CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Methods: The Adult Antiretroviral and Drug Resistance (Tshepo) study was a 3-year randomized clinical trial (RCT) following 650 ART-naïve adults from Botswana who initiated certain first-line NNRTI-based ART. Participants were stratified by baseline CD4 count (bCD4 <=200 and 201-350 cells/ μ l) and randomized to 2 different adherence strategy arms. The RCT was completed in 2007 and participants were subsequently transferred to the Botswana national ART program. Ten-year outcome data was abstracted from the national ART program. Results: 650 adults enrolled between 2002-2004, 451 (69.4%) of which were female. Forty-three percent had advanced WHO clinical disease (3 or 4) at enrollment. Median age and BMI [IQR] was 33.3 years [IQR 28.9 – 38.7] and 21.3 [19.2-24.3], respectively. During 3-year RCT follow-up 35 (5.4%) of 650 patients had at least one ART switch due to virologic failure at a mean 68.7 weeks (STD 33.3) and 37 (5.7%) of 650 patients died. As of March 2014, 538 (83%) patients were still receiving ART (mean duration of ART = 8.4 years); of these 268 (83.8%) had bCD4 <=200 and 270 (81.8%) had bCD4 201-350 (p=0.541). 172 (26.5%) of 650 patients had at least one confirmed virologic failure (VF) (Viral load>1000) at median 3.5 years (IQR 1.5 to 6.4); crude VF rate 4.2/100PY. Fifty-six (8.6%) patients had died at median 1.4 years (IQR 0.6 to 4.4); crude mortality rate of 1.2/100PY. Eighty-one (12.5%) of 650 patients were not traceable (unknown F/U status). There was no difference among cases of unknown F/U status or death by adherence arm, age strata and/or sex. As of 2014, among those receiving first-line ART, 50.3%were receiving TDF-based ART most others receiving ZDV/3TC-based regimens. More than one-third (38.7%) had been switched to 2 nd or 3 rd line protease inhibitor therapy-based ART. Conclusions: In summary, at greater than 10 years after enrollment, adult patients in a public ART programwhich emphasized education and lab monitoring, had excellent survival and retention (83%) years after completing a RCT. These results are encouraging for both the patients and for the potential impact of TasP in this setting. 991LB Use of Population Viral Load to Predict HIV-Incidence in a Hyperendemic Population in Rural KwaZulu-Natal, South Africa Frank Tanser 3 ;Tulio de Oliveira 3 ;Till Barnighausen 2 ; Deenan Pillay 1 1 University College London, London, United Kingdom; 2 Harvard School of Public Health, Boston, MA, US; 3 University of KwaZulu-Natal, Mtubatuba, South Africa Background: Population viral load is argued to be a good measure of HIV transmission potential in a population. However, calculation of this measure rarely includes HIV-infected individuals who have not accessed the health system for HIV diagnosis and treatment. Further, such approaches have not been applied to hyperendemic populations in Southern Africa. Here we provide results from a population viral load survey covering an entire population in a hyperendemic community participating in annual HIV testing in rural KwaZulu- Natal, South Africa. Methods: The study uses data from one of the most comprehensive demographic surveillance sites in Africa - the Africa Centre Demographic Information System. The site has conducted population-based HIV testing for over a decade. HIV testing takes place annually in all consenting resident individuals’ ≥ 15 years of age. We performed viral load measurements on blood-spots collected from all 2,420 individuals testing HIV positive in the population-based HIV testing. All individuals were geolocated to their exact homestead of residence. We used a two-dimensional Gaussian kernel of radius 3km to map viral suppression, geometric mean population viral load and dcalculated an index of transmission potential termed the population prevalence of detectable virus (PPDV) - proportion of the entire population (ie irrespective of HIV status) having a detectable viral load. We then followed up 11, 806 HIV-negative individuals and quantified in multi-variable survival analysis the effect of PPDV in the surrounding local community on the risk of HIV acquisition. Results: Overall median viral load was 6428 copies per ml. 30% (726) of all HIV positive individuals in the population were below the detectable limit for blood spots of 1550 copies/ml. Marked spatial heterogeneity in geometric mean population viral load (Figure 1a) and PPDV (Figure 1b) occurred across the surveillance area with clear evidence of spatial clustering of high viral loads. Every 1% increase in community level PPDV was independently associated with a 4% increase in individual risk of acquisition of infection (HR= 1.04, p<0.001).

Poster Abstracts

Geographical variations in geometric mean population viral load (a) and population prevalence of detectable virus [PPDV] (b) in 2011 derived by a two-dimensional standard Gaussian kernel of radius 3km. Formal spatial clusters of high viral loads (a) or high proportion of individuals with unsuppressed viral loads (b) identified by the Kulldorff spatial scan statistic are superimposed. Conclusions: Our findings reveal remarkable spatial variation in population viral load in this relatively homogenous population. The results show that even in a severely affected rural African population with a well-established HIV treatment programme, PPDV could play a role in targeting interventions to specific communities to reduce the overall rate of new infections. 992LB Phase 1 Safety & PK Trial of Polyurethane Tenofovir Disoproxil Fumarate Vaginal Ring Marla J. Keller 1 ; Pedro Mesquita 1 ; Mark A. Marzinke 2 ; RyanTeller 3 ; Bruce Frank 4 ; Mark Mitchnick 4 ; Peter L. Anderson 5 ; Craig Hendrix 2 ; Patrick F. Kiser 3 ; Betsy Herold 1 1 Albert Einstein College of Medicine, Bronx, NY, US; 2 Johns Hopkins University School of Medicine, Baltimore, MD, US; 3 Northwestern University, Evanston, IL, US; 4 Particle Sciences, Inc, Bethlehem, PA, US; 5 University of Colorado, Aurora, CO, US Background: The prodrug tenofovir disoproxil fumarate (TDF) is more potent than tenofovir (TFV) against HIV in vitro & has greater tissue permeability & cellular uptake. Sustained TDF delivery from a vaginal ring (VR) completely protected macaques frommultiple vaginal simian-HIV challenges. The objectives of this trial were to evaluate safety & PK of reservoir-type TDF & placebo polyurethane VRs when used continuously for 14 days (d). Methods: A randomized, single blind, placebo controlled trial of 30 women was conducted. Clinical safety was assessed. Swabs (2 proximal, 1 distal to VR) & plasma were collected 1, 3, 7 & 14 d after VR insertion & 2 & 7 d after removal for drug levels. Dried blood spots (DBS) were obtained after 7 & 14 d of VR use. Cervical tissue was collected for TFV, TFV- diphosphate (TFV-DP) & ex vivo HIV challenge studies. Tissue HIV LTR DNA in relative copy number was quantified by real-time PCR. To measure anti-HIV activity of luminal drug, Jurkat-Tat-CCR5 cells were challenged ex vivo with HIV-1 BaL in the presence of cervical swab eluants. p24 levels in supernatants were assessed by AlphaLISA.

585

CROI 2015