Alarmed researchers push for more aggressive treatment and monitoring

Rise of resistance must be attacked with vigilant testing, novel therapies

As reports of virologic breakthroughs begin to increase, AIDS researchers are wondering aloud if the honeymoon period for protease inhibitors is about to end, or whether new diagnostic tools and clinical fine-tuning can sustain the level of improvement seen in so many patients. It is clear, however, that clinicians must be more vigilant in detecting drug failure and responding with the right regimens, a group of top AIDS experts recently concluded.

"We are using these new drugs and getting benefit we have never seen before," said Michael Saag, MD, director of the University of Alabama’s AIDS Outpatient Clinic in Birmingham during a July teleconference on HIV treatment. "The question is, are we going to exhaust that benefit in large numbers of patients?"

The panel’s answer was a consensus that any resistance to the new drugs would not result in a return to morbidity and mortality rates seen five years ago. And yet new information coming out of a June conference on HIV resistance in St. Petersburg, FL, raised enough red flags to temper the enthusiasm of many AIDS researchers.

"I think the news from the meeting was fairly bad," said Brendan Larder, PhD, director of the clinical virology unit at Glaxo Wellcome. "It looks more and more like there are generalized patterns of resistance to all the protease inhibitors currently available."

John Mellors, MD, director of the HIV/AIDS program at the University of Pittsburgh, agreed. "It seems like treatment with one protease inhibitor certainly affects the response to the second protease inhibitor. The one exception appears to be nelfinavir, and the reason may be because we don’t have data on individuals who switch from nelfinavir to one of the other protease inhibitors. Every time we have looked at patients who have been treated with a protease inhibitor other than nelfinavir and then followed with a different protease inhibitor, the response looks fairly mediocre and is very worrisome, to be honest."

The extent of overlapping resistance patterns in the first generation of protease inhibitors, combined with the difficulty in following the regimens, has led some federal health officials to estimate that as many as one out of two patients could eventually fail on the new drug cocktails.

Such a sobering prognosis adds pressure to develop a second generation of protease inhibi tors and to follow patients more closely so failure can be detected as soon as possible. One new diagnostic tool that can help in that effort is the HIV resistance assay, which is presently used for research and could become commercially available in the next year. (See article on new assays on p. 113.) Another related set of tools is the more sensitive viral-load assays, one of which has been submitted to the Food and Drug Admini stration for commercial use. (See related article on p. 111.)

"Right now we are seeing with one eye: We are seeing with viral load," Mellors said. "We need to see with two eyes, and in my mind the second eye is drug susceptibility [testing]. So when in doubt, review the patient’s history and compliance with the regimen. One has to make a guess, and probably the best thing to do if someone is failing is to switch the entire regimen."

Relying on art of medicine

But even resistance testing, at least in the near future, will only be able to tell clinicians which drugs don’t work, not those that will work. Some of the panel members went so far as to recommend using novel combinations because, as Mellors put it, "our options are limited and we really don’t know what to do." Others urged arming patients with the most current knowledge about HIV pathogenesis so they are more likely to accept new treatment options.

"A big issue is general education. We have to educate about the natural history of disease," says Jeffrey Nadler, MD, director of clinical research at the University of South Florida College of Medicine in St. Petersburg. Nadler advocates treating HIV early, before the immune system is damaged, much like treating hypertension before a patient suffers a stroke. On the other hand, patients who are reluctant or distrusting of treatment should not be forced into it if they are not ready to make a commitment to complete therapy.

"We can make the recommendations, lay out the pros and cons," Nadler said. "Unfortunately, it’s a big investment in time, one that not only I as a physician make, but also our nurses and counselors and pharmacists, and we start therapy when the patient is ready."

But that investment will pay off, said Janet Mitchell, MD, MPH, chair of the department of obstetrics and gynecology at Interfaith Medical Center in Brooklyn. She referred to the high rate of HIV-positive pregnant women at Harlem Hospital who have agreed to treatment for preventing perinatal HIV transmission. "We had keeping people in care as our goal. Once you can keep them in care, you can monitor their disease, and over time they trust you and can see the benefits of pharmaceuticals even if they don’t in the beginning."

Indeed, the issue of compliance is so critical to avoiding drug failure that clinicians should offer drugs only if a patient is committed to taking them. A clinician should make a verbal contract with patients up front requiring the patient to inform the clinician when he or she is ready to make that commitment, Mellors said.

