PREVALENCE OF VIROLOGICAL SUPPRESSION AND PATTERNS OF ACQUIRED HIV DRUG RESISTANCE AMONG INDIVIDUALS ON FIRST-LINE ANTIRETROVIRAL THERAPY IN FISHERFOLK COMMUNITIES IN UGANDA

Abstract

Background: The UNAIDS targets sustained viral suppression among at least 90% of HIV individuals on antiretroviral therapy (ART), yet viral load (VL) monitoring is rarely done in resource-limited settings due to high costs. Objectives: We determined the prevalence of virological suppression, prevalence, and patterns of HIV-1 acquired drug resistance (ADR) and correlates of virological failure (VF) and ADR among Ugandan fisher-folk (FF) on first-line ART. Methods: We included HIV-1 infected participants aged ≥15 year on first-line antiretroviral therapy and stratified them by median duration on ART i.e. 6, 12, 24, 36 and ≥ 48 months. For VL testing, we implemented both a 10x10 matrix and a 5-mini-pool pooling strategies and a deconvolution algorithm to identify individuals with VF, defined as VL≥1000 RNA copies/ml. Individual samples with VF, were genotyped by Sanger sequencing, for HIV-1 ADR mutations. Correlates of VF and ADR were determined by logistic regression for p<0.05 and 95% confidence intervals. Results: Of 609 participants with median age of 35 years (IQR: 28-43), on first-line ART for a median duration of 22.7 (IQR: 8.9–36.4) months, 335 (55%) were female. The overall prevalence of virological suppression (VS) was 85.6% (521/609). We obtained sequences from 80 (90.9%) out of 88 participants with VF. Acquired drug resistance (ADR) was detected in 59 of 80 (73.6%) participants with successful resistance testing. Only being on ZDV+3TC+EFV/NVP regimen was significantly associated with VF, AOR 2.27 (1.37-3.74; p=0.001) while NNRTI mutations (p<0.001), being on ZDV+3TC+EFV/NVP regimen, AOR 6.42 (1.37-30.13, p=0.018) and VL≥10000 copies/ml, AOR 4.18 (1.25-14.05, p=0.02), and were correlates of ADR. Our pooling strategy for viral load testing saved 38.4% of VL testing cost and had a negative predictive value of 95.2% (samples with VL<1000 copies/ml as negatives). Conclusion: Prevalence of virological suppression among FF is high, but falls below UNAIDS 90% target, with higher prevalence of HIV-1 ADR among virological failures. A higher proportion of virological failures with no ADR suggests poor adherence. Integrated care strategies should include increased VL monitoring, introduction of potent PI-based 2nd-line regimen and adherence support. Adoption of pooling viral load testing could be useful in population monitoring of ART programmes.