Wednesday June 13 - the Population Approach Group in Europe
Audience Input for the PAGE 2008 Program .... P2-25 Erik Olofsen The
performance of model selection criteria in the ...... Since the PK of both TE and TD
could not be adequately characterised, it was ..... The second approach tests
whether the NPDE follow a N(0, 1) distribution after ...... Invest New Drugs 2005;
23: 225-34.
part of the document
Tuesday June 1218:30-20:30Reception at City Hall (for registered participants only)Wednesday June 1308:00-09:00Registration09:00-09:20Steen Ingwersen: Welcome and Introduction09:20-10:20Stuart Beal Methodology Sessionchair: Pascal Girard09:20-09:40Robert Bauer HYPERLINK "http://www.page-meeting.org/?abstract=1111" Advanced Population Analysis Features in the S-ADAPT/MCPEM Program09:40-10:00Fahima Nekka HYPERLINK "http://www.page-meeting.org/?abstract=1068" What Additional Information Can we Retrieve When Compliance is Accounted For? An explicit Compliance-Pharmacokinetic Formalism10:00-10:20Marc Lavielle HYPERLINK "http://www.page-meeting.org/?abstract=1192" The SAEM algorithm and its implementation in MONOLIX 2.110:20-11:50Coffee break, Poster and Software session IPosters in Group I are accompanied by their presenter11:50-12:30Oral contributions: Applicationschair: Mats Karlsson11:50-12:10Ashley Strougo HYPERLINK "http://www.page-meeting.org/?abstract=1092" Mechanism-based model of the effect of co-administration of exogenous testosterone and progestogens on the hypothalamic-pituitary-gonodal axis in men12:10-12:30Joe Standing HYPERLINK "http://www.page-meeting.org/?abstract=1086" Developmental Pharmacokinetics of Diclofenac for Acute Pain12:30-13:30Lunch13:30-14:10Nick Holford: Beginners Tutorial Modeling Disease Progression14:10-15:40Tea break, Poster and Software session IIPosters in Group II are accompanied by their presenter15:40-17:30Modeling Disease and Disease Progressionchair: Nick Holford15:40-16:10Claudio Cobelli HYPERLINK "http://www.page-meeting.org/?abstract=1226" Models of Glucose Metabolism and Control in Diabetes16:10-16:30Philippe Jacqmin HYPERLINK "http://www.page-meeting.org/?abstract=1194" Basic PK/PD principles of proliferative and circular systems16:30-16:50Kim Stuyckens HYPERLINK "http://www.page-meeting.org/?abstract=1185" Modeling Drug Effects and Resistance Development on Tumor Growth Dynamics16:50-17:10Kristin Karlsson HYPERLINK "http://www.page-meeting.org/?abstract=1191" Modelling of disease progression in acute stroke by simultaneously using the NIH stroke scale, the Scandinavian stroke scale and the Barthel index17:10-17:30Klaas Prins HYPERLINK "http://www.page-meeting.org/?abstract=1221" Integrated Modeling and Simulation of Clinical Response and Drop-out of D2 receptor agonists in Patients with Early Parkinsons Disease.
Thursday June 1409:00-10:20Lewis Sheiner Student Sessionchair: France Mentré, Nick Holford, René Bruno, Tim Sheiner09:00-09:25Karl Brendel HYPERLINK "http://www.page-meeting.org/?abstract=1085" Normalized Prediction Distribution Error for the Evaluation of Nonlinear Mixed-Models09:25-09:50S. Y. Amy Cheung HYPERLINK "http://www.page-meeting.org/?abstract=1083" Identifiability Analysis and Parameter List Reduction of a Nonlinear Cardiovascular PKPD Model09:50-10:15Radojka Savic HYPERLINK "http://www.page-meeting.org/?abstract=1087" Importance of Shrinkage in Empirical Bayes Estimates for Diagnostics and Estimation: Problems and Solutions10:15-10:25Presentation of Awards10:25-11:55Coffee break, Poster and Software session IIIPosters in Group III are accompanied by their presenter11:55-12:25Tim Sheiner: HYPERLINK "http://www.page-meeting.org/?abstract=1112" Planning To Communicate12:25-12:30Stacy TannenbaumPreview of American Conference on Pharmacometrics12:30-13:30LunchTutorial13:30-14:30Marc Buyse HYPERLINK "http://www.page-meeting.org/?abstract=1223" Validation of statistically reliable biomarkers14:30-16:00Tea break, Poster and Software session IVPosters in Group IV are accompanied by their presenter16:00-17:20Oral contributions: Applicationschair: Charlotte Kloft16:00-16:20Celine Dartois HYPERLINK "http://www.page-meeting.org/?abstract=1101" Impact of handling missing PK data on PD estimation explicit modeling of BLQ data in WinBUGS® reduced bias in the PD predictions - a preclinical example.16:20-16:40Teun Post HYPERLINK "http://www.page-meeting.org/?abstract=1098" Circadian rhythm in pharmacodynamics and its influence on the identification of treatment effects16:40-17:00Benoit You HYPERLINK "http://www.page-meeting.org/?abstract=1119" Kinetic models of PSA decrease after surgery in prostate tumor diseases as a help for clinician interpretation17:00-17:20Justin Wilkins HYPERLINK "http://www.page-meeting.org/?abstract=1113" A flexible approach to modeling variable absorption in the context of repeated dosing: illustrated with rifampicin18:00-24:00Social evening
Friday June 1509:00-10:00Methodology: Designchair: France Mentré09:00-09:20Joakim Nyberg HYPERLINK "http://www.page-meeting.org/?abstract=1160" Sequential versus simultaneous optimal experimental design on dose and sample times09:20-09:40Stefano Zamuner HYPERLINK "http://www.page-meeting.org/?abstract=1099" Optimal Design to Estimate the Time Varying Receptor Occupancy Relationship in a PET Experiment09:40-10:00France Mentré HYPERLINK "http://www.page-meeting.org/?abstract=1179" Software for optimal design in population pharmacokinetics and pharmacodynamics: a comparison10:00-10:05Preview of PAGE 200810:05-11:05Coffee break11:05-12:20Model Building Sessionchair: Dinesh de Alwis, Niclas Jonsson 11:05-11:30Jeroen Elassaiss-Schaap HYPERLINK "http://www.page-meeting.org/?abstract=1079" Interspecies Population Modeling Of Pharmacokinetic Data Available At The End Of Drug Discovery11:30-11:55Chantaratsamon Dansirikul HYPERLINK "http://www.page-meeting.org/?abstract=1142" Insulin secretion and hepatic extraction during euglycemic clamp study: modelling of insulin and C-peptide data11:55-12:20Massimiliano Germani HYPERLINK "http://www.page-meeting.org/?abstract=1123" A population PK-PD method for categorical data analysis of progesterone antagonist activity in the rabbit McPhails model12:20-12:30Steen Ingwersen: Closing Remarks12:30-12:45Audience Input for the PAGE 2008 Program
Software demonstrations
Bayer Technology ServicesInsightfulMango SolutionsNonLinear Mixed Effects ConsortiumSimCYPSystem for Population KineticsUSC*PACK - Roger Jelliffe HYPERLINK "http://www.page-meeting.org/?abstract=1211" The USC*PACK collection of BigWinPops software for nonparametric adaptive grid (NPAG) population PK/PD modeling, and the MM-USCPACK clinical software
Posters Wednesday Morning (group I)
Applications- Anti-infectives
P1-1 Julie Bertrand HYPERLINK "http://www.page-meeting.org/?abstract=1153" Influence of pharmacogenetic on pharmacokinetic interindividual variability of indinavir and lopinavir in HIV patients (COPHAR2 ANRS 111 trial) P1-2 Stefanie Hennig HYPERLINK "http://www.page-meeting.org/?abstract=1138" Tobramycin in paediatric CF patients - TCI or One dose fits all P1-3 Déborah Hirt HYPERLINK "http://www.page-meeting.org/?abstract=1107" Effect of CYP2C19 polymorphism on nelfinavir to M8 biotransformation in HIV patients. P1-4 Hui Kimko HYPERLINK "http://www.page-meeting.org/?abstract=1222" Population Pharmacokinetic Analysis To Support Dosing Regimens Of Ceftobiprole P1-5 Grant Langdon HYPERLINK "http://www.page-meeting.org/?abstract=1162" PK-PD modelling to support go/no go decisions for a novel gp120 inhibitor P1-6 Rocio Lledo HYPERLINK "http://www.page-meeting.org/?abstract=1097" Population Pharmacokinetics of Saquinavir in rats after IV and IP administration. An approach to Saquinavir/Ritonavir Pharmacokinetic interaction.
Applications- Biologicals/vaccines
P1-7 Balaji Agoram HYPERLINK "http://www.page-meeting.org/?abstract=1089" Application of mechanism-based population PKPD modelling in the rational selection of clinical candidates: an anti-IgE antibody example. P1-8 Lene Alifrangis HYPERLINK "http://www.page-meeting.org/?abstract=1150" Setting a Safe Starting Dose for a First-in-Man trial of a Monoclonal Antibody Based on Population PK-PD Predictions P1-9 Ekaterina Gibiansky HYPERLINK "http://www.page-meeting.org/?abstract=1159" Population Pharmacokinetics of Siplizumab (MEDI-507): Implications for Dosing P1-10 Ron Keizer HYPERLINK "http://www.page-meeting.org/?abstract=1205" Bioequivalence study of a C1-esterase-inhibitor product (Cetor®) with optimised sampling design P1-11 Wojciech Krzyzanski HYPERLINK "http://www.page-meeting.org/?abstract=1130" Pharmacodynamic Modelling of Recombinant Human Erythropoietin Effect on Reticulocyte Production Rate and Age Distribution in Healthy Subjects P1-12 Armel Stockis HYPERLINK "http://www.page-meeting.org/?abstract=1094" Population pharmacokinetics of certolizumab pegol
Methodology- Other topics
P1-13 Johan Areberg HYPERLINK "http://www.page-meeting.org/?abstract=1140" Simultaneous Population Pharmacokinetic Modelling of Parent Compound and Metabolite in Plasma and Urine for a New Drug Candidate P1-14 Martin Bergstrand HYPERLINK "http://www.page-meeting.org/?abstract=1201" A comparison of methods for handling of data below the limit of quantification in NONMEM VI P1-15 Robert Bies HYPERLINK "http://www.page-meeting.org/?abstract=1217" An MCPEM approach to understanding inter-animal and inter-treatment changes with in vivo striatal dopamine clearance in rats P1-16 Sophie Callies HYPERLINK "http://www.page-meeting.org/?abstract=1132" Modelling pharmacokinetic and pharmacodynamic properties of second generation antisense-oligonucleotides (ASOs) P1-17 Didier Concordet HYPERLINK "http://www.page-meeting.org/?abstract=1117" How to estimate population variance matrices with a Prescribed Pattern of Zeros? P1-18 Carine Crepin HYPERLINK "http://www.page-meeting.org/?abstract=1148" Elimination of anti-epileptic compounds in Marseille aquatic environment from private hospital effluent - modelling versus measurements P1-19 Mike Dunlavey HYPERLINK "http://www.page-meeting.org/?abstract=1076" Next-Generation Modeling Language P1-20 Iñaki F. Trocóniz HYPERLINK "http://www.page-meeting.org/?abstract=1078" Modelling Overdispersion and Markovian Features in Count Data P1-21 Clare Gaynor HYPERLINK "http://www.page-meeting.org/?abstract=1095" An assessment of prediction accuracy of two IVIVC modelling methodologies. P1-22 Ihab Girgis HYPERLINK "http://www.page-meeting.org/?abstract=1105" Parallel Bayesian Methodology for Population Analysis P1-23 Ihab Girgis HYPERLINK "http://www.page-meeting.org/?abstract=1182" Interactive Graphical Visualization Tool for Safety Data Screening P1-24 Thaddeus Grasela HYPERLINK "http://www.page-meeting.org/?abstract=1131" The Application of Systematic Analysis for Identifying and Addressing the Needs of the Pharmacometric Process P1-25 Serge Guzy HYPERLINK "http://www.page-meeting.org/?abstract=1070" Combining interoccasion variability and mixture within a MCPEM framework P1-26 Emilie Hénin HYPERLINK "http://www.page-meeting.org/?abstract=1193" Estimation of patient compliance from pharmacokinetic samples P1-27 Matt Hutmacher HYPERLINK "http://www.page-meeting.org/?abstract=1184" Comparing Minimum Hellinger Distance Estimation (MHDE) and Hypothesis Testing to Traditional Statistical Analyses a Simulation Study P1-28 Matt Hutmacher HYPERLINK "http://www.page-meeting.org/?abstract=1136" A new approach for population pharmacokinetic data analysis under noncompliance P1-29 Ivan Matthews HYPERLINK "http://www.page-meeting.org/?abstract=1109" Sensitivity analysis of a mixture model to determine genotype/phenotype P1-30 Chee Meng Ng HYPERLINK "http://www.page-meeting.org/?abstract=1096" A Comparison of Estimation Methods in Nonlinear Mixed-effect Model for Population Pharmacokinetic-pharmacodynamic Analysis P1-31 Chee Meng Ng HYPERLINK "http://www.page-meeting.org/?abstract=1135" A Systematical Approach to Bridge the Two-Stage Parametric Expectation Maximization Algorithm and Full Bayesian Three-Stage Hierarchical Nonlinear Mixed Effect Methods in Complex Population Pharmacokinetic/Pharmacodynamic Analysis: Troxacitabine-induced Neutropenia in Cancer Patients
P1-32 Stefaan Rossenu HYPERLINK "http://www.page-meeting.org/?abstract=1139" A mixture distribution approach to IVIVC modeling of a dual component drug delivery system P1-33 Alexander Staab HYPERLINK "http://www.page-meeting.org/?abstract=1124" How can routinely collected comedication information from phase II/III trials be used for screening of potential effects on model parameters? A case study with the oral direct thrombin inhibitor Dabigatran etexilate P1-34 Stacey Tannenbaum HYPERLINK "http://www.page-meeting.org/?abstract=1187" A Comparison of Fixed Dose-Controlled (FD) versus Pharmacokinetic Modified Dose-Controlled (PKMD) Clinical Study Designs P1-35 Michel Tod HYPERLINK "http://www.page-meeting.org/?abstract=1074" Steady-state Equation for the Bicompartmental Model with Gamma Absorption. Application to Mycophenolate PK in Renal Transplant Patients
Posters Wednesday Afternoon (group II)
Applications- CNS
P2-1 Emma Bostrom HYPERLINK "http://www.page-meeting.org/?abstract=1104" Blood-brain barrier transport helps explain discrepancies in in vivo potency between oxycodone and morphine P2-2 Marc-Antoine Fabre HYPERLINK "http://www.page-meeting.org/?abstract=1116" Population pharmacokinetic analysis in children, adolescents and adults with schizophrenia or bipolar disorder P2-3 Matt Hutmacher HYPERLINK "http://www.page-meeting.org/?abstract=1137" Modeling the Dropout from Longitudinal Adverse Event Data: Selecting Optimal Titration Regimens P2-4 Laura Iavarone HYPERLINK "http://www.page-meeting.org/?abstract=1175" Population PK/PD of Alprazolam in the Attenuation of ACTH Activation Induced by Cognitive Performance in Metyrapone-treated Healthy Volunteers P2-5 Maria Kjellsson HYPERLINK "http://www.page-meeting.org/?abstract=1158" Modelling Sleep Using Markov Mixed Effects Models P2-6 Frank Larsen HYPERLINK "http://www.page-meeting.org/?abstract=1202" Non-Linear Mixed Effects PK/PD Modelling of Acute Autoinhibitory Feedback Effects of Escitalopram (ESC) on Extracellular Serotonin (5-HT) Levels in Rat Brain P2-7 Gianluca Nucci HYPERLINK "http://www.page-meeting.org/?abstract=1177" Population pharmacokinetic modelling of pimozide and its relation to CYP2D6 genotype.
