Several recent studies suggest that cancer patients who receive blood transfusions at the time their tumors are surgically removed have a significantly greater chance of early recurrence and death from the disease than other patients.

Researchers said that preliminary findings, which indicate that blood transfusions may adversely affect the body's natural defenses and its response to tumor growth, could have major implications for the treatment of certain cancers and other diseases, as well as provide basic understanding about their development.

But, scientists cautioned, more studies are needed to determine whether blood transfusions or other factors are responsible for the striking differences among surgery patients treated for different forms of cancer.

At a meeting here which ended today at the American Association for the Advancement of Science, scientists reported a link between blood transfusions and subsequent deterioration in patients who underwent surgery to remove colon, lung, breast, cervical and soft-tissue cancers. Other studies, however, have found no such link.

"Many physicians are too liberal in giving transfusions," said Dr. Roger S. Foster , Jr., of the University of Vermont. "There are lots of reasons to use blood transfusions only when absolutely necessary . . . . If this link proves to be definitive, it will totally change blood banking practices."

"This is one more thing to keep in mind," said Dr. Paul I. Tartter, a surgeon at Mt. Sinai Medical Center in New York, who was the first to report a possible link between blood transfusions and poorer outcome in the incidence of cancer recurrence in surgery patients.

Cancer accounts for 460,000 deaths annually. Lung and colon cancers are the most common. While treatment frequently means surgery, there is wide variation in the degree to which blood transfusions accompany the operations.

Blumberg, in an April study in the British Medical Journal, found that colon cancer patients who got blood transfusions at the time of surgery had about a fivefold greater recurrence rate than patients who were not given transfusions. In a followup period of about four years, tumors recurred in 9 percent of those not given transfusions, compared with 43 percent of those with transfusions.

Foster, who directs the Vermont Regional Cancer Center in Burlington, reported that his group also found an "increased likelihood of death from colon cancer" in patients who got blood transfusions at surgery. While their five-year survival rate was 51 percent, it rose to 68 percent among colon cancer patients who did not receive transfusions, he said.

In patients who underwent therapeutic (as opposed to palliative) surgery for lung cancer, he found that the five-year survival for patients who received transfusions was 27 percent, compared with 44 percent for the patients who did not receive them, with the risk rising with the number of transfusions.

However, a Vermont study of breast cancer patients found no such difference. And researchers from the University of Michigan and the University of Melbourne in Australia said Thursday that they had not found a significant difference in their studies of colon cancer patients.

There was general agreement, based on human and animal studies, that blood transfusions can suppress activity in the immune system, reducing the body's defenses against infection and other foreign substances. But nobody knows exactly why.

In the 1970s, studies showed that when blood transfusions were given to kidney transplant patients before surgery, they were less likely to reject the organ.

Nobody knows exactly why the transfusions affect the immune system, but it is thought that the white blood cells are involved.

It is thought that frozen blood may have less effect on the immune system because the white blood cells are virtually destroyed in the freezing process.