Post-mortem autopsy of the brain is still the only positive way ofdiagnosing Alzheimer's disease (AD), and differentiating it fromother forms of senile dementia. The telltale fingerprints thepathologists find are beta-amyloid plaques and fibrillary tangles inthe brain centers of memory, cognition and function. (See BioWorldToday, Oct. 8, 1996, p. 5.)

To be sure, skilled geriatric neurologists, using neurobehavioral tests,get high scores _ better than 80 percent _ in accurately diagnosingAD in their patients. But 80 percent is far short of 100 percent. Thedifference means that many people with faltering or failed mentalprocesses get misdiagnosed as having AD, a progressive affliction forwhich there is no very effective treatment, let alone cure.

This automatically excludes them from consideration for the broadrange of other dementias, (some of which mimic AD), for whichuseful therapies are available.

For many chronic disorders, such as diabetes, simple blood testsexist. Given that populations in modern industrial societies livelonger, and become candidates for senile dementias, such a serum-sampling diagnostic for AD is urgently needed, and widely sought byresearchers.

One group, at the University of British Columbia, has developed ablood serum diagnostic test for AD, based on a seemingly irrelevantprotein.

Maverick Molecule Proves AD-Specific

Their paper in the November Nature Medicine bears the title: "Serumlevels of the iron-binding protein p97 are elevated in Alzheimer'sdisease." Its senior author, immunologist Wilfred Jefferies, is on thefaculty of the University of British Columbia, Vancouver.

He told BioWorld Today: "p97 is a molecule first identified 15 yearsago as a biochemical marker for melanoma in humans, where it ishighly expressed. It was cloned and sequenced at the time," hecontinued, "by molecular biologist Karl Eric Hellstrom for Bristol-Myers Squibb Co., [based in New York]. The protein was then betterknown as melanotransferrin."

Hellstrom, who is vice president of immunotherapeutics at Bristol-Myers Squibb in Seattle, gave Jefferies the cDNA clone for p97.With that, he was able to define the protein's function in irontransport, and also discover that it exists in two different forms, one acell-surface molecule; the other, a soluble one.

"We now encounter what we assume to be that soluble p97 in theserum of AD patients," Jefferies observed. He thereupon beganimmunostaining human brain tissues. "At the same time we started tolook at tissues of other neurodegenerative disorders _ Parkinson'sand Huntington's diseases, multiple sclerosis, amyotrophic lateralsclerosis."

He and the article's first author, Malcolm Kennard, determined thatthe p97 marker occurred in the plaque-associated microglial cells ofAD patients only.

"Then we put two and two together," Jefferies continued, "andsurmised that maybe the soluble form of the molecule could be foundin the cerebrospinal fluid (CSF) of AD patients, or even in theirblood serum."

Primed by these basic findings, Jefferies and his team proceeded toput their p97 marker to the test, assaying it in the blood and CSF ofthree separate cohorts of AD patients, 35 in all, matched with 32healthy controls.

As reported in next month's Nature Medicine, their nanogram-concentration, two-antibody, sandwich-like fluorescent assaydetermined not only that p97 could be measured in both CSF andserum, but that it apparently increased in concentration as the diseaseprogressed. Controls, on the other hand, had none of it.

However positive, Jefferies emphasized, these results are stillpreliminary and tentative. "What we haven't done," he pointed out,"is look at a larger number of neuropathologies, to find out if p97 isindeed specific to AD among all dementias. At this point, even withthe sample number that we've done, it's at the level of an observationthat has to be followed up in order to find out if it has clinicalrelevance."

He and his co-authors will do that by expanding their sample numberto several hundred. They will also compare serum elevations of theirmarker at different stages of AD, "to verify that it increases with theseverity of the disease."

He made the point that, "This test is completely different from thegenetic-disposition screens, which are based on looking atpresenilins, beta-amyloid fragments and APO-E. If ours does workout to be true, it could potentially be used in AD diseasemanagement, and differential diagnosis of senile dementias. And as ablood-serum test, it has a potential for doctors' offices."

"In clinical trials, he added, "it might provide an independentchemical marker that is more responsive than cognitive scores toquantitating therapeutic responses. So it could bring therapies on linefaster."

Jefferies also sees a suggestion of pre-symptomatic prognosis: "If ourstatistical analysis is correct, and there is some p97 elevation prior tosymptomatic display in AD, it's possible that one may be able to usethis test in identifying such patients a few years in advance of onset.Clinicians will tell you that at such a critical point, clinicalintervention could improve the outcome of treatment."

The University of British Columbia has two patents pending on thepotential AD diagnostic, and assigned exclusive worldwide rights to abiotech start-up company, Synapse Technologies Inc. in Vancouver.It was founded in 1992, the company's president, Bob Rieder toldBioWorld Today, with Jefferies as chief scientific officer.

Synapse's focus, Rieder added, "is diagnostic and therapeuticapplication of mechanisms for transport of materials across theblood-brain barrier."

Its present funder and partner, Jefferies said, is "MDS HealthVentures Capital Corp., Canada's largest VC firm."

Jefferies noted that Athena Neurosciences Inc., of South SanFrancisco, has begun to market an AD diagnostic test based onroutine sampling of patients' CSF, for biochemical analysis ofamyloid plaque and neurofibrillary tangle components. (SeeBioWorld Today, March 27, 1996, p. 1.)

"The point is," he said, "there's a place for different types of ADmarkers in the diagnostic community which fill different types ofniches, and have differing uses. They can probably co-exist as wellcommercially." n

-- David N. Leff Science Editor

(c) 1997 American Health Consultants. All rights reserved.