Lifespan

Books

Buccafusco J, ed. 2004. Cognitive Enhancing Drugs (Milestones in Drug Therapy). Birkhäuser, Basel.

Cabeza R, Nyberg L, Park D, eds. 2009. Cognitive Neuroscience of Aging: Linking Cognitive and Cerebral Aging. Oxford University Press, New York.

Craik F, Salthouse T, eds. 2007. The Handbook of Aging and Cognition, 3rd ed. Psychology Press, New York.

Dixon R, Backman L, Nilsson L, eds. 2004. New Frontiers in Cognitive Aging. Oxford University Press, New York.

Hof P, Mobbs C, eds. 2009. Handbook of the Neuroscience of Aging. Academic Press, New York.

Fillit H, Butler R, eds. 2006. Cognitive Decline: Strategies for Prevention, 2nd ed. Cambridge University Press, New York.

Levin ED, Buccafusco JJ. 2006. Animal Models of Cognitive Impairment. CRC Press, New York.

Marin D, Davis K, Aisen P. 1999. Alzheimer's Disease: Questions and Answers, 2nd edition. Merit Publishing International, Jupiter, FL.

Small G. 2003. The Memory Bible: An Innovative Strategy For Keeping Your Brain Young. Hyperion, New York.

Web Sites

Alzheimer's Drug Discovery Foundation
Information about various forms of research and drug discovery for AD, funding opportunities, and news from research and development.

Alzheimer’s Association
Statistics, basic information about the disease, as well as grant information and schedules for ICAD, the annual international meeting on dementia research.

Alzheimer Research Forum
Hosts many resources for scientists who study the molecular mechanisms of this neurodegenerative disease. See a presentation by speaker Lenore Launer on the interpretation of epidemiologic data and dementia and this webinar on whether researchers should develop “smart drugs” to stave off memory loss.

Journal Articles

Paul Aisen

Aisen PS. 2009. Interpreting biomarker data in therapeutic trials. J. Nutr. Health Aging 13: 337-338.

Andrieu S, Coley N, Aisen P, et al. 2009. Methodological issues in primary prevention trials for neurodegenerative dementia. J. Alzheimers Dis.16: 235-270.

Shaw LM, Vanderstichele H, Knapik-Czajka M, et al; Alzheimer's Disease Neuroimaging Initiative. 2009. Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects. Ann. Neurol. 65: 403-413.

George Bartzokis

Bartzokis G, Lu PH, Tingus K, et al. 2008. Lifespan trajectory of myelin integrity and maximum motor speed. Neurobiol. Aging. Oct 15. [Epub ahead of print]

Jack CR Jr, Lowe VJ, Weigand SD, et al. Alzheimer's Disease Neuroimaging Initiative. 2009. Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer's disease: implications for sequence of pathological events in Alzheimer's disease. Brain 132: 1355-1365. Full Text

Kochunov P, Thompson PM, Lancaster JL, et al. 2007. Relationship between white matter fractional anisotropy and other indices of cerebral health in normal aging: tract-based spatial statistics study of aging. Neuroimage. 35(2):478-487.

Jerry Buccafusco

Buccafusco JJ, Beach JW, Terry AV Jr. 2009. Desensitization of nicotinic acetylcholine receptors as a strategy for drug development. J. Pharmacol. Exp. Ther. 328: 364-370.

Buccafusco JJ, Powers JC, Hernandez MA, et al. 2007. MHP-133, a drug with multiple CNS targets: potential for neuroprotection and enhanced cognition. Neurochem. Res. 32: 1224-1237. Erratum in: Neurochem Res. 32: 1719.

Sood A, Beach JW, Webster SJ, et al. 2007. The effects of JWB1-84-1 on memory-related task performance by amyloid Abeta transgenic mice and by young and aged monkeys. Neuropharmacology 53: 588-600.

Steven Ferris

Carrillo MC, Blackwell A, Hampel H, et al. 2009. Early risk assessment for Alzheimer's disease. Alzheimers Dement. 5: 182-196.

De Santi S, Pirraglia E, Barr W, et al. 2008. Robust and conventional neuropsychological norms: diagnosis and prediction of age-related cognitive decline. Neuropsychology 22: 469-484. Full Text

Ferris S, Schneider L, Farmer M, et al. 2007. A double-blind, placebo-controlled trial of memantine in age-associated memory impairment. Int. J. Geriatr. Psychiatry 22: 448-455.

Prichep LS, John ER, Ferris SH, et al. 2006. Prediction of longitudinal cognitive decline in normal elderly with subjective complaints using electrophysiological imaging. Neurobiol. Aging 27: 471-481.

Howard Fillit

Fillit H. 2008. Drug discovery and the prevention of Alzheimer's disease. Alzheimers Dement. 4(1 Suppl 1): S26-S28.

