Neurobiologyof Behavioral Disturbances in Dementia: Implications for Clinical Practice

This symposium, held in February, 2002, in association with the 15th Annual Meeting of the American Association for Geriatric Psychiatry, presented an authoritative review of recent developments in the treatment of behavioral disturbances associated with dementia.

This program was supported by an unrestricted educational grant from Novartis Pharmaceuticals Corporation.

Pharmacological and Behavioral Interventions in Alzheimer’s Disease: Consensus and Controversies

Broad areas of consensus and a few specific areas of controversy characterize the literature and clinical evidence for the treatment of behavioral disturbances in Alzheimer’s disease and other dementias. Where there are controversies, commented Dr. Mustafa M. Husain, Program Chair and Associate Professor and Chief of Geriatric Psychiatry at the University of Texas Southwestern Medical Center, they are most typically the result of the dynamic flux of a rapidly growing knowledge base.

Much of the new data to guide the clinician focuses on use of cholinesterase inhibitors (ChE-I) in the management of behavioral disturbances, agents already known to produce long-term improvements in cognitive deficits. Some of the remaining clinical questions, now being evaluated are:

• Which ChE-I are most efficacious?
• Do pharmacological differences between ChE inhibitors translate into differences in clinical efficacy?
• Does early versus late initiation of treatment with ChE inhibitors affect long-term outcome?

Studies with Cholinesterase Inhibitors
Several controlled clinical studies with ChE inhibitors have served to point up the safety and efficacy of these agents
in alleviating behavioral symptoms of dementia.

Feldman and colleagues examined the effect of donepezil 5-10 mg on change from baseline of behavioral and neuropsychiatric symptoms per the Neuro-psychiatric Inventory (NPI) 12-item Total Scores (Feldman H et al. Neurology 2001). At week 18, they found no statistical difference between placebo and the active treatment group. However, by week 24, subjects receiving donepezil showed clinical improvement in their behavior.

In another small, 8-week open-label study of donepezil versus placebo in patients with Lewy Body dementia, Lancott reported that three patients showed improvement in agitation, irritability, apathy and depression. Two patients did not show much improvement, and one discontinued the drug because of adverse effects.

The ChE inhibitor, rivastigmine, was evaluated in a large (N=120) European study by McKeith and colleagues (McKeith I. Lancet 2000). The group looked at the effect of rivastigmine on behavior based on mean change in NPI score. Their results demonstrated that patients taking rivastigmine experienced significant improvement in apathy/indifference, anxiety, delusions and hallucinations versus placebo.

Similar findings emerged from a study by Bullock and colleagues (N=184), who reported improvements in 11 of 12 behavioral symptoms included in the NPI in patients taking rivastigmine, versus placebo. Only depression did not show statistically significant improvement from baseline in the active treatment group. Dr. Husain pointed out that symptoms usually treated with psychotropic agents, such as agitation, irritability, and disinhibition, were improved with rivastigmine without reducing or discontinuing usage, as in a nursing home study of psychotropic medications by Cummings (Cummings J, et al. APA Meeting Paper 2000 (Table 1).

This study is notable in light of 1987 OBRA regulations urging a reduction in use of neuroleptics and other psychotropic medications to treat behavioral disturbances in patients with dementia residing in nursing homes. The Cummings group found an almost 50% rate of reduction or discontinuation in use of neuroleptics in patients taking rivastigmine. Anxiolytic and antidepressant medication usage was similarly reduced or discontinued by almost 40% and 35%, respectively.

The effect of galantamine on behavioral symptoms, at three dose levels (8, 16, and 24 mg/day), and based on NPI Total Scores, was evaluated in a 5-month double-blind and placebo-controlled study in patients with AD (Tariot PN et al. Neurology 2000). Subjects taking 8 mg of galantamine did not have significant behavioral improvements versus placebo. However, patients taking galantamine 16 to 24 mg per day did not deteriorate in their behavioral symptoms from baseline.

Rivastigmine Long-term
Particularly relevant to a long-term, chronic disease are studies to evaluate efficacy and safety over a longer period. This objective was addressed in an open-label safety trial of rivastigmine use for behavioral disturbances. Over 96 weeks, patients taking rivastigmine 6-12 mg showed no statistically significant worsening of symptoms (Grace J et al. Int Psychogeriatr 2001).

An important treatment issue in progressive dementia relates to the timing of the initiation of therapy. In a rivastigmine study in which placebo-treated patients were crossed over to active treatment with rivastigmine and titrated to the highest tolerable dose, Rosler and co-investigators found that, while behaviors improved, they did not reach the levels of improvement of patients on active treatment from the start of the study (Rosler M. et al. BMJ 1999). Dr. Husain commented that it is becoming clearer that even a loss of 6 months can diminish the level of behavioral stabilization.” Data such as these argue for initiating therapy soon after the emergence of behavioral symptoms.

Summarizing areas of consensus, Dr. Husain noted that there is agreement among experts treating patients with Alzheimer’s disease (AD) in a number of areas:

• Behavioral interventions should be initiated with or without pharmacotherapy early in the management of AD, even given a current paucity of controlled studies to guide behavioral treatment.
• Atypical neuroleptics are preferable to traditional neuroleptics, primarily due to side effects of the traditional agents.
• Cholinesterase inhibitors have a role in the management of behavioral symptoms of AD.

