CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

645 SOCIOECONOMIC STATUS ASSOCIATES WITH ARTERIAL INFLAMMATION IN HIV Lili Zhang 1 , Amrit Narwan 2 , Nicki Naddaf 1 , Smruti Rahalkar 2 , Tomas Patrich 1 , Michael Osborne 1 , Steven G. Deeks 2 , Ahmed Tawakol 1 , Priscilla Hsue 2 1 Massachusetts General Hospital, Boston, MA, USA, 2 University of California San Francisco, San Francisco, CA, USA Background: Socioeconomic status (SES) is associated with higher mortality among individuals living with HIV. In the general population, lower SES associates with higher arterial inflammation (a key driver of atherosclerotic disease), and a greater cardiovascular disease risk. While higher arterial inflammation has been reported in treated HIV, the relationship between SES and arterial inflammation has not been studied. Methods: Men living with HIV were recruited from the SCOPE (Observational Study of the Consequences of the Protease Inhibitor Era), a clinic-based cohort of individuals receiving care in San Francisco. Arterial inflammation was measured using 18F-fluorodeoxyglucose (FDG-PET) positron emission tomography, as the uptake of FDG in the wall of the ascending aorta corrected for background. Zip-code-level SES measures were derived from the U.S. Census Bureau. Multivariable linear regression was utilized to assess the association between SES and arterial inflammation; mediation analysis was used to test whether systemic inflammation mediated that relationship. Results: Thirty-nine virologically-suppressed men living with HIV were studied (mean age of 50.5±11.1 years). The median CD4 count was 663cells/ mm3 (IQR: 399- 922); 82%were receiving antiretroviral therapies. Local median income inversely associated with arterial inflammation (standardized β [CI]:-0.400 [-0.757, -0.091], p=0.014, Fig 1A) after multivariable adjustment (age, Framingham risk score, statin use, antiretroviral use, current and nadir CD4 count). Similarly, after multivariable adjustment, % high school graduates inversely associated with arterial inflammation (-0.465 [-0.808, -0.161], p=0.005, Fig 1B) and CRP (-0.400 [-0.757, -0.091], p=0.014). Mediation analysis demonstrated the impact of SES on arterial inflammation is mediated by heightened systemic inflammation, namely: ↓ SES (as graduation rate) →↑ CRP →↑ arterial inflammation accounting for 44% of the total effect (p<0.05). Conclusion: In individuals living with HIV, community-level SES factors associate significantly with arterial inflammation, independently of traditional risk factors, statin therapy, and level of HIV disease control. The link between lower SES and arterial inflammation appears to be mediated by increased systemic inflammation. Strategies to recognize SES as a CV risk factor in HIV as well as targeted interventions may be helpful in reducing HIV-associated arterial inflammation as well as clinical CV events.

Poster Abstracts

646 INFLAMMATORY PHENOTYPES PREDICT PULSE WAVE VELOCITY CHANGE ON ART IN MALAWIAN ADULTS Christine Kelly 1 , Willard Tinago 1 , Alejandro A. Garcia 1 , Patricia Hunter 2 , Dagmar Alber 2 , Natasha Luckhurst 3 , Jake Connolly 3 , Francesca I. Arrigoni 3 , Raphael Kamng’ona 4 , Patrick W. Mallon 1 , Henry Mwandumba 4 , Sarah Walker 2 , Saye Khoo 5 , Nigel Klein 2 , for the REALITY study team 1 University College Dublin, Dublin, Ireland, 2 University College London, London, UK, 3 Kingston University, London, UK, 4 University of Malawi, Blantyre, Malawi, 5 University of Liverpool, Liverpool, UK Background: Inflammation has been linked to vascular dysfunction and increased risk of cardiovascular disease. In low-income settings, drivers of inflammation are multiple, with infectious and environmental factors contributing. We hypothesise that adult people living with HIV (PLWH) in sub-Saharan Africa starting ART with advanced immunosuppression can be stratified into inflammatory phenotypes that predict changes in vascular dysfunction on ART, as measured by pulse wave velocity(PWV). Methods: We recruited PLWH with CD4<100 cells/ul two weeks after starting ART in the REALITY trial (NCT01825031). PWV was recorded 2, 10, 24 and 42 weeks post ART. We measured markers of cell surface immune activation by flow cytometry and plasma inflammation markers by electrochemiluminescence at week 2. We identified inflammatory phenotypes using principle components analysis of 22 different markers, using linear mixed models to explore associations between inflammation clusters and change in PWV with ART. Results: In 260 of 279 PLWH with available biomarker data we identified three clusters representing 59 (cluster 1), 194 (cluster 2) and 7 (cluster 3) subjects (Figure 1A). Cluster 1 showed markedly higher CD4 and CD8 T cell expression of HLADR and PD1 vs clusters 2 and 3 (HLADR: CD4 86% vs 69%, CD8 84% vs 72%; PD1: CD4 69% vs 39%, CD8 54% vs 33% respectively; all p<0.0001). Although small, subjects in cluster 3 had significantly higher levels of inflammatory cytokine pathways (IL6, IFN γ , IP10, IL1RA, IL10), chemotaxis (IL8), systemic and vascular inflammation (CRP, ICAM1, VCAM1) and SAA (all p<0.001); and marginally lower pre-ART CD4 (17 vs 42 cells/mm3, p=0.08). Baseline PWV was statistically lower in cluster 3 (6.3m/s vs 7.6, p=0.009), but increased over 42 weeks (log change 0.1m/s vs -0.5, p=0.07, Fig 1B). In mixed models, IL1RA was independently associated with lower baseline PWV (log -0.32m/s per pg/ml higher, p=0.02) and attenuated decline in PWV by week 24 (change in log slope +0.39m/s per pg/ml higher, p=0.01). Cluster 3 also had lower adjusted baseline

CROI 2019 245