The AIDS Race

Can new drugs keep up with the wily virus?

A milliliter is a tiny amount of fluid, barely a drop. But in 1994, more than 650,000 AIDS viruses swarmed in every milliliter of Dawn Averitt's blood, seeking out and destroying her few surviving immune-system T cells. Desperate, she entered a clinical trial for a protease inhibitor. Within weeks doctors couldn't detect any virus in her blood, and her T cells started to rise.

About a year and a half later, her virus reappeared on lab tests, turning her into what researchers call a "treatment failure." Averitt, a leading AIDS activist, thought she knew what to expect: The virus would re-saturate her blood and continue the destruction of her immune system. That was the dominant theory, and, for a year and a half, she went to every doctor's visit expecting exactly that bad news.

But something curious happened. The level of HIV in her blood remained low, and she didn't lose any T cells. She felt fine, too. Her immune system wasn't getting any better, but it wasn't getting any worse. She was in a kind of limbo.

''You can look at my data,'' says San Francisco researcher Steven Deeks, ''and see a justification for any theory.''
Rick Gerharter
''You can look at my data,'' says San Francisco researcher Steven Deeks, ''and see a justification for any theory.''

So is the entire AIDS field, because it turns out that patients like Averitt are common. As many as half of HIV patients "fail" on the drug cocktails that have cut annual AIDS deaths for two years in a row. The drug-failure rate— inching up over time— is causing tremendous anxiety, as was apparent at a major AIDS meeting, the Sixth Conference on Retroviruses and Opportunistic Infections, which drew almost 3500 doctors, researchers, and advocates to Chicago last week.

"It's disturbing that as time goes on you have fewer people responding to therapy," says Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases. Indeed, every treatment failure can make the virus stronger, because HIV can develop drug-resistant mutations that allow it to escape not only the particular medications it has encountered but whole classes of drugs. In short, a patient's first shot is her best, but up to half of patients are failing their first regimen.

That's why Lawrence Deyton keeps thinking about the blip he noticed the other day. As director of AIDS services for the Department of Veterans Affairs, Deyton is responsible for the care of more HIV patients than anyone in the country— 18,600— and ever since the advent of protease inhibitors, all the signs at the VA have been good. But Deyton recently noticed that the length of hospital stays, which had been dropping, bumped up slightly: "I wondered if that would be the turning point." After all, he says, it's only a matter of time until the "virologic failures" turn into clinical failures— until disease and death go back up. A pessimistic view? "I'd put it this way," he answers. "It's realistic."

"This whole idea that things will come crashing down, it's just not what we're seeing," counters Roy Gulick, a physician and director of the Cornell HIV clinical trials unit at New York Presbyterian Hospital. Several prominent doctors concur. The Wall Street Journal dubbed the results of Gulick's most famous study— a three-year trial of the drug indinavir, which has failed to contain HIV in one-third of patients— "disheartening." But, Gulick says, "I take the opposite view. After three years, two-thirds of patients still have undetectable virus. And it's not like the whole field is going to stop. There's a new batch of drugs active against resistant virus."

This split made the Retrovirus Conference feel like two separate meetings, one warily optimistic, the other ominously pessimistic. And it was Dr. Steven Deeks, a physician and researcher at San Francisco General Hospital, who had the Rorschach test: a cohort of patients like Dawn Averitt, virologic failures who have retained strong immune systems and good health for almost two years. "You can look at my data," he says, "and see a justification for any theory you have."

Here's what Deeks has found: When drugs completely suppress the virus so that it can't be detected in blood, patients show the best immune-system regeneration, with their T-cell counts rising the highest. But when drugs only partially suppress the virus, or when they push it below the level of detection for a while but then let it creep back up, patients still show a sustained improvement. It's not as robust as the group in whom the virus is totally suppressed, but it's much better than untreated patients or those for whom the drugs don't work at all. These "partial responders" to treatment have held onto their gains for almost two years. What's more, preliminary studies show that they don't just have more T cells, but ones that are longer-living and less overtaxed than in untreated controls.

What's happening? The simplest explanation is that in partial responders the virus has mutated into a strain that is moderately resistant to the drugs. If so, classical Darwinian theory predicts bad news: The virus will continue to mutate, growing stronger in the presence of the drugs. Levels of virus will eventually rise, and the immune system will sink.

But there are other possibilities: Perhaps the drugs, through some unknown mechanism, directly fortify the immune system. Or maybe they render the virus less deadly so that it can replicate benignly without killing T cells— or, at least, without killing them as often. That's what a Canadian study found: In the test tube, protease inhibitors reduced HIV's ability to induce cell death, and this diminished virulence seemed to have nothing to do with the virus's capacity to replicate.

Next Page »