Patterns of HIV-1 Drug Resistance Observed Through Geospatial Analysis of Routine Diagnostic Testing in KwaZulu-Natal, South Africa

Abstract

HIV-1 drug resistance (HIVDR) impedes treatment and control of HIV-1, especially in high-prevalence settings such as KwaZulu-Natal (KZN) province, South Africa. This study merged routine HIV-1 genotypic resistance test (GRT) data with Geographic Information Systems coordinates to assess patterns and geographic distribution of HIVDR in KZN, among ART-experienced adults with virological failure. We curated 3133 GRT records generated between 1 January 2018 and 30 June 2022, which includes the early phase of dolutegravir (DTG) rollout, of which 2735 (87.30%) had HIVDR. Of the 2735, major protease, nucleoside, and non-nucleoside reverse transcriptase inhibitor mutations were detected in 41.24%, 84.97% and 88.08% of GRTs, respectively. Additional genotyping of HIV-1 integrase for 41/3133 (1.31%) GRTs showed that 17/41 (41.46%) had integrase strand transfer inhibitor resistance. Notably, of 26 patients on DTG with integrase genotyping, 9 (34.62%) had DTG-associated resistance mutations. Dual- or triple-class resistance was observed in four of every five GRTs. The odds of HIVDR increased significantly with age, with ≥60 years having 5 times higher odds of HIVDR compared to 18-29 years (p = 0.001). We identified geospatial differences in the burden of HIVDR, providing proof of concept that this could be used for data-driven public health decision making. Ongoing real-time HIVDR surveillance is essential for evaluating the outcomes of the updated South African HIV treatment programme.

Keywords: HIV-1; KwaZulu-Natal; South Africa; dolutegravir; drug resistance; genotypic susceptibility scores; geospatial analysis.

Figure 1

Public-sector healthcare facilities with HIV-1 genotypic resistance tests included by district in KwaZulu-Natal province, South Africa. Individual cross symbols pinpoint the Geographic Information Systems (GIS) coordinates of each healthcare facility that had requested a genotypic resistance test (GRT), the data of which had been included in the study. The asterisk denotes healthcare facilities that requested HIV-1 integrase testing. The cross symbols are coloured in pink, blue or orange based on whether the healthcare facility’s GIS coordinates are within a rural, peri-urban or urban subdistrict, respectively. The thin and thick black outlines represent the borders of the subdistricts and districts, respectively. Each district is illustrated in a different colour. The basemap of KwaZulu-Natal province was republished under a CC BY license with permission obtained from Carto Builder user Lilishia Gounder, original copyright 2024. Available at: https://pinea.app.carto.com/map/4d4c56c1-f82d-4409-b190-ea9ced309005 (accessed on 18 October 2024).

Figure 2

Flow diagram of HIV-1 genotypic resistance test records obtained and included in the final analysis. CDW, central data warehouse; INSTI, integrase strand transfer inhibitor; NHLS, National Health Laboratory Service.

Figure 3

Patterns of antiretroviral drug class resistance observed in 2735 genotypes with HIVDR obtained from KwaZulu-Natal province, South Africa. Please note that 41 genotypes included HIV-1 integrase testing, of which only 9 met the definition for 4-drug class resistance.

Figure 4

Specific mutations detected in 2735 genotypes with HIVDR obtained from KwaZulu-Natal province, South Africa. Mutations shown on the horizontal axis include “major” mutations observed in >6% of the genotypes with HIV-1 drug resistance, “major” as defined by Stanford HIV Drug Resistance Database or 2022 edition IAS–USA drug resistance mutations list [32,35]. HIVDR, HIV-1 drug resistance; INSTI, integrase strand transfer inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.

Figure 5

Antiretroviral drug susceptibility levels across KwaZulu-Natal province, South Africa. Districts and subdistricts are categorized as follows: (A) level of urbanization. Inverse distance weighted interpolation maps cumulatively reflect the drug susceptibilities for the following: (B) TDF, tenofovir; (C) DTG, dolutegravir; (D) LPV/r, lopinavir with boosted ritonavir; (E) EFV, efavirenz; (F) ETR, etravirine. Spectral colour change from blue to red reflects the drug susceptibility level as follows: S, susceptible; PLLR, potential low-level resistance; LLR, low-level resistance; IR, intermediate resistance; HLR, high-level resistance. The thin and thick black outlines represent the borders of the 44 subdistricts and 11 districts of KwaZulu-Natal (KZN) province, respectively. The basemap of KZN province was republished under a CC BY license with permission obtained from Carto Builder user Lilishia Gounder, original copyright 2024. Available at: https://pinea.app.carto.com/map/4d4c56c1-f82d-4409-b190-ea9ced309005 (accessed on 18 October 2024). * Please note that HIV-1 integrase testing was performed for 41 genotypes; the remaining 3092 genotypes that did not have integrase test requests were assumed to be susceptible to DTG for the purposes of creating the DTG interpolation map.