Patrick Hof

Dickstein DL, Kabaso D, Rocher AB, et al. 2007. Changes in the structural complexity of the aged brain. Aging Cell. 6: 275-284. Full Text

Kabaso D, Coskren PJ, Henry BI, et al. 2009. The electrotonic structure of pyramidal neurons contributing to prefrontal cortical circuits in macaque monkeys is significantly altered in aging. Cereb. Cortex. Jan 15. [Epub ahead of print]

Kreczmanski P, Heinsen H, Mantua V, et al. 2009. Microvessel length density, total length, and length per neuron in five subcortical regions in schizophrenia. Acta Neuropathol. 117: 409-421.

Morrison JH, Hof PR. 2007. Life and death of neurons in the aging cerebral cortex. Int. Rev. Neurobiol. 81: 41-57.

Robin Kleinman

Lu T, Pan Y, Kao SY, et al. 2004. Gene regulation and DNA damage in the ageing human brain. Nature 429: 883-891.

Lenore Launer

Cukierman-Yaffe T, Gerstein HC, Williamson JD, et al; Action to Control Cardiovascular Risk in Diabetes-Memory in Diabetes (ACCORD-MIND) Investigators. 2009. Relationship between baseline glycemic control and cognitive function in individuals with type 2 diabetes and other cardiovascular risk factors: the action to control cardiovascular risk in diabetes-memory in diabetes (ACCORD-MIND) trial. Diabetes Care 32: 221-226. Full Text

Kanaya AM, Lindquist K, Harris TB, et al. 2009. Total and regional adiposity and cognitive change in older adults: The Health, Aging and Body Composition (ABC) study. Arch. Neurol. 66: 329-335.

Saczynski JS, Jónsdóttir MK, Garcia ME, et al. 2008. Cognitive impairment: an increasingly important complication of type 2 diabetes: the age, gene/environment susceptibility—Reykjavik study. Am. J. Epidemiol. 168: 1132-1139.

Victoria Luine

Beck KD, Luine VN. 2009. Evidence for sex-specific shifting of neural processes underlying learning and memory following stress. Physiol. Behav. [Epub ahead of print]

Luine VN. 2007. The prefrontal cortex, gonadal hormones and memory. Horm. Behav. 51: 181-182.

Luine VN. 2008. Sex steroids and cognitive function. J. Neuroendocrinol. 20: 866-872.

David Lowe

Rezvani AH, Kholdebarin E, Brucato FH, et al. 2009. Effect of R3487/MEM3454, a novel nicotinic alpha7 receptor partial agonist and 5-HT3 antagonist on sustained attention in rats. Prog. Neuropsychopharmacol. Biol. Psychiatry 33: 269-275.

Timothy Salthouse

Salthouse TA. 2009. When does age-related cognitive decline begin? Neurobiol. Aging 30: 507-514.

Salthouse TA. 2007. Implications of within-person variability in cognitive and neuropsychological functioning for the interpretation of change. Neuropsychology 21: 401-411.

Tucker-Drob EM, Salthouse TA. Adult age trends in the relations among cognitive abilities. Psychol. Aging 23: 453-460.

Gary Small

Small GW, Kepe V, Ercoli LM, et al. 2006. PET of brain amyloid and tau in mild cognitive impairment. N. Engl. J. Med. 355: 2652-2663. Full Text

Small GW, Bookheimer SY, Thompson PM, et al. 2008. Current and future uses of neuroimaging for cognitively impaired patients. Lancet Neurol. 7: 161-172.

Braskie MN, Klunder AD, Hayashi KM, et al. 2008. Plaque and tangle imaging and cognition in normal aging and Alzheimer's disease. Neurobiol. Aging Nov 10. [Epub ahead of print]

Sponsorship

Presented by

Academy Friend

Accera

Media Partners

Alzheimer Research Forum

Cognitive Neuroscience Society

Academy of Cognitive Therapy

New York State Psychological Association

What is cognitive aging, and how do we establish clinically important endpoints?

Can we prevent brain aging?

When is the ideal time to intervene to prevent cognitive aging?

How should cognitive aging be defined?

Will preventing cognitive aging delay or prevent AD?

Will the overall impact of less smoking, more statin therapy, and increased weight in middle-aged people act to reduce or increase late-life dementia?

Why are diseases like Alzheimer's uniquely human?

Why is aging the biggest risk factor for AD?

Why does amyloid begin accumulating in the regions that are last to myelinate?

Could we treat dementia by protecting or repairing myelination?

Why does stress improve performance for female but not male rats on memory tests?

Should aging-related drug discovery be tested in models other than young male rats?

How do choline analogs act at the a-7 nicotinic receptor?

How can biomarkers be used to improve early diagnosis?

Is cognitive aging related to AD risk?

How can we validate biomarkers as surrogate endpoints in cognitive aging?

