Resistance Testing: A Look at the Data

Fall 2001

This article is part of The Body PRO's archive. Because it contains information that may no longer be accurate, this article should only be considered a historical document.

Resistance testing is fast becoming part of the standard of care for treating people living with HIV. It is yet another tool, along with viral load and CD4+ cell counts, to optimize the selection of anti-HIV therapies. Several past studies have shown that people using resistance tests to choose a regimen were more likely to achieve undetectable viral loads and have prolonged anti-HIV responses compared to people who chose a regimen simply based on viral load, CD4+ counts and their treatment history. Other studies have not shown a better result for people who used resistance tests. Now, several more recent studies show the complexities in interpreting the results of these tests and make it clear that some fine-tuning in interpretation is still required.

Genotypic Resistance Testing

Results from three separate studies illustrate the benefits of genotypic testing. The French Viradapt study followed 108 people, all of whom had previously taken anti-HIV therapies and had viral loads over 10,000 copies. Half of the participants received the best available therapy based on their previous treatment history, while the other half received the best available therapy based on results from a genotypic test. Neither group knew how treatment decisions were made for them. At the end of six months, those whose treatment decisions were based on the genotypic test experienced a more significant drop in viral load and were more likely to achieve undetectable viral loads compared to those who didn't get a resistance test (32% versus 14%).

The GART study, which was conducted in the United States, had very similar results. This study included 153 people, all of whom had experienced a return of measurable viral load while on a three drug regimen that included a protease inhibitor. Half of the participants chose a new regimen based on a genotypic test with expert interpretation. The other half selected a new regimen based on previous anti-HIV therapy history. After 12 weeks of the study, there was again a significant difference in response, favoring the group receiving the resistance test with expert interpretation.

The third genotypic testing study was conducted in Spain and known as the Havana Trial. This study was slightly different from Viradapt and GART. It included 274 participants, all of whom had been on prior anti-HIV therapies. Although participants were randomly assigned to receive either genotypic testing or no testing, they were also randomly assigned to receive either expert advice or no advice in helping to choose a new regimen. In other words, some of the participants received both resistance testing and expert advice, some received neither, and the remainder received only resistance testing or expert advice. The expert advice was given by a group of leading virologists and clinicians. At the end of the six month study, 58% of people receiving genotypic test information had undetectable viral loads compared to 42% of those who did not. People who received expert advice were also more likely to have undetectable viral loads compared to people who didn't get expert advice (59% versus 41%). Not surprisingly the group that received both genotypic testing and expert advice had the best overall response, with 69% having undetectable viral loads compared to 36% of those who received neither intervention. This study shows the importance of expert advice and its added benefit to resistance testing.


Phenotypic Resistance Testing

Results from a study known as VIRA3001 and conducted by Virco, developers of one of the phenotypic resistance tests, show the benefit of phenotypic testing. This study followed 273 people, all of whom had taken a three drug regimen containing a protease inhibitor and had viral loads above 2,000 copies. Almost all of the participants had taken Epivir (3TC) and AZT, and the majority had also taken Zerit (d4T). Additionally most of the participants had taken either Viracept (nelfinavir) or Crixivan (indinavir). Half of the participants made treatment decisions based on information from a phenotypic test while the other half made decisions based on treatment history. After 16 weeks of the study, people receiving the resistance test were more likely to have undetectable viral loads (45% versus 34%) and a more pronounced drop in viral load compared to those who didn't receive the test.

The California Collaborative Treatment Group's CCTG 575 study compared phenotypic testing to no resistance testing. 238 people, all of whom had taken anti-HIV therapy for an average of about three years, participated in the study. Most of the participants had taken one protease inhibitor-based regimen (mostly either Viracept or Crixivan), and most had not previously taken a non-nucleoside reverse transcriptase inhibitor. This study found no difference in response rates between the two strategies -- after one year, about 50% of the participants achieved viral loads below 400 copies and about 40% achieved below 50 copies. There may be several reasons for the lack of a difference seen in this study. Participants were generally not heavily pretreated and so had a lot of treatment options. Perhaps more importantly, the criteria used to determine resistance by the phenotypic test might have been flawed. As a result, participants who received the resistance test were far more likely to choose Zerit (d4T) and/or Videx (ddI) as part of their regimen because the test results showed that they were sensitive to these two drugs and that most were resistant to Ziagen (abacavir), whereas people without the resistance information were far more likely to choose Ziagen.

Since the start of this study, the field has rapidly evolved. We now know that there is significantly more cross-resistance (where resistance to one drug results in resistance to another) between AZT and d4T. Similarly, a smaller decrease in sensitivity to ddI than previously thought may result in the drug losing effectiveness, whereas the opposite may be true for Ziagen. As a result of this study, the criteria to determine resistance to ddI, d4T and Ziagen have been changed. Another study conducted in France, known as NARVAL, had similar findings and participants made similar treatment choices based on phenotypic resistance information.

Results from the CCTG study highlight the importance of establishing clinically relevant "cut offs" for the phenotypic tests to determine whether a particular drug will still be active against HIV (no resistance to the drug), whether it may have some moderate activity against HIV (low level resistance) or whether it isn't expected to have any activity against the virus (high level resistance). To date, these clinically relevant "cut offs" have only been established for two drugs, Ziagen and Kaletra (lopinavir), although it is likely that others, including ddI, d4T and tenofovir, may be established before the end of the year. To further add confusion to the matter, these "cut offs" are likely to be different for the two phenotypic resistance tests that are currently available because Virco's Antivirogram and ViroLogic's PhenoSense use different technologies to measure phenotypic resistance.

Ben Cheng is Director of Antiviral Advocacy at Project Inform in San Francisco.

This article was provided by AIDS Community Research Initiative of America. It is a part of the publication CRIA Update. Visit ACRIA's website to find out more about their activities, publications and services.


The content on this page is free of advertiser influence and was produced by our editorial team. See our content and advertising policies.