CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

875 THE RELATIVE IMPACTS OF ART AND HARM REDUCTION ON HIV INCIDENCE IN BRITISH COLUMBIA Bohdan Nosyk , Xiao Zang 1 , Jeong E. Min, Emanuel Krebs, Viviane D. Lima, M-J Milloy, Jean A. Shoveller, Rolando Barrios, Evan Wood, Julio S. Montaner BC Cntr for Excellence in HIV/AIDS, Vancouver, British Columbia, Canada Background: Access to antiretroviral therapy and harm reduction services have been cited as key contributors to the control of the HIV epidemic, however the specific contribution of the latter has been questioned due to uncertainty in the true efficacy of ART in prevention HIV transmission through needle sharing. We aimed to isolate the independent effects of the provision of opioid agonist treatment (OAT) and needle distribution (or harm reduction services), and the secondary preventive benefits of ART through needle sharing on HIV incidence in British Columbia, from 1996-2013. Methods: Using comprehensive linked population-level data, we populated a dynamic, compartmental transmission model to simulate the HIV/AIDS epidemic in BC from 1996- 2013. HIV incidence, mortality among PLHIV, and quality-adjusted life years (QALYs) were estimated. We further incorporated rates of OAT utilization and syringe distribution volumes to estimate their impact on the selected outcomes. We estimated scenarios designed to isolate the independent effects of ART on transmission via needle-sharing (assuming 50% (10%-90%) efficacy) and harm reduction services in reducing HIV incidence through needle sharing– both among PWID and at the population-level. Structural and parameter uncertainty was investigated. Results: We estimated that 3240 (2394-4562) incident HIV cases were averted between 1996 and 2013 as a result of the combined effect of the expansion of harm reduction services and ART coverage on HIV transmission via needle sharing. Decomposing the effects of these services, we estimate harm reduction services alone would have accounted for 77% (62%-95%) of averted HIV incidence, while ART alone would have accounted for 44% (10%-67%) of incident cases. Due to high distribution volumes, provision of sterile syringes predominantly accounted for incidence reductions attributable to harm reduction services, however OAT provided substantially greater QALY gains. Conclusion: The scale-up of harm reduction services had a profound impact on the course of the HIV epidemic in BC, with an impact on HIV incidence comparable to ART access. Harm reduction services such as needle distribution and OAT should be viewed as critical and cost-effective tools in a combination implementation strategy to reduce the public health and economic burden of HIV/AIDS. 876 ASSESSING UTILITY OF HIV INCIDENCE ASSAYS IN ESTIMATING INDIVIDUAL INFECTION TIMES Christopher D. Pilcher 1 , Shelley Facente 1 , Eduard Grebe 2 , Sheila Keating 3 , Reshma Kassanjee 4 , Matt A. Price 5 , Michael P. Busch 3 , Gary Murphy 6 , Alex Welte 2 , Travis C. Porco 1 1 Univ of California San Francisco, San Francisco, CA, USA, 2 Stellenbosch Univ, Stellenbosch, South Africa, 3 Blood Systems Rsr Inst, San Francisco, CA, USA, 4 Univ of Cape Town, Cape Town, South Africa, 5 Intl AIDS Vaccine Initiative, New York, NY, USA, 6 Pub Hlth England, London, UK Background: In HIV antibody-positive seroconverters, infection time can be estimated from negative and positive test dates. Even in frequent repeat testing situations (e.g. 3-12 monthly) this interval may not allow precise estimates. We explored whether modern tests for recent HIV infection could be used to narrow a 3-12 month interval and improve infection time estimates in HIV seroconverters. Methods: We examined a panel comprising 196 ART-naïve timepoints from 96 HIV seroconverters with first viremia documented and followed up to 2 years.We focused on three commercial assays that provide information on HIV recency: Sedia LAg Avidity assay (ODn); Architect Ag/Ab screening assay (S/CO) and Geenius supplemental test (recency index). Growth curves were derived by linear interpolation within the observed range, and extrapolation of a semiparametric growth curve outside the observed range for each subject. The distribution of infection times associated with a particular result was obtained by repeatedly randomly sampling the entire population of curves at some prior distribution of infection times: this sampling time frame ranged from 0 to 3, 6 or 12 months in alternative testing scenarios. Quantiles derived from the simulation procedure were then smoothed with splines (degrees of freedom, 5), and truncated for test values larger than the value for which the largest median time was observed. Results: The range of infection times seen at any test result was reduced by about half (42-56%) when compared to the prior distribution of plausible infection times, for all assays and scenarios. The Figure shows the median and 25-75%IQR of infection times associated with first Geenius results and a prior negative test in 3, 6 or 12 months prior. At lower assay values and in more frequent testing scenarios, the infection times associated with a given value were earlier and the IQR of these times was narrower. In the 3 monthly testing scenario, the average IQR was 22 days for LAg, 21 days for Geenius and 25 days for Architect. Conclusion: Commercial diagnostic and incidence assays can inform infection time estimates newly HIV Ab-positive individuals. For a given result, however, both the estimate and its precision depend on prior information from the testing history. When the plausible interval defined by the history is short, timing estimates can be made with precision similar to Fiebig staging (IQR11 days). These findings have implications for targeting HIV prevention, cure research and vaccine trial design.

Poster and Themed Discussion Abstracts

877 A MULTIPLEX ASSAY FOR MONITORING RECENT HIV-1 TRANSMISSION IN LOCALIZED OUTBREAKS Kelly Curtis , Ellsworth Campbell, Krystin A. Price, WilliamM. Switzer CDC, Atlanta, GA, USA

Background: Assays for determining recent HIV-1 infection are important public health tools for estimating HIV incidence. These assays provide a measurable distinction between recent and long-term infection due to differential antibody reactivity to specific HIV antigens. We developed a bead-based, multiplex assay which measures antibody binding and

CROI 2017 379