CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

687 ASSOCIATION OF SUBCLINICAL CMV DNA AND IMMUNOLOGIC MARKERS OF CARDIOVASCULAR DISEASE Ankita Garg, Sara Gianella, Masato Nakazawa, Stephen A. Spector University of California San Diego, La Jolla, CA, USA Background: HIV-CMV co-infected persons are at increased risk of cardiovascular disease (CVD) associated with persistent inflammation. Persons with high interferon (IFN)γ response to CMV have increased numbers of endothelium homing receptor (CX3CR1)- expressing cells that are associated with CVD. Here, we investigated the effect of subclinical CMV replication on these markers. Methods: 80 paired PBMC samples were collected from 40 CMV-seropositive, early HIV-infected men starting ART within a median of 3 mo from estimated date of infection, and achieved suppressed HIV RNA within a median of 3 mo on ART. PBMC were obtained >12 mo apart (1st sample a median of 34 wks post-ART initiation). CMV IFNγ response was determined by ELISPOT using a CMVpp65 peptide pool; CMV specific memory T cells were identified by flow cytometry. CMV and EBV levels were measured by ddPCR. Data were analyzed using a mixed effects regression model to predict associations between CMV shedding, IFNγ production and CX3CR1-expressing CD4+ and CD8+ T cells over time. Bayesian hierarchical models were used to quantify differences in CMV and EBV replication over time. Participants were classified as low (LR)- or high- responders (HR) according to IFNγ production (100 SFU/10 6 cells). Results: 26 (65%) participants were classified as HR and 14 (35%) as LR at the 1st time-point which did not change over time nor was influenced by CMV DNA levels (median SFU/10 6 cells at 1st/2nd time-point: HR: 383/308 vs LR: 21/41). Change in IFNγ levels over time was influenced by CMV levels (p<0.01), as individuals with a greater decline in IFNγ had increased levels of CMV DNA compared to those with low CMV. Higher CMV DNA was also associated with increased numbers of CD28+CD27-CD4+ T cells expressing CX3CR1 (p<0.001). Similarly, increased IFNγ production was associated with increased numbers of CMV-specific CX3CR1+CD28+CD27-CD4+ and CD8+ T cells (P<0.001). Using a similar interaction model, EBV was not associated with any of these findings. Conclusion: These findings demonstrate in HIV-CMV co-infected persons on suppressive ART that higher CMV levels and IFNγ responses are associated with a subset of CMV-specific memory T cells expressing CX3CR1, and that high and low IFNγ responders maintain their response category over time. Thus, we have identified a subgroup of HIV-infected CMV IFNγ HR with increased numbers of circulating T cells expressing CX3CR1 who may be at increased risk of CVD and other inflammatory diseases. 688 CYSTATIN C AND ATHEROSCLEROSIS IMAGING MARKERS IN HIV INFECTED PATIENTS Mitchell R. McClean 1 , Petra Buzkova 2 , Matthew Budoff 3 , Michelle Estrella 4 , Matthew Freiberg 5 , Howard Hodis 6 , Frank J. Palella 7 , Wendy Post 8 , Cecilia Shikuma 9 , Samir K. Gupta 1 1 Indiana University, Indianapolis, IN, USA, 2 University of Washington, Seattle, WA, USA, 3 University of California Los Angeles, Los Angeles, CA, USA, 4 University of California San Francisco, San Francisco, CA, USA, 5 Vanderbilt University, Nashville, TN, USA, 6 University of Southern California, Los Angeles, CA, USA, 7 Northwestern University, Chicago, IL, USA, 8 Johns Hopkins Hospital, Baltimore, MD, USA, 9 University of Hawaii, Honolulu, HI, USA Background: Reduced estimated glomerular filtration rate (eGFR) using serum creatinine (Cr) based estimates is associated with increased risk of cardiovascular disease (CVD) in HIV. Compared to eGFR using only Cr, serum Cystatin C (CC)-based eGFR is more predictive of CVD in the general population. However, it is unclear how CC-based eGFR compared to Cr-based eGFR in predicting subclinical CVD in HIV-infected persons. Methods: We conducted a cross-sectional analysis of data pooled from three large cohorts of HIV+ patients (VACS, MACS, Hawaii Aging) comparing associations between three CKD-EPI eGFR equations (Cr, CC, and Cr-CC) with intima media thickness (CIMT) at the common carotid artery and coronary artery calcium (CAC) scores using multivariable regression analysis. CIMT, CAC, and CC were centrally measured. eGFR and CIMT were analyzed as continuous variables. CAC scores were analyzed as a binary variable (detectable vs non-detectable calcification) and as a log 10 Agatston score in persons with detectable CAC. Statistical significance was defined as P<0.05. Results: We included HIV+ patients (97%male; mean (SD) age 52.34 (6.98) years; 35% black) who had CIMT (n=562) or CAC (n=296) measured. Depending on the formula used, 6.1-6.9% had eGFR <60 mL/min/1.73m2, and 32.8-51.6%

