Prevalence of HIV type-1 drug-associated mutations in pre-therapy patients in the Free State, South Africa

Background We aimed to characterize the molecular epidemiology of HIV type-1 (HIV-1) and the prevalence of drug- associated mutations prior to initiating highly active antiretroviral therapy (HAART) in the Free State province, South Africa. The Free State has a population of 3 million, an antenatal HIV prevalence of approximately 34% and a well established infrastucture for antiretroviral (ARV) provision. Methods HIV-1 polymerase genes were sequenced from 425 HAART-naive HIV-1-positive patients at voluntary primary healthcare HIV testing centres, who were subsequently attending district centres for assessment for commencing ARVs. Patients (>18 years) were sampled randomly with no exclusion for gender or clinical criteria. Sequences were analysed according to phylogeny and drug resistance. Results Phylogenetic clustering within the cohort was suggestive of multiple introductions of subtype C virus into the region. Drug resistance mutations (according to the International AIDS Society-USA classification) were distributed randomly across the cohort phylogeny with an overall prevalence of 2.3% in the sampled patients. When stratified according to CD4+ T-cell count, the prevalence of resistance was 3.6%, 0.9% and 1.2% for CD4+ T-cell counts <100, 200-350 and >500 cells/μ l, respectively, and was most common for non-nucleoside reverse transcriptase inhibitor resistance (3.1% in patients with CD4+ T-cell count <100 cells/μ l). We surveyed all drug-selected mutations and found further significant clustering among patients with low CD4+ T-cell counts ( P=0.003), suggesting unrecognized exposure to ARVs. Conclusions In the Free State population, there was a statistical association between low CD4+ T-cell counts and drug-associated viral polymorphisms. Our data advocate the benefit of detailed history taking from patients starting HAART at low CD4+ T-cell counts with close follow-up of the virological response.