Oral Communications August 2015 e3 CARDIOVASCULAR RISK AND ANDROGEN DEPRIVATION THERAPY FOR PROSTATE CANCER: SYSTEMATIC REVIEW AND META-ANALYSIS OF RANDOMISED CONTROLLED TRIALS AND OBSERVATIONAL STUDIES (METADTCR) L.-M. Scailteux; F. Naudet; Q. Alimi; S. Vincendeau; and E. Oger Rennes University Hospital, Rennes, France Introduction: Androgen deprivation therapy (ADT) is the corner- stone therapy in advanced prostate cancer management. Currently, to study cardiovascular risk, many studies pooled ADT modalities, sometimes with orchiectomy. Our objective was to compare the coronary and cerebrovascular risk (myocardial infarction, ischemic stroke) and the cardiovascular and overall mortality across the dif- ferent ADT modalities. PROSPERO Registration: CRD42014010598. Methods: We performed a literature search of randomized controlled trials (RCTs) and observational studies using MEDLINE and Embase since 1950 to July 28, 2014, without language restriction provided that they gave data on prostate cancer patients comparing one ADT modality to another or radiotherapy or total prostatectomy or pla- cebo. ADT modalities were GnRH agonists, GnRH antagonists, antiandrogens (steroidal or nonsteroidal), and newer drugs (abira- terone, enzalutamide). Orchiectomy was the relevant alternative. For observational studies pooling several ADT modalities, details on each ADT group were requested to the author. We will also perform a network meta-analysis including RCTs. Results: Among 3614 abstracts, 47 cohorts fulfilled inclusion criteria and 8 provided sufficient data to be analysed. One study gave details on cardiovascular death and 2 on only coronary and cerebrovascular risk, but cardiovascular events definition were too heterogeneous to be pooled (one study added arrhythmia and heart failure to ischemic heart disease). For the 6 other studies, as they did not compare the same modalities of ADT, they were not meta-analysed for overall survival. A total of 153 abstracts mentioning RCT fulfilled inclusion criteria, and data analysis is ongoing. Conclusions: Data from observational studies did not support con- sistent evidence that any particular ADT modality may increase car- diovascular risk or overall survival. We are conducting a nationwide population-based prospective cohort of 4 years thanks to the French Health Reimbursement Agency database, which allowed us to investi- gate more in depth the association between different ADT modalities and cardiovascular risk. CARBOXYLESTERASE 1 C.428G>A SINGLE NUCLEOTIDE VARIATION REDUCES HYDROLYSIS OF CLOPIDOGREL AND ENALAPRIL, BUT NOT THAT OF QUINAPRIL E.K. Tarkiainen; M.T. Holmberg; A. Tornio; M. Neuvonen; T. Launiainen; P.J. Neuvonen; J.T. Backman; and M. Niemi University of Helsinki and Helsinki University Hospital, Helsinki, Finland Background: Carboxylesterase 1 (CES1) hydrolyzes about 90% of the prodrug clopidogrel to inactive carboxylic acid metabolite and about 40% to 60% of the prodrugs quinapril and enalapril to their active metabolites. In vitro studies have shown that the CES1 c.428G>A (p.G143E, rs71647871) single-nucleotide variation (SNV) can markedly affect metabolism of clopidogrel and ACE inhibitors. Materials and Methods: We studied pharmacokinetics and pharmacodynamics of 600-mg oral clopidogrel, 10-mg oral quinapril, and 10-mg oral enalapril in 10 carriers and 12 noncar- riers of the CES1 c.428G>A SNV. Clopidogrel, its carboxylic acid acyl-β -D-glucuronide, and active cis 5-thiol metabolite plasma concentrations and platelet aggregation were measured for up to 12 hours. Quinapril and quinaprilat plasma concentrations were meas- ured for up to 24 hours and those of enalapril and enalaprilat for up to 48 hours. Results: Clopidogrel carboxylic acid to clopidogrel area under the plasma concentration-time curve from 0 h to infinity (AUC 0-∞ ) ratio was 53% smaller in CES1 c.428G/A carriers than in noncarriers (P = 0.009), indicating impaired hydrolysis of clopidogrel. Consequently, AUC 0-∞ of clopidogrel and its active cis 5-thiol metabo- lite were 123% (P = 0.004) and 67% (P = 0.009) larger in c.428G/A carriers than in noncarriers. Consistent with pharmacokinetics, aver- age inhibition of P2Y 12 -mediated platelet aggregation 0-12 h after clopidogrel intake and maximum observed platelet inhibition were 19 percentage points higher in c.428G/A carriers than in noncarri- ers (P = 0.036 and P = 0.041, respectively). AUC 0-∞ of enalaprilat was 20% lower in CES1 c.428G/A carriers than in noncarriers (P = 0.049). The CES1 c.428G>A genotype had no significant effect on quinapril pharmacokinetics. Conclusions: The CES1 c.428G>A SNV increased clopidogrel active metabolite concentrations and antiplatelet effects by reducing hydrolysis of parent clopidogrel to inactive metabolites. Therefore, the CES1 c.428A allele may increase clopidogrel efficacy and bleeding risk. The CES1 c.428G>A SNV decreased active enalaprilat concen- trations by reducing the hydrolysis of enalapril, but had no observ- able effect on quinapril pharmacokinetics. POPULATION PHARMACOKINETIC MODELING AND SIMULATIONS OF LONG-ACTING INTRAMUSCULAR RISPERIDONE ISM ® J. Winkler 1 ; E. Snoeck 1 ; J. Llaudó Garin 2 ; I. Ayani 2 ; J. Martínez-González 2 ; and I. Gutierro Adúriz 2 1 SGS Exprimo NV, Mechelen, Belgium; and 2 Laboratorios Farmaceuticos ROVI, S.A., Madrid, Spain Background/Introduction: The aim of the project was to develop a population PK model for a novel and long-acting intramuscular (i.m.) formulation (ISM ® ) of the atypical antipsychotic risperidone, including a relatively complex absorption profile combined with a previously published population PK disposition model of risperi- done and its metabolite 9-OH-risperidone after oral risperidone in schizophrenic patients. Material and Methods: A population PK analysis for risperidone ISM ® using Monolix software was conducted based on 1520 plasma samples from two single-dose studies, ROV-RISP-2009-01(17 healthy subjects) and PRISMA-1(48 schizophrenic subjects). Simulations were subsequently undertaken predicting the steady-state PK expo- sure of active moiety after risperidone ISM ® . Results: The PK disposition models of risperidone and 9-OH-risperidone consisted of two compartments for each com- pound. Four flexible depot compartments were included to describe the relatively complex absorption profile. Exponential models described IIV for the structural model parameters, and an addi- tive error model in the log domain described residual variability. The influence of the CYP2D6 genotype on the formation rate of 9-OH-risperidone was included in form of a mixture model fixing parameters based on the previously published population PK model. Goodness of fit plots indicated that the model described the data well. The residual error was relatively low with 18% to 19% CV. Simulations were subsequently undertaken, predicting steady-state PK exposure of active moiety after multiple doses of 37.5 to 150 mg risperidone ISM ® administered every 28 days for 12 weeks. In addi- tion, different possible dose de-escalation schemes with a starting dose of 100 mg risperidone ISM ® were simulated.