Speakers:
Jerry J. Buccafusco, Medical College of Georgia
Gary W. Small, University of California, Los Angeles
Paul S. Aisen, University of California, San Diego
David Lowe, Memory Pharmaceuticals
Allan Green, Allan M. Green Esq, LLC

Highlights

• Choline agonists show promise as neuroprotectants and cognitive enhancers.
• Choline agonists may work as receptor desensitizers rather than as a-7 nicotinic agonists.
• A cognitive aging biomarker might improve diagnostic accuracy and aid in early intervention.
• Testing disease-modifying treatments for AD presents major clinical trial design challenges.
• Challenges to cognitive aging treatments include defining cognitive decline and validating biomarkers.

The nicotine fix

Despite uncertainty about the neurobiology of the aging brain, efforts to develop treatments to target cognition are already well underway. More than 30 cognition-enhancing drugs are already in clinical trials, pointed out Jerry Buccafusco of the Medical College of Georgia, but problems have included unexpected side effects, inadequate animal models, narrow dose windows, and conflicts with other common medications of aging. He described his team's efforts to identify novel compounds that will act on multiple targets and have both cognitive enhancing and disease-modifying properties.

In transgenic mice, the choline analog JWB1-84-1 improves performance on the radial-arm task, a test of spatial working memory.

In an attempt to capture the neuroprotective and cognitive-enhancing benefits of nicotine without incurring its dangers, his group has worked with compounds based on choline, a low-potency agonist of the a-7 subtype of the acetylcholine nicotinic receptor. One such analog, JWB1-84-1, protects cells against amyloid damage in C. elegans, and shows properties as a neuroprotective agent and cognitive enhancer in transgenic amyloid-producing mice. It also improves working memory and reduces distractibility in a nonhuman primate. "I would say this compound would be a very good candidate for use in many cognitive impairment paradigms, or syndromes, plus for attention deficit disorder," said Buccafusco.

Choline analogs do not show signs of toxicity in rats or primates. And although they were synthesized as a-7 nicotinic receptor agonists, that does not seem to be how they affect cognition. Buccafusco proposed these and other nicotinic drugs actually desensitize the receptor, which in turn affects GABAergic interneurons. Indeed, results with four compounds suggest that desensitization is correlated with increased cognitive performance. "I never expected to see this in a million years," said Buccafusco. "We're convinced desensitization is the key, and can be considered a legitimate drug development target."

Seeing the brain age

Evaluating potential disease-modifying drugs will require excellent biomarkers, said Gary Small of the University of California, Los Angeles, who focused on potential applications of neuroimaging in the early detection of cognitive decline. Such biomarkers should also make it possible to intervene early on, when treatments would probably be more effective, to identify those patients most likely to decline rapidly, or to monitor treatment.

The imaging technique FDDNP-PET can differentiate Alzheimer's Disease (top row, center) from other neurodegenerative disorders.

Current guidelines suggest that all demented patients be screened with magnetic resonance imaging (MRI) or computed tomography (CT). Two molecular diagnostic probes currently being explored for AD are Pittsburgh Compound B (PiB), which is combined with positron emission tomography (PET) to visualize amyloid plaques, and 18F-FDDNP-PET, which depicts both amyloid and tau pathology. Unlike PiB, said Small, FDDNP can differentiate between frontotemporal dementia, AD, and progressive supernuclear palsy, and its signal changes with disease progression. Many other molecular imaging agents are currently under development, as are techniques such as monitoring brain activity with functional MRI, mapping glucose utilization with FDG-PET, and charting cortical thinning with structural MRI.

Validating a biomarker is a tricky process, cautioned Small. "These signals are complex," he said, and the relationship between detectable changes and underlying biology is not always straightforward. "PiB does not equal amyloid, FDDNP does not equal amyloid or tau, and atrophy does not equal disease progression." Nonetheless, such biomarkers could be extremely useful in practice and in clinical trials. Small described one recent study in which FDG-PET was used to visualize improved glucose metabolism in the brain in middle-aged people treated with a non-steroidal anti-inflammatory (NSAID), which correlated with improved executive function.

Some big problems

Developing a treatment than could actually slow or stop the progress of Alzheimer's Disease rather than just treating symptoms is very important, said Paul Aisen of the University of California, San Diego. "We have treatments that help—not hugely, but they help," he said. "But we have no disease-modifying therapy. The drugs improve symptoms, but they don't change the slope of the decline."

The drugs we have "improve symptoms, but don't change the slope of decline," said Aisen.

However, developing such a therapy may pose "a huge problem," he warned. Clinical trial design would be extremely challenging. To pass regulatory review, a disease-modifying drug for AD must show efficacy by improving clinically relevant symptoms, as well as arresting the process of decline in the longer term. "Disease modification is what we would like, but it's much harder to show the effect," he said. Such a trial would require hundreds or thousands of subjects, and last at least 18 months. It would also be a bit more of a gamble than the average phase 2 trial, as it would not generate an early efficacy signal. All of these problems make it inordinately expensive to test disease-modifying drugs for AD.

