Assessing the recency of HIV infections and estimating HIV incidence in two rural districts of Lesotho

Abstract

Background HIV recency assays are used to distinguish recently acquired infection from long-term infection among individuals newly diagnosed with HIV and thus identify hotspots of transmission. Additionally, since 2015, UNAIDS and the World Health Organisation (WHO) advise the use of HIV recency assay together with HIV viral load (VL) quantification as an incidence measure. Objectives Based on randomly selected samples of new HIV infections in two districts of Lesotho, we determined recent HIV infections, described and assessed risk factors for these recent infections, and estimated HIV incidence. Methods A total of 243 samples had been collected from newly diagnosed HIV- positive patients during a door-to-door HIV testing campaign from August 2018 to May 2019 in randomly selected rural villages of two districts in Lesotho (Butha-Buthe and Mokhotlong). The MSc candidate tested these samples for HIV recency using the Asanté HIV-1 Rapid Recency Assay and VL using the Roche Cobas System, according to the recent infection testing algorithm recommended by WHO. Moreover, she extracted the clinical and socio-demographic information from the trial database and conducted the descriptive (characteristics and incidence calculation) as well as inferential statistical analyses (risk factor analysis). Results Participants were recruited from August 2018 to May 2019. Among the 11,291 participants tested for HIV, 243 (2.2%) samples of new HIV diagnoses were available for recency and VL testing. Among the 243 samples, we found 18 (7.4%; 95% confidence interval 4.4%-11.5%) recent infections, while 225 (92.6%) were classified as long-term HIV infections. The majority of the participants were female (142 [58%]) with a median age of 37 years (interquartile range [IQR] 31-50 years). Although some characteristics showed a certain trend, only female gender was predictive for a recent HIV infection (adjusted odds ratio 3.74 (95% confidence interval 1.05-13.36); p-value 0.041). The estimated annual incidence was Ia = 0.29%. Conclusion During door-to-door testing among a general population sample in two rural districts of Lesotho, 7.4% of newly diagnosed HIV infections were recent (< 6 months) as well as a significant incidence rate. Women were more likely to have acquired HIV recently. Our results imply significant ongoing transmission in these rural communities.