Can interrupting their treatment benefit HIV-infected people?

Four years ago, the new, multidrug regimens against the human immunodeficiency virus held out a tantalizing hope: a cure for AIDS. Now, however, complete eradication of the virus with these life-prolonging treatments seems improbable. To further complicate matters, some patients taking the drug combinations have developed serious--potentially fatal--side effects.

Up to 900,000 people in the United States are now living with HIV. The retrovirus infects immune cells, over time wiping out a person's defenses against other infections. Despite the virus' ability to mutate rapidly, simultaneous treatment with three or more antiretroviral drugs has succeeded in beating back HIV concentrations in the blood of many patients. This in turn has reduced or eliminated many of the opportunistic infections--such as pneumonia, skin cancer, and tuberculosis--that ultimately kill AIDS patients. With these new drugs, about 80 percent of people survive for at least 10 years after becoming infected with HIV, European researchers reported in the April 1 LANCET. Before the advent of these combination therapies, only about 55 percent of people lived for 10 years or more after infection.

Yet this success has had a price. Researchers at a recent AIDS conference reported links between successful drug therapy and an increasing number of side effects. Among the most serious are liver failure and diabetes. Physicians are also observing lipodystrophy, abnormal accumulation of fat in the back and abdomen, and abnormally high concentrations of fats in the blood. Bone loss and heart attacks have also been tentatively correlated with anti-HIV therapies.

Over the past year, researchers have found that even when drugs suppress the virus to nearly undetectable amounts in a person's bloodstream, slowly replicating HIV may be hiding in sanctuaries such as the lymph nodes or genital tract. Perhaps the most disappointing news came last fall. In three patients who, according to the most sensitive assays, had no HIV in their blood, lymph nodes, or genital tract, the virus roared back soon after the patients stopped taking antiretroviral drugs, researchers at the National Institutes of Health in Bethesda, Md., reported in the Oct. 28, 1999 NATURE.

"Eradicating HIV has been an elusive goal. It is clear-cut that [multidrug treatment] is not enough," says Julianna Lisziewicz of the Research Institute for Genetic and Human Therapy in Washington, D.C.

If the battle against HIV can't yet be won with drugs, however, there's still hope for a standoff, she adds. Several years ago, against medical advice, one of Lisziewicz' patients stopped taking his medication, restarted, and then quit again. Against all expectations, this man's immune system has succeeded in keeping the virus under control, though not eliminating it, for 3 years so far.

Spurred by this unusual development and the mounting side effects of state-of-the-art drug therapies for HIV disease, Lisziewicz and other researchers have begun exploring the possible benefits of what they call structured treatment interruptions.

The research focuses on three major questions. Can interruptions in drug treatment preserve or strengthen a person's immune responses against HIV, reducing or even eliminating the need for drugs? Can treatment interruptions affect the virus' ability to resist drugs, making drugs more effective once a person resumes taking them? Finally, if interruptions don't have any benefit, do they at least cause no harm and therefore enable people experiencing side effects to take a much-desired break?

Preliminary evidence suggests that repeated treatment interruptions can strengthen some people's immune response to HIV. Most HIV-infected patients who stop taking antiretroviral drugs find that the virus quickly rebounds. However, several small studies presented at the recent Conference on Retroviruses and Opportunistic Infections in San Francisco suggest that some people who have interrupted their therapy more than once hold down their viral concentrations, or loads, longer in later drugfree intervals.

Researchers theorize that repeatedly exposing a patient to rebounds of virus sensitizes the person's immune system, just as repeated booster doses of vaccines protect healthy people against disease.

At the retrovirus meeting, researchers presented preliminary results of a European trial that will eventually enroll 120 patients. According to the experiment's design, therapy will stop for 2 weeks, then resume for 8 weeks, and the whole process will be repeated three times, says lead researcher Bernard Hirschel of the University Hospital in Geneva. Among 75 people who have completed the second treatment interruption, 10 had slightly smaller rebounds of virus in their blood than had been measured during the first interruption.