Dr. Husain concluded as well that early detection and early intervention with cholinesterase inhibitors have a beneficial long-term effect.

Diagnosis and Treatment of Dementia with Lewy Bodies: State of the Art

Geriatric psychiatry is “where some of the most interesting developments in clinical medicine are occurring,” according to Ian McKeith, Professor of
Old Age Psychiatry at the Institute for Ageing and Health at the University of Newcastle-upon-Tyne, UK. Dr. McKeith’s laboratory has generated much of the important work in dementia with Lewy bodies (DLB), a disease important in itself but also as a useful conceptual model to inform us about the other cholinergic deficit dementias.

Fundamental differences characterize DLB and Alzheimer’s disease, but the two are sometimes clinically difficult to distinguish. The distinction is crucial, however, because accurate diagnosis is important in developing an appropriate management and treatment strategy. Essentially, said Dr. McKeith, DLB is a pathological process that implicates abnormal cytoskeletal proteins, including ubiquitin and alpha-synuclein. “These biological foundations are fundamentally different from the abnormalities one gets in tau protein in Alzheimer’s disease.”

Consensus criteria for probable DLB were developed by McKeith and colleagues, who defined the condition as a “progressive cognitive decline that impairs occupational or social functioning (McKeith I. Neurology 1996). Atten-tional deficits with fronto-subcortical and visuo-spatial dysfunction are prominent in DLB. In addition to satisfying this definition, patients must display two of three essential features:

• Fluctuating cognition with pronounced variations in attention and arousal
• Recurrent visual hallucinations, typically well-formed and detailed
• Spontaneous motor features of Parkinsonism

What is known about DLB is contrasted by four “state of the art” questions that remain to be elucidated.

Question 1: Why do DLB patients typically have visual hallucinations?
Dr. McKeith commented that the visual hallucinations of DLB seem to be “a combination of attentional deficits plus visual perceptual impairments.”
Comparative SPECT scans of the brains of patients with AD and DLB reveal perfusion deficits to be more pronounced in the visual association cortex in DLB. Clinically, the deficit may be revealed on tests of copying and drawing. “If somebody can remember the day and the date and pass on recall tests, but can’t copy or draw,” said Dr. McKeith, “this is a good signpost that this may not be Alzheimer’s disease—it may be DLB.”

Question 2: Is DLB a rapidly progressive dementia?
Initial data collected by Dr. McKeith found that a group of Alzheimer’s disease patients survived 4.8 years after first clinical contact, whereas DLB patients survived for only 1.8 years (McKeith I et al. Psychol Med 1992). Rapid deterioration was therefore thought to be a characteristic of the disorder. More recent data show, however, that the natural history of cognitive decline is typically similar to AD (Ballard C et al. Arch Neurol 2001, Lopez OL et al. Int Psychogeriatr 2000).

Question 3: Can neuroimaging assist in diagnosis?
Standard structural imaging, such as coronal section MRI, is helpful, said Dr. McKeith, and may reveal a relative preservation of hippocampal structures in DLB, compared with an Alz-heimer’s disease brain.

But a much more useful, disease-specific, imaging test is the FP-CIT ligand, a cocaine derivative that, when injected into the bloodstream, is taken into the dopamine transporter site. This enables visualization of the activity of dopamine at the level of the striatum, which is usually clearly abnormal (reduced) in DLB but preserved in AD.

Question 4: Why do DLB patients respond well to cholinesterase (ChE) inhibitors?
There are several possible explanations for the response of DLB patients to ChE therapy. Dr. McKeith explained that in DLB and in Parkinson’s disease with dementia, there is a “double hit” on the cholinergic system. In addition to loss of basal forebrain cholinergic nuclei, as in AD, there is also brainstem nuclei degeneration in DLB.

Cortical cholinergic deficits are significantly greater in DLB patients than in Alzheimer’s disease patients, making them, as Dr. McKeith expressed it, sitting targets for cholinergic replacement (Perry EK et al. Neuroreport 1994) (Figure 1).

In a placebo-controlled randomized trial (McKeith I et al. Int J Geriatr Psychiatry 2000), the behavioral symptom most responsive to rivastigmine in DLB patients was apathy. Patients taking rivastigmine became more alert and responsive. Other behavioral symptoms affected in a positive way were visual
hallucinations, delusions, anxiety and irritability.

Dr. McKeith reiterated in his conclusion that DLB is not generally more rapidly progressive than AD. He noted that structural and functional imaging can be helpful in the differential diagnosis. Finally, cholinesterase inhibitors can produce short-and long-term benefits in cognition and behavioral symptoms.

Neural Pathways Mediating Behaviors in Dementia

One of the most exciting revelations emerging from research into the neurobiological bases for Alzheimer’s disease (AD), dementia with Lewy Bodies (DLB), and Parkinson’s disease (PD), is evidence for linkages among the disorders. Jeffrey Cummings, Director of the Alzheimer’s Disease Research Center at UCLA School of Medicine and Rose Professor of Neurology reviewed emerging evidence that, as he said “…the underlying abnormality among all degenerative dementias is the accumulation of abnormal protein products.” This commonality is beginning to link the dementias in ways that no one could have imagined just a few years ago.