P2-8 Gijs Santen HYPERLINK "http://www.page-meeting.org/?abstract=1145" Comparing treatment effect in depression trials: Mixed Model for Repeated Measures vs Linear Mixed Model P2-9 Armel Stockis HYPERLINK "http://www.page-meeting.org/?abstract=1093" Dose-response population modeling of the new antiepileptic drug brivaracetam in add-on treatment of partial onset seizures P2-10 Nathalie Toublanc HYPERLINK "http://www.page-meeting.org/?abstract=1108" Retrospective population pharmacokinetic analysis of seletracetam in epileptic and healthy adults P2-11 Maud Vernaz-Gris HYPERLINK "http://www.page-meeting.org/?abstract=1128" Pooled PK analysis of a new CNS drug, in healthy subjects P2-12 Sandra Visser HYPERLINK "http://www.page-meeting.org/?abstract=1196" Modeling the time-course of the antipyretic effects and prostaglandin inhibition in relation the analgesic effects of naproxen: a compound selection strategy P2-13 Katarina Vucicevic HYPERLINK "http://www.page-meeting.org/?abstract=1200" Population Pharmacokinetic Modelling of Amitriptyline in Depression PatientsP2-14 Mathilde Marchand HYPERLINK "http://www.page-meeting.org/?abstract=1212" Supporting the recommended paediatric dosing regimen for rufinamide using clinical trial simulation
Methodology- Model evaluation
P2-15 Mathilde Marchand HYPERLINK "http://www.page-meeting.org/?abstract=1147" Population PKPD modelling of biomarkers ANP and big ET-1 for two neutral endopeptidase inhibitors
P2-16 Julie Antic HYPERLINK "http://www.page-meeting.org/?abstract=1181" Evaluation of Non Parametric Methods for population PK/PD P2-17 Paul Baverel HYPERLINK "http://www.page-meeting.org/?abstract=1180" Evaluation Of The Nonparametric Estimation Method In NONMEM VI: Application To Real Data P2-18 Massimo Cella HYPERLINK "http://www.page-meeting.org/?abstract=1203" Scaling for function: a model-based approach to dosing recommendation for abacavir in HIV-infected children P2-19 Emmanuelle Comets HYPERLINK "http://www.page-meeting.org/?abstract=1120" Normalised prediction distribution errors in R: the npde library P2-20 Douglas Eleveld HYPERLINK "http://www.page-meeting.org/?abstract=1077" Is the expected performance of a target-controlled-infusion system influenced by the population analysis method P2-21 Carlos Fernandez-Teruel HYPERLINK "http://www.page-meeting.org/?abstract=1146" Simulations of bioequivalence trials using physiological-pharmacokinetic models with saturable and non-saturable hepatic clearance P2-22 Leonid Gibiansky HYPERLINK "http://www.page-meeting.org/?abstract=1106" Precision of Parameter Estimates: Covariance Step ($COV) versus Bootstrap Procedure P2-23 Céline Laffont HYPERLINK "http://www.page-meeting.org/?abstract=1227" Evaluation of model of Heart Rate during Exercise Tolerance Test with missing at random dropouts P2-24 Michael Neely HYPERLINK "http://www.page-meeting.org/?abstract=1082" Impact of Lopinavir Limit of Quantification (LOQ)-Censored Data Replacement on Population Pharmacokinetic (PK) Plasma and Saliva Modeling in HIV-Infected Children P2-25 Erik Olofsen HYPERLINK "http://www.page-meeting.org/?abstract=1198" The performance of model selection criteria in the absence of a fixed-dimensional correct model P2-26 Klas Petersson HYPERLINK "http://www.page-meeting.org/?abstract=1166" Semiparametric distributions with estimated shape parameters: Implementation and Evaluation P2-27 Mahesh Samtani HYPERLINK "http://www.page-meeting.org/?abstract=1220" PK/PD Model of Pegylated Thrombopoietin Mimetic Peptide in Healthy Subjects: Comparison of Verification Procedures for Assessing Model Predictability
Posters Thursday Morning (group III)
Applications- Coagulation
P3-1 Xavier Delavenne HYPERLINK "http://www.page-meeting.org/?abstract=1115" The use of an indirect response model to assess interaction between drugs: acenocoumarol and amoxicillin + clavulanic acid P3-2 Andreas Velsing Groth HYPERLINK "http://www.page-meeting.org/?abstract=1195" A Population PK/PD Model Assessing The Pharmacodynamics Of A Rapid-Acting Recombinant FVII Analogue, NN1731, In Healthy Male Subjects
Applications- CVS
P3-3 Stefanie Albers HYPERLINK "http://www.page-meeting.org/?abstract=1102" Population Pharmacokinetics and Dose Simulation of Carvedilol in Pediatric Patients with Congestive Heart Failure P3-4 Kevin Krudys HYPERLINK "http://www.page-meeting.org/?abstract=1125" Can Bayes Prevent QTC-interval prolongation? A challenge beyond random effects. P3-5 Céline Laffont HYPERLINK "http://www.page-meeting.org/?abstract=1161" Population pharmacokinetic analysis of perindoprilat in hypertensive paediatric patients P3-6 Divya Menon HYPERLINK "http://www.page-meeting.org/?abstract=1186" Pharmacokinetics / Pharmacodynamics of Intravenous Bolus Nicardipine in Adults Undergoing Cardiovascular Surgery
Applications- Oncology
P3-7 Azucena Aldaz HYPERLINK "http://www.page-meeting.org/?abstract=1210" New model for gemcitabine and its metabolites P3-8 Azucena Aldaz HYPERLINK "http://www.page-meeting.org/?abstract=1209" Comparison between NONMEM and NPAG for gemcitabine modelling P3-9 Gemma Dickinson HYPERLINK "http://www.page-meeting.org/?abstract=1144" Population pharmacokinetics of a novel anticancer drug, RH1, in terminal cancer patients P3-10 Yumi Fukushima HYPERLINK "http://www.page-meeting.org/?abstract=1121" Population Pharmacokinetic Analysis of Trastuzumab (Herceptin®) based on Data from Three Different Dosing Regimens P3-11 Iztok Grabnar HYPERLINK "http://www.page-meeting.org/?abstract=1199" Population Pharmacokinetics of Methotrexate in Children with Lymphoid Malignancy P3-12 Emma Hansson HYPERLINK "http://www.page-meeting.org/?abstract=1155" Modelling of Chemotherapy-induced Febrile Neutropenia using the Predicted Degree and Duration of Myelosuppression P3-13 Katharina Küster HYPERLINK "http://www.page-meeting.org/?abstract=1090" Matuzumab A Population Pharmacokinetic Model and its Evaluation P3-14 Ricardo Nalda-Molina HYPERLINK "http://www.page-meeting.org/?abstract=1206" Semi-mechanistic PKPD model for neutropenia using K-PD model in patients receiving high dose of chemotherapy P3-15 Ricardo Nalda-Molina HYPERLINK "http://www.page-meeting.org/?abstract=1204" Population Pharmacokinetics of Aplidin® (plitidepsin) in Subjects with Cancer P3-16 Dolors Soy Muner HYPERLINK "http://www.page-meeting.org/?abstract=1114" Population Pharmacokinetic Modeling Of Busulfan In Patients Undergoing Autologous Stem Cell Transplantation (ASCT) For Multiple Myeloma P3-17 Johan Wallin HYPERLINK "http://www.page-meeting.org/?abstract=1169" Predictive Performance of a Myelosuppression Model for Dose Individualization; Impact of Inter-Occasion Variability and Level and Type of Information Provided
Methodology- Algorithms
P3-18 Aris Dokoumetzidis HYPERLINK "http://www.page-meeting.org/?abstract=1129" An algorithm for proper lumping of systems of ODEs P3-19 Jeroen Elassaiss-Schaap HYPERLINK "http://www.page-meeting.org/?abstract=1188" Automation of Structural Pharmacokinetic Model Search in NONMEM: Evaluation with Preclinical Datasets P3-20 Serge Guzy HYPERLINK "http://www.page-meeting.org/?abstract=1214" Comparison between NONMEM and the Monte-Carlo Expectation Maximization (MC-PEM) Method Using a Physiologically-Based Glucose-Insulin Model P3-21 Robert Leary HYPERLINK "http://www.page-meeting.org/?abstract=1100" An evolutonary nonparametric NLME algorithm P3-22 Elodie Plan HYPERLINK "http://www.page-meeting.org/?abstract=1154" Investigation of performances of FOCE and LAPLACE algorithms in NONMEM VI in population parameters estimation of PK and PD continuous data P3-23 Benjamin Ribba HYPERLINK "http://www.page-meeting.org/?abstract=1143" Parameters estimation issues for complex population PK models: A Nonmem vs. Monolix study
Methodology- PBPK
P3-24 Frédérique Fenneteau HYPERLINK "http://www.page-meeting.org/?abstract=1071" A Physiologically Based Pharmacokinetic Model to Assess the Role of ABC Transporters in Drug Distribution P3-25 Mathilde Marchand HYPERLINK "http://www.page-meeting.org/?abstract=1213" Predicting human from animal PBPK simulation combined with in vitro data P3-26 Gianluca Nucci HYPERLINK "http://www.page-meeting.org/?abstract=1163" A Bayesian approach for the integration of preclinical information into a PBPK model for predicting human pharmacokinetics
Posters Thursday Afternoon (group IV)
Applications- Endocrine
P4-1 Silke Dittberner HYPERLINK "http://www.page-meeting.org/?abstract=1110" Determination of the absolute bioavailability of BI 1356, a substance with non-linear pharmacokinetics, using a population pharmacokinetic modelling approach P4-2 Daniel Jonker HYPERLINK "http://www.page-meeting.org/?abstract=1152" Pharmacokinetic modelling of the once-daily human glucagon-like peptide-1 analogue, liraglutide, in healthy volunteers and comparison to exenatide P4-3 Thomas Klitgaard HYPERLINK "http://www.page-meeting.org/?abstract=1224" Population Pharmacokinetic Model for Human Growth Hormone in Adult Patients in Chronic Dialysis vs. Healthy Subjects P4-4 Jean-Marie Martinez HYPERLINK "http://www.page-meeting.org/?abstract=1216" Population Pharmacokinetics of Rimonabant in Obesity P4-5 Klaas Prins HYPERLINK "http://www.page-meeting.org/?abstract=1219" Modeling of plasma aldosterone concentrations after prokinetic 5-HT4 receptor agonists: forming an integrated simulation framework for summary statistic and subject-level data
Applications- Other topics
P4-6 Neil Benson HYPERLINK "http://www.page-meeting.org/?abstract=1103" Utility of a mixed effects approach to defining target binding rate constants P4-7 Vincent Buchheit HYPERLINK "http://www.page-meeting.org/?abstract=1122" A dedicated SAS Programming Group working in a pharmaceutical Modeling & Simulation organization - Current role, experience and prospects P4-8 Joachim Grevel HYPERLINK "http://www.page-meeting.org/?abstract=1225" Evaluation of the population pharmacokinetic properties of lidocaine and its metabolites, MEGX, GX, and 2,6-xylidine, after application of lidocaine 5% medicated plaster P4-9 Déborah Hirt HYPERLINK "http://www.page-meeting.org/?abstract=1141" Pharmacokinetic-pharmacodynamic modeling of manganese in patients with acute alcoholic hepatitis after an IV infusion of mangafodipirP4-10 Eleanor Howgate HYPERLINK "http://www.page-meeting.org/?abstract=1156" Mechanistic Prediction of HIV Drug-Drug Interactions in Virtual Populations from in vitro Enzyme Kinetic Data: Ritonavir and SaquinavirP4-11 Roger Jelliffe HYPERLINK "http://www.page-meeting.org/?abstract=1072" A Population Model of Epidural Lidocaine P4-12 Viera Lukacova HYPERLINK "http://www.page-meeting.org/?abstract=1133" PK/PD modeling of Adinazolam effect of variability of absorption, PK and PD parameters on variability in PD response P4-13 Lynn McFadyen HYPERLINK "http://www.page-meeting.org/?abstract=1172" Maraviroc Exposure Response Analysis: Phase 3 Antiviral Efficacy in Treatment Experienced HIV+ Patients P4-14 Mark Peterson HYPERLINK "http://www.page-meeting.org/?abstract=1218" Calcium Homeostasis and Bone Remodeling: Development of an Integrated Model for Evaluation and Simulation of Therapeutic Responses to Bone-Related Therapies P4-15 Rogier Press HYPERLINK "http://www.page-meeting.org/?abstract=1084" Dose prediction of tacrolimus in de novo kidney transplant patients with population pharmacokinetic modellingP4-16 Catherine Mary Turner Sherwin HYPERLINK "http://www.page-meeting.org/?abstract=1134" A pharmacokinetic/pharmacodynamic model to determine optimal dosing targets for amikacin in neonatal sepsis P4-17 Anthe Zandvliet HYPERLINK "http://www.page-meeting.org/?abstract=1190" Dose individualization of indisulam to reduce the risk of severe myelosuppression
Methodology- Design
P4-18 Anthe Zandvliet HYPERLINK "http://www.page-meeting.org/?abstract=1189" Phase I study design of indisulam: evaluation and optimization
P4-19 Jeff Barrett HYPERLINK "http://www.page-meeting.org/?abstract=1075" Improving Study Design and Conduct Efficiency of Event-Driven Clinical Trials via Discrete Event Simulation: Application to Pediatric Oncology P4-20 Caroline Bazzoli HYPERLINK "http://www.page-meeting.org/?abstract=1176" Population design in nonlinear mixed effects multiple responses models: extension of PFIM and evaluation by simulation with NONMEM and MONOLIX P4-21 Marylore Chenel HYPERLINK "http://www.page-meeting.org/?abstract=1157" Comparison of uniresponse and multiresponse approaches of PopDes to optimize sampling times for drug-drug interaction studies: application to a Servier compoundP4-22 Marc-Antoine Fabré HYPERLINK "http://www.page-meeting.org/?abstract=1149" Selection of a dosing regimen with WST11 by Monte Carlo simulations, using PK data collected after single IV administration in healthy subjects and population PK modelling P4-23 Ivelina Gueorguieva HYPERLINK "http://www.page-meeting.org/?abstract=1171" Prospective application of a multivariate population optimal design to determine parent and metabolite pharmacokinetic sampling times in a Phase II study P4-24 Patrick Johnson HYPERLINK "http://www.page-meeting.org/?abstract=1127" Optimal dose and sample-size selection for dose-response studies P4-25 Marta Neve HYPERLINK "http://www.page-meeting.org/?abstract=1178" Population methods for dose escalation studies: an MCMC approach P4-26 Kayode Ogungbenro HYPERLINK "http://www.page-meeting.org/?abstract=1151" Sample Size Calculations for Repeated Binary Population Pharmacodynamic Experiments P4-27 Marcella Petrone HYPERLINK "http://www.page-meeting.org/?abstract=1183" Model-based sequential human PET study design for Optimal PK/RO assessment P4-28 Marc Pfister HYPERLINK "http://www.page-meeting.org/?abstract=1215" Optimal Sampling Design and Trial Simulation using POPT and NONMEM P4-29 Sylvie Retout HYPERLINK "http://www.page-meeting.org/?abstract=1164" Population designs evaluation and optimisation in R: the PFIM function and its new features
Abstracts
TOC \o "1-3" \h \z \u HYPERLINK \l "_Toc168370487" Oral Presentation: Applications PAGEREF _Toc168370487 \h 17
HYPERLINK \l "_Toc168370488" Claudio Cobelli Models of Glucose Metabolism and Control in Diabetes PAGEREF _Toc168370488 \h 17
HYPERLINK \l "_Toc168370489" Celine Dartois Impact of handling missing PK data on PD estimation explicit modeling of BLQ data in WinBUGS® reduced bias in the PD predictions - a preclinical example. PAGEREF _Toc168370489 \h 18
HYPERLINK \l "_Toc168370490" Philippe Jacqmin Basic PK/PD principles of proliferative and circular systems PAGEREF _Toc168370490 \h 20
HYPERLINK \l "_Toc168370491" Kristin Karlsson Modelling of disease progression in acute stroke by simultaneously using the NIH stroke scale, the Scandinavian stroke scale and the Barthel index PAGEREF _Toc168370491 \h 22
HYPERLINK \l "_Toc168370492" Teun Post Circadian rhythm in pharmacodynamics and its influence on the identification of treatment effects PAGEREF _Toc168370492 \h 23
HYPERLINK \l "_Toc168370493" Klaas Prins Integrated Modeling & Simulation of Clinical Response and Drop-out of D2 receptor agonists in Patients with Early Parkinsons Disease. PAGEREF _Toc168370493 \h 24
HYPERLINK \l "_Toc168370494" Tim Sheiner Planning To Communicate PAGEREF _Toc168370494 \h 25
HYPERLINK \l "_Toc168370495" Joe Standing Developmental Pharmacokinetics of Diclofenac for Acute Pain PAGEREF _Toc168370495 \h 26
HYPERLINK \l "_Toc168370496" Ashley Strougo Mechanism-based model of the effect of co-administration of exogenous testosterone and progestogens on the hypothalamic-pituitary-gonodal axis in men PAGEREF _Toc168370496 \h 28
HYPERLINK \l "_Toc168370497" Kim Stuyckens Modeling Drug Effects and Resistance Development on Tumor Growth Dynamics PAGEREF _Toc168370497 \h 30
HYPERLINK \l "_Toc168370498" Benoit You Kinetic models of PSA decrease after surgery in prostate tumor diseases as a help for clinician interpretation PAGEREF _Toc168370498 \h 31
HYPERLINK \l "_Toc168370499" Oral Presentation: Lewis Sheiner Student Session PAGEREF _Toc168370499 \h 33
HYPERLINK \l "_Toc168370500" Karl BRENDEL Normalized Prediction Distribution Error for the Evaluation of Nonlinear Mixed-Models PAGEREF _Toc168370500 \h 33
HYPERLINK \l "_Toc168370501" S. Y. Amy Cheung Identifiability Analysis and Parameter List Reduction of a Nonlinear Cardiovascular PKPD Model PAGEREF _Toc168370501 \h 36
HYPERLINK \l "_Toc168370502" Radojka Savic Importance of Shrinkage in Empirical Bayes Estimates for Diagnostics and Estimation: Problems and Solutions PAGEREF _Toc168370502 \h 39
HYPERLINK \l "_Toc168370503" Oral Presentation: Methodology PAGEREF _Toc168370503 \h 41
HYPERLINK \l "_Toc168370504" France Mentré Software for optimal design in population pharmacokinetics and pharmacodynamics: a comparison PAGEREF _Toc168370504 \h 41
HYPERLINK \l "_Toc168370505" Joakim Nyberg Sequential versus simultaneous optimal experimental design on dose and sample times PAGEREF _Toc168370505 \h 42
HYPERLINK \l "_Toc168370506" Justin Wilkins A flexible approach to modeling variable absorption in the context of repeated dosing: illustrated with rifampicin PAGEREF _Toc168370506 \h 44
HYPERLINK \l "_Toc168370507" Stefano Zamuner Optimal Design to Estimate the Time Varying Receptor Occupancy Relationship in a PET Experiment PAGEREF _Toc168370507 \h 46
HYPERLINK \l "_Toc168370508" Oral Presentation: Model Building Session PAGEREF _Toc168370508 \h 48
HYPERLINK \l "_Toc168370509" Chantaratsamon Dansirikul Insulin secretion and hepatic extraction during euglycemic clamp study: modelling of insulin and C-peptide data PAGEREF _Toc168370509 \h 48
HYPERLINK \l "_Toc168370510" Jeroen Elassaiss-Schaap Interspecies Population Modeling Of Pharmacokinetic Data Available At The End Of Drug Discovery PAGEREF _Toc168370510 \h 49
HYPERLINK \l "_Toc168370511" Massimiliano Germani A population PK-PD method for categorical data analysis of progesterone antagonist activity in the rabbit McPhails model PAGEREF _Toc168370511 \h 51
HYPERLINK \l "_Toc168370512" Oral Presentation: Stuart Beal Methodology Session PAGEREF _Toc168370512 \h 52
HYPERLINK \l "_Toc168370513" Robert Bauer Advanced Population Analysis Features in the S-ADAPT/MCPEM Program PAGEREF _Toc168370513 \h 52
HYPERLINK \l "_Toc168370514" Marc Lavielle The SAEM algorithm and its implementation in MONOLIX 2.1 PAGEREF _Toc168370514 \h 54
HYPERLINK \l "_Toc168370515" Fahima Nekka What Additional Information Can we Retrieve When Compliance is Accounted For? An explicit Compliance-Pharmacokinetic Formalism PAGEREF _Toc168370515 \h 56
HYPERLINK \l "_Toc168370516" Tutorial PAGEREF _Toc168370516 \h 57
HYPERLINK \l "_Toc168370517" Marc BUYSE Validation of statistically reliable biomarkers PAGEREF _Toc168370517 \h 57
HYPERLINK \l "_Toc168370518" Poster: Applications- Anti-infectives PAGEREF _Toc168370518 \h 59
HYPERLINK \l "_Toc168370519" julie bertrand Influence of pharmacogenetic on pharmacokinetic interindividual variability of indinavir and lopinavir in HIV patients (COPHAR2 ANRS 111 trial) PAGEREF _Toc168370519 \h 59
HYPERLINK \l "_Toc168370520" Stefanie Hennig Tobramycin in paediatric CF patients - TCI or One dose fits all PAGEREF _Toc168370520 \h 61
HYPERLINK \l "_Toc168370521" Déborah Hirt Effect of CYP2C19 polymorphism on nelfinavir to M8 biotransformation in HIV patients. PAGEREF _Toc168370521 \h 62
HYPERLINK \l "_Toc168370522" Hui Kimko Population Pharmacokinetic Analysis To Support Dosing Regimens Of Ceftobiprole PAGEREF _Toc168370522 \h 63
HYPERLINK \l "_Toc168370523" Grant Langdon PK-PD modelling to support go/no go decisions for a novel gp120 inhibitor PAGEREF _Toc168370523 \h 64
HYPERLINK \l "_Toc168370524" Rocio Lledo Population Pharmacokinetics of Saquinavir in rats after IV and IP administration. An approach to Saquinavir/Ritonavir Pharmacokinetic interaction. PAGEREF _Toc168370524 \h 65
HYPERLINK \l "_Toc168370525" Poster: Applications- Biologicals/vaccines PAGEREF _Toc168370525 \h 67
HYPERLINK \l "_Toc168370526" Balaji Agoram Application of mechanism-based population PKPD modelling in the rational selection of clinical candidates: an anti-IgE antibody example. PAGEREF _Toc168370526 \h 67
HYPERLINK \l "_Toc168370527" Lene Alifrangis Setting a Safe Starting Dose for a First-in-Man trial of a Monoclonal Antibody Based on Population PK-PD Predictions PAGEREF _Toc168370527 \h 68
HYPERLINK \l "_Toc168370528" Ekaterina Gibiansky Population Pharmacokinetics of Siplizumab (MEDI-507): Implications for Dosing PAGEREF _Toc168370528 \h 69
HYPERLINK \l "_Toc168370529" Ron Keizer Bioequivalence study of a C1-esterase-inhibitor product (Cetor®) with optimised sampling design PAGEREF _Toc168370529 \h 70
HYPERLINK \l "_Toc168370530" Wojciech Krzyzanski Pharmacodynamic Modelling of Recombinant Human Erythropoietin Effect on Reticulocyte Production Rate and Age Distribution in Healthy Subjects PAGEREF _Toc168370530 \h 71
HYPERLINK \l "_Toc168370531" Armel Stockis Population pharmacokinetics of certolizumab pegol PAGEREF _Toc168370531 \h 73
HYPERLINK \l "_Toc168370532" Poster: Applications- CNS PAGEREF _Toc168370532 \h 74
HYPERLINK \l "_Toc168370533" Emma Bostrom Blood-brain barrier transport helps explain discrepancies in in vivo potency between oxycodone and morphine PAGEREF _Toc168370533 \h 74
HYPERLINK \l "_Toc168370534" marc-antoine fabre Population pharmacokinetic analysis in children, adolescents and adults with schizophrenia or bipolar disorder PAGEREF _Toc168370534 \h 75
HYPERLINK \l "_Toc168370535" Matt Hutmacher Modeling the Dropout from Longitudinal Adverse Event Data: Selecting Optimal Titration Regimens PAGEREF _Toc168370535 \h 76
HYPERLINK \l "_Toc168370536" Laura Iavarone Population PK/PD of Alprazolam in the Attenuation of ACTH Activation Induced by Cognitive Performance in Metyrapone-treated Healthy Volunteers PAGEREF _Toc168370536 \h 78
HYPERLINK \l "_Toc168370537" Maria Kjellsson Modelling Sleep Using Markov Mixed Effects Models PAGEREF _Toc168370537 \h 79
HYPERLINK \l "_Toc168370538" Frank Larsen Non-Linear Mixed Effects PK/PD Modelling of Acute Autoinhibitory Feedback Effects of Escitalopram (ESC) on Extracellular Serotonin (5-HT) Levels in Rat Brain PAGEREF _Toc168370538 \h 81
