CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

classify each sample as recent or non-recent. Overall sample classification is determined by a majority voting system. Optimal classification cutoffs and the number of voting pairs were identified using a training set of 176 samples from 38 individuals and subsequently tested on the remaining 82 samples from 19 individuals. We compared these results to results from a standard Limiting Antigen Avidity (LAg-Avidity) protocol. Results: In the final model with 4 voting pairs, 79% (71/90) recent samples and all 168 of non-recent samples were correctly classified. In contrast, the LAg- Avidity protocol classified 43% (35/81) recent samples and all 165 non-recent samples correctly. Comparison of TSP vs. LAg-Avidity showed that the TSP approach captured a greater proportion of recent infections with a mean window period of 217 days (95% CI: 183-257 days; see figure). In contrast, the mean window period of the LAg-Avidity protocol is 106 days (95% CI: 76-146 days). Conclusion: We identified four 4 peptide pairs that outperformed the standard LAg-Avidity protocol for identifying samples from individuals with recent HIV infection. These peptides can be incorporated into a simple assay for field use. With a larger mean window period, the TSP classifier yields more precise incidence estimates without relying on viral load to identify non-recent samples. The TSP approach can also easily be applied to other populations and virus subtypes to identify novel peptide signatures for recent infection.

These included 27% (60/221) of the SC12-24 group, 2.7% (6/217) of the SC0-12 group, 0.15% (1/689) of those who enrolled HIV+ at PC12, and 0.17% (7/4022) of those who enrolled HIV+ at PC0 (most infected for >2 years). Use of a higher cutoff for the LAg assay (2.0 or 2.5 ODn) increased the proportion of the SC12-24 group classified as recently infected from 27% to 32% or 41%, respectively, but increased the FRR among those infected >2 years from 0.17% to 0.42% or 0.72%, respectively. In each study country and overall, the CSI estimates were nearly identical to observed incidence (Table). Conclusion: In this large community-randomized study, a widely-used CSI algorithm that included the LAg assay and HIV VL yielded accurate point estimates of incidence, despite high rates of viral suppression among those with both prevalent and incident infection. However, the CSI estimates were considerably less precise than observed incidence measured from cohort follow-up.

977 URINE TENOFOVIR LEVELS BY IMMUNOASSAY PREDICT HIV PROTECTION IN A LARGE PrEP TRIAL Randy Stalter 1 , Jared Baeten 1 , Deborah J. Donnell 2 , David Glidden 3 , Warren Rodrigues 4 , Guohong Wang 4 , Michael Vincent 4 , Matthew A.Spinelli 3 , Nelly R. Mugo 5 , Kelly Johnson 3 , Andrew Mujugira 1 , Mark A.Marzinke 6 , Craig W. Hendrix 6 , Monica Gandhi 3 , for the Partners PrEP Study Team 1 University of Washington, Seattle, WA, USA, 2 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 3 University of California San Francisco, San Francisco, CA, USA, 4 Abbott Labs, Abbott Park, IL, USA, 5 Kenya Medical Research Institute, Nairobi, Kenya, 6 Johns Hopkins University, Baltimore, MD, USA Background: New tools are needed to support PrEP use for individuals at high risk for HIV in sub-Saharan Africa, including objective adherence metrics that allow for provision of real-time feedback. Urine tenofovir (TFV) levels have been proposed as a marker of PrEP use that could be measured with low-cost, point-of-care (POC) antibody-based tests. We hypothesized that TFV levels in urine, measured via a recently-developed immunoassay, would be comparable to those in plasma, the gold standard for short-term PrEP adherence in clinical trials, and associated with protection from HIV. Methods: We measured TFV levels in stored urine samples collected from a randomly sampled cohort of HIV-negative men and women from the active PrEP arms in the Partners PrEP Study using enzyme-linked immunosorbent assay (ELISA) (lower limit of quantification [LLOQ] 1000 ng/mL). Date-matched plasma TFV concentrations were measured via liquid chromatography-tandemmass spectrometry (LC-MS/MS) with an LLOQ of 0.31 ng/mL. Using the same cohort and all HIV seroconverters on PrEP, we conducted a case-cohort analysis to assess association between recent urine TFV level >1500 ng/mL, a threshold which accurately classifies recent PrEP dosing, and protection from HIV. The 1500 ng/mL cut-off will be used for the first iteration of the POC assay. Estimates of the hazard contributed 722 and 91 urine samples, respectively; 39% of the cohort and 51% of cases were female. Detectable urine TFV levels showed 87% sensitivity (95% CI: 84-90%) and 73% (65-79%) specificity for detectable plasma TFV concentration, which is predictive of HIV protection. Using the urine level at first detection of seroconversion in the adjusted model, a urine TFV level >1500 ng/mL was associated with a 71% (95% CI: 24-89%; p=0.01) adjusted reduction in HIV risk. Conclusion: In a large completed PrEP trial, urine TFV levels measured via a novel immunoassay were predictive of protection from HIV. Detection of TFV in urine showed good sensitivity and specificity for detection of TFV in plasma measured via LC-MS/MS, an established metric of short-term PrEP adherence. The urine immunoassay has now been developed into a lateral flow assay which can provide results at the POC. Our findings suggest that a real-time assay to assess TFV levels in urine could be a valuable addition to existing objective metrics for PrEP adherence. ratio for the Cox model are adjusted for age, sex, and sexual behavior. Results: We included 292 participants in the cohort and 45 cases who

