CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Conclusions: The Abbott Combo test performed well when screening for HIV infection in a large, population-level survey in Africa. Special data and sample management procedures likely contributed to the high quality of initial HIV screening test results in this study. Quality assurance testing was helpful for confirming HIV infection status, especially for those with an initially reactive Abbott Combo test result. 524 HIV Testing in the US PrEP Demonstration Project: rEIA vs Antigen/Antibody vs RNA Oliver Bacon 1 ; EricVittinghoff 2 ; Stephanie Cohen 1 ; Susanne Doblecki-Lewis 3 ; Megan Coleman 4 ; Susan P. Buchbinder 1 ;Wairimu Chege 5 ; Richard Elion 6 ; Michael Kolber 3 ; AlbertY. Liu 1 1 San Francisco Dept of PH, San Francisco, CA, USA; 2 Univ of California San Francisco, San Francisco, CA, USA; 3 Univ of Miami, Miami, FL, USA; 4 Whitman-Walker Hlth, Washington, DC, USA; 5 DAIDS, NIAID, NIH, Bethesda, MD, USA; 6 George Washington Univ Sch of Med, Washington, DC, USA Background: Ensuring individuals are HIV-uninfected prior to pre-exposure prophylaxis (PrEP) initiation and continuation is critical to minimize the risk of developing resistance, but the optimal HIV testing algorithm for PrEP is not yet known. We evaluated the performance of rapid blood EIA (rEIA), lab-based p24-antigen/antibody (4 th generation), and RNA tests in detecting HIV infection at PrEP screening (n=635), and initiation and follow-up (n=557) in the MSM and transgender women (TGW) enrolled in the US PrEP Demonstration Project, an open label study of FTC/TDF PrEP in San Francisco, Miami, and Washington DC. Methods: MSM and TGW interested in PrEP underwent screening, followed by enrollment within 45 days, when PrEP was initiated. All participants (ppts) were HIV tested at screening, enrollment, and follow-up (weeks 4, 12, 24, 36, 48) with rEIA and 4 th generation tests. Pooled RNA was performed at all visits in San Francisco, with quantitative or qualitative RNA only at enrollment in DC and Miami, respectively. Any positive test was confirmed using local testing algorithms. Results: At screening there were 16 (2.5%) ppts with a reactive rEIA and/or 4 th generation test, 15 of whomwere confirmed to be HIV infected. At enrollment 3 ppts had HIV infection: all were rEIA (-), 4 th generation (-) and RNA (+), with viral loads of 120, 3343, and 51 copies/mL (all pre-treatment); 2 started ART as soon as the (+) RNA was known, and all viral loads were positive on repeat testing. Of 2 infections during follow-up, both were rEIA (+)/4 th generation (+) at the visit with first evidence of infection; both had low or undetectable drug levels at seroconversion and no evidence of resistance on standard and ultrasensitive genotyping assays. During follow-up, of 2680 rapid EIAs there were 6 false positives in 2 ppts, for a specificity of 99.78% and a positive predictive value (PPV) of 25%; of 2673 4 th gen tests there were 3 false positives in 2 ppts, for a specificity of 99.89% (PPV 50%). There were no false positive RNA tests at screening, enrollment or follow-up. Conclusions: Rapid EIA and Lab-based 4 th generation testing detected most HIV infection before PrEP initiation. Acute HIV infection should be ruled out with an RNA test before starting PrEP, if available, particularly in clients with recent exposure; low viral loads should not be assumed to be false positives. In this cohort, with high PrEP adherence, rapid EIA and lab based 4 th generation tests were adequate to detect HIV infection during follow-up. 525 Comparison of 4 HCV Viral Load Assays at High Viral Load Robert Ehret 1 ; Marcel Schütze 1 ; Andrew Moritz 1 ; HaukeWalter 2 ; Gunnar Schalasta 3 ; Annemarie Berger 4 ; Martin J. Obermeier 1 1 Med Cntr for Infectious Diseases, Berlin, Germany; 2 Medizinisches Infektiologiezentrum Berlin, Berlin, Germany; 3 Prof Gisela Enders & Kollegen MVZ GbR, Stuttgart, Germany; 4 Inst of Virology, Univ of Frankfurt, Frankfurt, Germany Background: Hepatitis C virus (HCV) load is the most important surrogate marker for HCV replication. The height of viral load can be predictive for treatment success. The combination of sofosbuvir and ledipasvir has proven to be highly effective in treatment of HCV infection. While standard treatment duration of this combination is 12 weeks, in case of genotype 1 and a viral load below 6 Mio IU/ml, the treatment duration can be reduced to 8 weeks, reducing adverse effects and lower the price of this highly effective but costly treatment option. While in the clinical trials the Roche high pure/Cobas Taqman (HPS/CTM) was used, this assay is not common in clinical routine. We therefore compared four different quantitative HCV assays, the Abbott RealTime HCV (ART), the Cobas Ampliprep/TaqMan HCV v2 (CAP/CTM2), Cobas TaqMan HCV v2 for use with the High Pure System, v2.0 (HPS/CTM2) and the new Hologic Aptima HCV Quant Dx (HAC) at high viral load. Methods: Seven clinical samples with different genotypes (2 x 1a, 1b, 2b, 3a, 4d and 4p) where tested with ART and diluted to a target concentration between 500.000 IU/ml and 3.000.000 IU/ml. All samples were tested in each assay in five replicates, resulting in overall 140 viral load measurements. Mean viral load and coefficients of variation were compared between all the assays. Results: As expected for the high viral load used for testing, the coefficients of variation (CV) were low. Calculated on the logarithmic values CV were between 0,29 and 1,84%. HPS/CTM2 results were 1.8 times higher than ART results for genotype 1 samples. CAP/CTM2 results for the same samples were in mean not different to the ART results. HAC were 1.3 times higher than HPS/CTM2 in the genotype 1 samples. HAC samples for all genotypes, except genotype 4, were quantified in mean 1.4 times higher than in HPS/ CTM2. Genotype 4 was under quantified by HPS/CTM2 and CAP/CTM2. ART results (genotype 4 excluded) were 1.2 times higher than CAP/CTM2. Conclusions: The HAC assay showed in genotype 1 the closest correlation to the HPS/CTM2. While the CAP/CTM2 shows better performance on genotype 4 than HPS/CTM2, detection in genotype 4 is still better with HAC and ART. Despite their calibration to the WHO HCV RNA international standards, the assays showed significantly different quantitative results in this high viral load range. Therefore, results obtained with assays used in clinical trial, cannot be easily translated to clinical routine.

Poster Abstracts

204

CROI 2016