Characterization of HIV drug resistance mutations among patients failing first-line antiretroviral therapy from a tertiary referral center in Lusaka, Zambia

Abstract

In settings of resource constraint, an understanding of HIV drug resistance can guide antiretroviral therapy (ART) at switch to second-line therapy. To determine the prevalence of such HIV drug resistance mutations (HIV DRM), we used an in-house sequencing assay in the pol gene (protease and partial reverse transcriptase) in a cohort of patients suspected of failing a first-line regimen, which in Zambia comprises two nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) and one non-nucleoside reverse transcriptase inhibitor (NNRTI). Our analysis cohort (n = 68) was referred to the University Teaching Hospital in Lusaka from November 2009 to October 2012. Median duration on first-line ART to suspected treatment failure was 3.2 years (IQR 1.7-4.7 years). The majority of patients (95%) harbored HIV-1 subtype C virus. Analysis of reverse transcriptase revealed M184V (88%), K103N/S (32%), and Y181C/I/V (41%) DRMs, with the latter conferring reduced susceptibility to the salvage therapy candidates etravirine and rilpivirine. Three patients (5%) had major protease inhibitor (PI) resistance mutations: all three had the V82A mutation, and one patient (Clade J virus) had a concurrent M46I, Q58E, and L76V DRM. HIV-1 genotyping revealed major and minor DRMs as well as high levels of polymorphisms in subtype C isolates from patients failing first-line antiretroviral therapy. Closer monitoring of DRM mutations at first-line failure can inform clinicians about future options for salvage therapy.

Keywords: HIV drug resistance; NNRTI; Zambia; diversity; pol.

Fig. 1

Sequences were uploaded onto the Stanford Drug Database program (http://sierra2.stanford.edu/sierra/servlet/JSierra), and both Drug Resistance Mutations (DRM) and polymorphisms were analyzed according to both the drug resistance profile on the curated website as well as the 2013 IAS guidelines. (A) Assessment of the overall prevalence of first line therapy patients with suspected treatment failure. (B) Prevalence of drug resistance mutations according to first line regimens prescribed.

Fig. 2

(A) All NRTI DRMs. (B) NNRTI-Efavirenz and Nevaripine DRMs. (C) NNRTI-Rilpivirine DRMs. (D) NNRTIEtravirine DRMs. Alphanumerical symbols refer to the amino acid position of the drug resistance mutation. The first letter represents the wild type residue, followed by the amino acid position, followed by the major/minor drug resistance mutation. In parentheses are polymorphisms at this genetic locus.

Fig. 3

The protease sequences obtained were aligned using the ClustalX multiple sequence alignment tool using the Clustal method. Phylogenetic trees were generated by using the neighbor-joining method with the PHYLIP software package (version 3.52c; Joseph Felsenstein, University of Washington). Sequences representative of HIV-1 clades were obtained from Los Alamos National Laboratories (www.LANL.gov) andrepresent global group M HIV-1 subtypes. Brackets indicate the clustering of clades, and branch numbers represent bootstrap values >75. Subtype nomenclature are denoted as the following: A.BC.xx, where “A” represents the HIV clade, “BC” represents the two letter country abbreviation, and “xx” represents the first two numbers of the sequence accession numbers on LANL. Country codes are represented as Ghana (GH), Nigeria (NG), Kenya (KE), Tanzania (TZ), Uganda (UG), Rwanda (RW), South Africa (ZA), Cameroon (CM), Democratic Republic of Congo (CD), Senegal (SN), Botswana (BW), Zambia (ZM), Malawi (MW), Ethiopia (ET), Somalia (SO), Central African Republic (CF), China (CN), Canada (CA), United States of America (US), Brazil (BR), Argentina (AR), Spain (ES), Cyprus (CY), Georgia (GE), India (IN), and Israel (IL). Patients within the study cohort are denoted as numerical values from 1 to 68.