CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

who were virologically-suppressed (<50 copies/mL) for ≥12 months. TFV-DP in DBS was quantified using a validated LC-MS/MS assay, and concentrations of 17 biomarkers of inflammation and immune activation were measured by electrochemiluminescence or ELISA. Log-transformed TFV-DP concentrations were analyzed using a mixed-effects model, providing estimates of percent change in plasma biomarker concentrations for every ½ log e increase in TFV-DP in DBS. Data are presented as median [Q1, Q3] or geometric mean [95% confidence interval]. Results: A total 123 visits from 69 participants (14 women, 19 Black, 8 Hispanic) with virologic suppression were analyzed. Median age and duration of HIV suppression were 48 [41, 57] and 4 [3, 7] years, respectively. Median time between visits was 6 [4, 8] months. The geometric mean TFV-DP for all visits was 1704 [1495, 1943] fmol/punch. After adjusting for age, gender, race, body mass index, tobacco use and statin use, TFV-DP in DBS was directly associated with biomarkers of inflammation (interleukin-8, tumor necrosis-α, serum amyloid A protein), monocyte activation (soluble CD14), T-cell activation (soluble CD27), and soluble intercellular and vascular cell adhesion molecules (sICAM-1 and sVCAM-1), as noted in the Table. Conclusion: In PLWH with long-term viral suppression on TDF-based ART, high TFV-DP in DBS was associated with higher biomarkers of endothelial activation, immune activation and inflammation. In contrast to the decrease in systemic inflammation/immune activation observed in viremic PLWH who initiate ART, these findings suggest that higher ART exposure could have a different effect in biomarkers of inflammation in chronically-suppressed individuals. Conversely, immune activation and inflammation could also influence TFV-DP in DBS. Further research is required to elucidate the mechanism and clinical significance of these findings.

and hypoxia, IL1 signaling, TGFB and CTLA4 signaling, and the fibrin-collagen formation system. RNASeq of classical monocytes revealed enrichment of additional pathways in PLHIV and HF, including hypoxia signaling, PI3K AKT MTOR activation, CXCR4 signaling, relaxin signaling, and TNFA/TGFB signaling. Our immunophenotyping analysis found that PLHIV with HF have significantly higher expression of CD57, a marker of senescence, on their CD4 T cells compared to non-HF PLHIV (p=0.01). Linear regression modeling integrating both data types identified upregulation of EGF/TGFB family genes as significantly associated with higher expression of CD57 on CD4 T cells. Conclusion: We have identified proinflammatory gene expression signatures that correlate with HF in PLHIV and constitute candidate biomarkers of HF in HIV alongside our immunophenotyping data. Our data provide a platform for future investigation of the inflammatory factors associated with chronic HIV infection over time and those that may promote HF risk. 659 HIV/HCV-SPECIFIC MARKERS AND ECHOCARDIOGRAPHIC PULMONARY ARTERY SYSTOLIC PRESSURE Courtney Zola 1 , Meredith S. Duncan 1 , Kaku So-Armah 2 , Kristina Crothers 3 , Adeel A. Butt 4 , Cynthia L. Gibert 5 , Joon W. Kim 6 , Joseph Lim 7 , Vincent Lo Re 8 , Hilary A. Tindle 1 , Matthew Freiberg 1 , Evan L. Brittain 1 1 Vanderbilt University, Nashville, TN, USA, 2 Boston University, Boston, MA, USA, 3 University of Washington, Seattle, WA, USA, 4 Weill Cornell Medicine, New York, NY, USA, 5 George Washington University, Washington, DC, USA, 6 James J. Peters VA Medical Center, Bronx, NY, USA, 7 Yale University, New Haven, CT, USA, 8 University of Pennsylvania, Philadelphia, PA, USA Background: Pulmonary hypertension is associated with increased mortality in those with HIV compared to matched, uninfected controls. In small cohorts, hepatitis C virus (HCV) co-infection appears to increase pulmonary hypertension risk. We hypothesized that markers of HIV/HCV disease activity would be associated pulmonary artery systolic pressure (PASP). Methods: We performed a cross-sectional study of participants from the Veterans Aging Cohort Study (VACS) enrolled April 2003 through October 2015 referred for an echocardiogram to examine the association between markers of HIV/HCV viral status and PASP. We performed multiple linear regression analysis to determine whether HIV/HCV mono-infection or coinfection were associated with higher PASP, adjusting for comorbidities with known PH associations and HIV/HCV status. We performed subset analyses, including markers of disease severity as follows: 1) restricted to HIV+ subjects to assess the association of HCV coinfection, higher HIV viral load, lower CD4+ T-cell count, and antiretroviral therapy (ART) with PASP levels and 2) restricted to those with chronic HCV infection to determine whether higher HCV viral load or interferon use was associated with higher PASP. Results: Among the 8,226 subjects in our sample, 2,194 (27%) had HIV only, 540 (7%) had HCV only, and 637 (8%) were HIV-HCV coinfected. In adjusted analyses, we did not observe an association between HIV mono-infection (β ˆ =0.19, 95% CI -0.52, 0.90), HCV mono-infection (β ˆ =0.04, 95% CI -1.19, 1.26), or HIV/HCV co-infection (β ˆ =0.71, CI -0.47, 1.88) with PASP. We observed a modest inverse association between CD4+ T-cell count and PASP (Table). Neither HIV nor HCV viral loads were associated with PASP. Those on “Other” ART regimens (i.e. not on NRTI+NNRTI or NRTI+PI) demonstrated reduced PASP. Interferon exposure was not associated with PASP among HCV-infected individuals. Conclusion: Contrary to reports from smaller, selected populations we did not observe an independent association between infection with HIV and/or HCV and higher PASP. Our sample of coinfected individuals is roughly six times larger than prior published cohorts. Lower absolute CD4+ T-cell count was inversely associated with PASP suggesting that a more intact adaptive immune system has a greater impact on PASP than viral replication. ART regimens may have variable effects on PASP, which requires further study.

