By David N. Leff

“Coming events cast their shadows before,” including many heart attacks.

Thus, neurologist David Holtzman, who focuses on Alzheimer’s disease (AD), observed: “AD is analagous to other illnesses, like heart disease. There you do a cholesterol test, and tell the patient, You’re at risk of getting a heart attack so you need to modify your diet and take this medicine.’ We don’t have anything like that in AD yet,” Holtzman continued. He practices research and clinical neurology at Washington University School of Medicine in St. Louis.

Riding the same analogy, Steven Paul, group vice president at Lilly Research Laboratories in Indianapolis, added: “The brain plaques that are neuronal hallmarks of AD are somewhat analogous to the plaques characteristic of arteriosclerosis. If you have a heart attack at age 65, the atherosclerotic process that caused that event probably started decades beforehand. Since we now know that Alzheimer’s pathology begins well before symptoms appear, we’re hoping it may be possible to develop a test that predicts the presence of amyloid plaques, and ultimately the risk of AD dementia similar to performing an angiogram to predict an impending heart attack.”

Paul and Holtzman are co-senior authors of a paper in today’s Science, dated March 22, 2002. It’s title: “Brain to plasma amyloid-b efflux: a measure of brain amyloid burden in a mouse model of Alzheimer’s disease.” The article’s first author is research fellow Ronald DeMattos.

“Our main finding,” Holtzman told BioWorld Today, “is that if we administer antibody against the amyloid-beta peptide into the blood of a mouse, antibody in the circulation accumulates, and that amount correlates strongly with the amount of plaque build-up in the brain. But if we don’t do anything to the animal other than draw its blood, the level of this amyloid-beta peptide in the circulation does not correlate with the amount of pathology in its brain.”

Mouse Test To Human: A Question Mark

“This is the first example of a blood test in a mammal,” he continued, “that can predict how much amyloid is in the brain. If the same finding holds up in a human which we don’t know yet then this would be very powerful because we know that the pathology of AD starts years before any symptoms arise. So as new drugs are being developed which hopefully are preventative, or at least good at delaying the onset of the illness you’d want to know to whom you should give those drugs. So you say to a patient, Okay, you’re at high risk of AD because the blood test shows an abnormal number of amyloid-beta molecules.”

First author DeMattos made the point that “the truly novel finding of our experiment is that a simple injection of murine monoclonal antibody altered the metabolism of amyloid-beta. Hopefully, we will be able to alter its metabolism in human patients, too. What the co-authors have shown in patients that don’t have any cognitive defect is that those individuals have no corresponding neuronal loss. They’ve got preclinical AD,” De Mattos said. “Those are the people we want to target therapeutically, and identify the quantity of peptide in their brains. Because the minute someone starts showing some dementia the very mildest form of AD they’ve already had substantial neuronal loss.”

Holtzman, Paul and their teams tested the diagnostic test on transgenic mice that mimicked human AD, behaviorally and cerebrally. “They only show AD pathology when they get a little older,” Holtzman pointed out. “The memory deficits that they have which aren’t present when they’re young become apparent when they’re older. And some memory impairments present, even when they’re young, get a little worse.

“We found that if we looked at these mice when they were about 1 year old,” he said, “all of them had some plaque deposition in the brain. But the amount of it was very variable. Some animals had a lot of plaques and others very few. We started wondering that if we looked at the amount of amyloid-beta in their blood, whether because there’s so much plaque variability there might be an equilibrium between amyloid-beta in the brain and in the blood, and whether the plaque modified that equilibrium. So if we gave the antibody, thinking that even though the amount of amyloid-beta in the blood wasn’t correlating with the plaque, maybe the antibody would facilitate a different equilibrium in the way the plaques modified amyloid beta metabolism.

“So we set up a cohort of 49 transgenic AD mice,” Holtzman recounted, “because we wanted there to be a large number of animals at different levels of neuronal pathology. That way if we did see differences in the levels that appeared in the blood, we’d have enough animals to know if there was a good correlation or not. Almost a year ago we started doing the actual in vivo experiments.”

Lilly’s Steven Paul provided the standard murine monoclonal antibody, directed at the amyloid-beta peptide’s central domain. That antibody draws amyloid-beta out of the brain and into the surrounding blood.

Monoclonal Captured Amyloid-Beta Targets

“What’s important about this particular monoclonal,” Holtzman observed, “is that it captures both the 40- and 42-amino-acid versions of the amyloid-beta peptide’s soluble form. At plaque-building time, this switches to insoluble.” Those 49 AD mice had a mutation in the APP gene, similar to the genetic abnormality found in some families with a strong history of AD. Within five minutes of antibody injection, amyloid-beta levels increased dramatically, and did correlate with the amount in the brain. The mice all developed plaques within a year though to varying degrees.

“This has obvious implications,” Holtzman observed, “for developing a similar blood test for brain amyloid load in humans. Though we will not be able to detect risk in someone who has not begun to accumulate amyloid, we hope to predict the disease well before AD symptoms appear. Such a noninvasive test replacing present post-mortem plaque detection could also distinguish individuals suffering from dementia caused by Alzheimer’s from those with other types of dementia. And it may help us eventually evaluate an individual’s response to particular medical therapies, now in development.”