HIV-1C dynamics and evolutionary trends in Botswana

[摘要] ENGLISH ABSTRACT : Introduction: HIV incidence estimates are critical for monitoring HIV transmission dynamics, and fordesign and evaluation of the impact of interventions. Biomarkers and assays for cross-sectionalsurveillance of HIV incidence are greatly needed because of the high costs and time needed tomaintain prospective cohorts to determine HIV incidence. New cross-sectional assays for estimation ofHIV incidence are attractive due to their improved performance and cost-effectiveness. In thisdissertation, methods for identification and characterization of recency of HIV infection are described.An in-depth review of HIV recency determination methods, including novel cross-sectional applicationof molecular methods, is given in 'From serological assays to genomics. Multi-assay approacheswere evaluated in order to increase the sensitivity and specificity of the commercial incidence assays inthe context of high treatment coverage and stable but high HIV prevalence in Botswana. A novelbiomarker based on HIV viral diversity was investigated as a complementary or standalone tool tocharacterize HIV recency. In this thesis, an innovative use of pairwise diversity and the time to themost recent common ancestor (tMRCA) in a heterosexual HIV-1 subtype C (HIV-1C) epidemic wereintroduced as novel approaches for HIV incidence estimation. We evaluated the properties of the newpotential tools for estimating time since infection, including their specificity and predictiveperformance in the context of the HIV-1C epidemic in Botswana.Methods: Characterization of HIV recency and novel biomarkers for estimation of HIV infectionincidence is based on application of immunologic and molecular methods:a) Evaluation of the long-term specificity (false recent classification rates) of serological tests forrecent infection, and algorithms for estimating HIV-1C incidence utilizing samples frompatients with known long-standing HIV infection.b) Application of within-host viral diversity for estimation of HIV-1C recency in Botswana usingsamples collected from patients with known time since seroconversion in the primary HIV-1Cinfection cohort.c) Investigation of intra-host viral pairwise diversity and the time to the most common recentancestor (tMRCA), as potential markers for HIV infection recency.Results: We estimated for the first time false recency rates (FRR) of the commercially available BEDand Limiting Antigen (LAg) assays in Botswana. We demonstrated that combined algorithms reduceFRR to the recommended < 2%. Including viral load in the assay algorithm resulted in an FRR of 0.4%for LAg. Analysis of the within-host viral pairwise diversity provided more accurate estimation of HIVrecency, as compared with the recommended LAg and BED using the receiver operator characteristicanalysis (ROC). We demonstrated that intra-host viral pairwise distances reduce misclassification andincrease the accuracy of serologic assays. tMRCA and intra-host viral pairwise distances correlatedwith time since HIV infection provide additional novel tools for reliable estimation of HIV recency.Conclusion: HIV infection recency can be determined cross-sectionally using a combination ofserological and molecular biomarkers. Including viral load and an assessment of prior exposure toARVs is critical for accurate estimation of HIV incidence. Intra-host pairwise diversity and tMRCAare able to predict time since HIV infection and can be used to improve accuracy in estimation of HIVinfection recency.

[发布机构] Stellenbosch University