HYPERLINK \l "_Toc168370539" Mathilde Marchand Supporting the recommended paediatric dosing regimen for rufinamide using clinical trial simulation PAGEREF _Toc168370539 \h 83
HYPERLINK \l "_Toc168370540" Gianluca Nucci Population pharmacokinetic modelling of pimozide and its relation to CYP2D6 genotype. PAGEREF _Toc168370540 \h 84
HYPERLINK \l "_Toc168370541" Gijs Santen Comparing treatment effect in depression trials: Mixed Model for Repeated Measures vs Linear Mixed Model PAGEREF _Toc168370541 \h 85
HYPERLINK \l "_Toc168370542" Armel Stockis Dose-response population modeling of the new antiepileptic drug brivaracetam in add-on treatment of partial onset seizures. PAGEREF _Toc168370542 \h 86
HYPERLINK \l "_Toc168370543" Nathalie Toublanc Retrospective population pharmacokinetic analysis of seletracetam in epileptic and healthy adults PAGEREF _Toc168370543 \h 87
HYPERLINK \l "_Toc168370544" Maud Vernaz-Gris Pooled PK analysis of a new CNS drug, in healthy subjects. PAGEREF _Toc168370544 \h 88
HYPERLINK \l "_Toc168370545" Sandra Visser Modeling the time-course of the antipyretic effects and prostaglandin inhibition in relation the analgesic effects of naproxen: a compound selection strategy PAGEREF _Toc168370545 \h 89
HYPERLINK \l "_Toc168370546" Katarina Vucicevic Population Pharmacokinetic Modelling of Amitriptyline in Depression Patients PAGEREF _Toc168370546 \h 91
HYPERLINK \l "_Toc168370547" Poster: Applications- Coagulation PAGEREF _Toc168370547 \h 92
HYPERLINK \l "_Toc168370548" Xavier Delavenne The use of an indirect response model to assess interaction between drugs: acenocoumarol and amoxicillin + clavulanic acid PAGEREF _Toc168370548 \h 92
HYPERLINK \l "_Toc168370549" Andreas Velsing Groth A Population PK/PD Model Assessing The Pharmacodynamics Of A Rapid-Acting Recombinant FVII Analogue, NN1731, In Healthy Male Subjects PAGEREF _Toc168370549 \h 94
HYPERLINK \l "_Toc168370550" Poster: Applications- CVS PAGEREF _Toc168370550 \h 95
HYPERLINK \l "_Toc168370551" Stefanie Albers Population Pharmacokinetics and Dose Simulation of Carvedilol in Pediatric Patients with Congestive Heart Failure PAGEREF _Toc168370551 \h 95
HYPERLINK \l "_Toc168370552" Kevin Krudys Can Bayes Prevent QTC-interval prolongation? A challenge beyond random effects. PAGEREF _Toc168370552 \h 97
HYPERLINK \l "_Toc168370553" Céline LAFFONT Population pharmacokinetic analysis of perindoprilat in hypertensive paediatric patients PAGEREF _Toc168370553 \h 99
HYPERLINK \l "_Toc168370554" Divya Menon Pharmacokinetics / Pharmacodynamics of Intravenous Bolus Nicardipine in Adults Undergoing Cardiovascular Surgery PAGEREF _Toc168370554 \h 101
HYPERLINK \l "_Toc168370555" Poster: Applications- Endocrine PAGEREF _Toc168370555 \h 103
HYPERLINK \l "_Toc168370556" Silke Dittberner Determination of the absolute bioavailability of BI 1356, a substance with non-linear pharmacokinetics, using a population pharmacokinetic modelling approach PAGEREF _Toc168370556 \h 103
HYPERLINK \l "_Toc168370557" Daniel Jonker Pharmacokinetic modelling of the once-daily human glucagon-like peptide-1 analogue, liraglutide, in healthy volunteers and comparison to exenatide PAGEREF _Toc168370557 \h 105
HYPERLINK \l "_Toc168370558" Thomas Klitgaard Population Pharmacokinetic Model for Human Growth Hormone in Adult Patients in Chronic Dialysis vs. Healthy Subjects PAGEREF _Toc168370558 \h 106
HYPERLINK \l "_Toc168370559" Jean-Marie Martinez Population Pharmacokinetics of Rimonabant in Obesity PAGEREF _Toc168370559 \h 107
HYPERLINK \l "_Toc168370560" Klaas Prins Modeling of plasma aldosterone concentrations after prokinetic 5-HT4 receptor agonists: forming an integrated simulation framework for summary statistic and subject-level data. PAGEREF _Toc168370560 \h 108
HYPERLINK \l "_Toc168370561" Poster: Applications- Oncology PAGEREF _Toc168370561 \h 110
HYPERLINK \l "_Toc168370562" Azucena Aldaz New model for gemcitabine and its metabolites PAGEREF _Toc168370562 \h 110
HYPERLINK \l "_Toc168370563" Azucena Aldaz Comparison between NONMEM and NPAG for gemcitabine modelling PAGEREF _Toc168370563 \h 111
HYPERLINK \l "_Toc168370564" Gemma Dickinson Population pharmacokinetics of a novel anticancer drug, RH1, in terminal cancer patients. PAGEREF _Toc168370564 \h 112
HYPERLINK \l "_Toc168370565" yumi fukushima Population Pharmacokinetic Analysis of Trastuzumab (Herceptin®) based on Data from Three Different Dosing Regimens. PAGEREF _Toc168370565 \h 113
HYPERLINK \l "_Toc168370566" Iztok Grabnar Population Pharmacokinetics of Methotrexate in Children with Lymphoid Malignancy PAGEREF _Toc168370566 \h 114
HYPERLINK \l "_Toc168370567" Emma Hansson Modelling of Chemotherapy-induced Febrile Neutropenia using the Predicted Degree and Duration of Myelosuppression PAGEREF _Toc168370567 \h 115
HYPERLINK \l "_Toc168370568" Katharina Küster Matuzumab A Population Pharmacokinetic Model and its Evaluation PAGEREF _Toc168370568 \h 116
HYPERLINK \l "_Toc168370569" Ricardo Nalda-Molina Population Pharmacokinetics of Aplidin® (plitidepsin) in Subjects with Cancer PAGEREF _Toc168370569 \h 117
HYPERLINK \l "_Toc168370570" Ricardo Nalda-Molina Semi-mechanistic PKPD model for neutropenia using K-PD model in patients receiving high dose of chemotherapy PAGEREF _Toc168370570 \h 118
HYPERLINK \l "_Toc168370571" Dolors Soy Muner Population Pharmacokinetic Modeling Of Busulfan In Patients Undergoing Autologous Stem Cell Transplantation (ASCT) For Multiple Myeloma PAGEREF _Toc168370571 \h 120
HYPERLINK \l "_Toc168370572" Johan Wallin Predictive Performance of a Myelosuppression Model for Dose Individualization; Impact of Inter-Occasion Variability and Level and Type of Information Provided PAGEREF _Toc168370572 \h 121
HYPERLINK \l "_Toc168370573" Poster: Applications- Other topics PAGEREF _Toc168370573 \h 123
HYPERLINK \l "_Toc168370574" Neil Benson Utility of a mixed effects approach to defining target binding rate constants. PAGEREF _Toc168370574 \h 123
HYPERLINK \l "_Toc168370575" vincent buchheit A dedicated SAS Programming Group working in a pharmaceutical Modeling & Simulation organization - Current role, experience and prospects PAGEREF _Toc168370575 \h 124
HYPERLINK \l "_Toc168370576" Joachim Grevel Evaluation of the population pharmacokinetic properties of lidocaine and its metabolites, MEGX, GX, and 2,6-xylidine, after application of lidocaine 5% medicated plaster. PAGEREF _Toc168370576 \h 126
HYPERLINK \l "_Toc168370577" Déborah Hirt Pharmacokinetic-pharmacodynamic modeling of manganese in patients with acute alcoholic hepatitis after an IV infusion of mangafodipir. PAGEREF _Toc168370577 \h 127
HYPERLINK \l "_Toc168370578" Eleanor Howgate Mechanistic Prediction of HIV Drug-Drug Interactions in Virtual Populations from in vitro Enzyme Kinetic Data: Ritonavir and Saquinavir. PAGEREF _Toc168370578 \h 128
HYPERLINK \l "_Toc168370579" Roger Jelliffe A Population Model of Epidural Lidocaine PAGEREF _Toc168370579 \h 129
HYPERLINK \l "_Toc168370580" Viera Lukacova PK/PD modeling of Adinazolam effect of variability of absorption, PK and PD parameters on variability in PD response PAGEREF _Toc168370580 \h 131
HYPERLINK \l "_Toc168370581" Lynn McFadyen Maraviroc Exposure Response Analysis: Phase 3 Antiviral Efficacy in Treatment Experienced HIV+ Patients PAGEREF _Toc168370581 \h 132
HYPERLINK \l "_Toc168370582" Mark Peterson Calcium Homeostasis and Bone Remodeling: Development of an Integrated Model for Evaluation and Simulation of Therapeutic Responses to Bone-Related Therapies PAGEREF _Toc168370582 \h 134
HYPERLINK \l "_Toc168370583" Rogier Press Dose prediction of tacrolimus in de novo kidney transplant patients with population pharmacokinetic modelling. PAGEREF _Toc168370583 \h 135
HYPERLINK \l "_Toc168370584" Catherine Mary Turner Sherwin A pharmacokinetic/pharmacodynamic model to determine optimal dosing targets for amikacin in neonatal sepsis PAGEREF _Toc168370584 \h 138
HYPERLINK \l "_Toc168370585" Anthe Zandvliet Dose individualization of indisulam to reduce the risk of severe myelosuppression PAGEREF _Toc168370585 \h 139
HYPERLINK \l "_Toc168370586" Poster: Methodology- Algorithms PAGEREF _Toc168370586 \h 140
HYPERLINK \l "_Toc168370587" Aris Dokoumetzidis An algorithm for proper lumping of systems of ODEs PAGEREF _Toc168370587 \h 140
HYPERLINK \l "_Toc168370588" Jeroen Elassaiss-Schaap Automation of Structural Pharmacokinetic Model Search in NONMEM: Evaluation with Preclinical Datasets PAGEREF _Toc168370588 \h 141
HYPERLINK \l "_Toc168370589" Serge Guzy Comparison between NONMEM and the Monte-Carlo Expectation Maximization (MC-PEM) Method Using a Physiologically-Based Glucose-Insulin Model PAGEREF _Toc168370589 \h 143
HYPERLINK \l "_Toc168370590" Robert Leary An evolutonary nonparametric NLME algorithm PAGEREF _Toc168370590 \h 145
HYPERLINK \l "_Toc168370591" Elodie Plan Investigation of performances of FOCE and LAPLACE algorithms in NONMEM VI in population parameters estimation of PK and PD continuous data PAGEREF _Toc168370591 \h 147
HYPERLINK \l "_Toc168370592" Benjamin Ribba Parameters estimation issues for complex population PK models: A Nonmem vs. Monolix study PAGEREF _Toc168370592 \h 148
HYPERLINK \l "_Toc168370593" Poster: Methodology- Design PAGEREF _Toc168370593 \h 150
HYPERLINK \l "_Toc168370594" Jeff Barrett Improving Study Design and Conduct Efficiency of Event-Driven Clinical Trials via Discrete Event Simulation: Application to Pediatric Oncology PAGEREF _Toc168370594 \h 150
HYPERLINK \l "_Toc168370595" caroline BAZZOLI Population design in nonlinear mixed effects multiple responses models: extension of PFIM and evaluation by simulation with NONMEM and MONOLIX PAGEREF _Toc168370595 \h 152
HYPERLINK \l "_Toc168370596" Marylore Chenel Comparison of uniresponse and multiresponse approaches of PopDes to optimize sampling times for drug-drug interaction studies: application to a Servier compound. PAGEREF _Toc168370596 \h 154
HYPERLINK \l "_Toc168370597" marc-antoine fabre Selection of a dosing regimen with WST11 by Monte Carlo simulations, using PK data collected after single IV administration in healthy subjects and population PK modelling. PAGEREF _Toc168370597 \h 156
HYPERLINK \l "_Toc168370598" Ivelina Gueorguieva Prospective application of a multivariate population optimal design to determine parent and metabolite pharmacokinetic sampling times in a Phase II study. PAGEREF _Toc168370598 \h 158
HYPERLINK \l "_Toc168370599" Patrick Johnson Optimal dose & sample-size selection for dose-response studies PAGEREF _Toc168370599 \h 159
HYPERLINK \l "_Toc168370600" Marta Neve Population methods for dose escalation studies: an MCMC approach PAGEREF _Toc168370600 \h 160
HYPERLINK \l "_Toc168370601" Kayode Ogungbenro Sample Size Calculations for Repeated Binary Population Pharmacodynamic Experiments PAGEREF _Toc168370601 \h 161
HYPERLINK \l "_Toc168370602" Marcella Petrone Model-based sequential human PET study design for Optimal PK/RO assessment PAGEREF _Toc168370602 \h 162
HYPERLINK \l "_Toc168370603" Marc Pfister Optimal Sampling Design and Trial Simulation using POPT and NONMEM PAGEREF _Toc168370603 \h 163
HYPERLINK \l "_Toc168370604" Sylvie Retout Population designs evaluation and optimisation in R: the PFIM function and its new features PAGEREF _Toc168370604 \h 164
HYPERLINK \l "_Toc168370605" Anthe Zandvliet Phase I study design of indisulam: evaluation and optimization PAGEREF _Toc168370605 \h 166
HYPERLINK \l "_Toc168370606" Julie ANTIC Evaluation of Non Parametric Methods for population PK/PD PAGEREF _Toc168370606 \h 168
HYPERLINK \l "_Toc168370607" Poster: Methodology- Model evaluation PAGEREF _Toc168370607 \h 170
HYPERLINK \l "_Toc168370608" Paul Baverel Evaluation Of The Nonparametric Estimation Method In NONMEM VI: Application To Real Data PAGEREF _Toc168370608 \h 170
HYPERLINK \l "_Toc168370609" Massimo Cella Scaling for function: a model-based approach to dosing recommendation for abacavir in HIV-infected children. PAGEREF _Toc168370609 \h 172
HYPERLINK \l "_Toc168370610" Emmanuelle Comets Normalised prediction distribution errors in R: the npde library PAGEREF _Toc168370610 \h 174
HYPERLINK \l "_Toc168370611" Douglas J. Eleveld Is the expected performance of a target-controlled-infusion system influenced by the population analysis method PAGEREF _Toc168370611 \h 175
HYPERLINK \l "_Toc168370612" Carlos Fernandez-Teruel Simulations of bioequivalence trials using physiological-pharmacokinetic models with saturable and non-saturable hepatic clearance PAGEREF _Toc168370612 \h 177
HYPERLINK \l "_Toc168370613" Leonid Gibiansky Precision of Parameter Estimates: Covariance Step ($COV) versus Bootstrap Procedure PAGEREF _Toc168370613 \h 178
HYPERLINK \l "_Toc168370614" Céline LAFFONT Evaluation of model of Heart Rate during Exercise Tolerance Test with missing at random dropouts PAGEREF _Toc168370614 \h 179
HYPERLINK \l "_Toc168370615" Mathilde Marchand Population PKPD modelling of biomarkers ANP and big ET-1 for two neutral endopeptidase inhibitors PAGEREF _Toc168370615 \h 181
HYPERLINK \l "_Toc168370616" Michael Neely Impact of Lopinavir Limit of Quantification (LOQ)-Censored Data Replacement on Population Pharmacokinetic (PK) Plasma and Saliva Modeling in HIV-Infected Children PAGEREF _Toc168370616 \h 182
HYPERLINK \l "_Toc168370617" Erik Olofsen The performance of model selection criteria in the absence of a fixed-dimensional correct model PAGEREF _Toc168370617 \h 184
HYPERLINK \l "_Toc168370618" Klas Petersson Semiparametric distributions with estimated shape parameters: Implementation and Evaluation PAGEREF _Toc168370618 \h 185
HYPERLINK \l "_Toc168370619" MAHESH SAMTANI PK/PD Model of Pegylated Thrombopoietin Mimetic Peptide in Healthy Subjects: Comparison of Verification Procedures for Assessing Model Predictability. PAGEREF _Toc168370619 \h 187
HYPERLINK \l "_Toc168370620" Poster: Methodology- Other topics PAGEREF _Toc168370620 \h 189
HYPERLINK \l "_Toc168370621" Johan Areberg Simultaneous Population Pharmacokinetic Modelling of Parent Compound and Metabolite in Plasma and Urine for a New Drug Candidate PAGEREF _Toc168370621 \h 189
HYPERLINK \l "_Toc168370622" Martin Bergstrand A comparison of methods for handling of data below the limit of quantification in NONMEM VI PAGEREF _Toc168370622 \h 190
HYPERLINK \l "_Toc168370623" Robert Bies An MCPEM approach to understanding inter-animal and inter-treatment changes with in vivo striatal dopamine clearance in rats. PAGEREF _Toc168370623 \h 192
HYPERLINK \l "_Toc168370624" sophie callies Modelling pharmacokinetic and pharmacodynamic properties of second generation antisense-oligonucleotides (ASOs). PAGEREF _Toc168370624 \h 193
HYPERLINK \l "_Toc168370625" Didier Concordet How to estimate population variance matrices with a Prescribed Pattern of Zeros? PAGEREF _Toc168370625 \h 194
HYPERLINK \l "_Toc168370626" Carine CREPIN Elimination of anti-epileptic compounds in Marseille aquatic environment from private hospital effluent - modelling versus measurements PAGEREF _Toc168370626 \h 196
HYPERLINK \l "_Toc168370627" Mike Dunlavey Next-Generation Modeling Language PAGEREF _Toc168370627 \h 197
HYPERLINK \l "_Toc168370628" Iñaki F. Trocóniz Modelling Overdispersion and Markovian Features in Count Data PAGEREF _Toc168370628 \h 198
HYPERLINK \l "_Toc168370629" Clare Gaynor An assessment of prediction accuracy of two IVIVC modelling methodologies. PAGEREF _Toc168370629 \h 200
HYPERLINK \l "_Toc168370630" Ihab Girgis Interactive Graphical Visualization Tool for Safety Data Screening PAGEREF _Toc168370630 \h 202
HYPERLINK \l "_Toc168370631" Ihab Girgis Parallel Bayesian Methodology for Population Analysis PAGEREF _Toc168370631 \h 203
HYPERLINK \l "_Toc168370632" Thaddeus Grasela The Application of Systematic Analysis for Identifying and Addressing the Needs of the Pharmacometric Process PAGEREF _Toc168370632 \h 205
HYPERLINK \l "_Toc168370633" Serge Guzy Combining interoccasion variability and mixture within a MCPEM framework PAGEREF _Toc168370633 \h 207
HYPERLINK \l "_Toc168370634" Emilie HENIN Estimation of patient compliance from pharmacokinetic samples PAGEREF _Toc168370634 \h 209
HYPERLINK \l "_Toc168370635" Matt Hutmacher A new approach for population pharmacokinetic data analysis under noncompliance PAGEREF _Toc168370635 \h 211
HYPERLINK \l "_Toc168370636" Matt Hutmacher Comparing Minimum Hellinger Distance Estimation (MHDE) and Hypothesis Testing to Traditional Statistical Analyses a Simulation Study PAGEREF _Toc168370636 \h 213
HYPERLINK \l "_Toc168370637" Ivan Matthews Sensitivity analysis of a mixture model to determine genotype/phenotype PAGEREF _Toc168370637 \h 215
HYPERLINK \l "_Toc168370638" Chee Meng Ng A Systematical Approach to Bridge the Two-Stage Parametric Expectation Maximization Algorithm and Full Bayesian Three-Stage Hierarchical Nonlinear Mixed Effect Methods in Complex Population Pharmacokinetic/Pharmacodynamic Analysis: Troxacitabine-induced Neutropenia in Cancer Patients PAGEREF _Toc168370638 \h 217
HYPERLINK \l "_Toc168370639" Chee Meng Ng A Comparison of Estimation Methods in Nonlinear Mixed-effect Model for Population Pharmacokinetic-pharmacodynamic Analysis PAGEREF _Toc168370639 \h 219
HYPERLINK \l "_Toc168370640" Stefaan Rossenu A mixture distribution approach to IVIVC modeling of a dual component drug delivery system PAGEREF _Toc168370640 \h 220
HYPERLINK \l "_Toc168370641" Alexander Staab How can routinely collected comedication information from phase II/III trials be used for screening of potential effects on model parameters? A case study with the oral direct thrombin inhibitor Dabigatran etexilate PAGEREF _Toc168370641 \h 221
HYPERLINK \l "_Toc168370642" Stacey Tannenbaum A Comparison of Fixed Dose-Controlled (FD) versus Pharmacokinetic Modified Dose-Controlled (PKMD) Clinical Study Designs PAGEREF _Toc168370642 \h 222
HYPERLINK \l "_Toc168370643" Michel Tod Steady-state Equation for the Bicompartmental Model with Gamma Absorption. Application to Mycophenolate PK in Renal Transplant Patients PAGEREF _Toc168370643 \h 223
HYPERLINK \l "_Toc168370644" Poster: Methodology- PBPK PAGEREF _Toc168370644 \h 224
HYPERLINK \l "_Toc168370645" Frédérique Fenneteau A Physiologically Based Pharmacokinetic Model to Assess the Role of ABC Transporters in Drug Distribution PAGEREF _Toc168370645 \h 224
HYPERLINK \l "_Toc168370646" Mathilde Marchand Predicting human from animal PBPK simulation combined with in vitro data PAGEREF _Toc168370646 \h 226
HYPERLINK \l "_Toc168370647" Gianluca Nucci A Bayesian approach for the integration of preclinical information into a PBPK model for predicting human pharmacokinetics PAGEREF _Toc168370647 \h 227
HYPERLINK \l "_Toc168370648" Software demonstration PAGEREF _Toc168370648 \h 228
HYPERLINK \l "_Toc168370649" Roger Jelliffe The USC*PACK collection of BigWinPops software for nonparametric adaptive grid (NPAG) population PK/PD modeling, and the MM-USCPACK clinical software PAGEREF _Toc168370649 \h 228
Oral Presentation: Applications
Claudio Cobelli Models of Glucose Metabolism and Control in Diabetes
Claudio CobelliUniversity of Padova, Italy
Diabetes is one of the major chronic diseases, i.e. together with its complications it accounts for more than 10% of national healthcare expenditure. Modeling can enhance understanding of this disease in quantitative terms and is becoming an increasingly important aid in diagnosis, prognosis and in planning of therapy. In recent years the scope of modeling in relation to carbohydrate metabolism and diabetes has seen dramatic expansion such that it is now being applied across the spectrum from populations of patients (public health) to whole-body, to organ and to molecular level. I will discuss recent developments mostly concentrating on whole-body and organ modeling. At whole-body level I will discuss models to assess the efficacy of homeostatic control , e.g insulin sensitivity, beta-cell function, hepatic insulin extraction and their interplay, and of system fluxes, from dynamic intravenous (IVGTT) or oral (OGTT/meal) experiments, possibly also including tracers. At organ level models to assess key processes like glucose transport and phosphorylation in skeletal muscle, a key target tissue, from PET tracer experiments will be presented. Aspects of model validation will be addressed. A variety of studies will be reviewed where models have provided unique information on nonaccessible parameters/signals, thus enhancing the understanding of the physiology and pathophysiology of glucose metabolism, like obesity and diabetes. Population modeling will also be discussed viz a viz individual modeling strategies . In addition to parsimonious "models to measure" there is the need to develop in silico large scale simulation models. These models can help when it is either not possible, appropriate, convenient or desirable to perform a particular experiment on the system. Thanks to a recent large scale clinical trial a new in silico model has been developed which is capable of generating realistic synthetic subjects. I will discuss its use in the context of a JDRF artificial pancreas project.
Oral Presentation: Applications
Celine Dartois Impact of handling missing PK data on PD estimation explicit modeling of BLQ data in WinBUGS® reduced bias in the PD predictions - a preclinical example.
Dartois C, Looby M, He H, Steimer J-L, and Pillai G.Modeling and Simulation, Novartis Pharma AG, Basel, Switzerland
Objectives: Purpose of this project was to compare 2 drugs, a lead compound and its backup on a pharmacodynamic endpoint based on their relative potency in animals.
Methods: Data came from 2 studies in which the 2 drugs were administered in single dose, some in cross-over and at different dose levels (from 0.1 to 10 mg and from 2 to 100 mg for the lead and its backup, respectively). PK and PD samples were taken at the same time, between 0.5 to 48h with non-balanced designs between the 2 studies. Exploratory analysis of the raw data was performed to highlight features of the design which could have an impact on modeling. Then, a non linear mixed effect approach was applied. Models were evaluated through a goodness of fit (GOF) and a visual predictive check.