"I think we can safely predict that structured treatment interruptions will induce remission in only a small number of patients," Hirschel says. So far, the interruptions don't seem to pose a danger, he adds.

Lisziewicz' recent research suggests that treatment interruptions may be more effective for patients on some drug regimens than on others. Those taking the drugs hydroxyurea and didanosine may be especially likely to respond well, she says. Hydroxyurea is a well-established cancer drug that interferes with cell division and perhaps makes cells less vulnerable to HIV. Didanosine blocks HIV replication. However, patients rarely take this drug combination because it isn't as effective as other multidrug therapies that are now the standard of care, Lisziewicz says.

She and her colleagues report that during an 8-week treatment break, none of the nine patients on the hydroxyurea-didanosine regimen showed increases in HIV blood concentrations large enough to require a patient to restart therapy. In contrast, HIV rebound prompted five of eight patients taking a break from other multidrug regimens to go back on their therapies before the end of the 8 weeks.

The relatively low potency of the hydroxyurea-didanosine treatment may actually be a benefit, Lisziewicz says. When multidrug therapy lowers virus loads to nearly undetectable concentrations, any immune response that might have been present is turned off because the immune system can't "see" the virus any more, she says.

If the virus can't be eliminated, what's needed is some condition in which HIV activates the immune system but can't replicate fast enough to overwhelm it, she says. Hydroxyurea-didanosine therapy's relatively low effectiveness at controlling HIV "was bad news a couple of years ago but is now good news," Lisziewicz maintains.

Other studies presented at the meeting suggested that some HIV-infected people who quit their medications don't experience an uncontrolled infection in the next few months. This challenges the idea that repeated cycles of structured treatment interruptions are critical to immune control of the virus.

An international team of researchers identified 238 patients who contracted HIV after 1996, began therapy, and then discontinued it for a variety of reasons but were nonetheless followed by their physicians for another 3 months. Thirteen of these patients showed low or undetectable blood concentrations of HIV at the end of that period.

Another study of 12 patients who quit taking antiretroviral drugs suggests that their HIV blood concentrations rose to pretherapy values and then plateaued.

Daniel R. Kuritzkes of the University of Colorado Health Sciences Center in Denver cautions that "although the results [of structured treatment interruptions] to date are tantalizing, they have to be considered equivocal at this point."

Whether or not drug holidays can boost immune response, there's some evidence that interrupting treatment can boost the chances that later drug therapy will control HIV. Generally, HIV that has developed drug resistance replicates less effectively than the drug-sensitive retrovirus. During treatment intermissions, when viral concentrations soar, the theory goes, the faster-growing, drug-susceptible forms of the virus will win out over the drug-resistant ones.

This idea receives support from a study by Steven G. Deeks of San Francisco General Hospital. At the recent meeting, he reported that 16 out of 18 people resistant to protease inhibitors--widely used and powerful anti-HIV drugs--showed a switch to a drug-susceptible virus within 9 weeks of being taken off therapy. Although this suggests that treatment breaks might be helpful, he and his colleagues caution that they could still isolate drug-resistant virus from blood samples of half of the patients tested so far.

Results of a more extensive experiment have been presented at several meetings by Veronica Miller of Johann Wolfgang Goethe University in Frankfurt, Germany. She followed just under 100 patients who were receiving treatment for HIV infections that had become drug-resistant. Half the patients were immediately switched to a new regimen containing five to nine medications; the other half took a 2-month break before starting on the intensive drug regimen. Once they were on the drug treatment, those who had taken the break were more likely than the others to have their viral concentrations fall below detection limits.

Of 39 patients who interrupted treatment and were tested for viral drug resistance, 26 showed a switch from drug-resistant to drug-sensitive virus. Kuritzkes cautions that if drug-resistant strains reemerge after new therapies are initiated, the benefit of the break would only be temporary.

Whether or not using structured treatment interruption is a successful, long-term approach for patients trying to fight off drug-resistant HIV will depend, in part, on how long a reservoir of drug-resistant virus persists, Kuritzkes says.

