Mixed results after trials for Alzheimer's vaccine

Small gains seen in some cognition

By Alice Dembner, Globe Staff  |  July 22, 2004

The first test in people of a vaccine to treat Alzheimer's disease produced mixed results, but it showed enough promise that companies are pressing ahead with other vaccines, researchers reported yesterday at the ninth International Conference on Alzheimer's Disease in Philadelphia.

Tests of the vaccine, made by Elan Corp. of Ireland and Wyeth of New Jersey, were halted in January 2002 when 18 of the 300 people in the trial developed a potentially fatal brain inflammation. But researchers, who continued to follow the patients, said yesterday that 59 people developed antibodies in response to the vaccine and scored better on a composite of nine tests of memory and attention. In addition, the vaccine appeared to reduce the amount of two problematic proteins associated with the disease in the brains and spinal fluid of some patients.

Separately, researchers reported that two other drugs designed to prevent Alzheimer's appeared safe in preliminary testing, and reduced levels of one of the proteins, beta-amyloid.

"These are small victories," said Zaven Khachaturian, former director of Alzheimer's research at the National Institutes of Health and a senior science adviser to the Alzheimer's Association. "Given the history of Alzheimer's disease, and the total hopelessness early on in the research, the small victories seem like a big deal."

The vaccine, which contains beta-amyloid, was designed to train each patient's immune system to recognize sticky clumps of the protein, or plaques, and create antibodies to attack them. Because safety problems halted the test, patients did not get a full course of the vaccine.

The vaccine results have been eagerly awaited since initial reports on small subsets of patients last year offered the first evidence that inoculation could slow the progress of Alzheimer's. The results on all 300 study subjects are more muddy.

The amount of tau, a second protein linked to Alzheimer's, dropped about 30 percent in the spinal fluid of 12 patients who made antibodies, but not at all in 12 patients who did not make antibodies, according to a presentation by Dr. Sid Gilman, a neurologist at the University of Michigan. This is the first time a drug has been shown to reduce tau levels in people.

But the results of cognitive tests were mixed. Although the mental function of patients who made antibodies to beta-amyloid stabilized or improved slightly on one set of tests, the patients did not show any change on a handful of other tests typically used in Alzheimer's, according to Gilman.

Autopsy results were similarly confusing. Autopsies on four patients showed a reduction in beta-amyloid plaques, but antibodies were detected in only three of those patients, said Dale Schenk, Elan's chief scientific officer.

Nonetheless, Elan and some competitors are pressing ahead with other potential vaccines.

"The safety issues aside, there's no question that AN-1792 [the vaccine] has a biological effect on the underlying pathology of Alzheimer's disease," said Schenk. "That is . . . good news for the overall approach."

Elan has just begun testing a manmade antibody to beta-amyloid in patients and is working in animals to develop what it hopes is a safe successor to AN-1792.

Khachaturian said the vaccine's apparent ability to eliminate plaques is scientifically exciting, but the lack of major improvement in the patients leaves its effectiveness in question.

In another presentation at the conference yesterday, researchers from Eli Lilly reported that a drug designed to prevent the formation of beta-amyloid was safely tested in 37 healthy adults and reduced blood levels of the protein.

Alice Dembner can be reached at dembner@globe.com.