Biomarkers such as volumetric MRI to detect hippocampal atrophy or PET molecular imaging might help alleviate the problem, by making it easier to determine early on whether a drug has efficacy. Monitoring biochemical biomarkers in cerebrospinal fluid is another possibility. But to develop a true surrogate biomarker—an index that is validated as a proxy for clinical efficacy—will require effective treatments.

Volumetric MRI shows the clear shrinkage of the hippocampus (in yellow box) with the passage of time—an observation that could be developed as a biomarker.

Extending his observations, Aisen questioned the functional significance of cognitive aging, and the possibilities of treating it, since the only way to justify treatment is to slow the process of decline. "Is there any reason to treat this gradual and functionally insignificant impairment that we call cognitive aging? I don't know," he said. He also cast doubt on the putative link between normal cognitive aging and AD, suggesting that the two may be distinct processes. Other challenges include establishing a surrogate biomarker for cognitive aging, which might be quite different than an AD biomarker, and documenting clinically significant improvement. "From a practical perspective in drug development," he remarked, "I think we have big problems."

The regulatory view

From an industry perspective, cognitive enhancing agents have many potential benefits, said David Lowe of Memory Pharmaceuticals. With an aging population, the potential market is clearly quite large. Cognitive enhancers might also be used in schizophrenia, Parkinson's disease, and other central nervous system disorders. Clinical trial methods and tools such as the Alzheimer's Disease Assessment Scale—cognitive subscale (ADAS-cog) originally developed for AD could be applied to cognitive-enhancement studies, although they are relatively blunt instruments, said Lowe. And there is no shortage of potential targets. One major stumbling block is the lack of regulatory clarity, he said: "It's a long [development] cycle, it's really expensive, and you want to be really sure there is a regulatory pathway." In general, therapeutics for cognitive aging are a high-risk, high-reward domain.

Ideally, a new therapeutic would have both disease-modifying and symptomatic effects, and Lowe mentioned the a-7 nicotinic agonists as one class of agents with such a "dual profile." Many aging-related biological changes remain unexplored, such as mitochondrial dysfunction, neurogenesis, the impact of neurotrophins, and choriod plexus function.

Of the many challenges to developing therapeutics for cognitive aging, Lowe mentioned the intractability of many targets, the lack of good animal models, the absence of a validated biomarker, and trial expense, among other considerations. Future approaches will probably include biological treatments, combined computer/drug therapy, better diagnostics, and treatment for comorbid conditions such as obesity and metabolic syndrome.

Regulatory and ethical issues stand in the way of developing a drug for cognitive aging.

Scrutinizing potential stumbling blocks that might stand in the way of developing a drug for cognitive aging, attorney Allan Green named several regulatory and ethical issues. First of all, the risk versus benefit tradeoff for such a treatment would need to be carefully calculated, since a drug taken to prevent cognitive aging would be given primarily to healthy people. Establishing validated endpoints might prove a challenge, although Green mentioned one example of a drug that was accepted based on its ability to slow pathological changes in the target organ, rather than a proven clinically meaningful change.

Citing the example of Provigil, which was approved for narcolepsy despite the fact that the precise mechanism by which it promotes wakefulness is not known, Green emphasized the importance of a clear definition of the condition, with a clear diagnosis and, ideally, a clinical test. The fact that narcolepsy was a well-defined clinical entity eased the process of approval, and Green suggested that professional societies should be as careful to define age-associated memory impairment.

Ultimately, while the news that cognitive aging is nearly ubiquitous—and, for now at least, virtually untreatable—is disheartening, the full picture of the aging brain is less daunting. Several speakers—many with grey hair themselves—alluded to the nonquantifiable benefits of wisdom. The aged brain may not be as quick or reliable as it is in youth, but it has other strengths. If brains were computers, pointed out organizer Howard Fillit, his college-aged child has much faster processing speed and much more impressive RAM, in comparison with his father. "But the hard drive is empty," he quipped.

From a neurobiological perspective, the fact that the aging brain is characterized by slow and gradual physiological and cognitive changes is also a blessing in disguise. This gentle trajectory of change suggests that if we can identify who might benefit most, early intervention might be effective—and that there's time to help old brains stay young.

Some signs of old age are hard to miss, such as wrinkled skin and grey hair. Cognitive changes are nowhere near as obvious. Nonetheless, just as our bodies shrink and weaken, our brains also change with age. Describing these changes, and considering how they might be prevented, were two of the main questions addressed at a daylong conference on therapeutics for cognitive aging held May 15, 2009 at the New York Academy of Sciences and cosponsored by the Alzheimer's Drug Discovery Foundation.

Nearly all of us experience cognitive aging, and the measurable decline in capacities is profound.