Poster Abstracts

976 EVALUATION OF CROSS-SECTIONAL HIV INCIDENCE TESTING IN THE HPTN 071 (POPART) TRIAL Ethan B. Klock 1 , Oliver Laeyendecker 2 , Reinaldo Fernandez 1 , Ethan A.Wilson 3 , Estelle Piwowar-Manning 1 , Sam Griffith 4 , B Kosloff 5 , Anneen Van Deventer 6 , Sarah Fidler 7 , Helen Ayles 8 , Peter Bock 6 , Deborah J. Donnell 3 , Richard J. Hayes 9 , Susan H. Eshleman 1 , for the HPTN 071 (PopART) Study Team 1 Johns Hopkins University School of Medicine, Baltimore, MD, USA, 2 NIH, Bethesda, MD, USA, 3 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 4 FHI 360, Durham, NC, USA, 5 London School of Hygiene & Tropical Medicine–Zambia, Lusaka, Zambia, 6 Stellenbosch University, Cape Town, South Africa, 7 Imperial College London, London, UK, 8 Zambart, Lusaka, Zambia, 9 London School of Hygiene & Tropical Medicine, London, UK Background: The Limiting Antigen Avidity (LAg) assay is used to estimate HIV incidence in cross-sectional surveys. Testing algorithms often include low HIV viral load (VL) as a marker of non-recent infection. We compared the accuracy of cross-sectional incidence (CSI) estimates to observed incidence in the community-randomized HPTN 071 (PopART) trial, where a substantial proportion of HIV+ participants were virally suppressed. Methods: HIV incidence was assessed in a Population Cohort (PC) 1-2 years after study initiation (between the PC12 and PC24 surveys). Observed incidence was based on confirmed seroconversion events between PC12 and PC24. The CSI analysis of the PC24 survey included 15,845 who remained HIV negative, 221 persons who seroconverted between PC12 and PC24 (SC12-24), 217 who seroconverted between PC0 and PC12 (SC0-12), 4,022 who were HIV+ at PC0; and 689 who enrolled HIV+ during the PC12 survey. The VL at PC24 was <1,000 copies/mL for 72.7% of HIV+ persons, including 31% (70/221) of the SC12-24 group. All HIV+ PC24 samples were tested using the Sedia LAg-Avidity assay. Recent infections were defined as having a LAg result <1.5 normalized optical density units (ODn) and HIV VL >1,000 copies/mL. The CSI estimate was determined using a mean duration of recent infection of 130 days (95% confidence interval [CI]: 117-143) and a false recent ratio (FRR) of 0%. Results: The LAg result was <1.5 ODn in 11.3% (582/5149) of all HIV+ persons; 74/582 had a VL >1,000 copies/mL and were classified as recently infected.

CROI 2020 367