Setting up of HIV-1 Antiretroviral resistance testing methods and their clinical applications.

Abstract

Following the World Health Organization (WHO)’s 3 by 5 initiative, antiretroviral treatment (ART) for HIV disease management has been scaled up rapidly in resource limited settings such as Zimbabwe. First line treatment options include a three antiretroviral (ARV) drug regimens compromising of two nucleoside reverse transcriptase inhibitors (NRTIs) and one non-nucleoside (NNRTI). There are fears that the use of SD NVP for PMTCT and rapid scale-up of ART will be followed by a rapid emergence of primary drug resistance in the population rendering the first line treatment options ineffective. Zimbabwe has instituted 90% coverage of SD NVP: ARVs became widely available in 2003 and currently about 200000 people are on treatment in both the

private and public sector. A cross-sectional study was conducted to determine the prevalence of primary drug resistance over a period of two years in a population where ART access is being rapidly scaled up.

This study was nested in a larger study on the HIV-1 Subtype C drug resistance and pathogenesis. Signed informed consents were obtained from eligible young pregnant (<25 years) women who were attending antenatal clinics in Chitungwiza,

Zimbabwe. Whole blood samples were collected in EDTA for CD4 counts, using Partec Cyflow, viral load using Roche HIV monitor Amplicor version 1.5, and estimation of period of infection using BED Calypte Assay as well as presence of HIV-1 drug resistance mutation using population sequencing. Sequence analysis for the prevalence of primary drug resistance was done using the Calibrated Population resistance software of the Stanford HIVDB and the Stanford Drug Resistance Mutations List. Sequences obtained in the analysis for drug resistance were also compared with other sequences from Southern Africa as well as other parts of the world where Subtype C is the dominant HIV-1 strain driving the epidemic.

Three hundred and three women with a mean age of 21 years (95% Confidence Interval, 21 to 22 years) were studied. The median CD4 count was 393 cells/uL, Interquartile range (IQR): 249-509 and among a subset of 107 women, median plasma

VL was 3.70 log10 copies/ml, IQR: 3.1-4.2. Thirty two percent (n=236) of the women were considered to be recently infected using the BED assay, (infected within 155 days of sample collection The median CD4 count in recently-infected women was significantly higher than in women with long term infections based on the BED results (p = 0.000). There were only two test specimens (0.85%) out of 236, with evidence of drug resistance. One specimen had the Y181C mutation associated with NNRTI resistance and a second had the I85V mutation associated with protease inhibitor (PI) resistance. There was no evidence of primary drug resistance in the recently infected women. No evidence of geographic clustering of sequences among subtype C Southern African sequences was

observed after phylogenetic analysis. Interestingly, subtype C sequences from India, Brazil, China and Ethiopia formed separate clusters distinct from the Southern Africa

sequences.

The lack of evidence of HIV-1 primary drug resistance in this population of young pregnant women suggests that the use of genotypic drug resistance data to select the most optimal treatment regimen may not yet be warranted. Currently there is no evidence of evolutionary compartmentalization of the HIV-1 Subtype C epidemic in Southern Africa. This may be indicative of the absence of selective pressure driving differentially adapted sub clusters.