Traditionally, neurologists have diagnosed Alzheimer disease (AD) based on clinical signs and symptoms, with confirmation of the telltale amyloid-b (Ab) plaques and tau tangles only upon autopsy. Knowing that earlier intervention may be crucial to slowing or even halting disease progression, researchers have long sought a marker that could be teamed with imaging technologies to highlight initial changes.
At the time of diagnosis, researchers estimate, nearly a third of the neurons in affected brain areas are already beyond salvation. "So finding markers that can help us identify patients prior to symptoms is really our big push now. Amyloid-b plaque deposition could be used as a biomarker, but little is known about its metabolism," said Anne Fagan, PhD, research associate professor of neurology at Washington University School of Medicine in St. Louis.
Generating much excitement at Neuroscience 2005, the Society for Neuroscience annual meeting held in Washington, DC, from November 12 to 16, was Pittsburgh compound B (PIB), developed at the University of Pittsburgh.1 According to several reports, PIB used with positron emission tomography (PET) is providing a window into the early pathogenesis of AD (see "Advances in Imaging for Diagnosis of Alzheimer Disease," Applied Neurology, May 2005, pages 19-23).
But having a marker in hand isn't the only requirement; knowing in whom to look is also important. "We wanted to follow very early patients to find the natural history of amyloid deposition, to look at those initial studies the way people have looked at autopsy data. That is, where do we expect the first deposition?" asked William Klunk, MD, PhD, associate professor of psychiatry at the University of Pittsburgh. Klunk coinvented PIB with Chet Mathis, PhD, professor of radiology and pharmaceutical sciences.
As is true for many neurodegenerative disorders, individuals with rare, early-onset inherited forms of AD provide the test cases because their genes are prophetic. At the meeting, Klunk reported on 2 such cases. The 37- and 38-year-old cousins had inherited variants of the presenilin 1 gene, indicating AD in their futures. Although both are still symptom-free, their scans revealed bright red striatal and precuneate regions against the gray matter.
The cases are accumulating, which validates the power of PIB. An image from a Down syndrome patient in whom symptoms had not yet developed also showed striatal pickup of PIB. In contrast, white matter areas had no amyloid deposition. In general, studies so far indicate that Ab deposition progresses from the cingulate cortex/precuneus and frontal cortex, to the parietal and temporal cortex and caudate, to the occipital cortex and sensory-motor cortex, consistent with initial effects on memory. These results parallel autopsy studies, Klunk said. Having such a time course will provide an invaluable tool for comparing the effectiveness of new therapies, he added.
Results of pilot studies are beginning to come in. Julie Price, PhD, associate professor of radiology and head of PET methodology at the University of Pittsburgh, reported a study of 15 individuals: 5 with AD, 5 with mild cognitive impairment (MCI), and 5 controls. Although the brains of all participants showed similar levels of nonspecific binding, "clear separation between controls and Alzheimer disease patients emerged. The MCI brains ranged from the control through the Alzheimer disease levels," Price reported.
Clearly, brain changes begin well before a person first loses his or her way home or falters in short-term memory tasks. Fagan sets the first inklings of AD back quite far, years or even a decade before MCI becomes noticeable. She reported a study on the relationship between in vivo plaques imaged with PIB and cerebrospinal fluid (CSF) levels of Ab 42, the variant of amyloid that consists of 42 amino acids.
The study collected images from 24 individuals, aged 48 through 83 years, at various stages of cognitive decline. "Regions of interest" were the prefrontal cortex, gyrus rectus, precuneus, and lateral temporal lobe, with the cerebellum serving as a control because it lacks uptake in all brains. Of the 24 participants, 7 showed PIB uptake in the regions of interest, in addition to having the lowest levels of Ab 42 in CSF.
"When I realized that everyone who was PIB-positive also had lower CSF Ab 42 levels, I had one of those 'aha!' moments that makes it so exciting to be a scientist," Fagan said, speculating that in early AD, transfer of Ab 42 from the brain to the CSF is impaired. "There's a sink in the brain. Having low CSF Ab 42 and being PIB-positive, therefore, may be useful markers," she added.
