Genotypic and Virtual Phenotypic Correlates of Virologic Response to Abacavir-Based Therapy

  • Genotypic (GT) and Virtual Phenotypic (vPT) Correlates of Virologic Response to Abacavir (ABC)-Based Therapy in the ZORRO Trial (ESS40009) (ThOrB1386)
    Authored by S. Hessenthaler, P. Ruane, G. Richmond, L. Yau, E.R. Lanier, J. Hernandez, A. Williams, T. Fralich
    View the original abstract

Dr. Peter Ruane from the Tower Medical Group in Los Angeles presented data on abacavir resistance from the ZORRO Trial. The trial was a community-based prospective study to assess the efficacy of adding abacavir to an antiretroviral regimen in previously HIV-drug-experienced patients. A sub-study of the trial presented here, attempted to characterize which mutations and how many were necessary to reduce the activity of an abacavir-containing regimen in previously NRTI-treated patients. The study measured viral load response at eight and 16 weeks.

Baseline HIV resistance genotype and virtual phenotype were used to guide therapy, assure the absence of abacavir resistance and predict treatment response to abacavir. A total of 180 patients were enrolled and had the following baseline data: a mean age of 40 years, 85 percent were male, 55 percent were Caucasian and 28 percent were African-American; the mean viral load was 4.3 log10/ml and the mean CD4+ count was 315/mm3.

Of the 180 patients, 145 completed 16 weeks of treatment. The baseline genotype revealed that 44 percent of patients had evidence of 3TC resistance and between 20-30 percent of patients had evidence of AZT or thymidine analogue mutations (TAMs). TAMs have historically been reported to cause AZT resistance.

There are a total of six such TAM mutations and the more mutations present, the higher the level of AZT resistance. We now know that these mutations also cause d4T resistance, and can impact the activity of abacavir and tenofovir (see below). In an intent-to-treat (ITT) analysis (this means using all patients enrolled in the study, whether they completed the study or not, in the final data analysis; this is thought to be a more rigorous way to analyze data than just using the data from patients who complete a study), it was found that patients were more likely to respond (a viral load of less than 400/ml) to the abacavir-containing regimen if they had no TAMs. Even if they had up to three TAMs, they still responded with at least a one log/ml (ten-fold) decline in viral load. At least four TAMs at baseline were necessary in order to see a lack of response to abacavir. This was also correlated with the virtual phenotype, in which more than three mutations were required to see a three- to four-fold decrease in susceptibility to abacavir. Additional predictors of response to abacavir in this study were: being of younger age, being male, having a higher CD4+ count and a fewer number of TAMs. It was unclear why being female would account for poorer responses to abacavir.

This study highlights a few important points. First it confirms previous data that demonstrated patients would respond to abacavir, despite the presence of important TAM mutations. A certain number and specific type of mutation set are required to inactivate abacavir. Thus, practitioners need to know what the particular pattern of mutations seen on a genotype report mean in order to accurately determine or predict whether abacavir will or will not have activity in a patient's next regimen. In addition, these results add to emerging data, which are helping to define what the increase in drug concentration needs to be (FC, fold-change) in order for a phenotype or virtual phenotype resistance assay to call a virus resistant to a drug. In this study and in others, a three- to four-fold increase in abacavir concentration is required to inhibit an abacavir-resistant virus compared to a virus with no mutations. This information helps practitioners better interpret phenotype resistance results.