A rare virus that causes cancer has been detected in some patients who have Acquired Immune Deficiency Syndrome, according to research to be published today. The discovery is the strongest clue yet to the cause of AIDS, a disease that destroys the body's immune system.

Researchers are unsure of whether this means that the virus causes the disease or simply affects patients after they become ill with AIDS.

Researchers at the Harvard School of Public Health, the National Cancer Institute (NCI) and the Pasteur Institute all found signs of human T-cell leukemia viruses in some patients with AIDS, as well as those with an illness that may precede the disease, according to research to be published today in the journal Science.

"It's a very strong lead. There hasn't been a major candidate thus far," said Dr. Peter Fischinger, the NCI's associate director. "Within six months to a year, we'll have a much better idea whether to rule this in or out."

"It's perhaps the most provocative, intriguing laboratory finding thus far," said Dr. James Curran, head of the Centers for Disease Control's AIDS task force. "However, these findings should not be prematurely overinterpreted as meaning this particular virus is the cause of AIDS . . . . Rather it is an important clue."

More than 1,410 male homosexuals, drug addicts, Haitians, sexual partners of AIDS patients, children, hemophiliacs and other recipients of blood products have contracted AIDS since the disease was first identified in 1981. So far, 541 of them have died. Their immune systems were destroyed, leaving them vulnerable to whatever infections or cancers came their way. There is no cure.

In the new research, a team headed by Harvard scientist Max Essex, working with CDC scientists, found a strong increase in T-cell leukemia virus antibodies in AIDS patients, a sign they had been infected with the cancer virus.

In tests of blood samples from 75 AIDS patients, they found signs of antibodies in 25 percent of the cases and weaker signals in another 10 percent. Half were clearly negative.

But in a group of matched control patients without the disease, they reported antibodies in only about 1 percent--one patient who was a friend of an AIDS victim. While more common in other parts of the world, T-cell leukemia virus infections in general are thought to occur in less than 1 percent of Americans.

In other studies, NCI's Dr. Robert Gallo and colleagues isolated the cancer virus from three patients with AIDS. They also found genetic material from the virus incorporated in some white blood cells of two out of 33 AIDS patients.

During the same period, French scientists found a related virus in a patient with an illness involving swollen lymph glands and fever that may be a forerunner of AIDS. The Harvard group found that one-fourth of patients with this illness showed signs of the virus.

Researchers suggested several possible explanations as to why antibodies or the virus itself were not found in more AIDS patients, if indeed the human cancer virus should turn out to be the cause.

"Our tests may not be sensitive enough," said Essex, adding that the virus could have done its damage earlier in the disease and disappeared or that the weakened immune systems of AIDS patients may not produce many antibodies. Gallo, who first isolated the cancer virus in humans in 1980, said the virus may be difficult to detect under his system of looking for it if the number of impaired cells is small.

In interviews, Gallo and Essex said the T-cell cancer virus is an important candidate as a possible cause of AIDS because both T-cell leukemia-type cancers and AIDS involve damage to T-cells, a type of white blood cell.

But while T-cell leukemia involves a proliferation of white blood cells that have gone awry, the opposite occurs with AIDS, said Gallo. He speculated that there could be a "subtle change" in the virus that affects whether it causes immune suppression or leukemia.

AIDS has been found in Haitians and Africans, while T-cell leukemia cancers are also more common in the Caribbean region and parts of Africa. On the other hand, there is no evidence of AIDS in southern Japan, where such cancers are also common.

Both human viral T-cell cancers and AIDS appear to be transmitted only after intimate personal contact or through blood products. And other distantly related viruses are known to depress the immune system as well as cause leukemia.

A fifth paper in Science by Essex's laboratory shows that a cat leukemia virus can cause both types of illness in animals. There is no evidence that this virus affects humans, Essex emphasized.