There is no consensus definition for cognitive aging. In some respects, it's easier to say what it isn't: it's not the debilitating dementia of Alzheimer's Disease (AD), nor is it even the significant deficits of the syndrome known as Mild Cognitive Impairment (MCI), which often anticipates full-blown dementia. But the gradual decline in mental capacity known as cognitive aging is neither subtle nor trivial. Beginning as early as age 20, the mental performance of normal healthy adults slowly deteriorates, as measured by a wide range of tests of memory, comprehension, cognitive function, and cognitive speed. The slide begins slowly, accelerating between middle age and old age. In a comprehensive overview of quantitative measures of cognitive aging, Timothy Salthouse of the University of Virginia dispelled the myth that this phenomenon is insignificant or rare: nearly all of us experience it, and the measurable loss of ability is profound.

Is it inevitable? Here, too, there is little agreement. In general, it's clearly the case that 70- and 80-year-olds generally think, speak, and react more slowly than their 18-year-old counterparts. But there is so much individual variability—including a few striking examples of people who do not seem to suffer much of a deficit at all—that it is difficult to conclude that cognitive aging is unstoppable.

The elderly brain

The biology of cognitive aging is poorly understood, although multiple changes during the lifespan seem to be involved. The cognitive problems that characterize neurodegenerative diseases are blamed on neuronal death, but other deficits may be important. Patrick Hof of the Mount Sinai School of Medicine described the profound changes to the synapses of living cells that occur in AD, and explored the altered physiology of cells that are weakened but not yet dead, focusing especially on the pyramidal cells that make long cortico-cortical connections. Victoria Luine of Hunter College discussed the influence of hormones on aging. She found that female rats experienced more severe cognitive aging on some measures, but performed roughly the same as males on others. The integrity of myelin, the fatty coating that sheathes neuronal processes and speeds conduction, may be an essential but frequently overlooked contributor to cognitive changes late in life, argued George Bartzokis of the University of California, Los Angeles. Common cardiovascular conditions in midlife such as diabetes and hypertension also accelerate the process of cognitive aging, found Lenore Launer of the National Institute on Aging.

Is cognitive aging a disease, a disorder, or merely an unpleasant fact of life? Steven Ferris of New York University School of Medicine argued that cognitive aging should be considered a syndrome, and treated as such. Presbyopia—short sight—is "normal" after a certain age, he pointed out, but we correct this problem with glasses rather than enduring it. A strong counterargument was articulated by Paul Aisen of the University of California, San Diego in his talk on clinical trial design. He questioned whether normal cognitive aging should be treated, and emphasized the difficulties in bringing a therapeutic for this indication to market. "Safety is an issue," he said during one of the roundtables. "When we're talking about treating the entire aging population potentially for the rest of their lives, it's going to be hard to get sufficient safety information to be comfortable that what we're doing has benefits that justify the risks," he said. "I remain somewhat skeptical that we'll ever find a pharmacological treatment for cognitive aging that will be justified."

Should we treat?

Can we do anything about cognitive aging—and should we? David Lowe of Memory Pharmaceuticals pointed out that the potential demand for a drug that alleviated the effects of cognitive aging would be enormous. Clearly, there is an unmet market need. The purpose of a cognitive aging therapeutic would need to be well-defined, pointed out organizer Howard Fillit of the Alzheimer's Drug Discovery Foundation, who suggested terms such as cognitive health or vitality as ways to frame the discussion. "What is the goal—is it maintaining cognitive vitality flatline, or just lifting the curve so that you slow the rate of progression?" he asked. Ferris suggested that if cognitive aging and AD are on a mechanistic continuum, a drug that delayed cognitive aging might prevent dementia in millions of people.

Therapeutics that might serve as cognitive aging treatments are already in development. Robin Kleiman of Pfizer Global Research and Development discussed the potential of phosphodiesterase inhibitors as enhancers of synaptic plasticity. Lowe mentioned translational work with compounds that show promise as both disease-modifying and symptom-treating agents, and Jerry J. Buccafucso of the Medical College of Georgia discussed the mechanism of alpha-7 nicotinic agonists as desensitizing agents of this subtype of acetylcholine receptor.

Many obstacles remain to bringing a drug for cognitive aging through clinical trials and to market. The lessons learned from the efforts to develop treatments for AD are instructive: A disease-modifying agent—one that might actually slow or stop the disease—works so slowly that proving efficacy in a clinical trial could be dauntingly time-consuming and expensive. Preventing cognitive aging will probably also require early intervention—which will require a way to identify the target population. For any hope of success, drug developers will need excellent biomarkers, argued Gary Small of the University of California, Los Angeles, who presented data on the use of newer molecular labels combined with neuroimaging techniques such as PET in diagnosing AD. Regulatory approval may also present a challenge, said Allan Green, a physician, lawyer, and research scientist with expertise in FDA oversight, given that "cognitive aging" is still poorly defined.

Please continue reading for a more detailed report on the conference.