Perhaps the most realistic hope for treatment interruptions is that they will offer a temporary haven from nasty side effects and the rigors of the treatment.

As long as interruptions prove feasible and safe, patients might benefit physically and emotionally by taking a break from their exhausting drug regimens, says Anthony S. Fauci of the National Institute of Allergy and Infectious Diseases in Bethesda, Md. "In and of itself, this would be an important advance," he says.

There are some potential risks to interrupting treatment, however. "This is very much experimental therapy," warns Lisziewicz. "Patients and doctors should not try it on their own. We don't yet have a recipe for what works--or doesn't work."

Stopping treatment all at once might leave some drugs in the bloodstream longer than others. The researchers are concerned that, under these conditions, the virus might develop resistance to the longer-lasting drugs.

Although restarting therapy has, so far, reined in the viral loads that rose during treatment interruptions, nobody knows what will happen during repeated drug holidays over long periods. Every time HIV concentrations rise, immune cells suffer, says David Ho of the Aaron Diamond AIDS Research Center in New York City. He warns that at some point the immune system may not recover, even after therapy begins again.

Despite all these unknowns, researchers are making concerted efforts to understand the implications of treatment interruptions--and to improve their chances of success.

"The immune system is quite effective in keeping some viruses dormant, such as hepatitis B, herpes zoster, and herpes simplex, even if it doesn't eliminate them completely," says Kendall A. Smith of the New York Hospital-Cornell Medical Center. He suggests that stimulating a person's immune response before stopping therapy may boost the likelihood that the immune system can keep HIV under control without drugs indefinitely.

Smith has tested a daily, low dose of IL-2--a signaling molecule that helps activate the immune system. Among nine patients who continued IL-2 injections while discontinuing anti-HIV drugs, the virus rebounded within about 5 weeks, he says. During the following 2 weeks, however, the viral load fell precipitously and then leveled off at about 10 percent of the rebound peak. As viral loads rose and fell, the concentration of immune cells responsible for removing the virus increased.

So far, the longest break that any of these patients has taken from anti-HIV therapy is 10 months. While still taking IL-2, all the patients restarted treatments, but after 2 to 10 months, four again discontinued therapy. "So far, the character of the relapse is attenuated," Smith says The peak viral concentrations were only a tenth those reached during the first break in therapy.

Strikingly, during IL-2 therapy, the blood concentrations of the types of white blood cells attacked by HIV decline briefly and only by about 25 percent as the virus rebounds, Smith reports. "Contrary to previous assumptions, chronic HIV infection does not preclude the development of effective immune reactivity to HIV," he says.

Several groups of researchers, including Smith's and Lisziewicz', are testing another way of boosting immunity to HIV. They're using vaccines to stimulate a patient's immune reaction against HIV rather than letting the viral loads rise.

In the first preliminary study of this strategy, Xia Jin of the Aaron Diamond AIDS Research Center injected a vaccine into four patients who were on antiretroviral drug therapy. After a month, they discontinued the drug therapy. In two of these patients, the viral infection increased more slowly than the researchers would have expected without the vaccine.

"This is a glimmer of hope for the potential use of therapeutic vaccination," Jin said.

"The work in structured interruptions of therapy is fascinating, but the numbers of patients in these studies are small," Deeks cautions. "If it works, it would be wonderful for the patients, but we need a lot more research."

According to Fauci, a pressing question is whether more than a small minority of patients will ever benefit from interrupted therapy.

Smith warns that if no cure for AIDS is found, but it becomes possible to restrict HIV infection to a chronic condition, new concerns will arise. After 2, 3, or even 4 decades of low HIV blood concentrations, patients might face an immune-system failure similar to that experienced by people who have progressed to AIDS, he says.

Yet regardless of potential problems down the road, any increased control of the disease by the immune system is a critical step forward because doctors can't rely solely on drugs that suppress HIV replication, he says.

"Right now, the disease is life-threatening, on the one hand, and the drugs that we have so far have life-threatening toxicities, on the other hand," he says. "It puts us between a rock and a hard place."