AIDS researchers have largely discarded their early hopes of developing a single drug effective enough to halt the spread of the deadly virus, and are searching instead for combinations that complement each other.

Right now it is not reasonable to look to a single 'magic bullet' to fight this epidemic or to talk about a cure," said Dr. Anthony S. Fauci, director of AIDS activities at the National Institutes of Health. "The model for our work is cancer treatment, where we attempt to assemble a potent mixture of drugs."

The quick approval of azidothymidine, or AZT, as the first treatment for acquired immune deficiency syndrome had led to expectations for other drugs that experts now consider unrealistic. While many major drug trials are under way, only AZT has proved effective in tests. The drug prevents human immunodeficiency virus (HIV), the AIDS virus, from multiplying and has improved the health of more than half of those who take it. But serious side effects, such as severe anemia, have forced others to stop using it.

The basic goal of AIDS drug researchers has been to create a series of safe drugs that people infected with the virus would be able to take for years without suffering disabling side effects. That will be essential for surviving AIDS in the future, because once the virus enters a person's genetic material he or she will remain infected permanently.

There is now enough knowledge about the HIV virus for scientists to envision a combination of drugs that will prevent it from reproducing, while boosting the body's immune system in order to defeat the infections that eventually kill AIDS patients.

Cancer researchers have long known that drugs that seem of little use by themselves can prove valuable when given in concert with others. In fact, doctors at the National Cancer Institute have already seen tentative evidence that alternating AZT with dideoxycytidine, or DDC -- a much more powerful and toxic chemical cousin -- seems to lessen the dangerous side effects of each drug.

DDC is at least 10 times better at killing the AIDS virus in a test tube than AZT. But in high doses in humans it often causes a condition called peripheral neuropathy, severe pain in the hands and feet of patients.

NCI researchers tried lowering the dose and alternating one week's worth of DDC with one of AZT. For several patients the change appears to have worked; a few have been on the course for almost six months without any of the problems associated with either drug.

"The results are preliminary but very exciting," said Dr. Samuel Broder, director of clinical oncology at NCI, and a leading AIDS researcher. "What we have shown is that alternating drugs and combining them may make the best sense in the long run as we look for a way to contain this epidemic."

Broder points to the history of leukemia research as a model for his work. Thirty years ago childhood leukemia was nearly always fatal. Today, the great majority recover.

"By 1960 it was clear that no one drug was going to cure childhood leukemia," he said. "What the pioneers in that field realized was that they should never simply throw away a drug when it didn't seem to work. The real advances came after people tried combining therapies that didn't work at all by themselves."

For those who can tolerate it, AZT has extended many lives, but it also has a tendency to damage bone marrow severely. In an attempt to minimize that side effect, researchers are about to initiate a trial in which AZT will be combined with a hormone, GM-CSF, that stimulates the growth of white blood cells found in bone marrow.

"What we are all realizing is that gaining control of this disease will be a long, slow incremental process," said Dr. Paul Volberding, chief of AIDS activities at San Fransisco General Hospital. "We have to take a systematic, rational approach to drug development. And we have to rely on the medical knowledge we have."

He and other AIDS researchers assume they will need an arsenal of drugs against the virus, since a drug's effect can vary widely depending on the health of the patient. Some people infected with HIV, for example, can remain healthy for years; others become desperately ill almost immmediately.

It may be possible to administer one powerful battery of drugs when a patient is first treated, another set for several months afterward and still a third, less toxic group, would be taken for years.

Researchers have identified more than a dozen spots on the AIDS virus that may be good drug targets. They include a variety of proteins, and at least two special genes that regulate the cell growth.

Perhaps the most promising future treatment involves the possibility of using genetically engineered copies of the molecular protein, called CD4, that serves as a receptor for the AIDS virus.

For the infection to begin, the outer envelope of the virus must lock into the CD4 receptors. Researchers hope to manufacture the fake CD4 receptors and give them to patients as a drug. The virus could then lock onto these elaborate decoys instead of the real thing. Since the AIDS virus can only replicate when it penetrates a cell's CD4, the fakes would, in effect, disarm the virus.

"We are working on some exciting ideas," said Broder. "But I don't want people to be misled. It takes quite a while to put an idea into pill form."

Despite its toxic potential, AZT will remain the major drug used to fight the virus until a new combination can be proven to be safe and effective. And with the risks so great for those infected, many physicians have begun to prescribe AZT to healthy patients who are infected even though the Food and Drug Administration has approved it only for sick people to use.

Until recently, only the sickest AIDS patients received AZT and other experimental drugs. But with the realization that early treatment holds the only hope for thousands of people, doctors have started to study the effect of AZT on healthy people who test positive for the AIDS virus.

"We need to find ways to intervene at an early stage as we have in other illnesses," said Dr. Jerome Groopman, who runs the AIDS program at Boston's New England Deaconess Hospital. "The real benefit of controlling high blood pressure, after all, comes before someone has a stroke. After the stroke you just won't see results that are as gratifying."

In addition to AZT, scientists are working on a variety of ways to stop the virus from multiplying in the body. Groopman and scientists at Genentech Inc., a biotechnology company in South San Francisco, have succeeded in creating a cellular protein that in laboratory experiments absorbs the AIDS virus, preventing it from reaching its target.

Their approach, described in this week's issue of Science, is one of a number of ways that are being tried to break the AIDS virus into its component parts and then disable them. Most of these approaches have not yet been fully tested in humans and none have been approved by the FDA.

For now, many of the doctors who must care for and advise hundreds of thousands of people infected with the AIDS virus have decided that they cannot in good conscience wait for definitive answers to early intervention studies before prescribing AZT, the one approved treatment.

"It's clear that people who start off healthy get sicker and sicker over time," said a prominent Washington AIDS doctor who asked that his name not be used. "How can we not use the one thing we have that seems to make a difference?"