Results: Exploratory analysis highlighted for the 2 drugs, a high variability in the absorption phase, and for the lead, paucity of data in the expected IC50 region. The lower limit of quantification (LOQ) of this drug was indeed approximately equal to IC50/2.5 and one quarter of data was below the LOQ (BLQ). Basic PKPD models for the 2 drugs, were analyzed using NONMEM, included 2 compartments and an Emax model, linked by common PD baseline. IIV variabilities were defined as log-normal. An inter-occasion effect was added on absorption parameters, an additive residual error model on log-transformed PK data. In the first approach, all BLQ data were discarded. Predictions of the missing data were often above LOQ and so induced an apparent bias in the PD estimation. It explained why a second approach was tested. The same PK models were fitted in Winbugs®, taken into account the different LOQ, with same considerations as in the first approach except for the inter-occasion effect, not included. Individual PK predictions were then introduced in the PD model fitted in NONMEM. For the 2 compounds, PKPD standard GOF of the 2 approaches were not really distinguishable. The second approach allowed to decrease significantly the number of missing data predictions above LOQ for the lead and in this way revealed this method to be more reliable. Other graphics showed the large impact of the second approach on the estimation of the PD parameters.
Conclusions: Handling missing data in PKPD modeling is a concrete question for which currently no perfect method is available [1]. More precisely, the handling of PK missing data may have a great impact on the PD estimations. In this project, the first method which was recommended despite its simplicity [2], used NONMEM® and discarded all PK data below LOQ. The second method used Winbugs®, in which BLQ data are identified and their influence on the likelihood is assessed. This second method which is also very simple to implement, offered largely more reliable PKPD results.
References: [1] Beal, SL. Conditioning on certain random events associated with statistical variability in PK/PD. JPP,2005(32)2.[2] Beal, SL. Way to fit a PK Model with some data below the quantification limit. JPP 2001, (28)5.
Oral Presentation: Applications
Philippe Jacqmin Basic PK/PD principles of proliferative and circular systems
Philippe Jacqmin (1), Lynn McFadyen (2) and Janet R. Wade (1)(1) Exprimo NV, Lummen, Belgium; (2) Clinical Pharmacology, Sandwich Laboratories, Pfizer Inc., UK.
Mathematical modelling is increasingly being applied to interpret and predict the dynamics of diseases (1,2,3). Within the infectious disease field it is commonly accepted that viral dynamics (VD) are characterized by a fundamental biological principle: the basic reproductive ratio (4) (RR0). RR0 is a derived model parameter that gives the average number of offspring generated by a single virus during its entire life span, in the absence of constraints. When RR0 is higher than 1, the system grows, when RR0 is lower than 1 the system goes to extinction. At RR0 of 1, production and elimination are in equilibrium and the system just survives. By extension and under certain assumptions, it can be proposed that proliferative systems such as bacteria, fungi and cancer cells share the same fundamental principle.
The goal of therapeutic agents used to treat proliferative systems is to bring RR0 below the break point of 1 and eradicate the disease. The inhibitor concentration (IC) that decreases RR0 to 1 can be called the reproduction minimum inhibitory concentration (RMIC). Assuming non competitive inhibition, it can be demonstrated that RMIC is equal to (RR0-1)*IC50 where RR0 is system specific and IC50 (concentration of inhibitor that gives 50% of the maximum inhibition) is compound specific. Across a population, the RMIC has a joint distribution arising from both RR0 and IC50. For a particular individual, when their IC is higher than their RMIC, the proliferative system goes to extinction. If IC is lower than the RMIC, the proliferative system will eventually grow (after a transient decrease in some cases: e.g. viral dynamics).
Based on these two basic model parameters: RR0 and RMIC, several PK-PD principles of proliferative systems can be derived, such as:
System survival (i.e. RR0INH=1) can occur at different levels of inhibition depending on RR0 (5): for example, when RR0=2, RMIC=IC50. When RR0=10, RMIC=9*IC50= IC90.
When in vitro and in vivo RR0 are different (which is often the case), in vitro and in vivo RMIC will also be different. Only when both in vitro and in vivo RR0 are known can in vitro RMIC be scaled to in vivo RMIC and use to predict efficacious inhibitor exposure.
Mechanistically, logistic regression of binary outcomes such as failure/success rates as a function of drug exposure is an expression of the RMIC distribution across the population.
Time of failure or success is a function of the IC/RMIC ratio: e.g. for failure (ratio is below 1), then the further below 1 the ratio is, the earlier the failure.
In order to be equally efficacious at steady state (i.e. same proliferation rate), two treatments (e.g. qd vs bid) should give rise to the same average RR0INH. This can easily be calculated using PK-PD simulations. Indirectly, it indicates that successful Cavg/RMIC, Cmin/RMIC and Cmax/RMIC ratios are PK (e.g. half-life) and schedule dependent. It also supports the concept of time above MIC (TAM) for time-dependent antibiotics (6).
Time varying inhibition of proliferative systems can be handled by calculating the equivalent effect constant concentration (7). (ECC) which is equal to IC50*INHavg/(1-INHavg) where INHavg is the area under the inhibition-time curve divided by the dosing interval.
These fundamental and derived basic PK-PD principles will be illustrated using a simple PK-PD model for proliferative systems and a more complex PK-PD model for viral dynamics (3,8,9). based on the Lotka-Volterra principle. Finally, some considerations on the application of these concepts to circular (disease) systems such as inflammation, allergy and bone turnover will be briefly mentioned.
References: [1]. Chan PL, Holford NH. Drug treatment effects on disease progression. Annu Rev Pharmacol Toxicol, 2001;41:625-659.[2]. Karlsson MO, Silber HE, Jauslin PM, Frey N, Gieschke R, Jorga K, Simonsson U. Modelling of glucose-insulin homeostasis in provocation studies; bridging between healthy volunteers and patients. In Measurement and kinetics of in vivo drug effect 2006; Advances in simultaneous pharmacokinetic/pharmacodynamic modelling'. Edited by M Dahnhof, DR Stanski and P Rolan. Leiden/Amsterdam Center for Drug Research.[3]. DeJongh J, Post T, Freijer J, DeWinter W, Danhof M, Ploeger B. Disease system analysis: distinguishing the disease status from the disease process. In Measurement and kinetics of in vivo drug effect 2006; Advances in simultaneous pharmacokinetic/pharmacodynamic modelling'. Edited by M Dahnhof, DR Stanski and P Rolan. Leiden/Amsterdam Center for Drug Research.[4]. Bonhoeffer S, Coffin JM, Nowak MA. Human immunodeficiency virus drug therapy and virus load. J. virol., 1997, 71, 3275-78.[5]. Rosario MC, Jacqmin P, Dorr P, Van der Ryst E, Hitchcock C. A pharmacokinetic-pharmacodynamic disease model to predict in vivo antiviral activity of maraviroc. Clin Pharmacol Ther 2005; 78:508-19.[6]. Craig W. Pharmacokinetic/pharmacodynamic parameters: rational for antibacterial dosing of mice and men. CID, 1998, 26, 1-12. HYPERLINK "http://www.journals.uchicago.edu/CID/journal/issues/v26n1/ja84_1/ja84_1.web.pdf" http://www.journals.uchicago.edu/CID/journal/issues/v26n1/ja84_1/ja84_1.web.pdf[7]. Rosario MC, Poland B, Sullivan J, Westby M, van der Ryst E. A pharmacokinetic-pharmacodynamic model to optimize the phase IIa development program of maraviroc. J Acquir Immune Defic Syndr. 2006 Jun;42(2):183-91.[8]. Perelson AS, Neumann AU, Markowitz M., Leonard JM, Ho DD, HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science, 1997, 271:1582-86.[9]. Funk G, Fisher M, Joos B, Opravil M, Gunthard H, Ledergerber B, Bonhoeffer S. Quantitatification of in vivo replicative capacity of HIV-1 in different compartments of infected cells. JAIDS, 2001, 26: 397-404.
Oral Presentation: Applications
Kristin Karlsson Modelling of disease progression in acute stroke by simultaneously using the NIH stroke scale, the Scandinavian stroke scale and the Barthel index
Kristin E. Karlsson(1), Justin Wilkins(1,3), Mats O. Karlsson(1) and E. Niclas Jonsson(1,2)(1) Dep. of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden, (2) Hoffman-La Roche Ltd, Basel, Switzerland, (3) Novartis Pharma AG, Basel, Switzerland
Objectives: The aim of this study was to develop a disease progression model jointly for three stroke scales; the NIH stroke scale, the Scandinavian stroke scale and Barthel index. The three stroke scales are assessing neurological and/or functional deficit in stroke patients.
Methods: Scores assessed on three occasions over a 90-day period in 772 acute stroke patients were used to model the time course of recovery using NONMEM VI. The patients were from the control arm of a double-blind, multinational, multicenter, placebo-controlled investigation of the efficacy of a novel acute stroke compound. At each measurement occasion, three discrete events, on each of the scales, were possible: attainment of a maximum score on the scale (full recovery), improvement or decline, or dropout [1]. Each of these possible events had a probability and a score change magnitude associated with it. A joint function for the dropout was used since the definition of dropout was withdrawal from the study, i.e. a dropout event will occur on all scales simultaneously. To accommodate the non-monotonic nature of these transitions, it was necessary to develop a strategy that considered both the longitudinal, continuous aspects and the probabilistic characteristics of the data simultaneously. Time-related variables, including baseline score, previous score, and time since the previous observation, were considered as predictors, as well as demographic covariates such as age.
Results: A model using the information from three stroke scales was developed which included a joint dropout model and correlation between the interindividual variabilities in the three scales.
Conclusions: In stroke studies several stroke scales are often used, e.g. one neurological and one functional assessment scale. With simultaneous modelling of these scales it is possible to combine information from the scales regarding for example correlations between interindividual variabilities between different scales. What is also important is that a simultaneous model enables the use of joint, i.e. scale independent, functions such as a dropout model. However, the scales can be used as predictors of the dropout event. This kind of model could utilise data across different scales collected in different trials.
Reference:[1] Jonsson F et al. A longitudinal model for non-monotonic clinical assessment scale data. J Pharmacokinet Pharmacodyn. 2005 Dec;32(5-6):795-815.
Oral Presentation: Applications
Teun Post Circadian rhythm in pharmacodynamics and its influence on the identification of treatment effects
Teun Post(1), Tomoo Funaki(2), Hiromi Maune(2), Henk-Jan Drenth(1)1 LAP&P Consultants BV, Leiden, The Netherlands, 2 Otsuka Pharmaceutical Co., Ltd., Japan
Objectives: Compound X is a aquaretic compound, being developed for the treatment of a specific kidney disease. Population PK-PD models were developed on phase II data to support dose selection for upcoming phase III studies. The treatment effects were evaluated based on the paramaters urine osmolarity (OSM) and urine volume. For the derived pharmacodynamic models, inclusion of a circadian rhythm was required to describe a non-stationary baseline with substantial variability. The objective is to demonstrate the quantification of treatment effects on pharmacodynamic parameters under influence of a circadian rhythm.
Methods: Phase II study data were available for 18 subjects, which were randomised to two treatment arms, comprising of three periods each. In the first period all subjects received a single dose of 15 mg and in the second period a single dose of 30 mg. In the third period, subjects in one arm received a 15 mg bid and in the other arm 30 mg qd dosing for 5 days. Full PK profiles were available for the first and second period and for days 1 and 5 in the third period. Urine osmolarity and urine volume were measured before, during and after treatment for all periods. As the cumulative urine volume (CUV) can be derived by integreating the urinary production rate (UPR) over time, a model was developed which was able to describe the effect on both UPR and CUV in one comprehensive model with a circadian rhythm implemented on the UPR. The concentration-response relationship of compound X on OSM and UPR-CUV was modelled using turnover models with a circadian rhythm on baseline.
Results: The population PK model adequately described the observed PK, as shown by a visual predictive check and a bootstrap analysis. Inclusion of circadian processes in the PK-PD model for OSM resulted in the identification of a clear Emax exposure-response relationship. In contrast, the EC50 could not be precisely identified for the UPR-CUV model due to remaining residual variability not explained by a circadian rhythm. Nevertheless, both models showed an adequate perfomance in a visual predictive check.
Conclusions: Implementation of circadian rhythm was needed to distinguish the treatment effect from the non-stationary baseline response for OSM. This shows the importance of addressing circadian rhythm when quantifying treatment effects. The developed PK-PD model for OSM was used to support selection of appropriate dosage regimens in upcoming phase III studies.
Oral Presentation: Applications
Klaas Prins Integrated Modeling & Simulation of Clinical Response and Drop-out of D2 receptor agonists in Patients with Early Parkinsons Disease.
E.H. Cox (1)(3), N.H. Prins (1), G. DSouza (2) & L. McFadyen (2)(1) Pharsight, Strategic Consulting Services, Lent, The Netherlands, (2) Pfizer, Clinical Research and Development, Sandwich, UK, (3) Current: J & J Pharmaceuticals, Beerse, Belgium
Objectives: This Modeling & Simulation (M&S) project aimed to establish a simulation framework for efficacy of the D2 receptor agonist sumanirole and its comparator ropinirole in the treatment of early Parkinson's disease (PD) to address drug development questions.
Methodology & Results: Available data consisted of longitudinal Unified Parkinson's Disease Rating Scale (UPDRS) II/III measurements from a total of 994 patients with early PD participating in two studies. The first was a 12-week placebo controlled fixed dose study (0, 2, 8, 24 & 48 mg) of sumanirole (7-week titration, 4-week maintenance and 1 week taper phases). The second study was a flexible-dose comparator study of placebo, sumanirole (d" 16 mg) and ropinirole (d" 24 mg) conducted over 40 weeks (13-week titration, 26 week maintenance, 1 week taper phases).
Using NONMEM, UPDRS response was modeled as a function of baseline UPDRS score, time and drug dose. A sizeable dropout due to tolerability issues or lack of efficacy was found in both studies. This was not consistently dose related or related to reported adverse events but correlated best with low response. Therefore response-related dropout was integrated into the UPDRS model. The dose-response relationship for sumanirole and ropinirole was best described by a power model which was validated and qualified through partial residual plots and posterior predictive checking. In the flexible dose study, no clear relationship was found between dose adjustment and lack of efficacy or adverse events. Without knowing the specifics for dose adjustment (or not) such flexible dose designs could not be simulated. Simulations under a fixed dose design indicated that in the expected patient population there is a reasonable probability that high doses of sumanirole (from 24 mg and upward) may be non-inferior on efficacy to ropinirole.
Conclusion: An integrated efficacy/dropout model for D2 receptor agonist treatment of early PD was established, constituting a suitable framework for simulation of drug response in patients with PD. Simulation of flexible dosing designs requires more specific information than protocol guidance on reasons for dosing decisions. The characterization of the sumanirole dose response was helpful in decision making.
Oral Presentation: Applications
Tim Sheiner Planning To Communicate
Tim SheinerPharsight Corporation
Objectives: As a discipline, Pharmacometics has matured to the point where no educated observer doubts that a skilled pharmacometrician can contribute valuable information to any drug development program. And yet, the general sense within the pharmacometric community is that the pharmaceutical industry is not taking full advantage of the value this discipline offers. The premise of this theoretical talk is that at least some of this inefficiency occurs because people from other disciplines simply do not understand what pharmacometricians are talking about. In addition to discussing the source of this communication problem, this talk argues that pharmacometricians who want to be effective must focus on this issue, and provides a conceptual framework to help them do so.
Methods: The message of this talk is that communicating effectively requires planning. Specifically, this planning requires appreciation of three crucial issues. First, one must accept that communication is more about the other than about oneself. Second, one must recognize and embrace the emotional aspect of communication even in professional settings. Third, one must realize that one gets better at communication just as one improves in any skill: by studying established techniques, adopting appropriate tools and practicing.
Conclusions: Pharmacometricians use complicated mathematics to produce probabilistic information. Many other important participants in the drug development process simply do not understand how to evaluate and utilize this information. Therefore, in order for Pharmacometrics to deliver its full value to the process of drug development, its practitioners must focus not only on producing information but also on how to communicate that information effectively.
References [1] Lesko, Paving the Critical Path: How can Clinical Pharmacology Help Achieve the Vision, CP&T, 81:2, 2/2007.[2] Mandema, et.al., Model-based Development of gemcabene, a new lipid-altering agent, The AAPS Journal, 7, E513-522.
Oral Presentation: Applications
Joe Standing Developmental Pharmacokinetics of Diclofenac for Acute Pain
JF Standing (1,2), RF Howard (1,3), A Johnston (4), I Savage (2), ICK Wong (1,2,3)(1) Great Ormond Street Hospital for Children, London, UK; (2) School of Pharmacy, University of London, UK; (3) Institute of Child Health, University College London, UK; (4) St Georges Medical School, University of London, UK.
Introduction: Diclofenac is an effective, opiate-sparing analgesic widely used for peri-operative pain in children [1] with single doses of 0.5-2mg/kg being used in clinical practice. Although no concentration/effect relationship has been ascertained, adult pharmacodynamic studies have shown a ceiling effect above 50mg [2]. Diclofenac's main therapeutic action is cyclooxygenase-2 (COX-2) inhibition, which occurs in a time-dependent manner in vitro [3]. There is no licensed paediatric oral formulation, and little published pharmacokinetic data in children.
Objectives: Investigate the ontogeny of diclofenac pharmacokinetics; recommend a suitable paediatric dose to achieve an equivalent effect of 50mg in adults.
Methods: Rich pharmacokinetic data in 30 adults given a 50mg dose was provided by the manufacturer of a new diclofenac oral suspension. A clinical pharmacokinetic study was carried out on a paediatric day-surgery ward. Before surgery each child (age 1-12y) was given a 1mg/kg dose by oral syringe. Serum samples were drawn on induction of anaesthesia, at the end of surgery and on removal of the venous cannula. A digital watch was provided to each child, by which dosing and blood sampling times were recorded.
Pooled adult and paediatric pharmacokinetic modelling was undertaken in NONMEM (version VI) using first-order conditional estimation with interaction. The absorption phase was modelled using single and dual absorption compartments, with combinations of first, zero, mixed first and zero, and transit absorption models [4] tested. Residual error was modelled separately for adult and paediatric data, assays having been performed in different laboratories. The final model was evaluated using predictive checks. Simulations were performed to predict paediatric dosing which gave similar exposure (AUC) to 50mg in adults, the rationale for this being that this is an effective dose in adults, the assumption that COX-2 concentration at the site of tissue injury is not developmentally different, and that enzyme drug exposure is important for analgesia [3].
Results and Discussion: A total of 558 serum diclofenac concentrations from 100 (70 paediatric, 30 adult, weight range 9-93kg) patients were used in the pooled analysis. Double peaks seen in raw plots were probably due to pH dependent diclofenac dissolution from solid particles in suspension. A one compartment model with dual absorption compartments was chosen. Allometric scaling of CL (wt^0.75) and VD (wt) [5] was added prospectively as part of the structural model in an attempt to delineate body size with age-related maturational variability. Geometric mean standardised CL of 52.9, 50.8 and 50.4 L/hr/70kg were estimated for patients aged 1-3, 4-12 and adults respectively, showing the allometric model adequately explains variations in diclofenac CL with age. This was confirmed by covariate analysis where there were no further significant influences on CL or VD. Allometric scaling is based on the observation that basal metabolic rate scales with wt^0.75 across species [6]. It must be noted the proposition that basal metabolic rate scales with wt^0.75 cannot be tested in humans (i.e. within a species) as a 9-fold adult weight range would be required to see significant differences with other proposed scaling factors such as wt^0.67 [6]. In this study wt^0.75 reasonably explained changes in diclofenac CL with size; it is envisaged future work will lead to a time when children can be considered small adults' in terms of drug dosing.
Of the simulated doses investigated, 1mg/kg gave paediatric AUC values divided by adult 50mg AUC ratios of 1.00, 1.08 and 1.18 for ages 1-3, 4-6 and 7-12 respectively. Dosing with body weight in a linear (1mg/kg) fashion produces almost a 20% difference in exposure between infants and children aged 7-12, which for diclofenac is unlikely to prove clinically significant either in terms of analgesia or adverse effects.
Conclusions: The allometric wt^0.75 model adequately explained variability in CL with age for diclofenac, which may indicate maturation of diclofenac ADME is complete by 1 year. This study has shown 1mg/kg to produce similar exposure in infants and children up to 12 years to 50mg in adults, and is acceptable for clinical practice; patients are unlikely to obtain further benefit from 2mg/kg. It must be noted that dosing in a linear fashion will lead to higher exposure in older children, which is unlikely to be of clinical significance for diclofenac, but could be for drugs with a narrower therapeutic index. Future work will investigate the influence of age and CYP2C9 genotype on the clearance of diclofenac to 4'-hydroxydiclofenac; and broader research on the development of a paediatric delineation factor'.
References: [1] Turner S, Longworth A, Nunn AJ, Choonara I. 1998. Unlicensed and off label drug use in paediatric wards: prospective study. British Medical Journal, 316:343-5.[2] McQuay HJ & Moore RA. 1998. Postoperative analgesia and vomiting, with special reference to day-case surgery: a systematic review. Health Technology Assessment 2:1-236, Winchester, UK.[3] Rowlinson SW, Kiefer JR, Prusakiewicz JJ, Pawlitz JL, Kozak KR, Kalgutkar AS, Stallings WC, Kurumbail RG, Marnett LJ. 2003. A novel mechanism of cyclooxygenase-2 inhibition involving interactions with Ser-530 and Tyr-385. Journal of Biological Chemistry, 46:45763-9.[4] Savic R, Jonker DM, Kerbusch T, Karlsson MO. 2004. Evaluation of a transit compartment model versus a lag time model for describing drug absorption delay. PAGE Abstract.[5] Meibohm B, Lear S, Pancetta JC, Barrett JS. 2005. Population pharmacokinetic studies in pediatrics: issues in design and analysis. The AAPS Journal, 7:E475-87.[6] Kleiber M. 1947. Body size and metabolic rate. Physiological Reviews, 27: 511-41.
Acknowledgements: Brian Anderson and Hussain Mulla gave invaluable advice on pharmacokinetic modelling, Rada Savic kindly provided code for transit absorption model. Funding was from Rosemont Pharmaceuticals Ltd, UK.
