Resistance test welcomed, with caution

Reliability of tests still a year or two away

Hailed as a breakthrough in HIV treatment, an HIV drug resistance test has been launched worldwide, sending out ripples of confusion as well as excitement among clinicians and patients. AIDS researchers caution, however, that resistance testing is in its infancy and has only a limited clinical usefulness at this time.

The value of a reliable HIV resistance test has grown considerably in the past few months as increased reports of drug failures make the headlines. In Atlanta, for example, as many as one-fourth of patients, many of them healthy at the time of treatment, have reported virologic breakthrough.

"We are seeing about a 25% to 30% failure rate, and that is after one year of therapy," says Anthony Braswell, MBA, MHA, executive director of AID Atlanta. "That is alarming because what a lot of places have done is suspend or cut funding to support services and are pumping it into drugs. Now those people are saying they need support services, and that safety net isn’t as strong as it was."

Defining drug failure

Most researchers define drug failure as a "significant" rise in viral load, with significant meaning a threefold increase. Or measured another way, this would mean a patient who has consistently undetectable virus that becomes detectable after two or more tests would be failing his regimen.

The exact reasons for drug failure are hard to discern. Noncompliance is a major factor, and one researcher at the resistance meeting reported findings from a military cohort that as many as 40% of drug failures were related to incomplete therapy. At the same time, researchers say the ability of the virus to mutate and escape drug pressure is probably accounting for more of the failures than they would like.

The first commercially available rapid HIV resistance assay is manufactured by Murex Technologies Corp. in Atlanta. Called LiPA HIV-1 RT test, it is only available for research use in the United States. The test uses multiple probe DNA technology to identify HIV resistance associated with the reverse transcriptase drugs AZT, ddI, ddC, and 3TC. The test allows HIV wild-type virus (drug-susceptible virus) and mutations — also known as polymorphisms — to be simultaneously detected at different codons of the reverse transcriptase enzyme. (A codon is a small sequence of DNA that codes for an amino acid.)

"What it does, for the first time, is it gives us a mechanism to determine whether patients are sensitive or resistant to a drug," said Jonathan Weber, MD, a professor of medicine at Imperial College in London, in a press release. "This will add another critical element for clinical decision making, and will inform people with HIV infection and their physicians on when and what drugs to change to in the event of drug failure."

Other researchers familiar with the test, however, caution that the use of genotypic resistance testing has limitations, and that patients have overblown its potential for steering them clear of resistance quagmires. They caution that HIV-resistance testing is about as far along as viral-load testing was three years ago.

Phenotypic or genotypic testing?

"These techniques will still remain experimental, even though they will be applied in clinical practice," says Brendan Larder, PhD, director of the clinical virology unit at Glaxo Wellcome. "We will need both values — genotypic and phenotypic — and put them together to say in the future what this means."

Indeed, researchers at Roche Molecular Systems are developing an HIV resistance test that, unlike Murex, is based on phenotypic instead of genotypic resistance, says John Sninsky, PhD, the company’s senior research director.

"Many people have questioned whether we really need resistance assays," he says. "We felt the question is, not do we need them but what kind should we have. . . . We don’t believe that genotyping is the way to go."

Larder explains that genotypic resistance testing has several advantages over phenotypic resistance testing. It is less expansive and more rapid because genetic analysis can be made without growing the virus, whereas phenotypic resistance testing generally requires growing virus. On the downside, genotypic tests are harder to interpret because the assay is looking only at a segment of the complete genetic sequence. He likened the resistance profile analysis to guessing at a phrase when knowing only a few letters.

With more than 180 mutations in the viral enzymes reverse transcriptase and protease, the development of resistance to one or several drugs is complex. More than one mutation may be required to develop resistance. Now that a second generation of antiretroviral drugs has become available, the relationship between genotypic and phenotypic resistance has become more fuzzy, Larder noted. Moreover, an individual who is experiencing virologic failure may not be failing all the drugs in the regimen.

Another limitation of the present assays is that they can identify resistance only on drugs one currently is taking. For example, a patient who was taking AZT for five years and then went off it for two years won’t be able to know whether his or her virus is resistant to AZT until the patient starts taking that drug again for at least several weeks.

"We don’t know the value of these assays in predicting the initial response, its magnitude or durability to therapy, or whether it’s initial therapy or salvage therapy," said John Mellors, MD, associate professor of medicine at the University of Pittsburgh School of Medicine. "So the real value of these tests in the long term may be really to distinguish virologic resistance from non-virologic drug failure."

Even with that important benefit, resistance testing may have to take a back seat to other funding priorities, Braswell says. "It is information that is good to know, but someone, especially at the federal level, is going to have to sit back and say, ‘This is the most information we can afford for you to know.’"