CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

±25.8 vs 23.8 ±16.6 U/dL, p=0.002) compared to control group. In PWH, log 10 IL-18 was associated with ALT (r=0.32, p=0.0001), AST (r=0.37, p=<0.0001), triglycerides (r=0.26, p=0.003), FIB-4 score (r=0.25, p=0.003), HIV plasma viral load (r=0.21, p=0.02), caspase-1 (r=0.31, p=0.0003), MCP-1 (r=0.32, p=0.0003), IL-6 (r=0.19, p=0.047), and LPS (r=0.22, p=0.03), and inversely associated with liver/spleen ratio (r=-0.23, p=0.02), HDL (r=-0.31, p=0.0003) and CD4+/CD8+ ratio (r=-0.2, p=0.02). The relationship between log 10 IL-18 with AST (β=33.5, p=0.0006) and ALT (β=37.9, p=0.001) remained significant after adjusting for age, gender, BMI, HIV RNA, and CD4+ count. In controls, log 10 IL-18 was also associated with ALT (r=0.37, p=0.004) and inversely associated with HDL (r=-0.27, p=0.04). Conclusion: We demonstrated significant relationships of IL-18 with liver transaminases and hepatosteatosis, suggesting the potential role of the inflammasome and IL-18 pathway in NAFLD progression in PWH. The relationship of IL-18 with LPS and MCP-1 may indicate IL-18's actions via known causes of NAFLD including intestinal microbial translocation and MCP-1/CCR2 signaling. Further studies are necessary to elucidate precise mechanisms involving IL-18 and inflammatory pathways in NAFLD development in PWH.

Poster Abstracts

547 EASL BIOMARKERS DIFFER IN PREDICTING NAFLD, NASH, AND FIBROSIS IN HIV +/– INDIVIDUALS Ricky Hsu 1 , Laurence Brunet 2 , Jennifer S. Fusco 2 , Gregory Fusco 2 1 New York University Langone Medical Center, New York, NY, USA, 2 Epividian, Durham, NC, USA Background: Fatty liver is a major health concern for people living with HIV as well as for the general US population. This study sought to compare 2 EASL- recommended biomarker-based risk scores with high sensitivity and specificity for diagnosing each of the 3 stages of fatty liver disease; NAFLD, NASH, and fat-induced fibrosis over 3 calendar periods. Methods: All HIV(+) and HIV(-) individuals in OPERA were included if all 6 scores could be calculated during one of the calendar periods of interest (2006- 2008, 2011-2013, 2016-2018) and they had no diagnoses of viral hepatitis, celiac, sclerosing cholangitis, or alcohol abuse. To mitigate outliers, average scores were obtained over each period to identify NAFLD (HSI NAFLD score > 36 or NAFLD Liver Fat score > -1.455), NASH (HAIR score >= 2 or Campos NASH score >= 5) and fibrosis (Fib-4 index > 2.67 or NAFLD Liver Fat score > 0.675). Results were age and sex standardized using the HIV(-) population as the standard and risk differences were estimated. Results: This study included 7,583 HIV(+) and 1,645 HIV(-) in 2006-2008; 25,347 HIV(+) and 65,903 HIV(-) in 2011-2013; and 46,229 HIV(+) and 100,699 HIV(-) persons in 2016-2018. Prevalence estimates varied substantially depending on the score used. HIV(+) persons were much more likely to have all biomarkers required for the 6 tests (>80%) than the HIV(-) persons (<25%). Among HIV(+) persons, after age/sex standardization, NAFLD prevalence increased over the years, ranging from 43-54%with HSI NAFLD score and 28-39%with NAFLD Liver Fat score; NASH prevalence remained stable, ranging from 16-17%with HAIR score and 3-6%with Campos NASH score; fibrosis prevalence remained stable, ranging from 3-4%with FIB-4 and 4-7%with NAFLD Liver Fat score (Fig). HIV(+) persons had a lower standardized prevalence of NAFLD and NASH than HIV(-) persons at most time points with either score (Fig). Conclusion: Despite similar published predictive values among EASL- recommended biomarker risk scores, calculated prevalence of NAFLD, NASH and liver fibrosis based on these scores differed significantly in the OPERA cohort. The selection of a study population among whom all scores could be calculated likely disproportionally included individuals at higher risk of fatty liver disease, thus overestimating the true prevalence especially among those without HIV. Further clinical validation of these scores is required before broad utilization in the staging of fatty liver disease.

548 AN RCT OF RALTEGRAVIR- VERSUS EFAVIRENZ-BASED ART IN HIV-HCV COINFECTION Thuy Le 1 , Chau V. Nguyen 2 , Thao P. Vu 3 , Ly T. Vo 4 , Thanh T. Nguyen 3 , Thu T. Nguyen 1 , Robert Paul 5 , Lishomwa C. Ndhlovu 6 , Dominic Chow 6 , Cecilia M. Shikuma 6 1 Duke University School of Medicine, Durham, NC, USA, 2 Hospital for Tropical Diseases, Ho Chi Minh City, Vietnam, 3 Oxford University Clinical Research Unit in Vietnam, Ho Chi Minh, Vietnam, 4 University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh, Vietnam, 5 University of Missouri, Columbia, MO, USA, 6 University of Hawaii at Manoa, Honolulu, HI, USA Background: Drug induced liver injury following initiation of ART is more common in HIV/HCV coinfected patients; however comparative data on hepatotoxicity of ARVs used in this population are lacking as HIV/HCV coinfected patients are largely excluded from clinical trials. We compared hepatotoxicity, virological, and clinical outcomes between raltegravir (RAL)- and efavirenz (EFV)- based ART in HIV/HCV coinfected patients starting 1st-line ART in Vietnam. Methods: This RCT allocated patients 1:1 to RAL/TDF/FTC or EFV/TDF/FTC between June 2014 and February 2017. Eligibility: HIV infection, ART-naïve, age ≥18, met Vietnam guidelines for ART (CD4 <500 cells/µl or WHO stage 3 or 4 disease), HCV infection (positive HCV antibody and HCV RNA), AST and ALT ≤80 U/L, creatinine clearance ≥60 ml/min, negative HBSAg, no evidence of decompensated cirrhosis, and not on rifampicin. We tested AST, ALT, bilirubin every month and CD4, HIV RNA, HCV RNA, fibroscan, and lipids at w0, 24, 48, and 72. We compared the rates of ALT and AST toxicity > grade 2 (primary outcome) and time to AIDS or death by arm using Kaplan Meier curves and log rank test. We compared the proportions of HIV RNA suppression at w72 by Chi-Square test. Results: We screened 207 and enrolled 80 participants (39 on RAL, 41 on EFV; median age 32; 88%male, 75%with history of IDU). EFV was associated with higher incidence of ALT and AST elevation (73.0% vs. 62.2%, P=0.14 and 61.8% vs. 42.5%, P=0.10, respectively). The majority of liver events occurred during the first 12 weeks. 5 patients (6%) died (2 in RAL arm died of TB; 3 in EFV arm died of TB, CNS infection, and suicide). 18 developed AIDS events (9 each arm). There were no significant differences in time to AIDS or death (P=0.94) or proportions of HIV RNA <150 copies/mL at w72 (87.9% in RAL, 85.7% in EFV, P=1.00). EFV was associated with a lower CD4 cell gain (170 vs. 224 cells/ µL, P=0.52), higher HDL gain (0.37 vs. 0.17 mmol/L, P=0.11), lower reduction

CROI 2020 196