Oral Presentation: Applications
Ashley Strougo Mechanism-based model of the effect of co-administration of exogenous testosterone and progestogens on the hypothalamic-pituitary-gonodal axis in men
Strougo, A(1), Elassaiss-Schaap, J(2), de Greef, HJMM(2), Drenth, H(1)(1) LAP&P Consultants BV, (2) PK-PD / M&S, Clinical Pharmacology & Kinetics, Organon
Objectives: Combined administration of testosterone (T) and progestogens has been investigated for the development of a male contraceptive method. Co-administration leads to suppression of the hypothalamic-pituitary-gonodal (HPG) axis, and consequently to reversible arrestment of spermatogenesis1,2. The aim of the current project is to develop a mechanism-based model of the homeostatic feedback mechanisms in the HPG axis to describe and quantify the suppressing effects on luteinizing hormone (LH), follicle stimulating hormone (FSH), and T following co-administration of T and one of the progestogens desogestrel (DSG) or etonogestrel (ENG).
Methods: The model was developed using data obtained from 288 healthy, male subjects in five different clinical trials 1,2,3,4, where progestogen (DSG or ENG) was orally or subcutaneous administered alone or in combination with intramuscular administration of T enanthate (TE) or T decanoate (TD). The modelling was performed using NONMEM V in three stages with increasing complexity. Firstly, the status of the HPG axis prior to treatment was described. This was followed by the description of the effect of DSG administration, and subsequently after co-administration of DSG/ENG and T.
Results: The homeostatic feedback relationships between T, LH, and FSH were implemented in the model with linked turn-over models. In this model framework, LH was assumed to stimulate the zero-order production rate constant (Kin) of T, and in turn, T was assumed to inhibit the Kin of LH and FSH. The inhibitive effect of DSG/ENG on LH and FSH formation appeared to be maximal in the current data set. Since the PK of both TE and TD could not be adequately characterised, it was described by an infusion thereby assuming constant T concentrations during treatment. This assumption resulted in overestimation of the T concentrations before steady state was achieved.
Conclusions: The model adequately described the time courses of LH and FSH and to lesser extent of T. Since arrestment of spermatogenesis is caused by suppression of LH and FSH, the next step is to link this mechanism-based model to the contraceptive effect, i.e. sperm-count. This mechanism-based modelling approach enables quantification of treatment effects on suppression of the HPG axis, by combining data of three hormones. Consequently, the developed model can be used in various stages of the development of male contraceptives.
References:[1] Wu, F.C., R. Balasubramanian, T.M. Mulders and H.J. Coelingh-Bennink, Oral progestogen combined with testosterone as a potential male contraceptive: additive effects between desogestrel and testosterone enanthate in suppression of spermatogenesis, pituitary-testicular axis, and lipid metabolism, J Clin Endocrinol Metab, 84: 112-22., 1999; [2] Anawalt, B.D., K.L. Herbst, A.M. Matsumoto, T.M. Mulders, H.J. Coelingh-Bennink and W.J. Bremner, Desogestrel plus testosterone effectively suppresses spermatogenesis but also causes modest weight gain and high-density lipoprotein suppression, Fertil Steril, 74: 707-14., 2000[3] Brady B.M., Amory J.K., Perheentupa A., Zitzmann M., Hay C., Apter D., Anderson R.A., Bremner W.J., Huhtaniemi I., Nieschlag E., Wu F.C.W., Kersemaekers W.M.: A multi-centre study investigating subcutaneous Etonogestrel implants with injectable Testosterone Decanoate as a potential long-acting male contraceptive. Human Reproduction 2006; 21: 285-294.[4] Cathy J. Hay, Brian M. Brady, Michael Zitzmann, Kaan Osmanagaoglu, Pasi Pollanen, Dan Apter, Frederick C.W. Wu, Richard A. Anderson, Eberhard Nieschlag, Paul Devroey, Ilpo Huhtaniemi, Wendy M. Kersemaekers: A multicenter phase IIb study of a novel combination of intramuscular androgen (Testosterone Decanoate) and oral progestogen (Etonogestrel) for male hormonal contraception. The Journal of Clinical Endocrinolgy & Metabolism 2005; 90: 2042-2049.
Oral Presentation: Applications
Kim Stuyckens Modeling Drug Effects and Resistance Development on Tumor Growth Dynamics
Kim Stuyckens, Stefaan Rossenu, Peter King, Janine Arts, Juan Jose Perez-Ruixo.Johnson & Johnson Pharmaceutical Research & Development. Beerse. Belgium.
Objective: The objective of this study is to develop a semi-mechanistic model to quantify tumor growth dynamics, the anticancer drug effects and the development of resistance.
Methods: U87 human glioblastome cell lines were implanted subcutaneously into mice. One week after tumor inoculation, 317 mice bearing a palpable tumor were selected and randomized into control and treated groups, which included an oral anticancer drug treatment at doses ranging from X to 40X mg/m2 and given as continuous (once daily or once weekly) or intermittent dosing (daily for 7 consecutive days on a 14 days cycle or daily for 3 consecutive days on a 7 days cycle). A total of 1699 measurements of tumor volumes were modeled using NONMEM. An exponential growth model described the tumor dynamics in nontreated animals. In treated animals, the tumor growth rate of sensitive cells was decreased by a factor proportional to both drug concentration and number of proliferating sensitive tumor cells as previously described1. A transit compartmental system was used to model the process of cell death, which occurs at later times. In addition, sensitive tumor cells that became resistant were less sensitive to drug concentration and follow an exponential growth model, similarly to what has been reported earlier2. In absence of pharmacokinetic data, a "kinetics of drug action" model, as described by Jaqmin et al3, was used to characterize the time course of tumor growth. The model was evaluated using visual predictive check.
Results: Typical value (%CV) of exponential tumor growth rate was 0.22 day-1 (11%). The drug effect on sensitive cells is 7 times higher than the effect on resistant cells. Damaged cells died after approximately 9 days and the resistance rate was estimated to be equal to 0.03 day-1. Visual predictive check confirmed that the model developed was suitable to describe the tumor cells dynamics in presence of anticancer treatment.
Conclusions: The integration of tumor growth data using modeling approach allows to characterize the dynamics of the tumor growth, to quantify the drug potency and to describe the development of drug resistant cells. This model can be used prospectively to optimize the design of future preclinical studies.
References:[1]. Simeoni, M et al. Clinical Cancer Research 2004; 64, 1094-1101.[2]. Chung, P et al. Antimicrobial Agents Chemother. 2006; 50: 2957-2965.[3]. Jacqmin, P et al. J. Pharmacokinetics and Pharmacodynamics. 2007; 34: 57-85.
Oral Presentation: Applications
Benoit You Kinetic models of PSA decrease after surgery in prostate tumor diseases as a help for clinician interpretation
Benoit You (1,2) , Paul Perrin (1,2), Philippe Paparel (1,2), Gilles Freyer (1,2), Olivier Colomban (1) Brigitte Tranchand (1,3), Pascal Girard (1)(1)EA3738 Faculté de Médecine Lyon Sud, Université de Lyon, Lyon, France (2)Hospices civils de Lyon, Lyon, France (3) Centre Léon Bérard, Lyon, France
Objectives: In order to characterize Prostate Specific Antigen (PSA) compartments releases (anatomy transitional and peripheral zones) and prognostic value of PSA decrease on patient's outcome, we built a kinetic model of PSA decline after prostate surgery in 109 patients with prostate tumor diseases.
Methods:
Patients: Fifty four Prostate Benign Hyperplasia patients treated with adenomectomy (PBH=adenoma developed within transitional zone; n=54, median age 62 years, creatinin clearance (CCL) 80 mL/min) and fifty five limited prostate cancer patients treated with radical prostatectomy (n=55; median age:69 years, CCR: 84 mL/min; 6.76 PSA assays/patient, differentiation score Gleason 7, pathology stage T2N0 to pT3N1) were included in a retrospective study.
Data: Data base involved 553 post-surgery values of serum PSA over a 4 years period (mean 3.35 PSA assays/patient in adenomectomy group and 6.76 in prostatectomy group) with a median follow-up of 97 days for adenomectomy group and 285 days for prostatectomy group.
Models: PSA declines were fitted according to multi-exponential models using NONMEM v5 with FOCE INTER. All parameters were considered with inter-individual variability (IIV) and patient's covariates were tested in models in order to reduce IIV.
Since the post-adenomectomy PSA showed a re-growth after few weeks linked to the PSA prostate residual peripheral zone production, a third exponential was added in the model to describe this phenomenon.Afterwards models were validated using visual predictive check.
Relationships between PSA decline profile and on one hand patients outcome (PSA biologic relapse: RLPS: 0=no, 1=Yes) and on the other hand 18 months relapse free survival (RFS%)) were determined using S-PLUS.
Results:
The best model was a bi-exponential one using multiplicative error.
After adenomectomy for PBH (n=54), PSA was fitted by:
PSA(t)= ((13.6-0.116*CCL)*EXP(-0.36*t))+4.54*EXP(-0.19*t)+0.75*EXP(+0.00024*t)
The final IIV of first, second and third exponential rates were 58%, 83% and 66% respectively.
According to this equation, we can infer the individual predicted PSA production by the transitional zone (median 0.126 ng/mL/ tissue gramme) which is close to the literature values evaluated by anatomopathology. Moreover, the PSA prostate peripheral production is estimated to be about 0.75ng/mL.
After radical prostatectomy for prostate cancer (n=55), PSA decrease (from Day 0 to 30 after surgery) was described by:
PSA(t)= ((8.58 -0.042*CCL)*EXP(-0.21*t)) + (1.84 + 1.87*RLPS)*EXP(-0.39*t)
IIV of first and second exponential rates were 33% and 53% respectively. This equation allowed us to calculate the PSA Area Under the Curve (AUC).Patients who presented a RLPS had an increased PSA AUC (33.16 vs 29.10 ng/mL.day, p=0.04).
Total PSA production was assumed to be the sum of: the PSA prostate transitional zone (adenoma) production, PSA prostate peripheral zone production and PSA cancer production. Consequently, the predicted PSA prostate cancer compartment production was estimated to be 0.038 ng/mL/gramme of cancer tissue. Relapse free survival was significantly better in patients with a small PSA cancer compartment production (linear regression, p=0.03).
Conclusions: Using modeling in 2 prostate tumor diseases, we not only determined prostate transitional and peripheral compartments production (results consistent with literature) but also PSA cancer compartment production. We showed the influence of renal function on PSA. Our results confirm the relationship between PSA decrease and cancer relapse. We planned to build a model describing relapse risk according to PSA AUC in order to give to physicians a tool for PSA interpretation after surgery.
Oral Presentation: Lewis Sheiner Student Session
Karl BRENDEL Normalized Prediction Distribution Error for the Evaluation of Nonlinear Mixed-Models
Brendel K1,2, Comets E1, Laffont C.M2, Mentré F1,3.(1)INSERM U738, Paris, France; University Paris 7, Paris, France;(2)Institut de recherches internationales Servier, Courbevoie, France;(3)AP-HP, Bichat Hospital, Paris, France.
Introduction: Although population pharmacokinetic and/or pharmacodynamic model evaluation is recommended by regulatory authorities, there is no consensus today on the appropriate approach to assess a population model. We have also shown in a recent literature survey [1] that model evaluation was not appropriately performed in most published population pharmacokinetic-pharmacodynamic analyses. In this context, we describe a new metric, that can be used for model evaluation in population PK or PD analyses. Our objectives were firstly to illustrate this metric by proposing different tests and graphs and secondly to evaluate it by simulation for a pharmacokinetic model with or without covariates.
Definition of the Normalized Prediction Distribution Error (NPDE): The null hypothesis (H0) is that data in the validation dataset can be described by a given model. Let MB be a model built from a dataset B and V a validation dataset.
Among the different approaches proposed in the literature to evaluate a population model, standardised prediction errors (computed in NONMEM through WRES) are frequently used, but they are computed using a first-order approximation. In this context, we developed a metric called Normalized Prediction Distribution Error (NPDE) based on the whole predictive distribution. For each observation, we define the prediction discrepancy as the percentile of this observation in the whole marginal predictive distribution under H0 [2]. The predictive distribution is obtained through Monte Carlo simulations. As prediction discrepancies are correlated within an individual, we use the mean and variance of predicted observations estimated empirically from simulations to obtain uncorrelated metrics [3]. NPDE are then obtained using the inverse function of the normal cumulative density function. By construction, if H0 is true, NPDE follow a N(0, 1) distribution without any approximation and are uncorrelated within an individual.
For WRES and NPDE, we use a Wilcoxon signed rank test to test whether the mean is significantly different from 0, a Fisher test to test whether the variance is significantly different from 1, a Shapiro-Wilks test to test if the distribution is significantly different from a normal distribution and a Kolmogorov-Smirnov test to test the departure from a N(0, 1) distribution. We have to consider sequentially the four tests to decide whether to reject a validation dataset.
a) Illustrative example
NPDE and WRES were applied to evaluate a one compartment model built from two phase II studies with zero order absorption and first order elimination. These metrics were applied on 2 simulated validation datasets based on the design of a real phase I study: the first (Vtrue) was simulated with the parameters values estimated in MB; the other one was simulated using the same model and a bioavailability multiplied by two (Vfalse).
Even on Vtrue, WRES were found to differ significantly from a normal distribution and NPDE followed a normal distribution. The mean was not significantly different from 0 for WRES and NPDE. On Vfalse, WRES and NPDE were not found to follow a normal distribution and showed a mean significantly different from 0. In conclusion, NPDE was able to appropriately evaluate Vtrue and reject Vfalse, while WRES showed less discrimination.
b) Evaluation of the type I error by simulations
The model used for simulations was a one compartment model with first order absorption built from two phase II and one phase III studies. We simulated with this model (without covariates) 1000 external validation datasets according to the design of another phase III study and calculated NPDE and WRES for these simulated datasets. We evaluated the type I error of the Kolmogorov-Smirnov test for these two metrics. The simulations under H0 showed a high type I error for the Kolmogorov-Smirnov test applied to WRES, but this test presents a type I error close to 5% for NPDE.
Evaluation of NPDE applied to a model with covariates: We considered here covariate models and investigate the application of NPDE to these models. We used covariates of a real phase III study and generated several validation datasets under H0 and under alternative assumptions without covariates, with one continuous covariate (weight) or with one categorical covariate (sex). We proposed two approaches to evaluate models with covariates by using NPDE. The first approach uses the Spearman correlation test or the Wilcoxon test, to test the relationship between NPDE and weight or sex, respectively. The second approach tests whether the NPDE follow a N(0, 1) distribution after splitting them into different groups of values of the covariates. Regarding the application of NPDE to covariate models, Spearman and Wilcoxon tests were not significant when models and validation datasets were consistent. When validation dataset and models were not consistent, these different tests showed a significant correlation between NPDE and covariates. We also find the same results by using the Kolmogorov-Smirnov test after splitting NPDE by covariates.
Conclusions: We assess here by simulation the statistical properties of NPDE for evaluation of a population pharmacokinetic model in comparison with WRES. We also evaluate the ability of NPDE to detect misspecification in the covariate model.The use of NPDE over WRES is recommended for model evaluation. NPDE do not depend on an approximation of the model and have good statistical properties. They can be viewed as a good alternative way of evaluation by looking at MC simulated predictions. NPDE thus appear as a good tool to evaluate population models, with or without covariates.
References: [1] Brendel K., Dartois C., Comets E., Lemenuel-Diot A., Laveille C., Tranchand B., Girard P., Laffont C.M., Mentré F. Are Population Pharmacokinetic and/or Pharmacodynamic Models Adequately Evaluated? Clin Pharmacokinet 2007; 46 (3): (2007).[2] Mentré F., S. Escolano S. Prediction discrepancies for the evaluation of nonlinear mixed-effects models. J Pharmacokinet Pharmacodyn. 33(3):345-367 (2006).[3] Brendel K., Comets E., Laffont C., Laveille C., Mentré F. Metrics for external model evaluation with an application to the population pharmacokinetics of gliclazide. Pharm Res. 23(9):2036-2049 (2006).
Oral Presentation: Lewis Sheiner Student Session
S. Y. Amy Cheung Identifiability Analysis and Parameter List Reduction of a Nonlinear Cardiovascular PKPD Model
Sau Yan Amy Cheung (1), James W. T. Yates (2), Oneeb Majid (3), Leon Aarons (4)(1) The Centre for Applied Pharmacokinetic Research (CAPKR), School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Manchester; (2) Astrazeneca R&D, Alderley Park, Macclesfield; (3) Medeval Ltd, Skelton House, Manchester Science Park, Manchester
Objective: Pharmacokinetic and pharmacodynamic modelling is becoming evermore sophisticated. Specifically there has been a drive to create more mechanistically relevant models. A particular feature of these models is that parameter values have to be inferred from observations of a small subset of the compartments in the model. In this talk, the problems caused by this and solutions, are examined.
Introduction: The distribution in the body and the effect on the cardiac system of an alpha 1A/1L partial agonist can be described by adapting a previously published nonlinear cardiovascular PKPD model [1]. This model comprises a linear two-compartment PK model with first order absorption and elimination and a PD model which characterises the regulation of the effects of the increased peripheral resistance induced by the constriction of blood vessels. A Hill-type Emax function is used to link the PK model to the reduction in total peripheral resistance in the PD of the model. The PD model consisted of four hemodynamic variables, total peripheral resistance (TPR), heart rate (HR), mean arterial pressure (MAP) and auxiliary control which described the arterial baroreceptor reflexes in the heart. Within the auxiliary control component a mechanism was incorporated for a first order delay. All these variables are joined together to form a closed loop control system to maintain the arterial pressure.
Experimentally the following measurements may be made: plasma concentration (Cp), mean arterial pressure and heart rate. The pharmacodynamic component of the model has 12 unknown parameters. These parameters can be divided into 4 types: three time constant parameters, three physiological parameters for setting the variables to their equilibrium levels, four control parameters related to heart rate and total peripheral resistance and two parameters for the Emax function. Some of these were found to be not well determined from parameter estimation using NONMEM version 5; these parameters were also found to have little influence during global sensitivity analysis. The sensitivity analysis of the PK-PD model was done using Simlab 1.1 and Matlab. Such a phenomenon suggested a potential unidentifiability status of these parameters within the model. Application of structural identifiability analysis [2] to the model was sought in order to verify the cause and factors of this phenomenon: whether it is due to the estimation technique or lack of structural identifiability of the model.
The goal of structural identifiability analysis is to evaluate the internal structure of a mathematical model based purely on the input-output responses, with the assumption of perfect noise free data. If all the unknown parameters in the model may be uniquely determined, then the model is globally uniquely identifiable. This means that the model is unique and the estimated parameters will be unique. If one or more of the parameters may take more than one of a finite number of values without affecting the goodness of fit then the model is locally identifiable. Finally, if there is one parameter that may take on an infinite number of possible values then the model is unidentifiable; this means there are infinite sets of parameters values that will fit to the model equally well. In this situation, re-design, reparameterisation or model reduction of the original model is necessary.
Methods: Generally, different methods are available for structural identifiability analysis of a nonlinear model such as the Taylor series expansion approach [3], the similarity transformation approach [4] and the differential algebra approach [5]. The nonlinear similarity transformation approach was considered in this study due to its robustness in handling complex models. For the analysis to apply, it is necessary for the model to be both controllable and observable. The nonlinear similarity transformation approach is not straightforward as a non-linear mapping is involved in the analysis; therefore a modified nonlinear similarity transformation approach was used. The new modified version makes use of the theorem [6] stating that if the differential equations of the model are polynomial in the state variables and the observation function is linear in terms of state variables, then it is sufficient to consider a linear map in the analysis. Therefore, to simplify the analysis, the PD model was rewritten in polynomial form and an extra state was added to ensure linear observation. The PK and PD parts of the model were analysed simultaneously. All the unknown parameters in the PK model were assumed to be known to decrease the complexity of the analysis and because the parameters were originally estimated in a sequential manner. Using this approach a linear mapping was deduced that altered the parameters in the model, but left the predicted time course for the observed variables unchanged.
Results: The analysis was conducted using MATHEMATICA (version 6). The parameters related to the control mechanism were found to be unidentifiable while all the equilibrium, time constant and core PD parameters were found to be globally identifiable. This result leads to an unidentifiable model which confirmed the findings from the sensitivity analysis and parameter estimation.
The reparameterisation [7], also known as parameter list reduction, method was then used as it may be applied using the similarity transformation that had already been deduced. In this method, the Taylor series of the similarity transformation criteria is calculated. It was found that the model is rank deficient by one and this means that the new parameterisation will have one less parameter than the old parameterisation. The model with one less parameter was used for a second structural identifiability analysis. It was confirmed that the new model parameterisation was now globally identifiable.
The improved identifiability of the model was then confirmed by a simulation study. A patient population was simulated in NONMEM and then refitted to confirm improved convergence of the estimation algorithm and that the estimated parameter values were comparable to those used for the simulation.
Conclusion: Structural identifiability analysis and parameter list reduction can be a helpful tools for the analysis of mechanistic PK-PD models. The example considered demonstrates how a model may be reduced in complexity while maintaining mechanistic relevance. The parameter list reduction technique allowed the similarity transformation to be used, allowing the two stages of analysis to be more computationally economical. Rewriting the model further reduced the complexity of the analysis required. This approach may be applied to a large class of models and so potentially allows the application of structural identifiability analysis to more complex PKPD models.
References:[1] Francheteau, P., Steimer, J.-L., Merdjan, H., Guerret, M. and Dubray, C., A Mathematical Model for Dynamics of Cardiovascular Drug Action: Application to Intravenous Dihydropyridines in Healthy Volunteers, Journal of Pharmacokinetics and Biopharmaceutics 21(5): 489-510, 1993.[2] Bellman, R. and Åström K. J., On Strutural Identifiability. Mathematical Biosciences 7:329-339, 1970.[3] Pohjampalo, H., System identifiability based on the power series expansion of solution. Mathematical Biosciences 41: 21-33, 1978.[4] Vajda, S., Godfrey, K. R., and Rabitz, H., Similarity Transformation Approach to Identifiability Analysis of Nonlinear Compartmental Models, Mathematical Biosciences 93(2): 217-248, 1989.[5] Margaria, G., Riccomagno, E., Chappell, M. J., Wynn, H. P., Differential algebra methods for the study of the structural identifiability of rational function state-space models in the biosciencs, Mathematical Biosciences 17:1-26, 2001.[6] Chappell, M. J., Godfrey, K. R., and Vajda, S., Global Identifiability of Parameters of Nonlinear Systems with Specified Inputs. Mathematical Biosciences. 102(1) : 41-73, 1990.[7] Chappell, M. J., Gunn, R. N., A procedure for generating locally identifiable reparameterisations of unidentifiable non-linear systems by the similarity transformation approach. Mathematical Biosciences. 148(1): 21-41, 1998.