Neurobiology of Alzheimer’s Disease (AD)
In AD, a condition whose neuropathology involves neuritic plaques and neurofibrillary tangles, there is an increasing consensus that the accumulation of beta amyloid in the brain is responsible for the cascade of changes that leads to cell death and the eventual evolution of symptoms recognized clinically.

In a recent study with psychotic and non-psychotic patients, Farber and co-investigators succeeded in demonstrating a link between one specific parameter of the neurobiology of AD and its neuropsychiatric manifestations (Farber NB et al. Arch Gen Psychiatry 2000). Dr. Cummings explained Farber’s findings: “…psychotic patients with AD had significantly more neurofibrillary tangles (NFT) in all of the neocortical areas that were examined—mid-frontal, superior temporal, and inferior parietal, compared to non-psychotic patients. They did not have more beta amyloid and they did not have more Lewy bodies.”

Cummings and colleagues also established that AD patients with agitation had many more NFT in the frontal cortex specifically, compared to patients without agitation (Tekin S et al. Ann Neurol 2001;49:355-61). The differentiation was similarly specific: there was no difference in the amount of Lewy bodies or the amount of plaque formation.

Neurotransmitter deficits in AD may be linked to symptoms. For example, the noradrenergic deficits of AD stem from locus ceruleus atrophy and are linked to depression. Serotonergic deficits have their source in raphe atrophy and are linked to depression and psychosis. Cholinergic deficits arise from nucleus basalis atrophy and are linked to apathy, hallucinations and a variety of other behavioral symptoms and cognitive deficits.

Cholinesterase inhibitors
The findings that cholinesterase inhibitors provide benefit in patients with cholinergic deficits are rational and well-documented. Rosler et al, in a study of the effect of rivistigmine on behavior, demonstrated an improvement in mood when the drug was introduced. Over time, few disturbances emerged when measured at week 52 and at week 104 (Rosler M et al. BMJ 1999). The data suggest that the rivastigmine decreased the emergence of new behavioral disorders at week 104 in a disease in which behavioral deficits become more prominent over time.

Neurobiology of Dementia with Lewy Bodies (DLB)
The neurobiology of DLB involves brainstem and cortical Lewy bodies, more plentiful in the limbic system and evolving as the disease evolves into the neocortex. DLB patients have cortical neuritic plaques similar to those seen in AD patients, but relatively few neurofibrillary tangles. There is a marked cholinergic deficit—greater than that of AD patients, Dr. Cummings noted. DLB is also characterized by both a dopaminergic and serotonergic deficit.

Cholinesterase inhibitors have shown efficacy in treating the behavioral defects of DLB, as demonstrated by studies such as one by Dr. McKeith’s group with rivastigmine that showed a significant improvement in neuropsychiatric symptoms as measured by the NPI 10 and NPI 4 (McKeith I. et al. Lancet 2000).

Neurobiology of Parkinson’s Disease (PD)
In addition to the well-known hallmarks of PD—bradykinesia, rigidity, and rest tremor—overt dementia is present in as many as 40% of patients. The dementia syndrome seems to evolve in concert with a cholinergic deficit.

Interestingly, neuropsychiatric symptoms in PD are more severe in patients with dementia than in those without the dementia syndrome, Dr. Cummings said. PD patients without dementia are characterized neurobiologically by degeneration of brainstem nuclei and a dopaminergic deficiency; PD patients with dementia have cortical Lewy bodies and a cholinergic deficit.

Neurobiologic Abnormalities: AD, DLB, PD
Perhaps more significant than their differences are the commonalities shared by AD, DLB, and PD. Shared are neurochemical abnormalities such as cholinergic deficits. Serotonergic deficits also characterize the three disorders, and noradrenegic deficits are present in most patients with AD, DLB, and PD.
Each disease also is most significantly associated with forms of protein aggregation. Beta amyloid accumulates in the brains of patients with AD and in patients with DLB. Alpha synuclein protein abnormalities are present in PD and DLB. Tau abnormalities are a secondary event in patients with AD.

How might these findings be explained in light of the general understanding of neuropathways in the brain, in light of its posterior and anterior “system”? Dr. Cummings explained that different areas of the brain have differential vulnerability to protein accumulation, such that the A-beta abnormalities of AD and DLB have their greatest impact on posterior structures and in the parietal cortex, where amyloid-opathy is most severe. On the other hand, the tau-opathies and the synuclein-opathies are disorders of the frontal, subcortical systems, so they tend to involve the frontal cortex, the caudate nuclei, pallidum, substantia nigra, or the thalamus.

Treatment implications of this new information suggest that, for protein aggreggation, promising treatments include secretase inhibitors, anti-aggregants, and agents affecting protein folding. For the transmitter deficits of all three disorders (AD, DLB, and PD with dementia), cholinesterase inhibitors have great promise.