Dr. Kennedy is professor in the Department of Psychiatry and Behavioral Sciences at Albert Einstein College of Medicine, and director of the Division of Geriatric Psychiatry at Montefiore Medical Center in the Bronx, New York.
Disclosure: Dr. Kennedy has received research support or honoraria from AstraZeneca, Eli Lilly, Forest, Janssen, Myriad, and Pfizer.

Please direct all correspondence to: Gary J. Kennedy, MD, Director, Department of Geriatric Psychiatry, MMC, 111 East 210th St, Klau One, Bronx, NY 10467; Tel: 718-920-4236; Fax: 718-920-6538; E-mail: gjkennedy@msn.com.



Present treatments for dementia are largely palliative, providing symptomatic but temporary improvement. In contrast, agents that promise disease modification offer the hope of prevention, arrest, or deceleration of decline. However, enthusiasm for these new agents must be tempered by acknowledgment of their inevitable limitations within the present standard dementia care.

Two recent reviews1,2 of treatments being developed for Alzheimer’s disease describe a substantial number of agents in randomized clinical trials, not only to demonstrate safety or determine effective dosing, but to test efficacy and tolerability as well. Although some of the medications are targeted at neurotransmitter systems, which may be the innocent bystanders in dementia pathogenesis, several address amyloid and tau pathologies thought to be central to neurodegeneration of the Alzheimer’s type. These latter agents offer the promise of preventing, arresting, or slowing decline through disease modification. However, disease-modifying agents will have to overcome even greater challenges faced by the present generation of medications for dementia if they are to be approved by the Food and Drug Administration. Moreover, they are not expected to be curative and their preventive potential depends on as yet unrealized preclinical diagnostics. Thus, they are unlikely to supplant symptomatic treatments. Without the continued evolution in collaborative models of dementia care, it is difficult to see how disease modifiers will ever meet their full public health potential. What follows is an effort to prepare practitioners and policy makers to better address the implications of disease modification in dementia (Table).




Disease Modification and Symptomatic Treatments Defined

The concept of disease modification is graphically portrayed in the accompanying Figure. With the start of a symptomatic treatment such as donepezil, galantamine, rivastigmine, or memantine, measures of benefit diverge initially from placebo but ultimately parallel the rate of decline of untreated patients. When withdrawn from treatment, within weeks patients who initially benefited perform no better than the placebo group. Thus, the effect is beneficial but palliative. In contrast, disease-modifying agents would either arrest or slow the progressive loss in quality of life. The result is an altered slope or velocity of decline which diverges at an acute angle from placebo. The larger the effect, the greater the divergence and the sooner the benefits become clinically apparent. Smaller effects may be reliably detected only after prolonged treatment. With the introduction of diagnostic tests to identify at-risk people prior to the onset of illness, disease modifiers may also be preventive.2


It is the mechanism of action that distinguishes symptomatic from modifying therapy. The present medications for mild, moderate, or severe Alzheimer’s disease either enhance cognitive performance or retard cognitive decline through the manipulation of neurotransmitter systems. They either boost the signal between cholinergic neurons or reduce the background noise from excessive neuronal excitation. Although the signal is enhanced, the receiver continues to fail at the same rate. In disease modification, the pathology being modified is upstream from events leading to the loss of neurons and neurotransmitters. Both the neurotransmitter signals and receivers continue to decline, though less rapidly. Disease modification does not imply that function will be restored permanently to premorbid levels. Hence, unless the modification occurs prior to neurotransmitter deficits, symptomatic agents would still have a role to play. By analogy, analgesics may be necessary both before and after an osteoarthritic joint is replaced with a mechanical prosthesis. Similarly, cholesterol-lowering agents are indicated both before and after myocardial infarction. Indeed, statin therapy is one possible model of disease modification in both coronary artery disease as well as Alzheimer’s dementia.1


How Would Disease Modification Be Measured?