Oral Presentation: Lewis Sheiner Student Session
Radojka Savic Importance of Shrinkage in Empirical Bayes Estimates for Diagnostics and Estimation: Problems and Solutions
Radojka M. Savic and Mats O. KarlssonDiv. of Pharmacokinetics and Drug Therapy, Dept of Pharmaceutical Biosciences, Faculty of Pharmacy, Uppsala University, Sweden
Introduction: Empirical Bayes estimates (EBEs), known also as POSTHOC estimates, provide modelers with individual-specific values, such as individual predictions (IPRED), estimates of interindividual (·) differences and residual error (IWRES) components. These estimates are conditional on individual data and on population parameters. Thus, whenever data becomes sparse or uninformative, the EBE distribution will shrink towards zero (·-shrinkage = 1-SD(·EBE,P)/ÉP, for given parameter P), IPREDs towards the corresponding observations, and IWRES towards zero (µ-shrinkage = 1-SD(IWRES)).(1) However EBEs are widely used in estimation process (FOCE), graphical diagnostics and recently as the points of support for nonparametric (NONP) distribution estimation in NONMEM VI.(1-3) In this work, we investigate how phenomenon of shrinkage affects these three aspects of population analysis, qualitatively and quantitatively. Additionally, solutions to some of these issues are suggested and explored.
Methods: (i) Shrinkage and EBE-based diagnostics This aspect was explored using PK (one compartment, first/zero order absorption, transit compartment model) and PD (Emax, indirect effect) models. Multiple datasets were simulated from these models such that estimated EBE distribution would results in gradually increasing extent of shrinkage for each simulation case. EBEs were estimated in NONMEM based on the true or a misspecified model. Standard errors (SEs) of EBEs were also estimated. Quantitative relationships between ·- and µ-shrinkage and informativeness of EBE-based diagnostics were evaluated.
(ii) Shrinkage and Estimation Method Pharmacokinetic data sets (100 sets for each condition) were simulated from a one compartment iv bolus model. Three scenarios with respect to EBE shrinkage magnitude were studied: (a) < 5% (low), (b) < 25% (medium) and (c) > 25% (high.). The extent of shrinkage was calculated after fitting the true model to this data and it was assessed by varying residual error magnitude (higher the residual variability greater the shrinkage) and with choice of sampling times. Each simulated data set was analyzed with the true model using FOCE and FO estimation method. To compare the estimated and the true parameter values, the relative estimation error and the absolute value of the relative bias (ARB) in parameter estimates were evaluated.
(iii) Shrinkage and NONP distribution estimation The same datasets from part (ii) were used to estimate nonparametric distributions. These distributions were evaluated at different percentiles and compared to the true ones using same statistics as described above. QQ plots and marginal cumulative density distribution plots were used as an additional tool to inspect the estimated distributions. Three methods for generating the support points in the presence of shrinkage were explored: (a) posthoc ·s based on the final parametric model (default NONMEM method); (b) posthoc ·s based on the final parametric model and inflated variances (inflation method), and (c) posthoc ·s from the final parametric model enhanced with n (here n=200) additional support points generated by simulation from the final model (simulation method).
Results: (i) Shrinkage and EBE-based diagnostics Already 20-40% of ·-shrinkage magnitude was sufficiently high to render EBE-based model diagnostics fundamentally misleading (hidden or falsely induced EBE-EBE correlations, distorted, hidden or falsely induced covariate relationships etc.) IPRED failed to detect structural model misspecification already at the µ-shrinkage magnitude of 20-30%. Similar extent of µ-shrinkage was sufficiently high to diminish the power of IWRES to identify the residual model misspecification. Estimation of EBE standard errors was valuable to determine the informative / uninformative EBEs.
(ii) Shrinkage and Estimation Method With FO, the ARB in estimated parameters ranged from 0-15 % no matter which study design was used to generate datasets. With FOCE and low shrinkage, ARB in parameter estimates was negligible (< 2%); with medium shrinkage ARB increased up to 5% while for the analysis of data with high shrinkage, ARB in parameter estimates approached the ARB value observed with FO.
(iii) Shrinkage and NONP distribution estimation Increased ARB in nonparametric distribution (up to 25%) was observed for studies with both medium and high extent of shrinkage as a consequence of rather sparse and restricted grid of support points used in the default nonparametric estimation. For the case with medium shrinkage, the bias was reduced down to < 5% using a new set of support points (EBEs) computed using the inflation method. However, when shrinkage had become substantial, EBEs would remain located around population mean regardless the variance used for inflation. To resolve this issue, the new "simulation" method was developed that uses a denser grid of support points at which the NONP distribution is to be evaluated. This approach was tested using datasets with low to high shrinkage extent. QQ plots showed an excellent agreement between the true and improved nonparametric distributions which was also confirmed by inspection of cumulative probability density functions.
Conclusions: The shrinkage phenomenon affects all three evaluated aspects of population analysis: EBE-based diagnostics are essentially of no value, parameter estimation bias using the FOCE method becomes similar to the FO method and estimated nonparametric distribution shows increased bias when the default method is used. For diagnostic purposes, it is desirable to report extent of µ- and ·-shrinkage to assess the relevance of graphs employing EBEs, IPRED and IWRES. SEs of individual ·s can provide additional information to allow informative diagnostics even in the presence of shrinkage. A new approach to obtain points of support for the nonparametric method has been developed that resulted in good estimation properties even in the presence of high shrinkage.
References:[1]. Savic RM, Wilkins JJ, Karlsson MO. (Un)informativeness of Empirical Bayes Estimate-Based Diagnostics. In: The AAPS Journal; 2006; Vol. 8, No. S2, Abstract T3360 (2006): American Association of Pharmaceutical Scientists; 2006.[2]. Sheiner LB, Beal SL. NONMEM Users Guide: NONMEM Project Group, University of California, San Francisco; 1992. [3]. Savic RM., Kjellsson MC., Karlsson MO. Evaluation of the nonparametric estimation method in NONMEM VI beta. PAGE 15 (2006) Abstr 937 [ HYPERLINK "/?abstract=937" http://www.page-meeting.org/?abstract=937 ].
Oral Presentation: Methodology
France Mentré Software for optimal design in population pharmacokinetics and pharmacodynamics: a comparison
F. Mentré (1), S. Duffull (2), I. Gueorguieva (3), A. Hooker (4), S. Leonov (5), K. Ogungbenro (6), S. Retout (1)(1) INSERM, U738, Paris, France; Université Paris 7, Paris, France; AP-HP, Hôpital Bichat, Paris, France. (2) School of Pharmacy, University of Otago, Dunedin, NZ. (3) Global PK/PD, Lilly Research Centre, Windlesham, Surrey GU20 6PH, UK. (4) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden. (5) GlaxoSmithKline Pharmaceuticals, Collegeville, PA 19426, USA. (6) Centre for Applied Pharmacokinetic Research, School of Pharmacy and Pharmaceutical Sciences, The University of Manchester, Manchester, UK
Introduction: Following the first theoretical work on optimal design for nonlinear mixed effect models, this research theme has rapidly grown both in methodological and application developments. There are now several different software tools that implement an evaluation of the Fisher information matrix for population PK and PD models and proposed optimization of the experimental designs. In 2006, the Population Optimal Design of Experiments workshop was created with a meeting every year in May (www.maths.qmul.ac.uk/~bb/PODE/PODE2007.html). This year at PODE07 a special session was organized to present different software tools for population PK/PD optimal design and to compare them with respect to their statistical methodology.
Objectives: ) To present the different software tools; 2) To compare the statistical methods implemented in these tools; 3) To report the conclusion of the PODE07 meeting with respect to future software development in population PK/PD design.
Methods: The software tools will be compared with respect to: a) their availability, b) required language, c) library of PK or PD models, d) ability to deal with multiresponse models and/or with models defined by differential equations, e) approximations made to compute the Fisher information matrix, f) optimisation criteria, g)optimisation algorithms, h) ability to optimize design structure, i) ability to deal with constraints in sampling times, j) availability of optimisation trough sampling windows, k) assessment of user specified designs, l) ability to deal with unbalanced multiresponse designs, m) ability to deal with correlations between random effects, o) provided outputs ...
Results: The five software tools discussed at PODE07 are (in alphabetical order): PFIM (S. Retout & F. Mentré), PkStaMP (S. Leonov), PopDes (K. Ogungbenro & I. Gueorguieva) PopED (A. Hooker), and WinPOPT (S. Duffull). Tables comparing the software with respect to the different aspects described in the method section will be reported. The conclusions of the PODE07 meeting regarding future software development for optimal design in population PK/PD will be presented.
Oral Presentation: Methodology
Joakim Nyberg Sequential versus simultaneous optimal experimental design on dose and sample times
Joakim Nyberg, Mats O Karlsson, Andrew HookerDepartment of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Background: To increase the efficiency of trials in drug development, optimal experimental design has been used to optimize the sampling schedule. This produces a sampling schedule that will give the most information possible, given a model [1]. In addition, optimizing other model dependent variables, e.g. dose, has recently been possible with some optimal design software [2].
Aim: To investigate different optimization strategies, simultaneous vs. sequential optimization, when optimizing over both sampling schedule and dose.
Methods: The PKPD model used is a one-compartment PK model with first-order absorption and linear elimination linked with a sigmoidal Emax PD model [3]. Dose and sampling times are varied between groups with one dose, 2 PK and 3 PD samples taken within each group. Initial dose and sample times are evenly spread across the design space. Optimization is performed in POPED [2] on 1-5 groups with 20 individuals each. Three different optimization strategies are used; (i) dose-sampling times sequentially, (ii) sampling times-dose sequentially and (iii) dose-sampling times at the same time. When using the sequential approach, e.g. dose first, the dose is fixed after optimization and the optimal sampling schedule is calculated with the fixed optimized dose.
Results: Different doses give different optimal sampling schedules and vice versa. When optimizing the dose first and then sampling schedule, the efficiency [4] in some cases was 55 % of the efficiency using the simultaneous optimization technique. Sequential optimization of time first has at minimum in our setup an efficiency of 75 % of the simultaneous approach. The coefficients of variance (CV) of the model parameters are increased in most of the parameters when optimizing in sequence, with some parameters this increase can be as much as 190 %.
Discussion: Optimizing over both dose and sampling times could change the optimal experimental design with respect to sampling schedules and hence a whole trial. The large differences in efficiency with different optimization strategies indicate that these strategies are of importance. Furthermore, the decrease in efficiency also reflects an over-all increase in CV of the model parameters indicating a higher uncertainty to the model.
References:[1]. Mentré, F., Mallet, A. and Baccar, D., Optimal design in random-effects regression models. Biometrika, 1997. 84(2): p. 429-442.[2]. Foracchia, M., Hooker, A., Vicini, P. and Ruggeri, A., POPED, a software for optimal experiment design in population kinetics. Comput Methods Programs Biomed, 2004. 74(1): p. 29-46.[3]. Hooker, A. and Vicini, P., Simultaneous population optimal design for pharmacokinetic-pharmacodynamic experiments. Aaps J, 2005. 7(4): p. E759-85.[4]. Retout, S., Mentre, F. and Bruno, R., Fisher information matrix for non-linear mixed-effects models: evaluation and application for optimal design of enoxaparin population pharmacokinetics. Stat Med, 2002. 21(18): p. 2623-39.
Oral Presentation: Methodology
Justin Wilkins A flexible approach to modeling variable absorption in the context of repeated dosing: illustrated with rifampicin
Justin J Wilkins (1,2), Radojka M Savic (2), Mats O Karlsson (2), Grant Langdon (1), Helen McIlleron (1), Goonaseelan (Colin) Pillai (3), Peter J Smith (1), Ulrika SH Simonsson (2,4)(1) Division of Clinical Pharmacology, Department of Medicine, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa; (2) Division of Pharmacokinetics and Drug Therapy, Department of Biopharmaceutical Sciences, Uppsala University, Uppsala, Sweden; (3) Modeling & Simulation, Clinical Development & Medical Affairs, Novartis Pharmaceuticals AG, Basel, Switzerland; (4) Clinical Pharmacology, AstraZeneca R&D, Mölndal, Sweden
Introduction and Objectives: A flexible transit compartment absorption model (TCAM), in which a delay in the onset of absorption and a gradually changing absorption rate were modeled as drug passage through a chain of hypothetical compartments before reaching the absorption compartment, has been demonstrated previously [1]. The original approach, which allowed for numerical estimation of the number of transit compartments in the chain, was extended to support its application to data arising from a multiple-dosing schedule. We present this approach, demonstrating its superiority over a range of other techniques, using the complex and variable absorption of rifampicin in pulmonary tuberculosis patients as an example.
Methods: Three datasets containing 2 913 rifampin plasma concentration-time data points, collected from 261 South African pulmonary tuberculosis patients receiving regular daily treatment for at least 10 days, were pooled. The pharmacokinetics were analyzed using nonlinear mixed-effect modeling. Various approaches to characterizing the absorption of rifampicin were tested, including nonlinear absorption, dual absorption compartments, lag time parameters, and the extended TCAM.
Results: The best fit was provided by the adaptation of the TCAM for multiple dosing, which proved superior to the other methods tested for describing the absorption phase. The adjusted TCAM successfully described atypical profiles in the data, produced a substantial improvement in overall goodness-of-fit, and was able to reproduce the central tendency and variability of the data in simulations.
Discussion: The model assumes a gradual increase in the absorption rate, which results in a smoother initial rise in the plasma concentration towards the maximum than that offered by the other approaches tested. The extended TCAM provides the same benefits as the previously-described method despite its fundamentally empirical nature, it offers a better approximation of underlying physiological processes, as well as desirable computational properties and adds support for multiple dose regimens, which are much more common in practice than single-dose scenarios. The extension of the TCAM to multiple dose data relies on the assumption that the complete dose has reached the absorption compartment before the next dose is given. Further, the pharmacokinetics of rifampicin were successfully described in a patient population.
Conclusions: The multiple-dose adaptation of the TCAM is shown to be a robust and flexible method for dealing with highly variable absorption.
References: [1] Radojka M. Savi, Daniël M. Jonker, Thomas Kerbusch, Mats O. Karlsson. Evaluation of a transit compartment model versus a lag time model for describing drug absorption delay. PAGE 13 (2004) Abstr 513 [www.page-meeting.org/?abstract=513].
Oral Presentation: Methodology
Stefano Zamuner Optimal Design to Estimate the Time Varying Receptor Occupancy Relationship in a PET Experiment
Stefano Zamuner, Roberto GomeniGlaxoSmithkline - CPK/MS Verona - Italy
Objectives: Positron emission tomography (PET) is one of the most effective imaging techniques used to investigate ligand-receptor binding in living brain. In a PET experiment the total number of subjects and the number of PET scans per individual are limited either for cost or for ethical reasons (no more that 3 scan per subject). On this basis, the definition of the experimental design may become a critical issue for the assessment of the population PK/RO relationship especially when a slow dissociation model characterizes the drug-receptor interaction. The purpose of this work is to define an optimal PET experiment strategy and to present a case study where this methodology is applied to evaluate the PK/ RO relationship for a novel 5-HT1A antagonist.
Methods: A mechanistic model was used to estimate the time-varying relationship between RO and PK. A receptor association-dissociation model was established from pre-clinical data:
dRO/dt=kon·Cp·(RT-RO)-koff·RO
Designs were optimized using a D-optimality criterion as implemented in the WinPOPT software (1).
A comparison of population optimal designs with an empirical approach for designing a PET experiment aimed to estimate the PK/RO relationship is presented. PK parameters obtained from a population PK model were considered as a known independent variable in the exploration of parsimonious PET experiment. A series of alternative designs were considered to explore the influence of: a) the PET scan time allocation, b) the number of subjects to elementary design and, c) the number of dose levels.
Results: All designs explored have a total of 32 samples (N=2 PET scans x 16 subjects). In the empirical design PET scans were performed at Tmax and trough levels (four doses with same PET scan time for each dose group).The efficiency criterion (2) for any given design showed that all the optimized designs improved the empirical ones (efficiency increased of at least 500%). Allocation of the appropriate time appeared to be the most critical factor to improve efficiency compared with number of groups/doses. Finally, a Monte Carlo simulation was used to assess the performance of optimal designs by estimating the kon and koff parameters (fixed/random effects) from simulated data. Performances were measured as bias, precision and accuracy. The optimized design provided more accurate and reliable model parameter estimates.
Conclusions: The results show that population D-optimal design provided more accurate and reliable model parameter estimates.
References: [1] Retout S, at al. Comput. Meth. Prog. Biomed. 2001; 65(2):141-51.[2] Duffull S, et al. J. Pharmacokinet. Pharmacodyn. 2005; 32:441-57.
Oral Presentation: Model Building Session
Chantaratsamon Dansirikul Insulin secretion and hepatic extraction during euglycemic clamp study: modelling of insulin and C-peptide data
C. Dansirikul, M.O. KarlssonDivision of Pharmacokinetics and Drug Therapy, Uppsala University, Uppsala, Sweden
Objectives: Insulin and C-peptide are co-secreted peptides from beta-cells on an equimolar basis. While 40-85% of insulin have been reported to be extracted by liver after secretion1-3, only a negligible amount of C-peptide is lost on a single pass through the liver. This analysis was performed to characterize the endogenous insulin secretion profile and its hepatic extraction when C-peptide data were utilized as a measure of pre-hepatic secretion.
Methods: Insulin and C-peptide concentration-time data were available from 15 healthy volunteers, who participated in a 24-hour euglycemic clamp study (without insulin injection). NONMEM VI with First Order Conditional Estimation (FOCE) method was used during the analyses. A flexible staircase of zero order input model4 was fitted to C-peptide concentration-time data whilst their disposition pharmacokinetics parameters (two-compartment model) were fixed to values from literature5. Different number of change-points was tested (2, 3, 4, and 5 change-points). Hepatic extraction of insulin was subsequently estimated using a sequential modelling approach. As such population parameters describing C-peptide secretion were fixed to their estimated values, and hepatic extraction of insulin was then estimated. During the estimation of hepatic insulin extraction, the Laplacian estimation method was used, in which it allowed the contribution of insulin concentration below the low limit of quantification (LLOQ) to the likelihood estimates.
Results: A zero order input model with 5 change-points was chosen to portray endogenous secretion profile of C-peptide. These change-points were at baseline, and 3, 6, 12, and 18 hours after study started. Secretion of C-peptide at baseline was 103 pmol/min. Its secretion then decreased over time up to 18 hours. Secretion was 90%, 77%, 63%, and 55% of the value at baseline, for the first four steps. After the last change-point (18 hours), secretion increased slightly to 61% of the value at baseline. Hepatic insulin extraction was estimated to be 56%. Inter-individual variability of hepatic extraction was not statistically significant to be included in the model (log-likelihood ratio test, at p-value of 0.05).
Conclusions: During a 24-hour euglycemic clamp, secretion of C-peptide and insulin changed over time. Hepatic insulin extraction was estimated to be similar to previous studies.
References:[1]. C Cobelli et al. Diabetes 37: 223-31 (1988).[2]. A Tura, et al. Am J Physiol Endocrinol Metab 281: E966-74 (2001).[3]. JJ Meier, et al. Diabetes 54: 1649-56 (2005). [4]. A Lindberg-freijs et al. Biopharmaceutics & Drug Disposition 15: 75-86 (1994).[5]. E Van Cauter, et al. Diabetes 41: 368-77 (1992).
Oral Presentation: Model Building Session
Jeroen Elassaiss-Schaap Interspecies Population Modeling Of Pharmacokinetic Data Available At The End Of Drug Discovery
Jeroen SchaapPK-PD/M&S, Clinical Pharmacology and Kinetics, Organon N.V., The Netherlands
Objectives: One of the challenges in pharmacokinetics is to predict pharmacokinetic profiles of new chemical entities in the transition from preclinical to clinical research. It is widely accepted that current allometric techniques provide insufficient quality in predictions of human PK. An extension of standard methods is non-linear modeling of all available preclinical PK data [1]. Application of this method in the pharmaceutical industry has been reported with extensive datasets, sometimes including human data [2,3,4]. We developed a interspecies nonlinear mixed-effects model for the preclinical pharmacokinetics of an (anonymous) compound as a test-case for feasibility of such an approach in an early stage of drug development.
Methods: The nonlinear population model was developed in NONMEM V. Model selection was performed on the basis of goodness-of-fit plots including Dedrick graphs, standard diagnostics in NONMEM output files and simulations sampled from the variance-covariance matrix. No covariates other than species and body weight were included in the dataset. The dataset contained 4 species (mouse, rat, dog, monkey), 2 routes of dosing (iv and po) and 2-3 richly sampled animals per route or n=3 samples per time point in mouse (one sample per animal).
Results: An interspecies population model was developed with reasonable success. Predictions were close to observed data but more importantly the model successfully facilitated project team discussions around hypotheses applied in the prediction of human pharmacokinetics and its uncertainty. The final model featured some extra parameters to capture the profile of one species (monkey); predictions were therefore augmented with scenarios for in/exclusion of these parameters. Limitations of the model were its instable bootstrap results and the amount of manpower spent, approximately 0.1 FTE. Such investment of resources is relatively large for an early stage of drug development. Of technical interest is that the assumption of multivariate normal distribution of uncertainty in structural model parameters seemed to hold only after lognormal specification of parameters.
Conclusion: Interspecies mixed-effect modeling was possible also with a pharmacokinetic dataset such as typically available at the end of drug discovery. The amount of manpower currently required however seems to limit routine application of this technique in the transition of preclinical to clinical research.
