CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

including end-stage renal disease (ESRD). We evaluated the association between coinfections and incident ESRD, independent of traditional risk factors. Methods: The British Columbia Hepatitis Testers Cohort includes ~1.7 million individuals tested for HBV, HCV or HIV, linked with laboratory and health-care administrative data. We defined ESRD through administrative codes for dialysis or kidney transplant. Individuals tested for all three infections, since 1990, were followed from the date of their last test (i.e baseline) until the earliest of i) incident ESRD, ii) death or iii) 12/31/2015. Fine and Gray models with adjustment for age, sex, ethnicity, alcohol and injection drug use (IDU), social/material deprivation, and history of diabetes and hypertension, were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of coinfections for ESRD, with death as a competing risk. Results: Of 524,186 individuals tested for all infections, we observed 3,762 incident ESRD events (1%) and 22,741 deaths (4%) during a median follow-up of 4 years (interquartile range [IQR]: 2, 9). The cohort was mostly female (54%), Caucasian (88%), and had a median age of 37 years (IQR: 29, 47). Additional comorbidities of hypertension (12%), alcohol or IDU (6%) and diabetes (3%) were uncommon. At least one chronic viral infection was found in 11%. The highest ESRD incidence rate (per 1,000 person-years) was in individuals with triple infection (27), followed by HCV/HIV (10), HBV/HIV (10), and HBV/ HCV coinfection (6). Cumulative incidence curves are shown in the Figure. In multivariate analysis, relative to those with no chronic infections, those with triple infection had the highest relative hazard for ESRD (HR 33, 95% CI: 28, 40), followed by coinfections: HCV/HIV (HR 18, 95% CI: 15, 21), HBV/HIV (HR 16, 95% CI: 12, 22), HIV monoinfection (HR 9.0, 95% CI: 7.6, 11), HBV/HCV coinfection (HR 8.0, 95% CI: 6.8, 9.5), and monoinfections: HCV (HR 6.0, 95% CI: 5.5, 6.7) and HBV (HR 4.5, 95% CI: 3.9, 5.2) Conclusion: HIV-infected individuals with chronic HBV and/or HCV coinfection were at highest risk of ESRD progression. Management of these syndemic conditions, particularly through prevention of HBV or HCV infection could reduce the risk of ESRD among HIV-infected individuals in clinical care.

security number with encrypted identifiers. We identified HIV+ candidates by fills of prescription medications exclusive to HIV treatment. After exploring potential mechanism(s) of missingness, we estimated HIV prevalence using multiple imputation by chained equations (MICE). We explored factors associated with ART regimens using logistic regression. Results: The pharmacy data linkage contained 91.0% (n=99,376) of all candidates in the national transplant registry in the study period. We identified 857 HIV+ candidates with an overall estimated prevalence of 0.95% (95% Confidence Interval: 0.89%-1.02%). HIV+ candidates were more often young (median [IQR]: 53 [48-59] vs. 56 [50-62]), African American (21.4 vs. 9.0%), and male (80.6 vs. 64.4%), with liver disease due to hepatitis C virus (33.5 vs. 26.1%) than HIV- candidates (p<0.001 for all). The use of pharmacoenhancers (PI/PEs) decreased over time (48.4% in 2007 to 20.0% in 2016) and were more likely to be used by African American candidates (aOR: 1.80, 95%CI: 1.18-2.74, p<0.01), adjusting for age, year, and sex. Conversely, integrase inhibitor (INSTIs) use increased over time (7.8% in 2008 to 52.3% in 2016) and were not associated with race (aOR: 1.02, 95% CI:0.62-1.64, p=.95) adjusting for age, year, and sex. Conclusion: We used a novel data linkage to identify a unique and previously unstudied population of HIV+ liver transplant candidates on the US waitlist. We found that the burden of HIV on the liver transplant waitlist was nearly 1%, ART use has shifted over time, and African Americans were almost twice as likely to be prescribed ART regimens containing pharmacoenhancers, which can interact with post-transplant immunosuppression.

Poster Abstracts

606 CHANGES IN LIVER STIFFNESS MEASUREMENT IN HIV AND HIV/HBV COINFECTED NIGERIANS Patricia A. Agaba 1 , Adovich Rivera 2 , Placid Ugoagwu 1 , Mohammad Muazu 1 , Jonathan Okpokwu 1 , Samuel Akpa 1 , Stephen McHenry 1 , Godwin Imade 1 , Oche Agbaji 1 , Chloe Thio 3 , Robert Murphy 2 , Claudia Hawkins 2 1 Jos University Teaching Hospital, Jos, Nigeria, 2 Northwestern University, Chicago, IL, USA, 3 Johns Hopkins University, Baltimore, MD, USA Background: There are limited data from sub-Saharan Africa on long-term liver fibrosis changes in HIV and HIV/HBV co-infected individuals. We assessed the effects of antiretroviral therapy (ART) on liver stiffness measurement (LSM) using transient elastography (TE) in HIV and HIV/HBV co-infected Nigerians and examined factors associated with liver fibrosis regression. Methods: ART-naïve HIV and HIV/HBV co-infected adults (>18 years) were enrolled into a longitudinal study of liver disease between 7/2011 – 2/2015 and followed annually for 6 years at Jos University Teaching Hospital, Jos, Nigeria. Changes in LSM over time were examined in a subset of subjects with ≥1 follow- up TE at Y3 and Y6. Predictors of a >30% decrease in LSM score during follow-up were assessed using Cox Proportional-Hazards models (CPH). Results: 232 HIV and 98 HIV/HBV co-infected subjects [71.2% female, median age 33.5 (IQR 12) yrs] were enrolled into the cohort. Among HIV/HBV co-infected subjects, median baseline HBV DNA was 1.67 [IQR 5.52] log10IU/mL and 6% were HBeAg seropositive. 79.4% initiated ART containing at least one HBV- active agent at enrollment and 100%were on ART at their last visit. Median duration of follow-up was 6.6 (4.1) yrs [HIV/HBV 6.4 (3.7), HIV 6.7 (4.8)]. 177/330 (54%) [45.7% HIV and 72.4% HIV/HBV co-infected] had a follow-up TE at Y3 and

605 PREVALENCE OF HIV AND PATTERNS OF ART USE AMONG US LIVER- TRANSPLANT CANDIDATES Ashton A. Shaffer 1 , Alvin G. Thomas 2 , Dorry Segev 1 1 Johns Hopkins University, Baltimore, MD, USA, 2 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Background: Despite the high burden of liver disease among HIV-infected (HIV+) individuals, the prevalence of HIV among candidates on the US liver transplant waitlist is unknown. Additionally, since the class of antiretroviral therapy (ART), particularly the use of pharmacoenhancers (protease inhibitors, cobicistat) may complicate post-transplant immunosuppression management, it is critical to understand which regimens are commonly used in this population.Therefore, we sought to estimate the prevalence of HIV and to describe patterns in ART use among US liver transplant candidates. Methods: We designed a retrospective cohort study (2007-2016) using pharmacy claims data (Symphony Health Solutions) linked to the national transplant registry (Scientific Registry of Transplant Recipients) using social

CROI 2019 229