CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Conclusions: A simple prognostic score calculated from information readily collected at clinical centres can accurately predict progression to LRD among HIV/HCV coinfected individuals. The score outperformed prediction based solely on fibrosis staging highlighting the importance of the other clinical elements. 638 Has Modern ART Reduced Endstage Liver Disease Risk in HIV-Hepatitis Coinfection? Marina B. Klein 1 ; Keri N. Althoff 3 ;Yuezhou Jing 3 ; Greg D. Kirk 3 ;Vincent Lo Re 2 ; Nina Kim 4 ; Mari Kitahata 4 ; ChloeThio 3 ; Michael J. Silverberg 5 ; Richard Moore 3 North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) 1 McGill University Health Centre, Montreal, Canada; 2 University of Pennsylvania, Philadelphia, PA, US; 3 Johns Hopkins University, Baltimore, MD, US; 4 University of Washington, Washington, DC, US; 5 Kaiser Permanente Northern California, Oakland, CA, US Background: HIV-infected adults are commonly coinfected with hepatitis B (HBV) and C (HCV) viruses and thus at risk for Endstage Liver Disease (ESLD). Whether safer, more effective modern ART has reduced ESLD rates is unknown. We estimated ESLD incidence since the introduction of combination ART by HBV and HCV co-infection status in the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD). Methods: HIV-infected adults participating in the 12 cohorts contributing to the NA-ACCORD who were observed for validated ESLD diagnoses from Jan 1, 1996 to Jan 1, 2010 were included. HBV was defined by a positive surface or e antigen test or detectable HBV DNA. HCV was defined as a positive antibody test or detectable HCV RNA. ESLD events (ascites, SBP, bleeding varices, encephalopathy, hepatoma) were validated using standardized screening and review protocols. Poisson regression models were used to estimate incidence rates (IR, per 1,000 p-years) and ratios (adjusted for age, sex, race and cohort; aIRR) with 95% confidence intervals ([,]) in the early (1996-2000), middle (2001-2005) and modern (2006-2010) ART eras. Results: 34,119 adults contributed 380 incident ESLD outcomes and >129,000 person-years. Overall, ESLD incidence was similar in the early, middle and modern ART eras. In all three eras, incidence rates were highest in HIV/HCV/HBV and lowest among HIV monoinfected with HIV/HCV/HBV having the highest aIRRs (see Table). No significant change in rates of ESLD was observed across the 3 time periods for any group. Comparing the early vs. modern eras, aIRR [95% CI] were: 1.6[0.8, 3.4]; 1.3[0.8, 2.1] and 0.5 [0.1, 1.9] for HIV/HCV, HIV/HBV and HIV/HCV/HBV respectively. Similarly, comparing middle vs. modern eras, aIRR [95% CI] were: 1.1[0.7, 1.6], 0.8[0.6, 1.2] and 0.5[0.2, 1.2]. Overall death rates were high in the early ART era which may have lead to an underestimation of ESLD risk in this period. However, death rates were similar in the middle and modern eras.

Poster Abstracts

Table. Incidence of ESLD by ART Era and Hepatitis Status in the NA-ACCORD Conclusions: Hepatitis virus co-infected adults are at markedly increased risk for ESLD compared those infected with HIV alone, with triply infected patients at greatest risk. No clear reduction in ESLD risk was observed over the three time periods. The continued high incidence of ESLD despite modern ART underscores the urgent need to specifically address HCV and HBV infections in HIV infected adults. 639 Marijuana Use Does Not Accelerate Liver Fibrosis in HCV/HIV-CoinfectedWomen Erin M. Kelly 1 ; Jennifer L. Dodge 1 ; Monika Sarkar 1 ; Audrey French 2 ; PhyllisTien 1 ; Marshall Glesby 3 ; Elizabeth Golub 4 ; Michael Augenbraun 5 ; Michael Plankey 6 ; Marion G. Peters 1 WIHS 1 University of California San Francisco, San Francisco, CA, US; 2 CORE Center/Stroger Hospital, Chicago, IL, US; 3 Weill Cornell Medical College, New York City, NY, US; 4 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, US; 5 State University of New York, Downstate Medical Center, Brooklyn, NY, US; 6 Georgetown University, Washington, DC, US Background: Cannabis (THC) use has been correlated with liver fibrosis progression in retrospective analyses of mono-infected chronic hepatitis C (HCV) patients, particularly in those with established fibrosis. We studied long-term effects of THC on fibrosis progression in women co-infected with HCV/HIV enrolled in Women’s Interagency HIV Study (WIHS), a prospective, multicenter, cohort of women with or at risk for HIV infection. Methods: Liver fibrosis was categorized according to APRI scores as mild (<0.5), moderate (0.5-1.5), or severe ( ≥ 1.5): women with severe fibrosis at entry into WIHS were excluded. THC and alcohol use were treated as continuous variables and quantified as average exposure over time in study until last follow-up or development of severe fibrosis. Associations between THC use and progression to severe fibrosis were assessed using Cox proportional hazards regression. Results: Among 670 HIV/HCV co-infected women (median follow-up: 5.1 (1.2-10.5) years), 323 (49%) reported no THC use; 209 (31%) reported ≥ weekly use; 134 (20%) 〈 weekly use; and 4 no THC data. Median APRI at entry were similar (0.53 vs 0.49 vs 0.50) in those who reported no THC use, 〈 weekly use and ≥ weekly use, respectively. Compared to women

401

CROI 2015