Speakers:
Timothy A. Salthouse, University of Virginia
Steven Ferris, New York University School of Medicine
Lenore J. Launer, National Institute on Aging

Highlights

• Cognitive aging is widespread and significant, beginning in mid-life or earlier.
• The relationship between cognitive aging and dementia is not clear.
• Skills and knowledge may compensate for age-related declines in cognition.
• Treatment and prevention of cognitive aging are major unmet needs.
• Barriers to developing treatments are not insurmountable.
• Smoking, high blood pressure, inflammation, diabetes and cardiovascular diseases in mid-life increase late-life risk of dementia.

Clearing away the myths

The observation that people often become a little absent-minded and forgetful as they get old isn't exactly controversial, but most of us think these deficits are fairly trivial, and set in only late in life. Many of us also secretly suspect we are exempt—that cognitive aging happens primarily to other people. "I'm going to convince you that all of those assumptions are wrong," said Timothy Salthouse, who presented convincing evidence that cognitive aging is not only commonplace and significant, but begins early in life.

Cognitive tests show that age has a profound effect on mental abilities that is not limited to memory. Analytical reasoning, spatial visualization, perceptual speed and pattern comparison all undergo age-related decline. On average, performance on just about any lab test that does not call upon crystallized knowledge peaks at about age 20, and drops steadily thereafter. These effects are not small. At roughly age 70, average performance on standardized scales of intellectual ability falls to about the 30th percentile of the average distribution. The important implication: preventing cognitive aging will require early intervention.

Two representative tests of cognitive aging suggest that the loss of cognitive ability between ages 18 and 80 is comparable to the gain between ages 8 and 18.

Furthermore, age-related declines are nearly ubiquitous. Those in the top quartile of performance fade at the same rate as those in the bottom for tests such as word recall. However, said Salthouse during the forum following his lecture, those near the top "have to fall a lot further" before becoming debilitated. This observation fuels the "brain reserve hypothesis," the idea that those who are born with or develop a more agile, physiologically stable brain are less likely to suffer from deficits later in life. While the relationship between normal aging and dementia is not clear, it may be that in dementia, the normal gradual rate of deterioration begins to accelerate.

Why, then, do most healthy older adults do just fine in day-to-day living? For one, there are large individual differences at every age. In addition, few of us need to perform at maximum ability to carry out the functions of everyday life. Cognitive agility is also just one element of overall performance; skills, personality, and experience also factor into performance. Finally, some things truly do get better with age. Emotion regulation improves with age, as does the store of knowledge. Performance on crossword puzzles, which relies upon stored knowledge, increases into the seventh decade of life.

The aging syndrome

Cognitive aging is a clinical syndrome, argued Steven Ferris of the New York University School of Medicine, who provided a rationale for treating the phenomenon. "This really is a problem, and we ought to do something about it," he said.

Ferris defined the syndrome as "the cognitive consequences of brain aging," involving a gradual decline in some but not all cognitive functions. Abilities are not lost, as they are in AD or MCI, but rather degraded. He mentioned other efforts to operationalize the phenomenon, most notably by the criteria of "age-associated memory impairment," proposed in 1986 by a National Institute of Mental Health working group. This definition included people at least 50 years old with an objectively confirmed complaint of memory decline in comparison to average performance, and without other problems such as MCI, depression, or disease. Applying these criteria can yield very large numbers: 47% of those in their 50s score at least one standard deviation below the mean on a standardized memory test; 85% of those in their 80s do so.

"There is a strong rationale for intervening," Ferris said. Reducing cognitive aging could improve quality of life, reduce the social impacts of aging, prolong working years, and improve productivity later in life. A treatment that slowed progression, rather than just treating symptoms, may also forestall true senility. "It seems obvious that if we could reduce the impact of brain aging or slow the process, it would inevitably reduce the prevalence of Alzheimer's Disease," he said.

Reducing cognitive aging could improve quality of life, prolong working years, and improve productivity.

Ferris rejected the argument that cognitive aging cannot be treated because it is not a disease; he pointed to other non-disease complaints such as headache, and other aging-related syndromes such as osteoporosis that are routinely treated. Other common objections include the sense that interfering with the brain is wrong, or that medicine should not improve upon normal biology. "There's nothing unethical about treating brain aging," Ferris said. Finally, some drug developers are concerned that the FDA would not approve a treatment indication that is not aimed at treating a disease, but Ferris dispelled this assertion as well. "It's not written anywhere that treatment targets have to be diseases," he said. "This is a concern, but it's not a showstopper."

About 20 agents have been explored as treatments for cognitive aging, but few showed benefits. The AD treatment memantine does not improve memory in the healthy aging population, but it does improve performance on a variety of measures of executive function and attention. Establishing outcome measures remains a major challenge of mounting studies in the future; Ferris proposed reducing the cost of trials with Web-based cognitive assessments.

Aging in the young

The final misunderstanding about cognitive aging is that it is a problem of the old. Much to the contrary, Lenore Launer of the National Institute on Aging suggested that some cortical changes begin to occur early in life, often fueled by vascular risk factors such as high cholesterol and high body weight. For example, the prevalence of tiny, diffuse ischemic lesions in the brain begins increasing at age 30. The significance of this insight? "It's never too early for prevention, but it can be too late," she said.

