Countries have meticulously designed and implemented antiretroviral treatment (ART) programs to control the human immunodeficiency virus (HIV) epidemic and contain disease progression into acquired immunodeficiency syndrome (AIDS). These ART programmes, particularly in resource-limited settings, are characterized by using standardized ART regimens. The significance of our research lies in its potential to enhance the effectiveness and sustainability of these ART programmes by monitoring and minimizing the further spread of HIV drug resistance (HIVDR). Our findings can significantly impact the management of HIVDR, which can affect the effectiveness of ART regimens and be a source of HIVDR transmission.
In many LMICs, the lack of HIVDR testing at treatment initiation or first-line regimen failure due to cost and capacity constraints is a significant challenge. These reasons underscore the ongoing need for continued and regular surveillance of drug resistance, which is crucial for the effective management of ART programmes. Nationally representative surveillance of HIVDR is essential for assessing the quality of ART programmes and guiding the selection of appropriate ART regimens.
Since 2019, the NICD and NHLS have been conducting nationally representative HIV drug resistance surveys. These surveys utilize remnant HIV viral load (VL) samples collected from public health facilities as part of routine programme monitoring, with drug-level testing serving as a proxy for treatment exposure.
The laboratory-based HIVDR surveys used the same sampling strategy among clients with VLs >1,000 copies/mL and were conducted in 2019, 2021, 2022, and 2023. Integrase inhibitor drug level and resistance testing was only performed from 2021 onwards.
In all HIVDR surveys, any drug level was detected in less than 60% of the specimens with a VL >1,000 copies/mL, with a significant drop to only 37% in 2023. This finding indicates that treatment failure is often due to sub-optimal treatment adherence. The proportion of samples with detectable protease inhibitor levels remained stable (4%-7%). As per changing treatment guidelines, the proportion of specimens with detectable efavirenz levels decreased from 43% in 2019 to 12% in 2023. In contrast, the proportion of specimens with detectable dolutegravir increased from 7% in 2021 to 18% in 2023.
The prevalence of any drug resistance, regardless of detectable drug levels, has shown a promising decline from 72% in 2019 to 54% in 2023. This decline and the similar trend in the proportion of NNRTI resistance in the survey population (71% in 2019 to 51% in 2023) bring a sense of optimism. The prevalence of PI resistance has remained stable, ranging from 2.2% to 4.1%, and INSTI resistance, although low, has steadily increased from 0.2% in 2021 to 2.3% in 2023.
Combining drug level results and resistance results show that the proportion of specimens with detectable NNRTI levels and NNRTI resistance remained above 84% from 2019 to 2023. PI resistance was found in about a third of samples with detectable PI levels, except for the 2021 survey, when we only detected PI resistance in 17.2% of samples with detectable PI levels.
The prevalence of INSTI resistance among samples with confirmed INSTI exposure increased from 2.7% in 2021 to 11.1% in 2022. However, no further increase was observed in 2023 (10.5%).
The observed HIVDR levels in this survey are similar to those observed before the roll-out of DTG; however, the overall prevalence of resistance appears to be declining in recent years. This decline is likely driven by less frequent NRTI and NNRTI resistance observations. The implications of this decline are significant, as it suggests that the current treatment strategies are effective in reducing drug resistance. However, a substantial proportion of patients with detectable drug levels remain positive for NNRTIs, indicating that the DTG roll-out was not fully implemented by May-June 2023, despite treatment guidelines recommending the unconditional switch to TLD for most patients by May 2022.
The prevalence of PI and INSTI resistance remains low, which aligns with the high genetic barrier of LPV/r and DTG and the recent introduction of DTG at a large scale. However, the confirmation of INSTI resistance levels of around 10% in the 2022 and 2023 surveys underscores the need for continued monitoring. Vigilance is crucial to tracking the development of INSTI resistance in patients with detectable DTG levels.
The sub-analysis of HIVDR resistance relative to the presence or absence of PIs or INSTIs indicates that screening tests for PIs and INSTIs drug levels could be used to triage specimens for HIVDR testing.
Despite the national representativeness of the survey, results should be interpreted cautiously, given the limitations of obtaining accurate treatment regimen information. In addition, all sub-analyses should be interpreted cautiously as the study was not powered to compare results among different age groups or by province. Also, because viral suppression may be higher amongst patients receiving DTG-based regimens, over-sampling of patients taking NNRTI-based regimens may have occurred. Should drug level testing prove to be an adequate measure as a proxy for ART exposure, the increase of patients without any detectable drug levels should be investigated, and investments could be made to improve adherence interventions; regular surveillance efforts are essential to continuously monitor the possible development of DTG resistance in the population. Detailed reports of all surveys are available on the NICD website.
HIV Drug Resistance Surveys :
The study team acknowledges the support of all the NHLS viral load laboratories who allowed us access to remnant specimens.
We would like to thank PEPFAR and CDC for the financial support as well as technical assistance.