As often happens in biomedical science, exceptions prove informative. Two PIB-negative patients with MCI turned out to have a non-Alzheimer brain disorder and a fondness for sleeping pills, respectively. The investigation also revealed the PIB-positive, low CSF Ab 42 profile in 3 cognitively normal individuals--something that has occasionally arisen among supposed controls in other studies.
Although the bioethics of these situations is challenging, use of PIB may reveal preclinical cases and provide the long-sought peek into the first stages of AD pathology. At the very least, said Fagan, "we can use PIB binding to rule out amyloid deposition as a contributing factor to cognitive impairment." In the future, a heads-up via PIB scans may be coupled with treatments that at least slow the course of the illness, if not cure it. Other talks at Neuroscience 2005 addressed some of these approaches.
Many studies have noted a lower incidence of AD among people who regularly take NSAIDs. This finding sparked several trials of the drugs. Results were disappointing in healthy individuals for rofecoxib (Vioxx, Merck), naproxen, and celecoxib (Celebrex, Pfizer). But these studies were helpful in indicating that any effect of these drugs on AD does not result from their anti-inflammatory activity. Instead, studies in transgenic mice and cell culture showed that 8 of the 20 NSAIDs examined decreased levels of gamma secretase, the enzyme that snips aberrantly sized Ab from the amyloid precursor protein (APP) that is embedded in neuronal cell membranes.2 So NSAIDs may find new life as "selective Ab-lowering agents" that modulate the dynamics of APP cleavage.
Sandra E. Black, MD, head of neurology at Sunnybrook and Women's College Health Sciences Centre of the University of Toronto, presented encouraging results on R-flurbiprofen (Flurizan, Myriad Genetics), which is also being studied as an anticancer agent. Black reported on the results of phase 2 clinical trials held at 31 sites in Canada and the United Kingdom that included a total of 207 patients who randomly received either low-dose (400 mg) or high-dose (800 mg) R-flurbiprofen or placebo for an initial 12 months. R-flurbiprofen targets alpha-secretase and leads to decreased amyloid deposition and increased learning and memory in transgenic mice. The human trials assessed cognition and activities of daily living (ADL).
Results were positive but not dramatic. By 9 months, patients receiving the higher dose scored 46% better on the cognitive tests and 45% better on ADL. However, among a subset of patients with mild AD who had high plasma levels of the drug, improvement in ADL was 62%, which reaches statistical significance. "So it looks like it helps a subset of the mildly affected," Black said. She reported on an additional 8 months of a blind study in which those given the low-dose regimen or placebo worsened and those given the high-dose regimen stabilized or improved. "It looks like a continuing benefit," she concluded. Phase 3 enrollment is ongoing.
Another existing drug class that may enter the Alzheimer market is the statins.3 Ironically, the NSAIDs were hypothesized to affect AD via their anti-inflammatory action but were found to target Ab formation directly. Statins, thought to intervene by lowering cholesterol synthesis, may instead lower AD risk via an anti-inflammatory route. However, studies so far have not shown effects on cognitive decline, possibly because of use of low doses and short-term studies.
At Neuroscience 2005, Ben Wolozin, MD, PhD, professor of pharmacology at Boston University, reported on a study of 20 brains from AD patients. Ten of the individuals had taken statins for at least a year (stopping less than a month before death) and 10 had not. Wolozin and coworkers used antibody staining of Ab and reactivity of microglia to assess the inflammatory response. "Results were striking. Patients on statins had no microglial reactivity, but patients not on statins had abundant microglial activity," he said. No effect was seen on amyloid. *
1. Klunk WE, Engler H, Nordberg A, et al. Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004;55:306-319.
2. Eriksen JL, Sagi SA, Smith TE, et al. NSAIDs and enantiomers of flurbiprofen target gamma-secretase and lower A-beta 42 in vivo. J Clin Invest. 2003;112:440-449.
3. Sparks DL, Sabbagh MN, Connor DJ, et al. Atorvastatin for the treatment of mild to moderate Alzheimer disease: preliminary results. Arch Neurol. 2005;62:753-757.