Looking at risk factors in mid-life often shows a different relationship between risk and disease than if the two are compared in the elderly. For example, low social engagement is known to predict dementia, but the full evidence suggests a more complex relationship. A significant falloff in social engagement from mid-age to old age is a stronger predictor than a consistently low level, said Launer. High levels of C-reactive protein, a measure of inflammation, is linked with a higher risk of mixed AD and cardiovascular disease later in life. Conversely, higher physical activity, high antioxidant intake, and lower BMI in mid-life are all associated with lower risk of dementia.

Untreated hypertension and ApoE4 status combine to greatly increase the risk for poor cognitive function, underlining the importance of treating high blood pressure.

Mid-life diseases may also predispose toward late-life dementia. High blood pressure in midlife—but not in old age—is predictive of poor cognition, perhaps because chronic untreated high blood pressure is associated with lower hippocampal volume and increased neuritic plaques. "If you're looking for a treatment for cognitive aging, treating blood pressure is already out there," she said. People who have hypertension and also possess the Apoe4 allele, a known risk factor for Alzheimer's Disease, have a dramatically increased risk for cognitive impairment. Similarly, diabetes seems to speed brain loss. Diabetics tend to have more small infarcts and lower brain volume, and the longer their diabetes goes untreated, the poorer the cognitive function in old age.

Together, the evidence that these physiological, disease-related, and behavioral risk factors in middle age predispose toward dementia late in life create a compelling rationale for early intervention. "I believe there's a lot of evidence for targeting people in middle age, and less evidence for targeting in later age," said Launer.

Speakers:
Patrick Hof, Mount Sinai School of Medicine
George Bartzokis, University of California, Los Angeles
Robin Kleiman, Pfizer Global Research and Development
Victoria Luine, Hunter College

Highlights

• Pyramidal neurons making long cortico-cortical connections are selectively vulnerable in AD and cognitive aging.
• Age changes the neuron by reducing spines, simplifying cell geometry, and decreasing dendritic arborization.
• Myelination in some parts of the brain peaks in middle age.
• Repairing myelination might be a tractable way to treat dementia.
• Gene expression relating to GABAergic transmission declines in old age.
• Phosphodiesterase inhibitors might boost synaptic transmission by increasing cGMP activity.
• Gonadal hormones affect performance in rats on a range of learning and memory tests.

Among old neurons

In the neurobiology of AD, much attention has been devoted to understanding cell death, but normal aging may involve more subtle alterations to select populations of neurons, explained Patrick Hof of Mount Sinai School of Medicine. Neuron loss is minimal in the healthy aging brain, and that's a good sign: "That means we can maintain those cells, and we can intervene." Understanding what is happening to cells that are "in transition"—neurons that are dying or vulnerable in other ways—may be essential to the neurobiology of cognitive aging.

Aging engenders profound changes in neuronal architecture. At right, the dendritic tree of a neuron from a 25-year-old macaque is impoverished in comparison with that of an 8-year-old macaque, left. It also has fewer contacts between cells.

Some of the first neurons to be affected in AD are layer III and layer V pyramidal cells that make long cortico-cortical connections, and Hof's research suggests the fate of this small population of cells may have a major impact on cognitive decline. Neuroanatomically, they are involved in important cognitive pathways that project from the prefrontal cortex to the superior temporal gyrus, from the STG to the inferior parietal cortex, and from area V1 to areas of the motor cortex, a pattern that correlates with the distribution of tangles in AD. Cells that are immunoreactive to a neurofilament protein epitope appear to be especially vulnerable, quickly accumulating tangles and shrinking with cognitive impairment. In these pathways, early intervention for AD might make a big difference, Hof said.

In cognitive aging, cells apparently undergo many changes short of outright death, including a decrease in synaptic density in dendritic spines of the hippocampus and particularly the cortex. Overall architectural changes include dendritic spine pathology; the older brain shows "a pruning of the branches." Old neurons also show fewer contacts between cells, a decrease in "mushroom" spines in favor of less stable thin spines, and a simplified cell geometry.

These morphological changes alter the electrical function of these cells; slice experiments with macaque brain show a dramatic increase in spontaneous action potentials, "as if the cells were trying to do something but cannot do a proper job," suggested Hof. Capacitative properties of these cells are also dramatically altered, changing the back-propagation of action potentials. "These cells really are not normal," he said. "They are not dying, but they are not normal."

Making the connection

Most neurobiologists focus on neurons, but it's the coating around our nerve cells that truly makes us human, said George Bartzokis of the University of California, Los Angeles. Myelin—the fatty sheath that insulates axons and speeds nerve conductance—is not uniquely human, but we have more and better cortical myelination than other creatures, providing us with stunning cognitive capacities. In his view, the brain is more like the Internet than a single computer—the connections are more important than individual processing nodes. "What's important is how wide the bandwidth is and how connected you are," said Bartzokis.

