Survival From HIV-1 Seroconversion in Southern Africa: A Retrospective Cohort Study in Nearly 2,000 Gold-Miners Over 10 Years of Follow-Up
April 26, 2007
Viktor von Wyl and colleagues1 looked at the Swiss Cohort to examine the emergence of antiretroviral resistance after patients began their first antiretroviral regimen that included either a non-nucleoside reverse transcriptase inhibitor (NNRTI) or a ritonavir (RTV, Norvir)-boosted protease inhibitor (PI).
The patients who were analyzed started their first antiretroviral regimen between January 1999 and December 2005. The regimen had to have included at least two nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) and a boosted PI or two NRTIs and an NNRTI. Patients had to have one additional HIV viral load of less than 50 copies/mL after achieving an undetectable viral load or a viral load measurement after 180 days without viral suppression.
In patients who had previously had undetectable viral load levels, viral failure was defined as either two consecutive viral loads of more than 500 copies/mL or one viral load value greater than 500 copies/mL followed by a stop or change in antiretroviral regimen. In those without complete suppression of viral load to undetectable levels since initiating therapy, viral failure was defined as one viral load value greater than 500 copies/mL after 180 days of continuous treatment.
ViroSeq and vircoTYPE resistance testing were performed on samples. Major mutations were defined using the International AIDS Society-USA guidelines.2 Minor mutations were not considered. Genotypic susceptibility scores (GSS) using the Stanford algorithm were determined for the number of mutations present. A score of less than 15 was considered susceptible, 15 to 59 indicated intermediate susceptibility, and more than 59 indicated high resistance. Lamivudine (3TC, Epivir) or emtricitabine (FTC, Emtriva) were considered separately from other NRTIs.
There were 1,929 patients who initiated therapy. However, 1,323 patients were in this analysis (518 on boosted PIs and 805 on NNRTI-containing regimens) and the primary analysis was an on-treatment analysis. Of the excluded patients, many had stopped treatment because of toxicities.
In both groups, the mean age was 38 years old. About 70% of participants were male. The median CD4+ cell count was about 200 cells/mm3 and the viral load was around 5.0 log10/mL. There were 69 cases of virologic failure (24 boosted PI and 45 NNRTI), of which 58 had resistance test data available (84%). Fifty-eight percent of failures were with subtype B virus.
The emergence of resistance was not associated with subtype. Ritonavir-boosted PIs included lopinavir (LPV; 75.5%), indinavir (IDV, Crixivan; 10.8%) and atazanavir (ATV, Reyataz; 8.9%). Of those on an NNRTI regimen, 91.4% received efavirenz (EFV, Sustiva, Stocrin). In addition, 85% of patients were receiving lamivudine or emtricitabine. The crude failure rate was not significantly different between NNRTI and boosted-PI groups. The crude adverse event rate was higher in the boosted-PI group compared to the NNRTI group. The NNRTI group had significantly more mutations than the boosted-PI group and had more antiretroviral classes affected. Resistance tests performed at the time of failure revealed that patients were more likely to be susceptible to lamivudine or emtricitabine and other NRTIs in the boosted-PI arm when compared to the NNRTI arm.
In this study, PI-containing regimens appear to have equal potency and have somewhat more adverse events, but lead to less resistance to other antiretroviral drugs in the same regimen when virologic failure occurs.
The numbers were small and therefore it was difficult to determine whether there was any selective advantage among the boosted-PI regimens used. Studies have suggested that the threshold for resistance development (or genetic barrier) is much higher for ritonavir-boosted PI regimens when compared to NNRTI regimens. In other words, it takes more mutations to affect the activity of boosted PIs compared to NNRTIs, whereas it may only take a single mutation to wipe out NNRTIs. And, as seen in this study, if an NNRTI regimen failed, it was much more likely that resistance would also be seen in the NRTI class of medications as well.
03.12.2007; Vol. 21; No. 5: P. 625-632; Judith R. Glynn; Pam Sonnenberg; Gill Nelson; Andre Bester; Stuart Shearer; Jill Murray