An analysis of 131 nucleoside analogue-experienced individuals followed for up to three years in the US study ACTG 364 has shown that individuals with higher baseline phenotypic susceptibility scores as calculated after using the Virologic Phenosense assay had a significantly lower likelihood of treatment failure during the follow-up period.
The study also found that hypersusceptibility to efavirenz was associated with a reduced risk of virologic failure, albeit with wide confidence intervals (hazard ratio 0.43, 95% CI 0.18-0.97).
Although the study was not designed to test the success of treatment regimens chosen on the basis of phenotypic susceptibility testing, one of the study’s authors, Dr Nicholas Hellman of Virologic, said in a press release issued today by the company that “This study reaffirms the predictive value of phenotypic tests.”
The findings are reported in the April 11 edition of AIDS.
Study design
ACTG 364 recruited 131 nucleoside reverse transcriptase inhibitor (NRTI-) experienced, protease inhibitor-naive patients with four years of mono or dual NRTI experience in ACTG 175 and 48 weeks of subsequent NRTI treatment in ACTG 302/303.
Participants in ACTG 364 received either d4T/ddI, ddI/3TC or d4T/3TC selected according to previous treatment experience, and were randomised to receive either nelfinavir or efavirenz. Thus ACTG 364 was not a randomised comparison of nucleoside analogue backbones, nor a comparison of drug selection by treatment history or phenotypic resistance testing. 75% of participants received two new NRTIs, whilst the remainder received one new NRTI and one recycled drug.
Phenotypic susceptibility was assessed in 131 of 156 individuals with baseline viral load above 2,000 copies/ml, and long-term treatment response was analysed in these individuals at 16, 24 48, 96 and 144 weeks. This group of patients had a median CD4 cell count of 336 cells/mm3 and median viral load of 4.17 log10 copies/ml.
A phenotypic susceptibility score (PSS) was calculated in two ways: as an either/or score based on whether a sample showed a reduction in susceptibility below the cut-off point for a particular drug (dichotomous), and as a continuous score that also reflected the degree of reduction in susceptibility above the cut-off point.
Using a continuous PSS was found to yield a highly significant result, with every 1 unit increase in PSS resulting in a 46% reduction in the risk of virologic failure (95% CI 0.38-0.76) (p<0.001). A PSS of 3 or more was associated with the lowest risk of virologic failure at all time points after week 16 (although the number of patients still on the assigned treatment fell from 93 at week 24 to 43 at week 144).
However, the quantity of new drugs adopted at the beginning of ACTG 364 appeared to have a stronger effect by multivariate analysis. Each new drug resulted in a 66% reduction in the risk of virologic failure (95% CI (0.22-0.50, p<0.001).
On the question of whether phenotypic susceptibility testing and the calculation of phenotypic susceptibility testing is ready for widespread adoption, the authors of the study are more restrained than Virologic. They say that “data from additional clinical trials will be needed to translate these results into a definitive algorithm.”
The study also observed that efavirenz hypersusceptibility, a correlate of AZT resistance in this population, was associated with a reduced risk of virologic failure. These results confirm findings from previous, smaller studies.
Katzenstein DA et al. Phenotypic susceptibility and virological outcome in nucleoside-experienced patients receiving three or four antiretroviral drugs. AIDS 821-830, 2003.