The development of measures for disease modification in dementia has involved at least three elements. First, a human biomarker at or near the cause of the illness needed to be identified. Second, an animal model of the illness in which the biomarker could be experimentally manipulated needed to be developed. Third, agents which reduce the pathogenic effects of the biomarker in animals need to show the same or similar actions in humans but with clinical significance. It is clinical significance which has to be defined if the agent is to be prescribed. However, the definition of meaningful clinical significance is not straightforward. Although FDA-approved medications for dementia reliably demonstrate cognitive benefits superior to placebo, a significant minority of treated patients experience no benefit. These agents may delay the emergence of behavioral problems, but their capacity to reduce them once present is unreliable.3,4 Further, a significant minority of providers find the benefits neither socially meaningful nor economically valuable.5,6

Based on analyses of medical expenditures and the increase in life expectancy between 1960 and 2000 in the United States, Cutler and colleagues7 found that the value extracted from dollars spent was substantial. However, further expenditures are expected to yield decreasing value, meaning that increased spending will likely be accompanied by smaller and smaller increases in the life span. Thus, the quality of existence in the latter part of life takes on greater importance. If disease modification can extend the quality of life (eg, independence, productivity) for the patient and family caregivers, increased life span need not be the measure by which value is determined. Indeed, pushing dementia-related disability to the end of the life span would be nearly as desirable as the more distant goal of outright prevention. Complicating the equation is the amount of time necessary to demonstrate a reduction in the slope of cognitive decline or quality of life that must be demonstrated to argue for the presence of modification. As a result, an accounting of the social and economic benefits will follow at some distance from the expense of initiating disease modification.


Unanticipated Consequences

Post and Whitehouse’s8 cautionary examples of dementia treatment following the introduction of donepezil at first reading appear not to apply to disease-modifying therapies. In the cases cited, improved cognition following cholinesterase therapy was accompanied by a return of obsessive worry, agitation, painful and unwelcome insight, and awareness of limitations. For the patients’ caregivers, burden increased rather than declined and the quality of life for both parties was degraded rather than enhanced. In an accompanying editorial, Sachs9 advised that while the vignettes cited by Post and Whitehouse8 are exceptions rather than the norm, they are nonetheless familiar to physicians who treat patients and their families coping with dementia. The cases highlight the need to communicate clear, realistic goals of treatment. If symptomatic improvement in one domain is associated by symptom exacerbation in another, the balance of harm and benefit may not favor continued treatment. Because the cholinesterase inhibitors are not thought to increase the life span of people with dementia, they are not seen as prolonging their disability or exacerbating suffering except in the rare cases cited by Post and Whitehouse.8

However, if disease modification does indeed prolong life by delaying cognitive decline, the expected delay in loss of independence and emergence of behavioral disturbances may not result in a lesser duration of dependency and discomfort. Thus, the question remains of when to stop treatment whether it is merely symptomatic or genuinely disease modifying. Similarly, once the process of neuronal death is well underway there may be little gained from modification agents even if they were to arrest the process entirely. Again, the goals of care must be made explicit to insure that modified disease means more than prolonged dependency.


Disease Modification and Diagnosis With Biomarkers

Symptomatic treatment is recommended across the array of common dementias including Alzheimer’s, mixed (Alzheimer’s disease with vascular components), and Parkinson’s dementia.10 However, for disease modification to be effective, particularly for prevention, precise diagnosis is needed to distinguish Alzheimer’s disease from the prevalent but less frequent vascular and Parkinson’s dementias. Dubois and colleagues11 recommend new research diagnostic criteria for Alzheimer’s disease which would require the presence of significant impairment in episodic memory plus at least one abnormal biomarker as detected by either structural neuroimaging with magnetic resonance imaging, functional imaging with positron emission tomography, cerebrospinal fluid analyses of b-amyloid, or hyperphosphorylated tau proteins. If any of the agents currently in clinical trials demonstrate clinically significant disease modification as well as favorable alteration in biomarkers, the research criteria proposed by Fillit and colleagues10 may well take their place in standard clinical practice. Biomarkers sensitive to early effects or lack thereof from disease modifiers would also be useful for the determination of treatment failure. Because disease modification will require long-term treatment, minimizing the expense and disappointment when the treatment fails is critical.

