CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

1 The Johns Hopkins Univ, Baltimore, MD, USA, 2 Northwestern Univ, Chicago, IL, USA, 3 Los Angeles Biomed Rsr Inst at Harbor–UCLA Med Cntr, Torrance, CA, USA, 4 Univ of Pittsburgh, Pittsburgh, PA, USA, 5 Harbor–UCLA Med Cntr, Torrance, CA, USA Background: Coronary artery disease (CAD) is more common in HIV-infected (HIV+) compared to HIV-uninfected (HIV-) men in the Multicenter AIDS Cohort Study (MACS). Cumulative HIV RNA burden and associated inflammation may contribute to CAD. We assessed the relative prognostic value of viral copy-years (VCY), a cumulative viral load (VL) metric, to other VL measures on presence of CAD. Methods: MACS participants age 40-70 years without history of coronary revascularization were evaluated for subclinical CAD by cardiac CT angiography (CTA) from 2010- 2013. Our sample included 223 HIV+men with CTA measurements, a baseline VL measurement at or within 6 months prior to combination antiretroviral (cART) initiation, and subsequent VL measurements no more than 2 years apart. VCY since cART initiation was calculated using the trapezoidal rule to estimate the area under the curve between VL measurements. The association of VCY with subclinical CAD was compared to four alternative VL metrics since cART initiation: 1) peak VL, 2) baseline VL, 3) detectable VL (>50 copies/mL) at most recent measure before CTA (yes/no), and 4) any detectable VL within 5 years before CTA (yes/no). We used logistic regression to estimate odds ratios for two outcomes: coronary artery stenosis ≥50% (obstructive CAD) and non-calcified plaque. Models included age, race, MACS site, cohort, smoking status, ASCVD risk score, and nadir CD4+ T cell count. Model fit was assessed by the Akaike Information Criterion (AIC). Results: Studied men were 42% non-white, median age 52 years, a median 10.4 years from cART initiation to CTA (IQR 7.7, 13.8). The prevalence of stenosis ≥50% and non- calcified plaque were 16% and 62%, respectively. With the exception of most recent VL, each VL exposure metric was statistically significantly (p <0.05) associated with coronary artery stenosis ≥50% (Table 1). Both each log10 increase in VCY and detectable VL during the previous 5 years were associated with more than a two-fold increased odds of coronary artery stenosis (OR 2.4 CI: 1.4,4.0 and OR 2.6, CI: 1.1,6.2, respectively). Regression using VCY demonstrated the best fit based on AIC. None of the VL measures were associated with non-calcified plaque presence. Conclusion: Cumulative viral load was strongly associated with obstructive CAD in HIV+men. Other metrics of viremia, particularly the widely used most recent VL measure, may underestimate the role of ongoing exposure to viral replication in the development of coronary artery disease in HIV+ individuals.

Poster and Themed Discussion Abstracts

625 HIGHER CARNITINE LEVELS ARE ASSOCIATED WITH SUBSEQUENT MYOCARDIAL INFARCTIONS IN HIV Arjun Sinha 1 , Peter W. Hunt 1 , Jeffrey N. Martin 1 , Yifei Ma 1 , Heidi M. Crane 2 , Daniel Drozd 2 , Peter Ganz 1 , Stanley Hazen 3 , Priscilla Hsue 1 1 Univ of California San Francisco, San Francisco, CA, USA, 2 Univ of Washington, Seattle, WA, USA, 3 Cleveland Clinic, Cleveland, OH, USA

Background: HIV infection is associated with an increased risk of myocardial infarction (MI); however, the pathophysiology is incompletely understood. HIV alters the gut microbiome. Choline, carnitine, betaine, and trimethylamine N-oxide (TMAO), are small molecules that are metabolized or produced by the gut microbiome and TMAO is associated with MIs among adults without HIV. We have shown that carnitine and betaine are independently associated with carotid artery intima-media thickness in HIV-infected subjects, but TMAO and choline are not. We assessed the hypothesis that these gut microbiota-associated small molecules are independently predictive of MI in HIV-infected adults. Methods: This was a nested case-control study of HIV-infected individuals with suppressed viral load (VL) on antiretroviral therapy (ART) within the US based 8-site CNICS network. The cases had adjudicated and confirmed Type 1 MI from 2001-2012. The controls were matched by incidence density sampling to each case by calendar time, age, gender, race, duration of VL suppression, and CD4 count. Plasma levels of TMAO, Betaine, Carnitine, and Choline were measured at Cleveland Clinic using stable isotope dilution liquid chromatography tandemmass spectrometry. Plasma samples for cases and controls were collected prior to the event date. Associations between the small molecules and MI were assessed using conditional logistic regression. Results: There were 36 cases and 69 controls. The median age was 49 years (46, 58) and 77%were male. The median time of VL suppression was 3 months (1, 5) and the median CD4 count was 562 cells/mm3 (381, 809). The two groups had similar proportion of hypertension, T2DM, and active smoking. Cholesterol levels were similar as well but the cases had higher median triglycerides (184 mg/dL vs 146 mg/dL, p=0.05). After adjusting for triglycerides, elevated carnitine levels were strongly associated with MI (OR=4.95 for the top quartile (>33 µM) versus quartiles 1-3; (95% CI [1.29, 18.95], p=0.02, see Table). The other small molecules did not have a significant association with MI. Cholesterol and triglyceride levels as well as smoking were not associated with carnitine levels. Conclusion: Carnitine is independently predictive of MI in treated and suppressed HIV-infected individuals and appears to be independent of TMAO. This finding suggests that the mechanism of atherosclerosis in HIV is distinct from uninfected individuals and unique interventions may be indicated in HIV to reduce CV risk.

CROI 2017 267