By Lisa Seachrist

Washington Editor

WASHINGTON — An FDA advisory panel has endorsed the agency's plan to approve AIDS drugs based on their ability to suppress HIV replication.

The Antiviral Drugs Advisory Committee this week agreed that a durable suppression of virus as detected by assays of HIV RNA predicts clinical benefits for patients and could serve as a surrogate endpoint for antiretroviral agents.

"We have in front of us extraordinarily strong data," said panel chair Scott Hammer, associate professor of medicine at Harvard Medical School. "But in order to fully understand the implications of viral load on clinical outcomes, we must insist on a strong commitment to Phase IV studies."

Jeff Murray, medical officer with the agency's division of antiviral drug products, noted that the committee's endorsement should alleviate any fears that companies currently conducting clinical trials have about the acceptability of the surrogate endpoint.

"We've already moved to this endpoint in the agency," Murray said. "This should reassure sponsors that this endpoint is acceptable. Companies planning on AIDS drug trials can look to this meeting for guidance."

Murray noted that the agency may sponsor informal workshops to help companies design clinical trials using RNA as a surrogate marker, but that it is unlikely the agency will create a guidance document on the issue.

The agency's decision to move toward measures of viral RNA in the blood — or viral loads — as a marker for tracking the progression of HIV disease comes on the heels of studies showing that patients whose viral loads are suppressed below the detection limits of current assays live longer and aren't as likely to develop opportunistic infections.

In June, the Public Health Service published a new draft set of guidelines for treating HIV-infected adults and adolescents that noted treatment should be started early in the course of this disease and should aim to reduce plasma HIV RNA levels as low as possible, to that not detectable by current assays.

However, the usefulness of the surrogate has yet to be firmly established in clinical trials, so the agency requested the committee give its opinion on whether they should proceed in accepting measures of HIV RNA as a surrogate endpoint and how the endpoint should be used.

"We are asking the committee to help us decide how we are going to use viral loads and CD4 counts in evaluating drugs," said David Feigal, director of the office of drug evaluation IV at the FDA. "Many diseases are already treated by monitoring disease progression with laboratory tests; are we torturing ourselves too much not to realize that is what we are doing here?"

The committee unanimously agreed that the time has come to use viral load as a surrogate marker, prompting John Modlin, a physician with the Dartmouth/Hitchcock Medical Center, to note, "We are passed the issue of if, but on to the issue of how we are going to use these measures."

However, the panel also agreed that there were still many questions to be answered about what measures of viral load mean in determining whether a patient is responding to therapy or failing therapy. For example, the committee examined data from several pharmaceutical and biotechnology companies that showed that for patients who responded to therapy, it took anywhere from four to 24 weeks for them to have their viral loads drop below the detectable levels of the current commercially available assay — 400 copies/milliliter.

Newer, more sensitive tests are beginning to allow researchers to detect viral load as low as 20 copies/ milliliter. The committee noted that it needed more information about the implications of detecting such minute levels of virus and recommended the agency base clinical trial decisions on the less sensitive test that is currently in use. Even with that test, the panel urged caution in switching drug regimens.

"I see many patients that bounce around the detectable levels and are clinically fine," said Wm. Christopher Mathews, professor of clinical medicine at the University of California at San Diego Medical Center. "We need to be sure that we don't prematurely decide a therapy isn't benefiting a patient and switch them off the therapy."

Judith Feinberg, associate professor of clinical medicine at the University of Cincinnati, echoed Mathews concerns, noting, "We don't know if it is in the best interest of patients to switch drug regimens at 400 copies. We've only got 11 drugs and they often have cross resistances."

Given the limited number of drugs, as a patient begins to fail therapy, the Public Health Service guidelines recommend switching at least two of the three drugs in the regimen he or she is using. Hammer said it is very important to be sure that "patients aren't worse off after the clinical study than they would have been had they never participated."

In order to assure that patients wouldn't receive sub-optimal care, the committee agreed it would be important to allow for some flexibility in trial design.

"We need to be in sync with what clinicians really need to do," said Wafaa El-Sadr, director of the division of infectious diseases at Harlem Hospital Center. "Clinical trials need to reflect what is good clinical practice."

The committee also suggested the agency look closely at whether viral load means different things to different subpopulations of AIDS patients — for example, children and very advanced patients who have used many different therapies.

Murray noted that the agency is likely to use a 500 copy/milliliter viral load as the indication that a therapy is failing and continue with its plans to label a drug effective if it reduces viral load by 0.5 log in 16 to 24 weeks and in combination with other agents suppresses viral load for 48 weeks. *