The prevalence of pre‐treatment and acquired HIV drug resistance in Vietnam: a nationally representative survey, 2017–2018

Abstract Introduction Monitoring the population‐level emergence and transmission of HIV drug resistance (HIVDR) is necessary for supporting public health programmes. This study provides a nationally representative prevalence estimate of HIVDR in people initiating antiretroviral therapy (ART) and estimates of acquired HIVDR and viral load (VL) suppression in people who have received it for 12 or ≥48 months in Vietnam. Methods The study was conducted between September 2017 and March 2018 following World Health Organization guidance. Thirty ART clinics were randomly sampled using probability proportional to size sampling from a total of 367 ART clinics in the country. Results and Discussion In total, 409 patients initiating ART were enrolled into the survey of pre‐treatment HIVDR. The prevalence of any pre‐treatment HIVDR was 5.8% (95% CI 3.4–9.5%), and the prevalence of non‐nucleoside reverse transcriptase inhibitor resistance was 3.4% (95% CI 1.8–6.2%). Four hundred twenty‐nine patients on ART for 12±3 months and 723 patients on ART for ≥48 months were enrolled into the surveys of acquired HIVDR. The prevalence of VL suppression (defined as <1000 copies/ml) in patients on ART for 12±3 and ≥48 months was 95.5% (95% CI 91.3–97.8%) and 96.1% (95% CI 93.2–97.8%), respectively. Among individuals with viral non‐suppression, any HIVDR was detected in 11/14 (weighted prevalence 74.3%) of those on ART for 12±3 months and in 24/27 (weighted prevalence 88.5%) of those receiving ART for ≥48 months. Conclusions This nationally representative study of HIVDR found high levels of VL suppression among those on ART for 12 and ≥48 months. Overall, high levels of VL suppression at both time points suggested good adherence among patients receiving ART and quality of treatment services in Vietnam. Clinical Trial Number Not applicable


I N T R O D U C T I O N
Antiretroviral therapy (ART) was first introduced in Vietnam in 1996 and is provided free of charge. As of December 2020, there was an estimated 215,000 people living with HIV in Vietnam and about 70% of them were receiving ART [1]. The first nationally representative survey of acquired HIV drug resistance (ADR) in Vietnam was conducted in 2014 among 365 individuals receiving ART for at least 12 months using a cross-sectional design adopted from contemporary World Health Organiation (WHO) guidance [2,3]. The prevalence of viral load (VL) suppression (defined as VL <1000 copies/ml) was 95.1% (95% CI 92.3-96.9%), and the prevalence of drug resistance to any antiretroviral (ARV) drug was 4.6% (95% CI 2.8-7.5) [2]. In 2019, WHO reported HIV drug resistance (HIVDR) data from 44 nationally representative HIVDR surveys from 24 low-and middle-income countries, including Vietnam. In this work, we report prevalence estimates of pre-treatment HIV drug resistance (PDR) among people starting ART and the prevalence of VL suppression and ADR among those taking ART.

M E T H O D S 2.1 Study design and sampling
We conducted a nationally representative cross-sectional study of PDR and ADR in adults receiving ART for 12±3 months (ADR12) and ≥48 months (ADR48+). At the end of 2016, Vietnam had a total of 367 ART outpatient clinics (OPCs) providing ART to 105,938 adults. Of the 263 OPCs in operation for ≥48 months (representing 92.4% of individuals on ART), 25 were sampled to contribute to the PDR, ADR12 and ADR48+ month surveys. Of the remaining 104 OPCs in operation for <48 months, five OPCs were sampled to contribute to the PDR and ADR12 months study. This sampling approach maximized overlap in sites contributing to each of the three time points [4]. Adults (≥18 years of age) presenting to the sampled clinics were consecutively screened for eligibility and enrolled until the pre-determined target sample size for each study time point at each site was achieved.

Study population
The PDR survey included people living with HIV (PLHV) presenting to sampled clinics for ART initiation for the first time or re-initiation of ART if they were on it for less than 1 month and had stopped it for more than 3 months. For ADR surveys, PLHV who have been receiving ART for 12 (±3) months (ADR12) or for ≥48 months were enrolled. All participants in PDR and ADR surveys were 18 years of age or older and provided written informed consent.

Sample size
Sample size calculation followed procedures described by WHO for two-stage cluster sampling for combined PDR and ADR surveys [3,5]. The overall sample size was 405 patients for the PDR survey and was 435 and 725 patients for the ADR12 and ADR48+ surveys, respectively.

Survey procedure
All participating OPCs consecutively enrolled eligible patients until reaching the target site-specific sample size or passing 6 months of the study start date, whichever came first. Whole blood specimens were collected for plasma VL measurement in both PDR and ADR survey populations. Collecting and processing of specimens followed WHO laboratory guidance [6]. Specimens with VL ≥1000 copies/ml were genotyped. Specimens were tested for HIVDR at one of two laboratories designated by WHO for the purpose of HIVDR surveillance. HIV-1 RNA was quantified by the Roche Cobas AmpliPrep/Cobas TaqMan HIV-1 assay (Roche Molecular Systems, Inc., Branchburg, NJ, USA) in both National Institute of Hygiene and Epidemiology and Pasteur Institute in Ho Chi Minh city with a lower limit of quantification of 20 copies/ml. HIVDR genotyping of HIV-1 reverse transcriptase and protease was performed using standard population sequencing. Sequencing was performed using ABI 3130XL system using the Big-Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystem, Foster City, CA, USA). The Stanford HIV drug resistance algorithm (HIVdb), version 8.4, was used to predict drug resistance and characterize subtypes. Sequences classified as having low-, intermediate-or high-level HIVDR by Stanford HIVdb were classified as resistant. Sequences classified as potential low-level resistance or susceptible were classified as susceptible. The outcome "any HIVDR" was defined per WHO guidance as resistance to any nucleoside reverse trascriptase inhibitor (NRTI), efavirenz (EFV) or nevirapine (NVP) and any ritonavir-boosted PI (PI/r) [7].

