"Since the majority of the approximately 4 million Americans who have AD become afflicted over the age of 70, the main genetic causes for the disease have yet to be identified," said NIMH grantee, Rudolph E. Tanzi, Ph.D., of Harvard's Massachusetts General Hospital, the senior author of the study. People with AD occurring under age 70 make up less than 10 percent of the AD population.

The gene under study, the Apolipoprotein E-4 (apoE-4) gene, appears in half to two-thirds of all Alzheimer's patients, compared to less than one-third of the general population, suggesting that it plays a significant role in the disease process. But the new report shows that despite its prevalence, the gene is only a strong predictor of risk for the disease in individuals under 70, especially those who have inherited two copies of it -- one from each parent -- versus one copy.

Steven E. Hyman, M.D., Director of the National Institute of Mental Health, whose Genetics Initiative funded the study, said, "These results should steer scientists elsewhere for genetic clues to the later-onset, more common form of this mysterious, mind-destroying disease."

The newly published research is remarkable for its sample size of 679 individuals in 310 families, which was large enough for the researchers to separate out families by age at onset of AD symptoms. Dr Tanzi said, "It remains critical to search for additional genes involved in the development of AD beyond age 70."

In fact, Dr. Tanzi reported such evidence at a July 1996 AD meeting in Japan and again last week at a science symposium in Colorado. He said that the NIMH Genetics Initiative data provided preliminary evidence, as yet unpublished, that there may be late-onset AD genes on chromosomes 12 and 3.

The newly published research also shows that in the studied families, several individuals over 80, who were Alzheimer's-free, carried two copies of the apoE-4 gene. So, said Dr. Tanzi, the apoE-4 test is clearly not a reliable predictor of the disease. This finding is consistent with the findings of two national panels that examined the practical considerations regarding genetic assessment for AD using apoE-4 within the last 2 years. However, said Dr. Tanzi, genetic research carries the greatest promise to ultimately clarify mechanisms leading to the development of AD. "In the future, reliable genetic tests for AD could be used to identify those who are most likely to develop the disease," he said.

Dr. Tanzi's Harvard collaborators on the study, published in the February issue of Neurology, were lead authors, Deborah Blacker, M.D., Sc.D., and Jonathan Haines, Ph.D., along with principal investigator Marilyn S. Albert, Ph.D.

NIMH is one of the 16 institutes that make up the National Institutes of Health, part of the Public Health Service within the Department of Health and Human Services.

The report published in the February issue of Neurology was funded by the NIMH Genetics Initiative. Samples for this study came from three sites: Massachusetts General Hospital, affiliated with Harvard Medical School in Boston; the University of Alabama School of Medicine in Birmingham; and The Johns Hopkins Medical School in Baltimore.

For this study of Alzheimer's Disease, families were recruited from memory disorder clinics, nursing homes, and the communities surrounding the three sites. They were required to contain at least two living blood relatives with memory problems. One member of the pair was required to meet standard criteria for probable AD, and the other for possible or probable AD. Autopsies, which are the only way to definitively diagnose AD, have been performed on 113 of the initial subjects, with 112 (99.1%) meeting neuropathologic criteria for definite AD.

Steven O. Moldin, Ph.D., Chief, Genetics Research Program and Scientific Coordinator, NIMH Genetics Initiative has said, "Given our lack of knowledge of exactly how many genes are involved in AD and how they interact, it becomes very difficult to determine exactly how large a genetic sample should be. Experience with schizophrenia and bipolar disorder as well as with other genetically complex disorders like type I diabetes, suggests that sample sizes of at least 400 to 500 families ultimately may be required. The NIMH is currently exploring the various options for the growth and enrichment of this scientific resource."

Genetic research offers an opportunity to determine how AD develops and to develop a genetic test to identify individuals with the disease. Identification of persons with mild cognitive complaints who are at risk for AD would allow physicians to apply existing antidementia treatments before extensive brain damage develops, thereby slowing the progression of AD and substantially reducing the number of years of disability associated with the disease. This is especially important because no currently foreseeable treatment for AD would reverse permanent neuronal damage. Early detection strategies can provide reassurance to people whose age-related memory complaints are benign or help in the diagnosis of other more treatable causes of dementia. In addition, patients may wish to know their prognoses while their mental faculties are intact so that they can plan for their future care. The currently available genetic test involving the gene apoE-4 has been deemed an unreliable predictor of AD, emphasizing the need for more research into the genetics of this complex disorder.