A top scoring pairs classifier for recent HIV infections

Area of Science:

• Epidemiology
• Immunology
• Biotechnology

Background:

• Accurate HIV incidence estimation is vital for monitoring epidemics and evaluating prevention strategies.
• Current cross-sectional incidence assays, like the BED and LAg Avidity assays, can be biased by viral suppression, leading to overestimation.
• The increasing use of antiretroviral treatment necessitates new algorithms unaffected by viral load.

Purpose of the Study:

• To develop and validate a novel HIV incidence classifier using antibody responses alone.
• To assess the performance of this new classifier, focusing on its mean window period and accuracy.
• To provide a tool for more reliable HIV incidence estimation in the context of widespread treatment.

Main Methods:

• Utilized a phage display system to quantify antibody binding to over 3300 HIV peptides.
• Developed a classifier based on top-scoring peptide pairs to identify recent HIV infections.
• Validated the classifier using plasma samples from individuals with known seroconversion dates.

Main Results:

• The novel peptide-pair classifier demonstrated a mean window period of 217 days (95% CI: 183-257).
• This is significantly longer than the LAg-Avidity assay's mean window period of 106 days (95% CI: 76-146).
• The classifier accurately distinguished recent from non-recent infections, outperforming the LAg-Avidity assay in classifying recent samples.

Conclusions:

• The developed antibody-based classifier provides a robust method for HIV incidence estimation, independent of viral load.
• Its longer mean window period and improved accuracy offer advantages over existing assays, particularly in populations with high treatment coverage.
• This tool can enhance the accuracy of HIV epidemic monitoring and intervention impact assessment.