Data management and statistical analysis
Data were double entered using Epi Data 3.0 (Odensk, Denmark) and statistical analysis was performed using STATA version 14 (STATA Corp., College Station, TX, USA) following WHO methods. The prevalence of VL suppression and HIVDR were estimated using STATA's survey (SVY) commands [3,5]. Data were adjusted by number of patients who initiated therapy (PDR), on ART for 12 months (ADR12) or for at least 48 months (ADR48+) in the year prior to the survey initiation, observed clinic-level patient accrual, number of patients screened and the number of individuals with sequences genotyped. For ADR12 and ADR48+ surveys, the prevalence estimates of HIVDR were further adjusted by clinic-specific data on retention and clinic-specific data on unadjusted virological suppression.

Ethics
The study was reviewed and approved by the Institutional Review Board in Biomedical Research, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam (Approval number: IRB-VN01057-14/2017). In the ADR survey, genotyping was performed only for patients with non-suppressed VLs (VL >1000 copies/ml). Viral non-suppression was documented in 18/429 patients (4.2%) receiving ART for 12 months and 32/723 patients (4.4%) receiving ≥48 months. copies/ml, with any PDR detected in 22/340. After adjusting for the survey design, the nationally representative prevalence of any PDR HIVDR was 5.8% (95% CI 3.4-9.5%) ( Table 2). In specimens with any HIVDR, non-nucleoside reverse transcriptase inhibitor (NNRTI) resistance was predicted in 15/22 (68.2%); one had NVP resistance, and 14 had NVP and EFV resistance. NRTI resistance was predicted in 13/22 (59.1%), and 6/22 (27.3%) had resistance to both NNRTI and NRTI. The nationally representative prevalence of PDR to the NNRTI drug class was 3.4% (95% CI 1.8-6.2%) ( Table 2). We performed genotyping for 333 specimens and no PI resistance was detected in this survey. The prevalence of any PDR was higher in people reporting prior ARV drug exposure compared to those without (20% vs. 5.2%; p = 0.02).

R E S U LT S A N D D I S C U S S I O N
Since the introduction of the WHO's PDR survey method in 2014, 12 countries have reported national prevalence estimates of HIVDR with any HIDR prevalence estimates ranging from 8.2% to 23.4% [8][9][10][11]. For Vietnam, this was the first nationally representative survey of PDR using the 2014 WHO method. The prevalence of NNRTI resistance (3.4% [1.8-6.2%]) in this PDR survey was higher than that observed (1.6%, 8/490) in the survey using the WHO's cohort design in four sentinel ART clinics with 501 patients initiating ART from 2009 to 2010 [12]. The prevalence of any PDR in this survey (5.8%, 95% CI 3.4-9.5%) was similar to the over-all prevalence of any PDR in Asia as observed in a 10year meta-analysis of 16,088 genotype from the region (from 1996 to 2016), which had increased from 2.6% (95% CI 1.4-4.9) (NNRTI PDR 0.7%, 95% CI 0.2-2.8%) before 2005 to 5.5% (95% CI 3.3-8.6) (NNRTI PDR 4%, 95% CI 2.1-6.7%) in 2014-2016 [12]. However, we reported a lower overall prevalence of PDR than the recent report of 14.7% (95% CI:9.8-21.4%), NNRTI PDR of 9% and PI PDR of 5% in the survey in three ART clinics in HCMC, Vietnam in 2016 [14].

Acquired HIV drug resistance ≥48 months after ART initiation (ADR48+)
A total of 723 patients from 25 clinics were included in the ADR48+ survey (Table 1).
The prevalence of HIVDR among patients on ART for 12 months (3.0%) was similar to that observed in people receiving ART for at least 48 months (3.4%) (p = 0.39 by Fisher's exact test) ( Table 1). This is the second nationally representative ADR survey in Vietnam and we replicated the low prevalence of ADR (3% for ADR12 and 3.4% for ADR48+), which was seen in 2014 among people on ART for at least 36 months (4.6%) [2]. Among countries which completed national ADR surveys between 2014 and 2018, the prevalence of ADR among people on ART for 12 ±3 months ranged from 4.3% to 17.1% and from 4.6% to 28.3% for those receiving ART for ≥48 months [15].
Vietnam has initiated the transition to dolutegravir containing regimens as preferred first-and second-line regimens since 2020 [16]. A third round of HIVDR surveys was conducted in 2020, and the data are being analysed.
Our report has limitations. Firstly, our results were not reported in a timely manner. However, the data were included in the global report and Vietnam has considered these results and decided to continue the first-line ART regimen containing NNRTIs and keep improving the quality of ART service. Secondly, we were not able to evaluate the HIVDR when the VL was less than 1000 copies/ml. However, we assumed that the prevalence of HIVDR among patient with suppressed VL was low and this issue is recognized as a technical limitation in other studies [17].

C O N C L U S I O N S
Vietnam achieved very high level of HIV VL suppression among people on ART and maintained moderate prevalence of HIVDR in patients starting ART and low prevalence of HIVDR in patients receiving ART.