CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

1003 INNATE DYNAMIC RANGE OF ABBOT PRISM HIV O PLUS IS ADEQUATE FOR RECENCY STAGING Eduard Grebe 1 , Marion Vermeulen 2 , Tinus Brits 2 , Ronel Swanevelder 2 , Genevieve Jacobs 2 , Michael P. Busch 3 , Alex Welte 1 1 Stellenbosch University, Stellenbosch, South Africa, 2 South African National Blood Service, Johannesburg, South Africa, 3 Blood Systems Research Institute, San Francisco, CA, USA Background: Tests for ‘recent’ HIV infection have found significant application in major population-level incidence (and trend) estimation projects. Recency assays are often modifications of diagnostic assays designed to extend the dynamic range and facilitate the application of recency discrimination thresholds. This raises the prospect that diagnostic platforms with high innate dynamic range may allow adequate recency staging to support incidence estimation. We investigated the suitability of the Abbott PRISM HIV O Plus assay for recency staging using data from the South African National Blood Service, which uses the assay for routine screening, by estimating the Mean Duration of Recent Infection (MDRI) for a range of recency discrimination thresholds. Methods: We estimated MDRI by fitting regression models for the probability of testing recent as a function of time since infection. In the context of blood donations, repeat observations from seroconverting donors are not routinely available, and time since infection is often very imprecisely known. We developed a method that does not depend on a ‘point estimate’ of infection time, but averages over unobserved (possible) infection times in repeat donors, and utilises a single PRISM signal-to-cutoff ratio (S/CO) from the positive donation. The analysis utilised specimens from 3,675 seroconverting repeat blood donors, with a median inter-donation interval of 341 days. Intervals were adjusted according to the sensitivity of screening assays. Reproducibility of the MDRI estimate was assessed by resampling the data in 10,000 bootstrapping iterations. Results: The figure shows MDRI values against S/CO recency discrimination thresholds. At plausible thresholds, the MDRI ranges from approximately 100 days to over 300 days, with coefficients of variation ranging from 4% to 9%. At an S/CO threshold of 80 the MDRI is estimated at 117 days (95% CI: 107-130) and at a threshold of 100, at 252 days (206-282). Conclusion: These MDRI estimates are comparable to that of the widely-used LAg assay when assessed using similar data from the same population. This is particularly promising given that the PRISM platformwas not designed for staging. The results suggest that the platform’s performance is sufficient to support estimating incidence in first-time blood donors using routinely- available screening data, without the need for additional recency testing.

1004 MEASUREMENT OF ANTIBODY EPITOPE SIGNATURES TO DETERMINE RECENCY OF HIV INFECTION Kathryn E. Stephenson 1 , Lisa M. Howe 1 , Dan Barouch 1 , Steven G. Deeks 2 , Sheila M. Keating 3 , Christopher D. Pilcher 2 , Timothy J. Henrich 2 1 Beth Israel Deaconess Medical Center, Boston, MA, USA, 2 University of California San Francisco, San Francisco, CA, USA, 3 Blood Systems Research Institute, San Francisco, CA, USA Background: One of the challenges in HIV epidemiology is the generation of reliable estimates of HIV incidence in cross-sectional surveys. Most tests for recent HIV infection measure either the magnitude or avidity of HIV-specific antibodies: both parameters are sensitive but not specific. Assays that quantify the epitope diversity of HIV-specific antibodies are a potential strategy to determine recent HIV infection. The hypothesis is that the greater the duration of HIV infection, the greater the epitope binding diversity of HIV-specific antibodies. We therefore used a novel assay, the global HIV peptide microarray, to define antibody epitope signatures associated with different stages of HIV infection. Methods: Plasma samples from 3 groups were obtained from the Consortium for the Evaluation and Performance of HIV Incidence Assays (CEPHIA): ART-naïve individuals with recent HIV infection (<12 mos, N=17), ART-naïve individuals with non-recent infection (>12 mos, N=17) and ART-suppressed individuals with non-recent infection (N=15). Samples were incubated with the peptide microarray, a multiplex platform to measure antibody binding to 6,654 peptides from across the HIV proteome and covering the majority of global sequences. Positive peptide responses were tallied by protein, with fewer positive responses denoting a narrower antibody epitope signature. Results: ART-naïve individuals with recent HIV infection had significantly fewer positive HIV Env peptides than those with non-recent infection (median of 50 vs. 138, P=0.0002). The expansion of positive responses was observed at the V3 loop of Env, which is involved in coreceptor engagement. Positive Env responses were also significantly and strongly correlated with the duration of HIV infection (days from detectable infection) among ART-naïve participants (P<0.0001, R=0.67 by Spearman test). This positive association persisted when ART-suppressed participants were included in the analysis (P=0.0078, R=0.39). Conclusion: Recent HIV infection was associated with a narrow epitope signature, while chronic HIV infection (even among persons with low HIV RNA) was associated with a broader and more diverse epitope signature. This data suggests that the global HIV peptide microarray can be used to measure epitope signatures associated with recent and non-recent HIV infection, independent of HIV RNA. Out data suggest that further optimization of the microarray in larger cohorts may allow for high throughput applications.

Poster Abstracts

CROI 2018 384