CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

661

Abacavir (ABC) Use and Risk of Recurrent Myocardial Infarction (MI) Caroline Sabin 1 ; Lene Ryom 2 ; Ole Kirk 2 ; Antonella d’Arminio Monforte 3 ; Christian Pradier 4 ; RainerWeber 5 ; Andrew N. Phillips 1 ; Francois Dabis 6 ; Jens D. Lundgren 2 ; for the D:A:D Study Group 1 Univ Coll London, London, UK; 2 Rigshospitalet, Univ of Copenhagen, Copenhagen, Denmark; 3 Infectious Diseases Clinic, San Paolo Hosp, Univ of Milan, Milan, Italy; 4 Nice Univ Hosp, Nice, France; 5 Univ Hosp Zurich, Zurich, Switzerland; 6 INSERM U897, ISPED, Univ de Bordeaux, Bordeaux, France Background: Whilst several studies have reported that current exposure to ABC is associated with an increased risk of MI, associations with the risk of a subsequent MI have not been investigated. Methods: We considered the rate of recurrent MI among 816 D:A:D participants who experienced an MI during study follow-up and who remained under follow-up at 28 days post-MI. Follow-up was considered from 28 days post-MI to the date of next MI, death, 1/2/2014 or 6 months after last clinic visit. Poisson regression models considered associations between recurrent MI and exposure to ABC (use at initial MI, current post-MI exposure and cumulative exposure), before and after adjusting for year and age. Results: Included individuals were largely male (91.4%), infected with HIV through sex between men (59.6%) and had a median age of 51 years (inter-quartile range [IQR] 44-58) at initial MI. 80.1% of participants were current/ex-smokers. Median CD4 at initial MI was 503 (IQR 340-728) cells/mm 3 and 66.8% had a HIV RNA <50 cps/ml. 415 people (50.9%) had received ABC prior to initial MI for a median of 3.1 years (IQR 0.1-13.9). Of the 277 (34.0%) still on ABC at initial MI, 204 (73.7%) subsequently stopped it at a median of 337 (0-3616) days post-MI. There were 102 recurrent MIs over 3863 person-years (PY, rate 2.64/1000 PY, 95% confidence interval [95%CI] 2.13-3.15). Rates of recurrent MI were 2.75 (1.90-3.60) and 2.57 (1.93-3.21)/1000 PY in those who were and were not on ABC at initial MI, and 3.47 (2.37-4.57) and 2.31 (1.75-2.88)/1000 PY in those who were and were not currently receiving ABC post-MI. Whilst neither cumulative exposure to ABC nor receipt of ABC at initial MI were associated with recurrent MI risk, current post-MI exposure was associated with an increased risk (Table). With the exception of age (1.02 (1.00-1.04)/5 years), there were no significant associations between demographic/lifestyle factors and recurrent MI. Earlier initial MI was, however, associated with an increased risk (1999-2001 vs. 2011-2013: 12.32 (5.17-29.37)). The association between recent ABC use and recurrent MI risk was similar after controlling for age, but attenuated after controlling for calendar year (Table). Conclusions: Whilst we found some evidence that use of ABC post-MI was also associated with an elevated risk of a recurrent MI, this appeared to be largely explained by greater use of ABC in those with an MI in the earlier years of the study.

662 Cardiovascular Risk Profile of Abacavir and Tenofovir Independent of HIV Infection Michael Emerson 1 ; Erica Smyth 1 ; Mark Nelson 2 1 Imperial Coll London, London, UK; 2 Chelsea and Westminster Hosp NHS Fndn Trust, London, UK

Background: Observational and clinical studies suggest that abacavir sulphate (ABC), a component of antiretroviral therapy (ART) may be associated with a reversible increased risk of myocardial infarction (MI) and increased platelet aggregation in HIV positive patients. It is not clear whether increased cardiovascular risk is driven pharmacologically by ABC or pathophysiologically by HIV and co-morbidities. Since MI is platelet-driven, our hypothesis was that ABC increases cardiovascular risk via pharmacological modulation of platelet aggregation. Methods: The direct effect of ART on platelets, independent of HIV infection, was determined by assessing aggregation of isolated human platelets from non-infected volunteers in the presence of approximate Cmax concentrations of metabolites of ABC or tenofovir disoproxil fumarate (TDF) . The ability of ABC and TDF to interrupt inhibition of platelet aggregation by the endogenous negative regulator nitric oxide (NO) was also compared. In addition, platelet thromboembolismwas assessed in the circulation of animals treated with ABC or TDF allowing for assessment of the effects of ART in the presence of endogenous NO. Results: Tenofovir (TFV, metabolite of TDF that is converted to active metabolite intracellularly) significantly inhibited isolated platelet aggregation in vitro, however, no effect was detected with carbovir triphosphate (CT, active metabolite of ABC). CT blocked NO-mediated inhibition of platelet aggregation, in contrast no significant effect was observed for TFV. Administration of ABC to mice significantly enhanced platelet thromboembolism 30 mins after treatment. The effect of ABC in mice dissipated within 4 hours indicating a reversible effect. No effect was observed following treatment of mice with TDF at any time point. Conclusions: The increased cardiovascular risk associated with ABC in patient studies may be mediated by reversible pharmacological modulation of platelet and endothelial function rather than by HIV infection or differences in confounding factors between patient groups. In contrast, TDF exerts effects upon platelets that would not be expected to increase the incidence of platelet-driven events such as MI. 663 Abacavir Induces Platelet-Endothelium Interactions Through Endothelial P2X7 Receptors Cesar Rios-Navarro 1 ; Samuel Orden 1 ;Victor Collado-Diaz 1 ; María Ángeles Martínez-Cuesta 1 ; JuanV. Esplugues 2 ; Angeles Alvarez 1 1 Univ of Valencia, Valencia, Spain; 2 Univ de Valencia, Valencia, Spain Background: Controversy surrounding the epidemiological association of Abacavir (ABC) with cardiovascular diseases is fuelled by the lack of clear evidence concerning the underlying mechanisms. We have reported that ABC induces leukocyte-endothelial cell interactions. This is relevant, as thrombus formation also occurs as a result of the interplay between platelets and endothelial cells. Given the chemical similarity between then drug and purinergic mediators (ATP, ADP and AMP), we have now evaluated the role of ATP- receptors on platelet-endothelial cell interactions induced by ABC. Methods: Human umbilical vein endothelial cells (HUVEC) and washed platelets were treated with clinical concentrations of ABC (0.5 - 5 µg/ml) and platelet-endothelial interactions were evaluated using an in vitro flow chamber system. To determine the cell type involved, each one was treated individually with ABC. To assess the implication of purinergic receptors, cells were pre-treated prior to administration of ABC (5 µg/ml) with non-selective (Suramin) or selective P2X7 (A804598) or P2X2/3 (A317491) ATP-receptor antagonists in the case of HUVEC, and with selective P2Y1 (MRS2500) or P2Y12 (Clopidogrel) or P2X1 (NF449) ADP or ATP-receptor antagonists in the case of platelets.

Poster Abstracts

272

CROI 2016