A breakthrough in risk identification through personal genomics12 might justify the prescription of disease-modification agents prior to any phenotypic evidence of illness. However, a statistically significant association between a personal genetic variation and an illness does not mean that it is clinically significant. Simply knowing about one’s risk status does not always lead to better health decisions. Some patients may over-interpret a relative increase in risk as absolute certainty that disease will follow. Others, finding they are at relatively less risk than expected, may forego health-promoting behaviors. As a result, patient education coupled with provider counseling would be required to ethically address the uncertainties of risk and benefit. The costs of a personal genome, even if current prices fall precipitously, will add expense and inevitable concerns about inequities in care and insurance coverage. Finally, at what threshold of risk should medication be started? Further, what evidence will physicians use to declare a disease-modifying treatment is ineffective or has lost efficacy after initial benefit? These are scientific questions of probability but the emotional weight attached to them will be personal to the patient. Thus, tact, time, and counseling will be necessary to help patients and their families make informed, personalized decisions.



Advances in preclinical diagnosis may transform disease modification into disease prevention. However, in the interim, disease modification, if shown to be efficacious, will impose substantial burdens before the desired benefits are evident. Disease-modifying agents may fail to demonstrate their full potential unless they are incorporated into a model of disease-modifying care. The collaborative model is supported by several randomized controlled trials but is not the current standard.13-15 Nonetheless, if disease-modifying agents delay the onset of dementia-related disability until the end of the life span, the economic value would be substantial even if they cannot cure or prevent the disease. In summary, enthusiasm for the next-generation medications should be tempered by an awareness of the implications of disease modification for dementia care. PP



1. Grossberg GT, Pejovic V, Miller ML. Current strategies for the treatment and prevention of Alzheimer’s disease. Primary Psychiatry. 2007;14(8):39-54.
2. Christensen DD. Changing the course of Alzheimer’s disease: anti-amyloid disease-modifying treatments on the horizon. Prim Care Companion J Clin Psychiatry. 2007;9(1):32-41.
3. Howard RJ, Juszczak E, Ballard CG, et al. Donepezil for the treatment of agitation in Alzheimer’s disease. N Eng J Med. 2007;357(14):1382-1392.
4. Yaffe K. Treatment of neuropsychiatric symptoms in patients with dementia. N Engl J Med. 2007;357(14):1441-1443.
5. Birks JS, Harvey R. Donepezil for dementia due to Alzheimer’s disease. Cochrane Database Syst Rev. 2003;(3):CD001190.
6. Kmietowicz Z. NICE proposes to withdraw Alzheimer’s drugs from NHS. BMJ. 2005;330(7490):495.
7. Cutler DM, Rosen AB, Vijan S. The value of medical spending in the United States, 1960-2000. N Engl J Med. 2006;355(9):920-927.
8. Post SG, Whitehouse PJ. Emerging antidementia drugs: a preliminary ethical view. J Am Geriatr Soc. 1998;46(6):784-787.
9. Sachs GA. Dementia and the goals of care. J Am Geriatr Soc. 1998;46(6):782-783.
10. Fillit HM, Doody RS, Binaso K, et al. Recommendation for best practices in the treatment of Alzheimer’s disease in managed care. Am J Geriatr Pharmacother. 2006;4(suppl A):S9-S24.
11. Dubois B, Feldman HH, Jacova C, et al. Research criteria for the diagnosis of Alzheimer’s disease: revising the NINCDS-ADRDA criteria. Lancet Neurol. 2007;6(8):734-746.
12. McGuire AL, Cho MK, McGuire SE, Caulfield T. The future of personal genomics. Science. 2007;317(5845):1687.
13. Callahan CM, Boustani MA, Unverzagt FW, et al. Effectiveness of collaborative care for older adults with Alzheimer disease in primary care: a randomized controlled trial. JAMA. 2006;295(18):2148-2157.
14. Vickrey BG, Mittman BS, Connor KI, et al. The effect of a disease management intervention on quality and outcomes of dementia care: a randomized controlled trial. Ann Intern Med. 2006;145(10):713-726.
15. Mittelman MS, Roth DL, Clay OJ, Haley WE. Preserving health of Alzheimer caregivers: impact of a spouse caregiver intervention. Am J Geriatr Psychaitry. 2007;15(9):780-789.