New evidence suggests that genetic factors may play a much greater role than previously thought in determining who gets Alzheimer's disease, the irreversible brain deterioration mainly associated with old age.

Studies done at Johns Hopkins Medical Institution have suggested that most cases of Alzheimer's disease may be caused by a so-called autosomal dominant gene, carried by either males or females. This means that the gene and the disease in a latent form may be present in half of those persons who have one parent possessing the gene and in three fourths of the offspring when both parents have the gene.

But because the onset of the illness is usually in late old age, only about one out of six persons with a parent, brother or sister with Alzheimer's could expect to live long enough to develop it.

According to Dr. John C. Breitner, coauthor of the Hopkins study, "the risk of developing Alzheimer's increases rapidly in late old age, and preliminary results suggest it may reach 50 percent of those possessing the gene who reach 90."

The delay in gene expression may depend on variables including possible trigger viruses or other environmental factors -- even, some studies suggest, a head injury.

Scientists have known for years that the family histories of some Alzheimer patients have presented a picture of this kind of classical inheritance, but these have been considered rare.

Last year, however, the Hopkins team headed by Dr. Marshall Folstein and Breitner (now with the Mount Sinai School of Medicine in New York) published a study of 3,500 nursing home patients. They screened for patients with dementing diseases and among them found 150 who met diagnostic criteria for probable Alzheimer's disease. Of those who could be studied, about 75 percent developed problems by their fourth year of illness in the use of language or in coordinating the movements of such everyday tasks as dressing.

When family pedigrees were constructed on these patients, a strong tendency was found for their first-degree relatives (parents, brothers and sisters) to develop Alzheimer-like dementia. The chance of these relatives becoming affected by age 90 (if they lived that long) was over 50 percent, or seven times higher than relatives of nondemented nursing home patients.

There are other examples of illnesses transmitted by dominent genes, including Huntington's disease, the terminal, degenerative disease of brain and body that killed Woody Guthrie. Because Huntington's disease begins so much earlier in life -- somewhere between 35 and 40 -- it shows up much more often in people with the genetic predisposition than in Alzheimer's.

At a recent meeting here of the American Psychosomatic Society, Breitner said the work at Hopkins "looked as if we'd demonstrated a specific group of individuals with classic symptoms who had a relatively pure Mendelian dominant condition" -- the classical pattern of heredity.

Because "75 percent of those ill for four years developed agraphia the inability to write a simple sentence, the criterion used in the study to distinguish Alzheimer's from other similar dementias , this suggests that such cases are the large majority Alzheimer's patients, and that Alzheimer's is, in the main, a genetic condition."

In answer to a question at the conference about how widespread the gene for the familial Alzheimer's is, Breitner said: "That's the scary part. There have been estimates of the population gene frequency as between 5 and 10 percent, but since we're diploid organisms with two complete sets of chromosomes, one set from each parent and therefore have a double chance of having the gene, somewhere between 15 and 20 percent of us may be walking around with a gene for Alzheimer's that would express itself and cause the disease if we lived long enough."

Breitner and his colleagues at Mount Sinai and the Bronx Veterans Administration Medical Center plan to publish a second study soon on a completely different population. The new study was conducted as part of a national Alzheimer's Disease Research program -- Mount Sinai and Hopkins are two of five such national centers -- and used EEGs, CAT scans and other sophisticated neurological determinants for diagnosis in place of the more primitive writing test, considered insufficient by some colleagues.

Although the study is not yet complete, Breitner reported at the meeting that a preliminary analysis confirms the Hopkins findings. A new and similarly designed study is also under way at Hopkins.

And an Italian epidemiological study of Alzheimer patients in communities, nursing homes and hospitals in Italy, scheduled to be published early next month, is expected to say that the single most compelling risk factor for Alzheimer's is the presence of a parent, brother or sister with the disorder.

Linking genetics and Alzheimer's is a concern even to those who are doing the studies because they fear that people who have close relatives with Alzheimer's may be unnecessarily alarmed.

And how important the genetic factor actually is has not been fully accepted by the experts.

Neurobiologist Dr. Katherine L. Bick, deputy director of the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) of the National Institutes of Health, says that "we know that if you have relatives who also have the disease, the increased risk is four to five times higher, but that is also true of diabetes and tuberculosis in which a genetic predisposition plays a part. There are a lot of chunks of evidence, but how much like Huntington's Alzheimer's disease is, is still a research question. It is very difficult at this point to say to anyone, 'Aha, your children are at risk.' "

However, Breitner and his colleagues point out that one of the reasons why only a few family histories demonstrate the "profound heritibility" they suspect is present is the generally late onset of the illness. "It stands to reason," Breitner said, "that if this were a genetic condition, there would be many individuals who might bear the gene and be predisposed or predestined to get it who would die first from another cause."

They suggest that the newer studies will bring broader acceptance to the hypothesis if the preliminary findings stand.

On the other hand, Breitner does not eliminate the possibility of other than genetic factors operating. "Why does it take 80 years to manifest? Clearly there must be the opportunity to have the environment have some say in how and when the gene is to be expressed. Yet other than head injury, no environmental risk factors have been turned up so far by epidemiological studies."

Alzheimer's disease was named by the German psychiatrist Emil Kraepelin after his colleague Alois Alzheimer, often credited with first describing the illness. However Breitner likes to point out that the British physician James Cowles Prichard described a group of Alzheimer patients in 1837 -- 70 years before Alzheimer described his. "Kraepelin," speculates Breitner, "probably didn't read English." NIH Studies

The National Institute on Aging is conducting several studies on senile dementias including Alzheimer's. For information, phone Angela Moore, 496-4754 or write National Institute on Aging, NIH, Bethesda Md. 20205.