Myelination in the frontal and temporal lobes continues to increase well into adulthood, peaking between age 45 and 50. These parts of the brain, last to fully myelinate, are also the first to accumulate amyloid, and they also subserve the cognitive functions that are the first to suffer in AD. Myelin integrity correlates strongly with finger-tapping speed, a proxy for the pace of action potential in the brain, and both decline with age.

In the frontal (left) and temporal lobes (right), myelination increases well into adulthood.

Myelin breakdown is accelerated by the ApoE4 allele, and Bartzokis described this variant not as a specific risk factor for AD, but "a risk factor for repairing your brain," whether the damage is from neurodegenerative disease, stroke, or other illness. Those with other variants of this allele are better able to repair damaged myelin.

Focusing on myelination may suggest tractable approaches to treating dementia, suggested Bartzokis. Unlike trying to replace lost neurons, the dropoff in myelination "is a very fixable problem." Oligodendrocytes, which create the myelin, increase naturally with age, and making new myelin and repairing the damage is something the brain does normally, and could potentially be manipulated.

The aging transcriptome

In describing the "aging transcriptome" of the human prefrontal cortex, Robin Kleiman of Pfizer Global Research and Development noted that while gene expression is heterogeneous in middle-aged people, by roughly age 73 it generally assumes a typical pattern, including downregulation of genes involved in synaptic transmission and vesicular transport, along with an increase in stress response and inflammation genes.

In the aging transcriptome, proteins involved in GABAergic transmission are downregulated.

One striking transition is the downregulation in proteins involved in GABAergic transmission, suggesting decreased GABAergic inhibitory control over the PFC with age. GABAergic neurons play an important role in synchronizing the firing patterns of pyramidal neurons and the excitatory activity of the PFC, and research in schizophrenia suggests that the deficits seen in cognition and working memory relate to GABAergic cortical deficits.

In the aging rodent, Kleiman's group at Pfizer has found synaptic plasticity deficits. The induction of long-term potentiation (LTP)—a phenomenon of enduring electrochemical sensitization typically observed in hippocampal neurons that is believed to be crucial in memory—is normal. But maintenance of LTP, dependent on the cyclic nucleotide cascade, is impaired. The deficit is reversible with phosphodiesterase inhibitors, which hydrolyze cyclic AMP (cAMP). This represents an "interesting therapeutic approach," said Kleiman, since different PDEs are expressed in different regions of the brain. In particular, PDE9 has a high affinity for cyclic GMP; inhibitors of this PDE have been effective in LTP on slices from older rats. Cyclic GMP treatments have also been shown to reverse the LTP inhibition caused by amyloid ß-42 deposits. The hypothesis to test, said Kleiman, is that PDE9 inhibitors, by restoring plasticity mechanisms, might stabilize synapses in the face of insults such as toxic Aß-42.

The sex effect

Approaching the biology of cognitive aging from a different perspective, Victoria Luine of Hunter College described the differing impact of male and female sex hormones on cognitive function and aging. Learning and memory can be tested in rodents with the radial arm maze, in which food is hidden in one arm; repeat visits to empty arms are an index of forgetfulness. Luine prefers tasks such as the object recognition or object placement tests, in which the animal's response to new or rearranged objects is gauged as an index of recall. These tests specifically measure working memory, don't rely on reinforcements or punishments, and are easy for aged rats to tolerate.

Young female rats (left) have high apical spine density in the prefrontal cortex as compared to aged (right).

In general, older animals do not perform as well on these tests as young ones, but there are many subtle differences. Female rats of any age do not perform as well as males do on the radial arm maze, but both sexes do equally well on object recognition, in which the naturally curious rodent's exploratory reaction to old and new objects in its cage is monitored as a sign of memory.

Test performance reflects anatomical changes: In a Golgi analysis, impaired older female rats with low serum estradiol show a decrease of apical spine density in the hippocampus and frontal cortex. Ovarectomized young female rats also show poorer performance on the object placement task as well as decreases in spine density, changes that were reversed by estradiol. For males, castration and the resultant loss of testosterone impairs performance on the object-placement test, but did not change spine density in the hippocampus. The upshot is that gonadal hormone levels, spine density, and recognition memory seem to be related in females, but not males.

Life experience has a strong influence on this effect. Female rats who have had many litters of pups perform better on both object-placement and object-recognition tasks than their nulliparous counterparts, perhaps as a result of the increase in brain-derived neurotrophic factor (BDNF) relating to pregnancy. Chronic stress also has unexpected effects, improving performance on the radial arm task and object placement task in young females while depressing it in young males. One important conclusion of these observations is that the current model for drug discovery, which predominantly tests compounds in young male rodents, may be wholly inadequate for research into cognitive aging. "We have to look beyond the normal male," suggested Luine.