With patients who are at low risk of progression and unenthusiastic about therapy, deferring treatment is the wise choice — and one that statistically has a relatively good prognosis, he noted. A look-back study of gay men found that for those who had less than 5,000 copies of HIV RNA and CD4 counts above 500, the risk of progression is only less than 5% at three years, and 10% to 15% at nine years.

It is important to closely monitor even this group of patients with regular viral-load testing, Mellors added. "Regardless of whether we initiate or delay therapy, we must re-evaluate," he cautioned. "Just as we don’t take an individual blood pressure on one occasion and not check it again, we can’t do the same with HIV. We need to measure the viral load, and monitor it regardless of their decision."

Putting new guidelines into practice

Reacting to the new Public Health Service guidelines for HIV treatment, the panel discussed early vs. delayed treatment, and how to respond to a virologic breakthrough or a failing regimen. Underpinning the decision process is the recently validated recognition that the virus is driving disease progression, and that the greater the amount of viral replication, the faster the rate of CD4 decline and progression to disease. Conversely, Mellors noted, it is believed but not proven that a greater rate of virus destruction leads to greater rate of CD4 destruction, and consequently, an earlier burnout of the immune system.

Martin Markowitz, MD, staff investigator at the Aaron Diamond AIDS Research Center in New York, advocated aggressive early therapy based on patients treated at the institute shortly after they became infected. Analysis of those patients has revealed that for recent seroconverters, the pool size of latently infected cells in certain compartments is about one log, or 100 times, lower than in patients with chronic infection.

Also noteworthy are the center’s results on treating sicker patients, those with high viral loads and low CD4 counts. Treatment with three drugs often is not adequate to bring viral levels below detection, leading Markowitz and his colleagues to predict that four drugs may be necessary in this population to prevent resistance, similar to methods for treating multidrug-resistant tuberculosis.

One of the biggest issues facing clinicians is how to balance the need to avoid resistance with other clinical considerations. Mellors reported that a recent meta-analysis has shown that the degree of viral load suppression necessary to prevent disease progression is much less than that required to prevent resistance. In general, lowering virus to less than 5,000 copies can prevent clinical endpoints; but to avoid resistance, and thus have a durable response, a lower level is required, he said.

‘Undetectable’ vs. ‘suppressed’

How low? That’s an important question, and one that raises the problem of referring to undetectable virus and complete suppression as if they were synonymous. Undetectable virus means virus is undetectable using current assays. Complete suppression, on the other hand, is a biological phenomenon in which no virus is replicating. Most commercial assays are sensitive down to about 400 copies. A patient with 300 copies would be considered to have undetectable virus, and yet virus still would be present and replicating, Markowitz noted.

"You can keep patients at a level that allows less than complete suppression, and we know that the accumulation of mutations will take that viral load and make it climb," he said, adding that even with small amounts of replication, all regimens of currently approved drugs are doomed to failure.

At the same time, the meta-analysis on viral load and disease progression predicts that the failure would take years to result in disease, Saag noted. "If we can sustain somebody at 200 [copies] we could probably say, ‘Hallelujah, they have 10 years of disease-free time.’"

For patients with healthy immune systems and low viral loads, three-drug regimens are usually adequate to bring virus down to nearly complete suppression. With sicker, drug-experienced patients, the situation is less predictable, the panel noted. Echoing an increasingly familiar refrain, Markowitz said he advocates hitting hard rather than light. "If you don’t get what you want and then you start adding drugs, that is the worst thing to do — to spare the patient up front and then find yourself behind the eight ball."

Many clinicians don’t understand that the number of treatment options is more limited than expected, despite the fact that 11 antiretroviral drugs are now on the market and can be used in various combinations. Because many of the drugs have overlapping resistance patterns, one drug company researcher’s mathematical calculations estimate that, after initiating triple-drug therapy, a patient has only two other effective combinations to switch to.

In some cases, the best option for patients who are drug-experienced and not achieving undetectable levels may be to just buy time, said Nadler. Several new protease inhibitors are coming through the pipeline and may have different resistance profiles. The panel noted, however, that it will take not just one new drug, but rather a combination of two or three new drugs to have a significant impact on drug-experienced patients.