CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

for HPgV-2 antibodies on the ARCHITECT instrument, followed by molecular characterization. All viral sequences were classified by phylogenetic analysis. Results: Overall, HCV RNA was detected in 321 (2.58%; 95% CI: 2.31-2.87%) specimens, with slightly higher prevalence in the HIV positive individuals (2.94%; 95% CI: 2.5-3.46%) than in the HIV negative individuals (2.35%; 95% CI: 2.04-2.71%). Notably, the median age of HCV RNA negative individuals was 28 years while the median for positive individuals was 54. Phylogenetic classification of sequences from N=97 specimens identified HCV genotypes 1 (20%), 2 (17%), and 4 (63%). HPgV-2 antibodies were detected in N=28 (10.61%; 95% CI: 7.44-14.9%) of the HCV RNA positive specimens, with higher prevalence in the HCV-HIV co-infected group (13.08%; 95% CI: 7.96-20.77%) than the HCV mono-infected group (8.92%; 95% CI: 5.39-14.41%). HPgV-2 RNA was detected in N=6 specimens by RT-PCR and/or next generation sequencing, including N=2 HIV co-infected specimens. Classification of the obtained HPgV-2 sequences indicates they are closely related to strains identified previously in the United States. Conclusion: In the largest HCV surveillance study to date in Cameroon, we find that the prevalence of HCV is relatively low in our study population compared to the results of smaller previous studies. The discovery of HPgV-2 in South Cameroon expands the geography of this virus to the African continent, indicating it may be more widespread than previously appreciated. 629 PREVALENCE AND CORRELATES OF HPGV INFECTION AMONG PWID WITH HIV INFECTION IN INDIA Nandagopal Paneerselvam 1 , Thongadi R. Dinesha 1 , Shanmugam Saravanan 1 , Allison M. McFall 2 , Pachamuthu Balakrishnan 1 , Aylur K. Srikrishnan 1 , Sunil S. Solomon 2 , Shruti H. Mehta 2 1 YR Gaitonde Center for AIDS Research and Education, Chennai, India, 2 Johns Hopkins Hospital, Baltimore, MD, USA Background: Human Pegivirus (HPgV) shares overlapping transmission routes with HIV as it is transmitted through percutaneous and sexual routes. Previous studies have shown that HPgV infection is associated with reduced disease progression among HIV-infected individuals. However, most of these data are from resource-rich settings with non-clade C HIV infection. We estimate the prevalence of HPgV infection and its association with disease progression among HIV-infected people who inject drugs (PWID) in Chennai, India. Methods: We tested stored plasma samples from 154 HIV-infected PWID who were recruited and followed as part of a community-based cohort of PWID in Chennai, India (the Chennai HIV, HCV and EeRal study (CHHEERS)) from 2012- 2016. Samples from the baseline study visit were screened for HPgV infection using Real Time PCR. We compared baseline characteristics of persons who were positive and negative for HPgV using the Fisher’s exact test for dichotomous or categorical characteristics and the Mann-Whitney rank sum test for continuous characteristics, with a particular focus on HIV disease parameters including CD4, HIV plasma viral load, and antiretroviral therapy (ART). Results: All PWID in this cohort were male and the median age was 40 (interquartile range [IQR]:36–43). The median CD4 cell count was 357 (IQR: 226- 489), nearly half (47%) were currently taking ART, of whom 38% had suppressed HIV viral load. The prevalence of HPgV infection was 23% (95% confidence interval: 17-31%). There were no statistically significant differences between HPgV positive and HPgV negative persons at baseline in terms of HIV viral load (80% and 78%with detectable viral load, respectively ;p = 0.92) and CD4 cell counts (median 357 and 347, respectively, p=0.77) even after accounting for ART. However, at 24 months of follow-up the median CD4 count among those on ART was significantly higher among HPgV positives compared to those who were negative for HPgV(p=0.01) (Figure 1). Conclusion: We observed a high prevalence of co-infection with HPgV among a cohort of HIV-infected PWID in India. The HPgV infected patients exhibited a superior response to ART with better recovery of CD4cells. Further evaluation of these relationships in a larger sample size might shed light on the role of HPgV infection among PWID with HIV infection.