Poster Abstracts

658 TRANSCRIPTOMIC BIOMARKERS OF HEART FAILURE IN PEOPLE LIVING WITH HIV Cheryl Cameron 1 , Chris T. Longenecker 1 , Brian Richardson 1 , Michael Fang 1 , Jonathan Buggey 2 , Sadeer Al-Kindi 2 , Michael Cartwright 1 , Carmen Nichols 1 , Pearline Cartwright 1 , Jackelyn Golden 1 , Mark Cameron 1 1 Case Western Reserve University, Cleveland, OH, USA, 2 University Hospitals Cleveland Medical Center, Cleveland, OH, USA Background: Compared to HIV-uninfected controls, people living with HIV (PLHIV) have a 25-80% higher risk of heart failure (HF), including both reduced ejection fraction HF (HFrEF) and preserved ejection fraction HF (HFpEF). HF is therefore a prevalent condition in HIV that may increase substantially as the HIV-infected population ages. However, very little is known about HF risks in the current ART treatment era and the goal of this study was to identify a peripheral immune cell transcriptomic signature of HF in PLHIV. Our hypothesis is that functional genomic variation related to chronic inflammatory responses may drive HF risk in PLHIV. Methods: As part of a case-control study within the Case Center for AIDS Research prospective clinical cohort (AIDS 125; UH IRB 01-98-55), we performed total RNA sequencing and immunophenotyping of PBMCs and classical monocytes obtained from (a) 10 HIV+ subjects with adjudicated incident HFrEF; (b) 6 HIV+ subjects with adjudicated incident HFpEF; and (c) 16 age- and gender-matched control subjects without HF. Low input libraries were generated using Kapa RNA Hyper kits and sequenced on an Illumina NextSeq 550 (75 bp, paired-end, 30 million reads/sample). Differentially expressed genes were found by two-group t test (P≤0.05) and organized into top pathways by P value (P≤0.05) via gene set variation analysis (GSVA). Results: RNASeq analysis of PBMCs revealed enrichment of potentially therapeutically targetable pathways in PLHIV and HF, including adipogenesis

CROI 2019 250