References:[1] H. Boxenbaum. Interspecies scaling, allometry, physiological time, and the ground plan of pharmacokinetics. J.Pharmacokinet.Biopharm. 10 (2):201-227, 1982.[2] V. F. Cosson, E. Fuseau, C. Efthymiopoulos, and A. Bye. Mixed effect modeling of sumatriptan pharmacokinetics during drug development. I: Interspecies allometric scaling. J.Pharmacokinet.Biopharm. 25 (2):149-167, 1997.[3] K. Jolling, J. J. Perez Ruixo, A. Hemeryck, A. Vermeulen, and T. Greway. Mixed-effects modelling of the interspecies pharmacokinetic scaling of pegylated human erythropoietin. Eur.J.Pharm.Sci. 24 (5):465-475, 2005.[4] T. Martin-Jimenez and J. E. Riviere. Mixed-effects modeling of the interspecies pharmacokinetic scaling of oxytetracycline. J.Pharm.Sci. 91 (2):331-341, 2002.
Oral Presentation: Model Building Session
Massimiliano Germani A population PK-PD method for categorical data analysis of progesterone antagonist activity in the rabbit McPhails model
Francesca Del Bene, Massimiliano Germani, Maurizio Rocchetti, Alex De Giorgio-Miller, Nick Pullen, Peter Bungay, Chris Kohl & Piet van der GraafACCELERA - Nerviano Medical Sciences (Italy) & Pfizer Global Research & Development (United Kingdom)
Objectives: The McPhail's test [1] of endometrial differentiation and thickening is a commonly used method for preclinical assessment of progesterone antagonist activity in vivo [see, for example, [2]). However, as far as we know, a method for integrated PK/PD analysis that allows for a comparison of the dynamic effects of compounds in this model has not yet been described, possibly due to the categorical nature of the endpoint generated. Therefore, the objective of this study was to develop an ordered categorical population PK-PD analysis methodology for describing the effect of progesterone antagonists in a McPhail's preclinical model of endometrial thickening. Three compounds were tested and analysed for this purpose.
Methods: The efficacy of three compounds was evaluated after repeated administrations (once or twice daily for four consecutive days) of different dose levels (n=4to four rabbits/dose). McPhail's test results were expressed as a score from 0-4 depending on the degree of endometrial tissue differentiation and thickening. Different sparse pharmacokinetic sampling procedures were adopted for the three compounds, dividing the animals into two groups of two rabbits each with a number of samples ranging from four to eleven depending on the investigated compound.
Results: Due to the limited number of animals available, the efficacy data, expressed as McPhail's score, were reduced to a binary PD variable with value equal 1 and 0 according to a McPhail's score d" 2 and >2, respectively. Subsequently, for each compound, the individual cumulative unbound AUC values obtained from population PK analysis were related to the PD variable applying a logistic regression model implemented in NONMEM in order to estimate the probability of observing an outcome less or equal than 2.With the aim of minimizing the number of parameters and enhancing statistical power, the PK-PD categorical model was further modified and implemented in Matlab for performing a joint analysis of all the compounds. The AUC values related to the 80% and 90% of probability to obtain the outcome of interest (McPhail's score d" 2) were calculated and showed for each compound.
Conclusions: We have developed a method that allows for a simultaneous analysis and comparison of PK-PD relationships of series of compounds in the McPhails model of progesterone antagonist activity. The method has been implemented in NONMEM and Matlab and should provide a quantitative basis for the rational selection of preclinical candidates for further clinical development.
References: [1] McPhails, M.K. (1934). J. Physiol. 83: 145-156.[2] Kurata, Y. et al. (2005). J. Pharmacol. Exp. Ther. 313: 916-920.
Oral Presentation: Stuart Beal Methodology Session
Robert Bauer Advanced Population Analysis Features in the S-ADAPT/MCPEM Program
Robert J. BauerXOMA (US) LLC
Objective: To demonstrate advanced population analysis and simulation abilities of S-ADAPT.
Methods: S-ADAPT is a Fortran 95 open-source, free program distributed by U. of Southern California, Biomedical Simulations Resource department (USC, BMSR), and has been successfully used to analyze clinical data for Raptiva, consisting of six differential equations and 16 model parameters [1]. The S-ADAPT Program provides an environment for performing population analysis of data, with or without covariates, using complex PK/PD models with extensive simulation tools. The program provides several interface types: Interactive command line, interactive menu, or execution of a series of commands from a script file, allowing complete batch-processed control [2]. The command line allows one to evaluate algebraic expressions, and store the results in user defined variables. The run-time environment also includes a built-in database for storing and retrieving data, and maintaining analysis results. A population analysis validation system is available that generates simulated data sets, analyzes each data set, and outputs bias and precision statistics. The open-source code may be compiled by Intel or Compaq Visual Fortran for Microsoft Windows, Intel compiler for Linux, and g95 (MingW), a free FORTRAN compiler. The user fills out a FORTRAN model file from a template, providing the code necessary to describe the PK and PD model functions, residual error functions, parameter transformations, covariate models for the population parameters, and differential equations if needed. Template model files for basic PK models based on the various Advan/Trans algorithms in NONMEM are available, to be used as is, or modifiable by the user. Data may be imported and used in NONMEM format. Nonlinear mixed effects population analysis may be performed at two or three hierarchical stages (incorporating prior information), and maximization methods include iterative two-stage and Monte-Carlo Parametric Expectation-Maximization (MCPEM) methods [3]. Deterministic and Monte-Carlo algorithms may be simultaneously used to increase efficiency of the analysis while retaining accuracy. Inter-occasion variability, population mixtures, and below quantification limit (BQL) data may also be modeled in S-ADAPT. The population analysis provides standard error analyses, post-hoc analyses, and graphical viewing of post-hoc results [2,3,4]. In addition, two or three-stage hierarchical Bayesian analysis may be performed in S-ADAPT, to provide quantile ranges for the estimation of the population parameters. Extensive capabilities for importing and exporting data and/or analysis results are provided, including easy export of data, dosing information, initial parameters values and prior information to WinBUGS data files. S-ADAPT also has the ability to distribute the computation effort of a single analysis or multiple analyses across several computers to reduce analysis time.
Results: Comparisons of results from S-ADAPT with NONMEM and WinBUGS have been performed, as well as validation analyses using multiple sets of simulated data [3,4,5]. S-ADAPT's performance was very stable, and provided population means, inter-subject variances, and their standard errors with little bias. S-ADAPT tended to perform slowly with simple one and two compartment PK models, but performed more efficiently with more complex PK/PD models involving differential equations.
Conclusions: The S-ADAPT program using MCPEM methods offers a robust and versatile environment for PK/PD modeling and population analysis.
References:[1] Ng CM, Joshi A, Dedrick R, Garovoy M, Bauer R. Pharmacokinetic-pharmacodynamic-efficacy analysis of efalizumab in patients with moderate to severe psoriasis. Pharmaceutical Research. 2005;22(7):1088-1100.[2] S-ADAPT/MCPEM User's Guide [computer program]. Version 1.52. Berkeley, CA.; 2006. HYPERLINK "http://bmsr.usc.edu/Software/Adapt/sadapt.html" http://bmsr.usc.edu/Software/Adapt/sadapt.html.[3] Bauer RJ, Guzy S. Monte Carlo parametric expectation maximization (MC-PEM) method for analyzing population pharmacokinetic/pharmacodynamic data. In: D'Argenio DZ, ed. Advanced Methods of Pharmacokinetic and Pharmacodynamic Systems Analysis. Vol 3. Boston: Kluwer Academic Publishers; 2004:135-163.[4] Bauer RJ, Guzy, S, and Ng, C. A survey of population analysis methods for complex pharmacokinetic and pharmacodynamic models with examples. AAPS Journal 2007; 9(1) Article 7, E60-E83.[5] Girard P, Mentre F. A comparison of estimation methods in nonlinear mixed effects models using a blind analysis. PAGE Meeting, Pamplona, Spain. 2005. Abstract 834.
Oral Presentation: Stuart Beal Methodology Session
Marc Lavielle The SAEM algorithm and its implementation in MONOLIX 2.1
M. Lavielle (1,2), H. Mesa (3,2), F. Mentré (4,5)(1) INRIA Futurs, Paris, France; (2) University Paris 5, Paris, France. (3) University of La Habana, Cuba. (4) INSERM, U738, Paris, France ; (5) Université Paris 7, Paris, France ; AP-HP, Hôpital Bichat, Paris, France.
Introduction: The statistical model for most population PK/PD analyses is the nonlinear-mixed effects model (NLMEM). As opposed to linear models, there are statistical issues to express the optimisation criteria for these nonlinear models so that first approximation methods (FO and FOCE) based on linearization of the model were proposed. It is well known that these methods have several methodological and theoretical drawbacks. They are also very sensitive to initial estimates which make lot's of run to failed to converge with a waste of time for the modeller. Population analyses are now used not only to provide mean estimates but also to make model selection, hypothesis testing, simulations and predictions based on all the estimated components: better estimation methods are therefore needed.
The SAEM (Stochastic Approximation EM) algorithm avoids any linearization and is based on recent statistical algorithms. This algorithm is a powerful tool for Maximum Likelihood Estimation (MLE) for very general incomplete data models. The convergence of this algorithm to the MLE and its good statistical properties have been proven. The SAEM algorithm is implemented in the free MONOLIX software that can be downloaded at http://www.monolix.org. It is possible to download the full Matlab version of MONOLIX 2.1 and/or only a compiled version of the software that does not require Matlab.
Objectives: 1) To present the new release of MONOLIX (version 2.1); 2) To illustrate the estimation capabilities of the algorithm on some difficult examples; 3) To discuss how MONOLIX could be extended to other incomplete data problems
Methods/ Results:
1) Version 2.1 of MONOLIX has major improvement compare to the previous one: a) a library of PK and PD models, some defined with differential equations, b) ability to deal with data below limit of quantification, c) lot's of interactive graphical outputs...
2) This version of MONOLX was tested on several data sets. For some of which it was difficult to get the FOCE algorithm to converge (models with Michaelis-Menten elimination and multiple doses, models with lag time, etc...). Results and demonstrations will be presented.
3) The SAEM algorithm can handle very general non linear mixed effect models:
left-censored data,
models defined by stochastic differential equations,
multi-responses models,
mixtures of distributions,
inter-occasion variability,
missing covariates,
time to event data,
count data,
dropouts,
...
Indeed, all these models are statistical models that include a set of observations and a set of non observed data. SAEM requires the computation of the conditional distribution of these non observed data and their simulation at each iteration. Some of these models are already implemented in version 2.1 of the MONOLIX software and other extensions will be discussed.
Acknowledgements: Version 2.1 of MONOLIX was developed with the financial support of Johnson & Johnson Pharmaceutical Research & Development, a Division of Janssen Pharmaceutica N.V.
Oral Presentation: Stuart Beal Methodology Session
Fahima Nekka What Additional Information Can we Retrieve When Compliance is Accounted For? An explicit Compliance-Pharmacokinetic Formalism
F. Nekka, J. Li and F. FenneteauUniversite de Montreal
Objectives: Compliance reflects the patient drug input and has to be considered as a history of drug intake. Quality of compliance should be judged through its impact on drug therapy. As such, the compliance phenomenon has to be explicitly formalized and properly characterized in terms of fate of drugs. Having this in mind, we assessed the effect of compliance and extracted its main characteristics.
Methods: We formalized the notion of randomness in drug intake behavior and explicitly included it into a pharmacokinetic model to characterize its effect on the concentration variation. For this, we adopted a stochastic approach to model non-compliance behavior using a general distribution which can be accessible by recording patient drug history
Results: We proved that the stochastic feature of drug intake, previously treated as a nonsystematic noise to a deterministic model, is in fact an inherent component which has to be considered as a part of the mechanistic model. Indeed, in the case of poor compliance, we can retract the source of what is generally classified as pure noise. Thus, the variable compliance behavior adds additional variation to the regular oscillation curve that we have properly characterized.
Conclusions: The benefit of this approach is evident since the knowledge of the origin of the variation will allow an objective intervention. We have developed a general formalism of the compliance phenomenon that can be adapted to different situations, i.e., different compliance patterns in various pathologies.
References: [1] B. Vrijens and J. Urquhart. New findings about patient adherence to prescribed drug dosing regimens: an introduction to pharmionics. Eur. J. Hosp. Pharm. Sci. 11(5):103-106 (2005).[2] J. Li and F. Nekka, A Pharmacokinetic Formalism Explicitly Integrating the Patient Drug Compliance, Journal of Pharmacokinetics and Pharmacodynamics, vol. 34, no. 1, pp. 115-139, 2007.
Tutorial
Marc BUYSE Validation of statistically reliable biomarkers
Marc BuyseIDDI, Belgium
DefinitionsBiomarkers play an increasing role in the development of new cancer treatments. A biomarker is defined as "a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention". The term biomarker covers characteristics measured at baseline as well as those measured repeatedly over time, before, during or after treatment. Clinical data, laboratory data, imaging data, gene expression and proteomic data can all be considered potential biomarkers.
Uses of biomarkersBiomarkers can be useful as:
prognostic factors that predict the outcome of individual patients in terms of a clinical endpoint
predictive factors that predict the effect of a specific treatment on a clinical endpoint in groups of patients
surrogate endpoints that replace a clinical endpoint of interest
Biomarkers can be used to stratify the patients at entry in clinical trials, to select the patients eligible for clinical trials, to monitor patients and guide treatment decisions, or to substitute for a clinical endpoint in the evaluation of the effects of new treatments.
Requirements for biomarkersThe clinical literature is replete with examples of the use of biomarkers, but in many cases these have not been properly identified and validated, resulting in a large number of false claims, inappropriate trial designs, and ultimately sub-optimal use of biomarkers for patient management.
Requirements for prognostic biomarkers: The baseline value of the biomarker, or changes in the biomarker over time, should be correlated with the clinical endpoint in untreated or in treated patients
Requirements for predictive biomarkers: The baseline value of the biomarker, or changes in the biomarker over time, should be correlated with the effect of treatment on the clinical endpoint
Requirements for surrogate biomarkers: The difference in the biomarker values between two randomized treatments should be correlated with the difference in the clinical endpoint
Study designsDifferent study designs are required for the identification of biomarkers. Case-control or cohort studies are sufficient to identify prognostic biomarkers, large randomized trials are needed to identify predictive biomarkers, and multiple randomized trials are needed to identify surrogate biomarkers. In all cases, the biomarker should be validated either through cross-validation (internal validation in the discovery set) or in different trials (external validation in a confirmatory set).
Validation criteria The criteria used to validate biomarkers include classification measures (sensitivity and specificity, ROC curves), treatment effect measures (odds ratios, hazard ratios) association measures (correlation coefficients, information theory based measures), and prediction measures (the surrogate threshold effect). These various criteria all have advantages and limitations that will be illustrated with several examples in the field of cancer.
Poster: Applications- Anti-infectives
julie bertrand Influence of pharmacogenetic on pharmacokinetic interindividual variability of indinavir and lopinavir in HIV patients (COPHAR2 ANRS 111 trial)
J. Bertrand (1), X. Panhard (1), A. Tran (2), E. Rey (2), S. Auleley (1), X. Duval (1, 3, 5), D. Salmon (4), J.M. Tréluyer (2), F. Mentré (1) and the COPHAR2- ANRS 111 study group(1) INSERM, U738, Paris, France ; Université Paris 7, Paris, France ; AP-HP, Hôpital Bichat, Paris, France ; (2) AP-HP, Hôpital Cochin, Département de pharmacologie clinique EA3620, Paris, France ; (3) AP-HP, Hôpital Bichat, Service des Maladies Infectieuses B, Paris, France ; (4) Université Paris 5, AP-HP, Hôpital Cochin, Service de Médecine interne Paris, France ; (5) CIC 007, Hôpital Bichat, Paris, France
Objectives: To evaluate the effect of genetic polymorphisms on indinavir and on lopinavir pharmacokinetic (PK) variability in HIV-infected patients initiating a protease inhibitor (PI) containing highly active antiretroviral therapy (HAART) and to study the link between concentrations and short term response or metabolic toxicity.
Methods: Among the patients enrolled in the COPHAR 2 - ANRS 111 trial, 34 and 38 PI naive-patients initiating lopinavir/ritonavir (LPV) or indinavir/ritonavir (IDV), respectively, were studied. At week 2, 4 blood samples were taken before and up to 12 hours following the drug intake to measure PI plasma levels. Genotyping for CYP3A4, 3A5, and MDR1 (exon 21 and 26) was performed. For each PI (LPV and IDV), a population PK model was developed in order to estimate the model parameters and to analyse the genotypes influence. The SAEM algorithm implemented in the Monolix software version 2.1 was used [1, 2]. For each patient, mean (Cmean), minimum (Cmin) and maximum (Cmax) plasma concentrations were derived from the estimated individual PK parameters. For each PI, the link between plasma level and HIV RNA level variation between day 0 and Week 2 or variations of metabolic levels between day 0 and Week 4 were studied using Spearman nonparametric correlation tests.
Results: A one-compartment model with first-order absorption and elimination best described LPV and IDV concentrations. No significant genetic effect was found on LPV pharmacokinetics. For IDV, patients *1B/*1B for CYP3A4 gene had an IDV absorption divided by 3.70 compared to *1A/*1B or *1A/*1A genotypes (p= 0.02). With respect to link between PK and short-term efficacy, Cmean and Cmin were positively correlated to HIV RNA variation (p=0.02, p=0.03, respectively) in the IDV group. No significant relationship was found between LPV or IDV concentrations and metabolic toxicity. However, in the IDV group, patients with the *1A/*1B or *1A/*1A genotypes had significantly higher increase in triglycerides during the 4 weeks of treatment (p=0.02).
Conclusions: This study points out the role of genetic polymorphisms on IDV pharmacokinetic variability and confirms the link between IDV exposure and short term efficacy. No such effect could be found, in this sample, for LPV.
References:[1] Kuhn, E. & Lavielle, M. Maximum likelihood estimation in nonlinear mixed effects models. Comput. Stat. Data Anal. 49, 1020-1038 (2005).[2] HYPERLINK "http://software.monolix.org/" http://software.monolix.org/.Poster: Applications- Anti-infectives
Stefanie Hennig Tobramycin in paediatric CF patients - TCI or One dose fits all
S. Hennig(1), R. Norris(2) C. M. J. Kirkpatrick(1)1.School of Pharmacy, The University of Queensland, Brisbane, QLD 4072, Australia. 2. Australian Centre for Paediatric Pharmacokinetics, Mater Pharmacy Services, Brisbane, QLD 4101, Australia.
Objectives: The aim of this study was to i) develop a population pharmacokinetic model for tobramycin in paediatric cystic fibrosis (CF) patients ii) investigate the influence of covariates, and iii) use the quantified random and predictable components of variability to asses the need for target concentration intervention for tobramycin (TCI).
Methods: Retrospective demographic, dosing and concentration data was collected from 35 cystic fibrosis patients (21 female) aged 0.5 - 17.8 years old from whom 318 tobramycin plasma concentrations were obtained during standard clinical care monitoring. A nonlinear mixed-effect modelling approach (software: NONMEM V, G77, FOCE+I) was used to describe the population pharmacokinetics of tobramycin. Simulations were performed with NONMEM using weight based dosing with a weight from a covariate distribution model and analysed using S-Plus.
Results: A two-compartment model best described the tobramycin data, with population parameter estimates for clearance of central compartment (CL) of 6.37 L/h/70 kg; volume of central compartment (Vc) of 18.7 L/70 kg; inter-compartmental clearance of 0.393 L/h and volume of peripheral compartment of 1.32 L. The inclusion of total body weight allometrically as covariate reduced the random component of between subject variability (BSV) in CL from 50.1% to 11.7% and in Vc from 62.2% to 11.6%. The between occasion variability on CL was estimated in the final model as 6.5%. With the "one dose fits all" approach about 1/3 of the patients would be at risk of over-exposure. A once daily dose of 10 mg/kg tobramycin was found to provide the best compromise between success and over-exposure.
Conclusions: This study provides the first pharmacokinetic model of once-daily IV tobramycin for the use of TCI in paediatric CF patients. Simulations showed that one dose does not fit all and TCI and dose adjustment is required.
Poster: Applications- Anti-infectives
Déborah Hirt Effect of CYP2C19 polymorphism on nelfinavir to M8 biotransformation in HIV patients.
D. Hirt, F. Mentré, A. Tran, E. Rey, S. Auleley, X. Duval, D. Salmon, J.M. Tréluyer, and the COPHAR2- ANRS study group.Hôpital Cochin-Saint Vincent de Paul
Objectives: To evaluate the effect of CYP2C19 polymorphism on nelfinavir and M8 pharmacokinetic variability in HIV infected patients and to study the link between pharmacokinetic exposure and short term efficacy and toxicity.
Methods: One hundred and twenty nelfinavir and 119 M8 concentrations were measured in 34 naive-patients enrolled in the COPHAR 2-ANRS 111 trial. Two weeks after initiating the treatment, 4 blood samples were taken 1, 3, 6 and 12 hours after drug administration. Genotyping for CYP3A4, 3A5, 2C19 and MDR1 (exon 21 and 26) was performed. A population pharmacokinetic model was developed in order to describe nelfinavir and M8 concentration time-courses and estimate inter-patient variabilities. The influence of individual characteristics and genotypes were tested in the population model using a likelihood ratio test. Maximum, minimum and mean individual concentrations derived from the estimated individual parameters were linked to short-term efficacy and toxicity using Spearman nonparametric correlation tests.
Results: A one-compartment model with first-order absorption, elimination and metabolism to M8 best described nelfinavir data. M8 was modelled by an additional compartment. Mean pharmacokinetic estimates and the corresponding inter-subject variabilities (%) were: absorption rate 0.17 h-1(99%), absorption lag time 0.82 h, apparent nelfinavir clearance 52 L/h (49%), apparent nelfinavir volume of distribution 191 L, M8 elimination rate constant 1.76 h-1 and the parameter expressing nelfinavir to M8 transformation that could be estimated (CLm/Vm) 0.39 h-1 (59%). This parameter CLm/Vm was divided by 1.98 in AG or AA patients for CYP2C19*2 compared to GG patients. With respect to link between pharmacokinetic and short term metabolic toxicity, nelfinavir minimum plasma concentrations and mean concentrations were positively correlated to increase of glycemia (p=0.03, p=0.02). Mean concentrations was also positively correlated with triglycerides increase (p=0.04).