630 HEPATITIS C VIRUS AND INCIDENT TYPE 1 AND TYPE 2 MYOCARDIAL INFARCTION IN HIV Jessica Williams-Nguyen 1 , Robin M. Nance 1 , Sara Lindström 1 , Susan Heckbert 1 , Inga Peter 2 , Matthew Budoff 3 , W. C. Mathews 4 , Joseph J. Eron 5 , Peter W. Hunt 6 , Richard D. Moore 7 , Michael J. Mugavero 8 , Mari Kitahata 1 , Michael Saag 8 , Heidi M. Crane 1 , Joseph Delaney 1 1 University of Washington, Seattle, WA, USA, 2 Icahn School of Medicine at Mt Sinai, New York, NY, USA, 3 Los Angeles Biomedical Research Institute at Harbor–UCLA Medical Center, Torrance, CA, USA, 4 University of California San Diego, San Diego, CA, USA, 5 University of North Carolina Chapel Hill, Chapel Hill, NC, USA, 6 University of California San Francisco, San Francisco, CA, USA, 7 Johns Hopkins University, Baltimore, MD, USA, 8 University of Alabama at Birmingham, Birmingham, AL, USA Background: Persons living with HIV (PLWH) are at increased risk of cardiovascular events including Type 1 (atheroembolic) and Type 2 (oxygen supply-demand mismatch) myocardial infarction (MI), potentially due in part to chronic inflammatory processes. Evidence is inconclusive on the role of chronic coinfection with hepatitis C virus (HCV), common in PLWH, in modulating the risk of MI. This analysis sought to understand whether HCV in PLWH is associated with increased MI risk or differences in risk by MI type. Methods: The CFAR Network of Integrated Clinical Systems (CNICS) is a multi-center HIV cohort with comprehensive clinical data including centrally adjudicated MI. We estimated the association between HCV and time to incident Type 1 MI (T1MI) and Type 2 MI (T2MI) using Cox regression models with multiple imputation to accommodate missingness in covariates. Estimates were adjusted for age, sex, site, race, ethnicity, smoking, substance use, measures of cholesterol, diabetes, treated hypertension, statin use, nadir CD4 count and HIV viral load. Results: Among 24,755 PLWH, 2,280 (9.2%) were positive for HCV infection at baseline and there were 332 T1MI and 328 T2MI during a median of 4.2 years of follow-up. HCV was not associated with overall MI (adjusted hazard ratio (aHR) 1.20, 95% Confidence Interval (CI): 0.94-1.52). In subtype analysis, HCV was associated with risk of T2MI (aHR 1.51, 95% CI: 1.12-2.05) but not T1MI (aHR 0.86, 95% CI: 0.57-1.28). In further analyses examining adjudicated causes of T2MI, HCV was not associated with T2MI attributed to cocaine use (aHR 1.04, 95% CI: 0.41-2.60) but was associated with a 2-fold greater risk of T2MI attributed to sepsis (aHR 2.01, 95% CI: 1.22-3.29). Conclusion: HCV status in PLWH is associated with T2MI but not with classical, plaque-rupture-event T1MI. In particular, we found an association between HCV status and T2MI attributed to sepsis suggesting that HCV-infected PLWH have a higher risk of developing sepsis or T2MI as a complication of sepsis. This could be explained by consequences of HCV infection, such as liver dysfunction and chronic inflammation, or differences in sepsis risk factors between HCV-infected and uninfected individuals. Our analysis of MI subtypes allowed us to obtain a more detailed understanding of the cardiovascular risks associated with HCV in PLWH and may thereby facilitate future treatment and prevention efforts.

Poster Abstracts

CROI 2018 233