Evaluation of multi-assay algorithms for cross-sectional HIV incidence estimation in settings with universal antiretroviral treatment

Table 2 Multi-assay algorithms

MAA	Criteria used to classify samples as “recent”	Mean window period
LAg + VL	LAg-Avidity < 1.5 OD-n and VL > 1000 copies/mL	130 (118, 142) [8]
LAg + BR	LAg-Avidity < 2.8 OD-n and BioRad AI < 40%	119 (94, 144) [20]
LAg + PepPair^a	LAg-Avidity < 2.5 OD-n plus log₁₀(peptide pair 4) < − 0.02	102 (69, 159)

The table shows the three multi-assay algorithms (MAAs) evaluated in this report. The mean window period is shown in days; 95% confidence intervals are shown in parentheses
^aThe LAg + PepPair MAA includes the log₁₀ ratio of antibody reactivity to two HIV peptides as a biomarker of HIV infection duration. Pep_93621 is a 15-amino acid peptide located in gp160; antibody reactivity to this peptide increases over the course of HIV infection. Pep_92687 is a 22-amino acid peptide located in the gag-polyprotein; antibody reactivity to this peptide decreases over the course of infection. Antibody reactivity to these peptides was measured using the MSD U-PLEX assay. The value included in the MAA (log₁₀[peptide pair 4]) was calculated as: log₁₀([MSD U-PLEX assay result for upgoing gp160 peptide 93621]/[MSD U-PLEX result for downgoing gag peptide 92687]). Additional file 1 provides additional information on the development and performance of the LAg + PepPair MAA

ISSN: 1471-2334