CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

TUESDAY, FEBRUARY 24, 2015 Session P-K1 Poster Session

Poster Hall

2:30 pm– 4:00 pm ART: Adherence, Adherence, Adherence 555 Self-Reported Versus Blood-Tested ART Intake to Estimate ART Coverage in South Africa Helena Huerga 1 ; GillesVan Cutsem 2 ; LubbeWiesner 3 ; Malika Bouhenia 1 ; Jihane Ben Farhat 1 ; Emmanuel Fajardo 2 ; Ruggero G. Giuliani 2 ; David Maman 1 ;Thomas Ellman 2 ; Jean-François Etard 1 1 Epicentre, Paris, France; 2 Médecins Sans Frontières, Cape Town, South Africa; 3 University of Cape Town, Cape Town, South Africa Background: Reliable information on ART intake is essential to estimate ART coverage and interpret HIV resistance. In the context of a population-based HIV survey in Mbongolwane and Eshowe, KwaZulu-Natal, South Africa, we assessed self-reported versus blood-tested ART intake agreement and we estimated ART coverage using both measurements. Methods: Cross-sectional population-based survey. Persons aged 15-59 years were eligible. Face-to-face interviews were followed by rapid HIV testing on site and blood collection for ART testing and other HIV-related tests. Qualitative ART testing for nevirapine, efavirenz and lopinavir covered all ARV regimens in the public sector and detected drugs down to 0.2 m g/ml. ART coverage was defined as the proportion of HIV positive on ART among those eligible according to National Guidelines. Results: In total, 5649/6688 (84.5%) individuals were included. Overall HIV prevalence was 25.2% (95%CI: 23.6-26.9). Self-reported vs blood-tested ART intake agreement was 91.9% (kappa=0.84). In total, 58/741 (7.8%) participants with a positive blood test declared not taking ART and 52/712 (7.3%) self-reporting ART intake had a negative blood test (table). ART intake agreement was similar in women and men: 91.9% (kappa=0.84) vs 92.0% (kappa=0.84) and lower in those aged <25 years compared to ≥ 25 years: 88.4% (kappa=0.70) vs 92.6% (kappa=0.85). Among individuals <25 years: 15/64 (23.4%) with positive blood test declared no ART and 12/61 (19.7%) with negative blood test self- reported ART intake. ART coverage was 75.0% using blood-tested ART and 72.1% using self-reported ART. Viral load was ≥ 1000 cp/ml in 17/57 (29.8%) with negative self-report and positive blood-test and 39/52 (75.0%) with positive self-report and negative blood-test. Resistance to ARV was found in 33 (61.1%, 95%CI: 47.2-73.4) of the 54 individuals with viral load ≥ 1000 cp/ml who self-reported ART intake. Of them, 12 (36.4%) had a negative ART blood test. Self-reported and blood-tested ART intake among HIV positive participants Conclusions: Self-reported and blood-tested ART agreement was high in the overall population but lower among young people. Stigma related issues may explain some of the measurement differences found in young people. Overall ART coverage estimations were not affected by the measurement discrepancies. ART self-reporting can be a good tool for overall ART coverage estimations. However, blood testing for ART is preferable when assessing ART intake in young people and should be taken in consideration when interpreting ART resistance. 556 Real-Time Electronic Adherence Monitoring and Risk of Viral Rebound Jessica E. Haberer 1 ; Nicholas Musinguzi 2 ; Mark Siedner 1 ;Yap Boum 3 ; PeterW. Hunt 4 ; Jeffrey Martin 4 ; David R. Bangsberg 1 1 Massachusetts General Hospital, Harvard Medical School, Boston, MA, US; 2 Mbarara University of Science and Technology, Mbarara, Uganda; 3 Médecins Sans Frontières, Mbarara, Uganda; 4 University of California San Francisco, San Francisco, CA, US Background: While sustained adherence to antiretroviral therapy (ART) is known to be critical for viral suppression, the risk of viral rebound associated with any given adherence interruption is poorly understood. We used real-time adherence monitoring and HIV RNA measurement during adherence interruptions to understand the characteristics of interruptions that predict viral rebound. Methods: The Ugandan AIDS Rural Treatment Outcomes Study (UARTO) is an observational cohort of adults initiating ART. Procedures include 4-monthly assessment of HIV RNA and adherence monitoring with a wireless device that transmits a cellular signal with each opening (Wisepill). If no signal is received for 48+ hours, research staff visit participants’ homes unannounced to ascertain the cause and collect blood for HIV RNA measurement. For participants with prior viral suppression (<400 copies/ml), we fit generalized estimating equation logistic regression models to identify predictors of rebound viremia during adherence interruptions. Results: We monitored 470 participants between June 2011 and January 2014; 72%were women, median age was 35 years (IQR 28-42), baseline CD4 was 198 cells/mm 3 (IQR 110-294), pre-ART HIV RNA was 5.0 log 10 copies/ml (IQR 4.5-5.5) and median duration of prior viral suppression was 13 months (IQR 3-48). Overall median adherence was 97% (IQR 90-100%). We observed 4,447 interruptions lasting 48+ hours; median duration was 2.8 days (IQR 2-4.2). Participants had a median of 1 interruption per month (IQR 1-1). The following interruptions were excluded from analysis: 3,303 (74%) had no blood drawn for HIV RNA assessment (i.e., participant restarted ART prior to investigation, participant declined, technical cause of the interruption), 447 (10%) were missed, and 127 (3%) had detectable HIV RNA prior to the interruption. Of the remaining 570 (13%) interruptions, 14 (2%) revealed new viral rebound. Predictors of viral rebound are shown in the table. No interactions were seen. All but 9 interruptions (561/570; 99%) ended with the investigation and 11/14 (79%) had re-suppressed HIV RNA on subsequent testing.

Poster Abstracts

Predictors of viral rebound

358

CROI 2015