had eGFR <90 mL/min/1.73m2. Mean (SD) CIMT was 0.79 (0.15) mm. Of the 296 with CAC measures, 145 (49%) had CAC, with a mean (SD) log 10 Agatston score of 1.85 (0.79). After adjusting for traditional CVD risk factors, demographics, and HIV parameters, each 10 mL/min/1.73m2 lower eGFR by Cr-CC, by Cr, and by CC was associated, respectively, with an 8.3µm (95% CI, 0.6-16µm; P=0.035), a 7.6µm (95%CI, 0.0-1.51µm; P=0.052), and a 5.4µm (95% CI, -1.5-12.3µm; P=0.124) higher CIMT. Each 10 mL/min/1.73m2 lower CC-eGFR was associated with higher log 10 Agatston score of 0.0941 (95% CI 0.0151-0.173; P=0.022), but associations with other eGFR formulae did not reach significance. eGFR was not associated with detectable CAC. Conclusion: In this group of HIV+ patients who were predominantly male and with eGFR ≥60mL/min/1.73m2, lower eGFR was significantly associated with higher CIMT when using the Cr-CC formula and with higher CAC scores using the CC formula. GFR estimating formulae incorporating cystatin C may identify HIV+ patients with subclinical CVD and for whom there is a greater need of aggressive CVD risk reduction. Angsana Phuphuakrat , Artit Wongsa, Sirawat Srichatrapimuk, Boonrat Tassaneetrithep, Sasisopin Kiertiburanakul, Somnuek Sungkanuparph Mahidol University, Bangkok, Thailand Background: Despite undetectable plasma viral load, HIV-infected individuals receiving antiretroviral drug (ARV) have chronic immune activation and persistent low-grade inflammation. This has been associated with increased risk of cardiovascular diseases. Pitavastatin is a newer statin that has less drug- drug interactions with ARV compared with older statins. Thus, it is a preferred drug for the treatment of dyslipidemia in patients with HIV. Data regarding anti-inflammatory effect of pitavastatin in HIV-infected patients are limited. We studied the effects of pitavastatin on atherosclerotic-associated cellular inflammatory biomarkers in virologically-suppressed HIV-infected individuals. Methods: This study was an exploratory analysis of atherosclerotic-associated inflammatory cellular biomarkers. The study was a substudy of a randomized, double-blind, crossover study that evaluated the effect of pitavastatin versus placebo in HIV-infected dyslipidemic patients, who received atazanavir/ ritonavir-based antiretroviral agents (ClinicalTrials.gov NCT02442700). Patients were randomized to receive 12 weeks of pitavastatin 2 mg/day or placebo, followed by 2 weeks of washout period and 12 weeks of another treatment arm. Blood collected at 12 weeks of treatment were analyzed for atherosclerotic- associated cellular inflammatory biomarkers on a flow cytometer. Comparisons of the biomarkers between patients receiving pitavastatin and placebo treatment were performed by Wilcoxon signed ranks test. Results: Twenty-four HIV-infected individuals were included. Median (interquartile range; IQR) age of the patients was 46 (41-56) years and 14 (58%) patients were men. Median (IQR) baseline CD4+ lymphocyte counts was 662 (561-836) cells/mm 3 . As compared to placebo, treatment with pitavastatin resulted in significantly lower the proportions of patrolling (CD14DimCD16+) monocytes (p=0.018) and PD1+CD4+ T cells (p=0.029). However, no significant difference in the proportions of HLA-DR+CD38+ CD4+ T cells, HLA-DR+CD38+ CD8+ T cells, PD1+CD8+ T cells, and Treg was found. Conclusion: This preliminary study shows that pitavastatin lowers the proportions of patrolling monocyte and PD1+CD4+ T cells in virologically- suppressed HIV-infected individuals. Further study on effects of pitavastatin to prevent cardiovascular diseases in HIV-infected individuals should be investigated. 690 STATIN COVERAGE IN AN HIV COHORT: COMPARISON OF ATP III, ACC/ AHA, AND NLA GUIDELINES Matthew E. Levy , Alan Greenberg, Manya Magnus, Naji Younes, Amanda D. Castel The George Washington University, Washington, DC, USA Background: Dyslipidemia is a major cardiovascular disease risk factor that is highly prevalent among HIV-infected populations. Statin coverage has been examined among HIV-infected patients using 2004 Adult Treatment Panel III (ATP III) and 2013 American College of Cardiology/American Heart Association (ACC/AHA) guidelines, yet the impact of 2014 National Lipid Association (NLA) guidelines has yet to be examined. We investigated statin eligibility, prescribing practices, and clinical responses using these three guidelines. 689 EFFECTS OF PITAVASTATIN ON ATHEROSCLEROTIC-ASSOCIATED BIOMARKERS IN PEOPLE WITH HIV

Poster Abstracts

CROI 2018 256