%0 Journal Article %A Eduard Grebe %A Alex Welte %A Jake Hall %A Sheila M. Keating %A Shelley N. Facente %A Kara Marson %A Jeffrey N. Martin %A Susan J. Little %A Matthew A. Price %A Esper G. Kallas %A Michael P. Busch %A Christopher D. Pilcher %A Gary Murphy %A on behalf of the Consortium for the Evaluation and Performance of HIV Incidence Assays (CEPHIA) %T Diagnostic, Infection Timing and Incidence Surveillance Applications of High Dynamic Range Chemiluminescent HIV Immuno-Assay Platforms %D 2017 %R 10.1101/132332 %J bioRxiv %P 132332 %X Background Custom staging assays, including the Sedia HIV-1 Limiting Antigen Avidity EIA (LAg) and avidity modifications of the Ortho VITROS anti-HIV-1+2 and Abbott ARCHITECT HIV Ag/Ab Combo assays, are used to identify ‘recent’ infections in clinical settings and for cross-sectional HIV incidence estimation. However, the high dynamic range of chemiluminescent platforms allows differentiating recent and longstanding infection on signal intensity, and this raises the prospect of using unmodified diagnostic assays for infection timing and surveillance applications.Methods We tested a panel of 2,500 well-characterised specimens with estimable duration of HIV infection with the three assays and the unmodified ARCHITECT. Regression models were used to estimate mean durations of recent infection (MDRI), context-specific false-recent rates (FRR) and correlation between signal intensity and LAg measurements. A hypothetical epidemiological scenario was constructed to evaluate utility in surveillance applications.Results Over a range of MDRIs (reflecting recency discrimination thresholds), a diluted ARCHITECT-based RITA produced lower FRRs than the VITROS platform (FRR ≈ 0.5% and 1.5% respectively at MDRI of 200 days) and the unmodified diagnostic ARCHITECT produces incidence estimates with comparable precision to LAg (RSE ≈ 17.5% and 15% respectively at MDRI of 200 days). ARCHITECT S/CO measurements were highly correlated with LAg ODn measurements (r = 0.80) and values below 200 are strongly predictive of LAg recency and duration of infection less than one year.Conclusions Low quantitative measurements from the unmodified ARCHITECT obviate the need for additional recency testing and its use is feasible in clinical staging and incidence surveillance applications. %U https://www.biorxiv.org/content/biorxiv/early/2017/05/03/132332.full.pdf