CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Conclusions: PCSK9 is increased in HIV-infected individuals. Traditional risk factors, HCV, and Lp(a)>90 nmol/L were also independently associated with PCSK9. Future studies should explore whether PCSK9 inhibition may be used to treat dyslipidemia and statin resistance in HIV+ patients. 732 Enhanced, Not Inhibited Monocyte Cholesterol Efflux Characterises Untreated HIV Jane A. O’Halloran 2 ;Therese Herlihy 2 ; Alan Macken 2 ; Louise Rainford 2 ; John S. Lambert 2 ; Gerard J. Sheehan 2 ; Niall G. Mahon 2 ; Leo P. Lawler 2 ; PatrickW. Mallon 2 1 University College Dublin, Dublin, Ireland; 2 University College Dublin, Dublin, Ireland; 3 University College Dublin, Dublin, Ireland; 4 University College Dublin, Dublin, Ireland; 5 University College Dublin, Dublin, Ireland Background: Dyslipidaemia in untreated HIV infection is characterised by reduced high density lipoprotein cholesterol (HDL) and increased risk of cardiovascular disease (CVD). In vitro, HIV impairs monocyte cholesterol efflux (MCE) onto apolipoprotein A1 (ApoA1) via the ATP-binding cassette transporter A1 (ABCA1) potentially explaining lower HDL. We aimed to determine if MCE was inhibited in untreated HIV in vivo. Methods: Using a novel, dynamic ex vivo assay, we compared MCE in HIV positive (HIVpos) subjects not on antiretroviral therapy (ART) and HIV negative (HIVneg) controls matched for age, gender, race, smoking and hepatitis C status. Monocytes were isolated from fasting blood and monocyte intracellular cholesterol (MIC) was measured by florescence and corrected for total cell count before and after cholesterol loading (T=0, 2, 4, 6, 24 hours post loading). MCE was calculated as a ratio of extracellular (supernatant) cholesterol to MIC (EC T : MIC T ) with an additional 24 hr measure in the presence of ApoA1 (EC apoA1 : MIC apoA1 ). Changes in MCE were correlated with lipids and carotid intima-media thickness (C-IMT). Data are median [IQR]. Comparisons were made using non-parametric analyses. Results: We recruited 50 HIVpos subjects (52% homosexual, 36% heterosexual; CD4+ 410[268, 588] cells/mm 3 ; log HIV RNA 4.01[3.52, 4.78] copies/ml) and 50 matched controls. The HIVpos group had significantly lower total, low density lipoprotein and HDL cholesterol but similar triglycerides and C-IMT (table 1). There was no significant between-group difference in fasting MIC (HIVpos 2.0 [1.6, 2.4] versus HIVneg 1.8 [1.6, 2.4] pg/cell, p =0.53) or post cholesterol loaded MIC (HIVpos 7.1 [5.5, 9.6] versus HIVneg 6.7 [5.1, 9.7], p =0.39). However, MCE was significantly and consistently greater in the HIVpos group over time (table 1). The addition of ApoA1 increased EC 24 :MIC 24 in both groups, with no between-group difference observed. Higher HDL correlated with lower EC T : MIC T ratio (T 2 r =-0.41, T 4 r =-0.34, T 6 r =-0.28, T 24 r =-0.29, all p ≤ 0.005). Neither C-IMT nor HIV RNA correlated with MCE.

Table 1. Baseline Characteristics and Monocyte Cholesterol Efflux by HIV status Conclusions: These data suggests that untreated HIV is characterised by enhanced rather than decreased MCE, with higher MCE correlating with lower HDL. This unexpected finding may reflect up-regulation of MCE pathways compensating for any potential negative effect of HIV on ABCA1-medicated cholesterol efflux. Further research into the pathways involved, the effects of ART and the impact of these findings on CVD pathogenesis is required. 733 Rosuvastatin vs Protease Inhibitor Switch for Hypercholesterolemia: Randomised Trial Frederick J. Lee 7 ; Polyana Monteiro 3 ; David Baker 1 ; Mark Bloch 2 ; Robert Finlayson 8 ; Richard Moore 5 ; Norman Roth 6 ; Jennifer F. Hoy 4 ; Esteban Martinez 3 ; Andrew Carr 7 1 East Sydney Doctors, Sydney, Australia; 2 Holdsworth House Medical Practice, Sydney, Australia; 3 Hospital Clinic, University of Barcelona, Barcelona, Spain; 4 Monash University/Alfred Hospital, Melbourne, Australia; 5 Northside Clinic, Melbourne, Australia; 6 Prahran Market Clinic, Melbourne, Australia; 7 St. Vincent’s Hospital, Sydney, Australia; 8 Taylor Square Private Clinic, Sydney, Australia Background: Optimal management of hypercholesterolaemia in adults receiving a ritonavir-boosted protease inhibitor (rPI) is unknown. The two proven options, statin therapy and rPI switching to a lipid-neutral alternative, have not been compared since the advent of current-generation statins and PIs, despite multiple studies (mostly industry- sponsored) focussed upon rPI switching. Methods: Adults on rPI-based therapy with plasma viral load <50 cp/mL for ≥ 6 months, total cholesterol ≥ 5.5 mmol/L (213 mg/dL) and elevated cardiovascular risk (Framingham score ≥ 8% at 10 years or family history of premature cardiac disease), and not on lipid-lowering therapy, were randomised to open-label rosuvastatin 10 mg/day or to switch rPI, stratified by cholesterol (> or ≤ 7.0 mmol/L [272 mg/dL]) and rPI type (atazanavir or other); all subjects received standardised diet/lifestyle advice. The primary endpoint was change in fasting total cholesterol at Week 12 (ITT analysis, Wilcoxon rank-sum test); we hypothesized that rosuvastatin would be more effective. Final analyses are reported. Results: Baseline characteristics (mean [SD] or n [%]) of the 43 subjects were: age 55 (8.5) years; n=42 (98%) male; n=41 (95%) white race; total cholesterol 6.2 mmol/L (240 mg/ dL). Baseline rPI types were lopinavir (n=22; 51%), atazanavir (n=12; 28%) and darunavir (n=9; 21%). Within the switch group, the most common rPI substitutes were raltegravir (n=10; 50%) and rilpivirine (n=6; 30%). All subjects adhered to strategy through Week 12. By Week 4, rosuvastatin resulted in greater declines in total cholesterol (24.1% vs. 7.3%, p <0.001), LDL cholesterol, and total:HDL cholesterol ratio, than rPI switch. These changes were maintained at Week 12 ( Table ), with trends to greater declines in cardiovascular risk scores, despite similar weight changes. Conversely, rPI switch caused greater falls in VLDL cholesterol and triglycerides, but more study drug-related adverse events (11 vs. 1, p =0.001 [ χ 2 ]; mostly grade 1 nausea or diarrhoea). HIV viral load rose to 360 cp/mL in 1 rosuvastatin subject by Week 12, but was <50 cp/mL 3 months later without therapy change. One serious adverse event occurred in each arm, both unrelated to study drug/switch. No grade 3-4 laboratory adverse event was seen. No adverse event led to discontinuation of rosuvastatin or switch drug.

Poster Abstracts

451

CROI 2015