Alzheimer's Club

"For 14 years, neurologist Allen Roses has argued that a gene discovery he made about Alzheimer's disease would lead to new treatments for the mind-robbing illness. His prediction may finally be coming true.

Roses, a research executive at drugmaker GlaxoSmithKline Plc, provided evidence yesterday that the company's best-selling diabetes pill Avandia may slow the effects of Alzheimer's. At a science meeting in San Diego, he presented a 500-patient study showing Avandia enhanced memory when compared with a placebo.

The study supports a hypothesis that the disease, which affects 4.5 million Americans, is caused by some of the same biological malfunctions that are involved in diabetes. The Glaxo researchers found that only people with a particular gene linked to a protein that transports cholesterol through the blood responded to the treatment.

"It is one of the first studies to indicate that a drug works for some people and not others based on a test of their genetics,'' Roses, 62, said before his presentation at the University of San Diego School of Medicine. The results, he said, suggest ``a totally new approach to treating the disease.''

If the Glaxo results are reproduced in future studies, doctors would have to conduct a gene test on patients before prescribing Avandia, Roses said. Glaxo and other drugmakers are researching ways to determine who will benefit most from drugs by analyzing patients' genetic profiles.

Preliminary Findings

Avandia generated $2 billion in worldwide sales last year and was London-based Glaxo's second-biggest drug, behind the asthma medicine Advair. In a statement released by spokesman Rick Koenig, Glaxo said the study was "exploratory'' and Avandia isn't approved for Alzheimer's.

The research involved a longer-lasting version of the drug called Avandia XR that isn't yet approved in U.S. for treating diabetes. Koenig said Glaxo is planning further tests of Avandia XR and no Alzheimer's disease trials of the drug are yet scheduled.

"The findings are too preliminary to apply to clinical practice without confirmation in subsequent clinical trials,'' Koenig said in the statement.

The new study will be published in The Pharmacogenomics Journal in a few weeks. The findings may help explain a mystery Roses set in motion with his gene discovery in 1992.

Brain Glucose

The drug works by improving the body's use of the hormone insulin, which moves glucose from the bloodstream into the body cells. Glaxo's research indicates Avandia's impact on Alzheimer's disease isn't related to the drug's effect on insulin. Instead, the drug appears to increase the availability of blood sugar in the brain, which may extend the life of neurons involved in memory, Roses said in the interview.

Alzheimer's disease is the most common form of dementia among the aged in the U.S., according to the U.S. National Institute of Aging. The best-selling drug in the U.S. for Alzheimer's is Aricept, sold by Pfizer Inc., with sales of $308 million in 2004. The product doesn't affect the underlying cause of the disease, which even when treated by the drug typically causes severe degeneration of mental functions after five to 10 years.

The U.S. institute says the precise cause of Alzheimer's is unknown. One idea is that brain cells are killed by clumps of so-called amyloid plaques that are often seen in X-ray images of the brains of deceased victims.

A Clue

Another notion was first greeted with skepticism by scientists when uncovered in the early 1990s by a genetics lab run by Roses at Duke University in Durham, North Carolina. Roses and his research colleagues found that people who are born with a version of a gene that makes a natural substance called apolipoprotein E, or APOE, are at a high risk of developing Alzheimer's disease. APOE carries cholesterol through the blood.

Research by other scientists over the years confirmed that people who inherit one type of the gene, called APOE4, are at higher risk than others of developing Alzheimer's, according to the U.S. institute of aging.

In 1997, Roses joined Glaxo to direct its genetics program and continue his APOE work. In 2002, Glaxo hired his wife, Ann Saunders, who helped lead the initial APOE research at Duke. In 2003, researchers at the Gladstone Institute in San Francisco provided Saunders a critical clue.

The Gladstone lab published an experiment showing that the No. 4 version of the APOE blood protein interfered with the ability of neurons to make use of glucose in the brain. This inability to access glucose might be killing off crucial brain cells, the Gladstone and Glaxo researchers guessed.

"We decided to try drugs like Avandia because they appear to promote glucose production by neurons,'' Roses said.

New Analysis

To test the hypothesis in people, Glaxo undertook a study of 500 patients with mild forms of Alzheimer's who were being treated in Europe and New Zealand. After six months, Avandia showed no benefit when compared with a placebo. When Glaxo separated patients on the basis of their genetics, it found that those who didn't have the type 4 variant of the gene showed a ``statistically significant'' improvement in memory, Roses said.

"The first analysis of the drug showed it failed,'' Roses said. ``Only after we re-analyzed the results based on patients' genetics did we uncover a strong response.''

Roses said Glaxo is seeking U.S. Food and Drug Administration approval for future studies that would include DNA analysis of test subjects. Glaxo said it may undertake larger trials after it gets an agreement from the FDA.

Source: Michael Waldholz. Glaxo Diabetes Drug Avandia May Help Treat Alzheimer's Disease. Bloomberg.com (7 Dec 2005) [FullText]