Conclusions: CYP2C19*2 polymorphism influences nelfinavir - M8 pharmacokinetics.
Poster: Applications- Anti-infectives
Hui Kimko Population Pharmacokinetic Analysis To Support Dosing Regimens Of Ceftobiprole
Hui Kimko, Steven Xu, Bindu Murthy, Mahesh Samtani, Partha Nandy, Richard Strauss, Gary NoelJohnson & Johnson Pharmaceutical Research & Development; Raritan, NJ; Titusville, NJ
Objectives: Ceftobiprole is a first-in-class broad-spectrum cephalosporin with activity against methicillin-resistant staphylococci. Dosage adjustment strategy was supported by a population analysis approach.
Methods: Ceftobiprole plasma pharmacokinetic (PK) data from healthy volunteers and patients in eight phase 1 (n=162), one phase 2 (n=27) and two phase 3 (n=414) clinical trials were combined to identify factors contributing towards inter-individual variability in PK of ceftobiprole. NONMEM was used to develop a PK model with statistically significant covariates. To test clinical relevance of the covariates, the percentage of time the concentrations of ceftobiprole above MIC during a dosing interval (%T>MIC) was estimated, which is the PK/PD index of ceftobiprole.
Results: A three-compartment model with first-order elimination provided the best fit for the ceftobiprole plasma- concentration time-profile. In the final population PK model, clearance (CL) was a function of creatinine clearance and health status (healthy volunteers vs. patients); volume of distribution in the central compartment (V1) was a function of body weight and health status; volume of distribution in a shallow peripheral compartment (V2) was a function of gender and health status; volume of distribution in a deep peripheral compartment (V3) was a function of gender. Of these covariates, only the renal function was identified as the clinically relevant factor using %T>MIC. Age (i.e., ³ 18 years old) was neither statistically nor clinically influential when creatinine clearance was included as a covariate. The influence of other statistically significant covariates such as gender and body weight on PK was negligible with suggested dosing regimen adjustments based on degree of renal function. Exploratory analyses with race and concomitant medications as covariates suggested no change in PK of ceftobiprole due to these factors.
Conclusion: The population PK analyses support the proposed labeling dosing for ceftobiprole and dosing adjustments based on creatinine clearance only. No adjustments to ceftobiprole dosing appear to be warranted for age, race, gender, body weight or the assessed concomitant medications.
Poster: Applications- Anti-infectives
Grant Langdon PK-PD modelling to support go/no go decisions for a novel gp120 inhibitor
PLS Chan, E van Schaick, G Langdon, J Davis, T Parkinson, L McFadyen
Objectives: PF-00821385 is a specific inhibitor of HIV-1 gp120 mediated cell-cell fusion. Inhibitors of HIV-1 gp120 represent a novel mechanistic approach with the promise of activity against both CXCR4 and CCR5-using B-clade viruses. The aim of this analysis was to describe the population pharmacokinetics (PK) of PF00821385 in healthy subjects and to use this to update a Pharmacokinetic-Pharmacodynamic Disease (PK-PD-Disease) model based on preclinical data to predict a likely oral dose to achieve a 1.5 log drop in viral load (VL) following 10 days of treatment.
Methods: Single doses of PF-00821385 were dosed orally to 24 healthy male volunteers. The maximum tolerated dose was 1300 mg. A total of 969 concentrations were available for modelling and population PK parameters were estimated using NONMEM.Using only scaled preclinical data (animal PK and in vitro IC90 data) a PK-PD-disease model previously developed to describe viral load changes for CCR5 antagonists was adapted to evaluate, by simulation, possible clinical doses of PF00821385. Trial Simulator 2.2 (Pharsight Corp.) was used. Simulations for gp120 antagonists are complicated by high variability in assay sensitivity, thus two IC50 targets were utilised viz., low of 121 ng/mL and high of 489 ng/mL. These represent scaling of the median and upper limit of sensitivity of IC90 values for PF-00821285 in a panel of clinical virus isolates (Virologic Phenosense assay).New simulations were performed with the model updated with human PK data.
Results: A two-compartment disposition model with first-order absorption was used to fit the log-transformed plasma concentrations. The population estimated rate of absorption (ka) was 0.597 h-1 with an elimination half-life during the initial phase of 0.3 hours. If the "true" IC50 for PF-00821385 was low (121 ng/mL), the minimum predicted B.I.D. dose required for 1.5 log drop in VL was 319 mg. If the "true" IC50 was high (489 ng/mL), the minimum predicted Q.D. and B.I.D. doses were above 1300 mg. Further simulations were performed with theoretical assumptions on ka changes to predict likely doses for controlled-release formulations. A minimum predicted dose of 1000 mg was required for IC50 of 489 ng/mL.
Conclusions: The predicted dose required to target 70% of Clade B virus isolates was close to or above the maximum tolerated dose of PF-00821385. A controlled release formulation could improve the PK characteristics, but would only be of utility in the treatment of a narrow spectrum of viral sensitivities. Based on this outcome, further development of PF-00821385 will not be pursued.
Poster: Applications- Anti-infectives
Rocio Lledo Population Pharmacokinetics of Saquinavir in rats after IV and IP administration. An approach to Saquinavir/Ritonavir Pharmacokinetic interaction.
R. Lledó-García (1), M. Merino Sanjuán (1), L. Prats (1), A. Nácher (1), V.G. Casabó (1)(1) Department of Pharmacy and Pharmaceutical Technology, University of Valencia, Valencia, Spain.
Objectives: The protease inhibitor (PI) saquinavir (SQV) is characterized by a low and variable oral bioavailiability [1], which can be increased by the addition of ritonavir (RTV). This combination is currently used in the management of AIDS. Although the interaction between these two drugs has already been reported in previous studies [2], so far the roles of liver and intestine in SQV first pass metabolism have not been clarify. This study aimed to assess SQV pharmacokinetic disposition profile when IV administered and assess the hepatic first pass metabolism when IP administered.
Methods: 48, 24 and 12 mg doses of SQV were IV and IP administered to rats subjected to jugular vein cannulation. In addition, 24 and 6 mg dose of SQV/RTV were IV administered. Plasma samples were analysed for drugs content by HPLC/UV. As a first approach to data, a non compartmental analysis (WinNonLin) was performed. Subsequently, a stepwise population pharmacokinetic approach was performed using NONMEM. Throughout the development of the model the interaction between drugs was assessed.
Results: The non-compartmental analysis shows an insignificant hepatic first pass metabolism where an important one was expected, being likely located at the gastrointestinal tract. In the population approach, a two-compartment model considering a plasma protein dynamic binding and an elimination process from central compartment following Michaelis-Menten kinetics, were used to describe SQV disposition processes. Moreover, the IP incorporation process was described by a precipitation of the drug in the IP cavity, being dissolution limitative factor for absorption. The main PK parameters are shown: Vc (0.244L), plasma proteins binding and unbinding constants resulted in Kb (2.71x10-06 L/mg.h) and Ku (0.137h-1), respectively; Vm (56.7 mg/h), Km (57.7 mg/L), being the last multiplied by 4.88 when RTV is coadministered. IP Ka (1 h-1), limited by the drug dissolution.
Conclusions: Data analysis showed non linear disposition processes for SQV and located drugs interaction at the elimination process, so that RTV inhibits SQV metabolism. Opposite to the expected result, IP data showed that SQV low bioavailability was not mainly due to its hepatic first pass metabolism, escaping most part of the drug from liver. However, further studies are being conducted, involving an oral administration required to properly describe the processes involved in SQV presystemic losses.
References:[1]. Guiard-Schmid, J.B., J.M. Poirier, J.L. Meynard, P. Bonnard, A.H. Gbadoe, C. Amiel, F. Calligaris, B. Abraham, G. Pialoux, P.M. Girard, P. Jaillon and W. Rozenbaum, High variability of plasma drug concentrations in dual protease inhibitor regimens. Antimicrob Agents Chemother, 2003. 47(3): p. 986-90.[2]. Shibata, N., W. Gao, H. Okamoto, T. Kishida, K. Iwasaki, Y. Yoshikawa and K. Takada, Drug interactions between HIV protease inhibitors based on physiologically-based pharmacokinetic model. J Pharm Sci, 2002. 91(3): p. 680-9.Poster: Applications- Biologicals/vaccines
Balaji Agoram Application of mechanism-based population PKPD modelling in the rational selection of clinical candidates: an anti-IgE antibody example.
Balaji Agoram, Steven Martin, Piet van der Graaf.Pfizer, Inc. UK
Objectives: Design, selection and development of clinical candidates should, optimally, incorporate mechanistic PKPD knowledge gained from previous experience with the same therapeutic targets. We have illustrated this idea using reported PKPD analyses on omalizumab, a humanised monoclonal antibody for the treatment of asthma and allergic rhinitis.
To characterise the relationship between in vitro potency and the in vivo efficacy profile of anti IgE antibody omalizumab using simulated data generated from reported PKPD models for omalizumab. To evaluate possible in vitro changes to the molecule to improve its in vivo PD profile.
Methods: A PKPD model of omalizumab was gathered from literature1 and implemented in the nonlinear mixed effects modelling package, NONMEM. With this PKPD model, deterministic and stochastic simulations were performed using mean and uncertainty distributions of the parameters to characterise the relationship between in vitro affinity and PK parameters on the in vivo time-course of effect profile.
Results: Sensitivity analysis indicated that a 5-fold increase in in vitro affinity is likely to translate into increased efficacy and/or reduced dose size. Beyond this limit, further increases in affinity are unlikely to result in additional clinical benefit. The clinical efficacy appears limited by serum half-life of the compound. This was confirmed by the sensitivity analysis on the serum half-life. Increasing half-life at same potency resulted in increased efficacy.
Conclusions: The mechanism-based PKPD approach has provided a framework to quantify the nonlinear relationship between in vitro/in vivo affinity and clinical potency/efficacy for anti-IgE antibodies and can be used for efficient selection of follow-on candidates. The model allows for prediction of in vivo dose-response relationships on the basis of in vitro characteristics and hence for rational and efficient compound design. For example, the model predicts that large (>10-fold) increases in in vitro affinity (which may be difficult to achieve) do not necessarily translate into increased clinical efficacy. This suggests that additional, alternative, improvements, such as decreased susceptibility to non-specific clearance, might be worth exploring.
References:[1]. Meno-tetang, and Lowe, 2005, Basic & Clin Pharmacol. Toxicol. 96 182-192
Poster: Applications- Biologicals/vaccines
Lene Alifrangis Setting a Safe Starting Dose for a First-in-Man trial of a Monoclonal Antibody Based on Population PK-PD Predictions
L. Alifrangis (1), P. André (2), R.V. Overgaard (1), C. Sola (2), A. Tisserant (2), N. Wagtmann (1), F. Romagne (2), S.H. Ingwersen (1).(1) Novo Nordisk A/S, Copenhagen, Denmark; (2) Innate Pharma, Marseille, France.
Objectives: The disastrous outcome of a recent phase 1 trial with a monoclonal antibody (mAb) (1) has highlighted the need for improved methods for determining safe starting doses for First-in-Man trials. Here, we used a population Pharmacokinetic (PK)-Pharmacodynamic (PD) model to predict a safe starting dose for the First-in-Man trial of a human IgG4 mAb specific for inhibitory Killer cell Ig-like Receptors (KIR) expressed on human NK cells. By blocking the interactions of KIR with its HLA-C ligands on target cells, the anti-KIR mAb (designated 1-7F9) facilitates NK-mediated killing of cancer cells.
Methods: An in vivo PK-PD model was established in a transgenic mouse strain expressing the human KIR. Using NONMEM V and population methods, a sequential approach was applied, first modelling PK followed by PD. Receptor occupancy was used as a measure of PD. The structural PD parameters of the mouse PK/PD model were combined with predicted PK parameters of IgG's in humans to devise a PK/PD model predicting the relationship between dose, resulting plasma concentration profile and KIR-receptor occupancy profile in humans.
Results: The PK of Anti-KIR(1-7F9) after i.v. administration in the transgenic as well as wild-type B6 mice was modelled by a 2-compartment model, combined with a 3rd saturable distribution compartment. The presence of the KIR antigen did not affect clearance or distribution of the Anti-KIR mAb. In vivo, the dissociation constant (Kd) changed from an initial 0.004 ug/ml to 0.1 ug/ml over time. Kd showed a good in vitro - in vivo correlation for the transgenic mice as well as a good interspecies correlation in vitro for KIR-transgenic mice vs humans. Based on the PK-PD model, a 0.3 ug/kg dose given to humans was predicted to result in 61% receptor saturation as the peak value with a rapid decay of saturation at later time points. Hence, this was suggested as the starting dose for the first phase 1 trial.
Conclusions: Using a combination of population PK-PD methods and in vitro-in vivo comparisons, a cautious starting dose based on all available scientific information has been suggested. The PK-PD model indicated that a surprisingly low dose of the Anti-KIR mAb may result in substantial receptor occupancy.
Reference: [1] Suntharalingam, G et al.(2006). N Engl J Med 355:1018-1028.
Poster: Applications- Biologicals/vaccines
Ekaterina Gibiansky Population Pharmacokinetics of Siplizumab (MEDI-507): Implications for Dosing
Gibiansky E. (1), J. Janik (2), D. Mahony (2), K. Kaucic (1), L. Hammershaimb (1), G. Robbie (1).(1) Medimmune Inc., Gaithersburg, USA; (2) Metabolism Branch, Center for Cancer Research, National Cancer Institute, NIH, Bethesda, USA
Objectives: Siplizumab (MEDI-507), a humanized IgG1k class monoclonal antibody which targets CD2 expressing T- and NK-cells, is being evaluated in an ongoing open label Phase I dose-escalation trial in patients with CD2-positive lymphoproliferative diseases. The aim of this report is to describe a population pharmacokinetic model of siplizumab and simulations of alternative doses/regimens performed for optimization of dosing.
Methods: 490 serum siplizumab samples were collected from 25 patients who received 0.4-4.8 mg/kg of siplizumab as 1- 3 consecutive daily doses every 14 days for 1-8 cycles. The population pharmacokinetic analysis was performed using NONMEM. Linear and nonlinear one and two compartment models were evaluated. Simulations were performed to identify doses and dosing regimens that would maintain drug concentrations above target levels necessary for saturation of CD2 receptors.
Results: Siplizumab pharmacokinetics was described by one-compartment model with 2 parallel mechanisms of clearance, linear and Michaelis-Menton (MM) elimination. The half-life of the linear portion of elimination (presumably, FcRn mediated) was 31 days, consistent with expectations, while MM elimination (target-mediated clearance) had a short half-life (38 hours at concentrations below Km of 15 ¼g/mL) suggesting suboptimal target saturation. Pharmacokinetics did not change over time, which was expected, since no immunogenicity was observed. Simulations of various dosing regimens suggested that doses > 4.8 mg/kg weekly are necessary to maintain drug concentrations above target level (2xKm, 30 µg/mL, the level at which half-life of target-mediated clearance increases 3 times) for 90% of patients.
Conclusions: Siplizumab concentrations follow one-compartment kinetics with linear FcRn-mediated and nonlinear target-mediated clearance. The short half-life of target-mediated clearance suggests that tissue target saturation is suboptimal at doses/regimens studied thus far. Dose escalation was accelerated based on simulations, with weekly doses > 4.8 mg/kg, to maintain concentrations above the target level and to possibly increase target saturation in the tissues.
Poster: Applications- Biologicals/vaccines
Ron Keizer Bioequivalence study of a C1-esterase-inhibitor product (Cetor®) with optimised sampling design
RJ Keizer(1), E van Twuijver(2), JJ Marcar(2), PFW Strengers(2), ADR Huitema(1)(1) Slotervaart Hospital, Amsterdam. Dept. of Pharmacy & Pharmacology; (2) Sanquin Plasma Products, Amsterdam
Objectives: Cetor is a concentrate of a highly purified C1-inhibitor concentrate prepared from human fresh frozen plasma, used in the treatment of hereditary and acquired angioedema (HAE/AAE). A change in the manufacturing process required a bioequivalence study to demonstrate that the change had not affected the pharmacokinetics of the product. The design of the clinical study was optimised by the use of trial simulations.
Methods: A population pharmacokinetic model built from retrospective data (9 patients) was used to optimise the sampling design and to investigate the bioequivalence of the two products. Data from total antigen and functional protein assays were simultaneously analysed in the pharmacokinetic model by defining functional protein as a fraction of total antigen (Ffunc). Trial-simulations were performed with a sample size of 10 patients and both a full (n=14) and a reduced (n=8) sampling design were tested. A randomised cross-over design, with an interval of 1 week between the administration of the two products, was used. The power of the study to detect a type I error of 0.05 was assessed using simulations of three scenarios, assuming the PK characteristics of both products are equal (a), Ffunc being 25% lower (b) or CL being 20% higher (c) than the reference compound, respectively. In the actual clinical trial, 13 HAE patients were included. Relative differences in PK properties induced by the adaptations in production process were estimated for CL, V, Ffunc. NONMEM V.1.1 was used in the PK analysis.
Results: For a full sampling schedule the power to detect differences in product characteristics were 92,2% and 47.8% for scenarios b and c, respectively. For the reduced sampling schedule these were 86.1% and 40.3. Only an increase in sample size led to an increase in power of the study. However, this could not be implemented due to the limited number of HAE patients. Data from the actual clinical trial were described with a linear one-compartment. No significant differences were found in PK characteristics and all confidence intervals of the relative differences in PK parameters were between 0.8 and 1.25.
Conclusions: Trial-simulation was successfully used to optimise the design of a bioequivalence study of two C1-esterase-inhibitor products. A reduced sampling design only had minimal influence on the power of the study. The results of the clinical trial showed that the adaptations in the production process did not lead to changes in PK parameters.
Poster: Applications- Biologicals/vaccines
Wojciech Krzyzanski Pharmacodynamic Modelling of Recombinant Human Erythropoietin Effect on Reticulocyte Production Rate and Age Distribution in Healthy Subjects
Wojciech Krzyzanski (1) and Juan Jose Perez Ruixo (2)(1) Department of Pharmaceutical Sciences, University at Buffalo, 547 Cooke Hall, Buffalo, NY 14260. (2) Clinical Pharmacology, Johnson & Johnson Pharmaceutical Research & Development, Division of Janssen Pharmaceutica, Turnhoutseweg 30, B-2340, Belgium.
Objectives: To evaluate the effect of rHu-EPO on reticulocyte production rate and age distribution in healthy subjects.
Methods: Four pharmacokinetic (PK) and pharmacodynamic (PD) non-linear mixed effect models were used to describe the time course of natural cells, and their age distribution, as a function of the hematopoietic growth factor concentrations. The models account for a) stimulation of production of progenitor cells in bone marrow (Model A); b) shortening of differentiation and maturation times of early progenitors in bone marrow (Model B); c) the combination of the two previous mechanism of action (Model C), and d) the stimulation of production of progenitor cells in bone marrow and the increase of maturation times of the circulating reticulocytes (Model D). Published data collected from 87 subjects who received a single subcutaneous dose of rHu-EPO (dose range: 20 - 160 kIU) in 3 phase I studies were used to develop and validate the PKPD model [1,2]. Model evaluation was examined using goodness of fit plots, non-parametric bootstrap and posterior predictive check.
Results: The previously published PK model describing rHu-EPO serum concentrations in healthy subjects was adopted and the individual PK parameter estimates fixed and used in all PD models [3]. If the value of the objective function (OFV) was used as a criterion of goodness of fit, the models can be ordered A > B > C >> D with Model D corresponding to the lowest OFV different from Model C by 150. Although the numbers of estimated parameters (fixed effect + interindividual variability + residual variability) were ordered inversely 9 < 11 = 11 < 13, the notable change in the OFV (Akaike Information Criterion) indicated Model D as the best. Model validation evidenced accurate and precise prediction of model parameters and the time course of percentage of reticulocytes. Model D attributed the observed increase in reticulocyte counts to the stimulation of production of progenitor cells in bone marrow as well as transient increase in the mean maturation time of the circulating reticulocytes.
Conclusions: A semi-physiological model quantifying the hematopoietic growth factor effects on production rate of precursor cells and its age distribution were developed. The most successful model indicated an indirect and delayed effect of rHu-EPO on the age distribution of the circulating reticulocytes that augments its stimulatory effect on the bone marrow precursor cells.
References: [1] Krzyzanski W, Perez-Ruixo JJ, An assessment of recombinant human erythropoietin effect on reticulocyte production rate and lifespan distribution in healthy subjects. Pharm. Res. (2007).[2] Cheung WK, Goon BL, Guilfoyle MC, et al., Pharmacokinetics and pharmacodynamics of recombinant human erythropoietin after single and multiple subcutaneous doses to healthy subjects. Clin. Pharmacol. Ther. 64: 412-23 (1998).[3] Olsson-Giskeskog P, Jacqmin P, Perez-Ruixo JJ, Population pharmacokinetics meta-analysis of recombinant human erythropoietin in healthy subjects. Clin. Pharmacokinet. 46: 159-173 (2007).Poster: Applications- Biologicals/vaccines
Armel Stockis Population pharmacokinetics of certolizumab pegol
Etienne Pigeolet(1), PhilippeJacqmin(2), Maria Laura Sargentini-Maier(1), G. Parker(1) and Armel Stockis(1).(1) UCB Pharma, Braine l'Alleud (Belgium) and Slough (UK); (2) Exprimo, Lummen (Belgium).
Objectives: Certolizumab pegol (CZP) is a pegylated Fab´ fragment of a humanized anti-TNF antibody. The aim of the analysis was to identify demographic and physiologic determinants of the disposition of CZP, in Crohn's Disease (CD) patients.
Methods: We evaluated 10275 plasma concentration-time records from 1580 subjects of whom 80% were patients with Crohn's disease, 15% rheumatoid arthritis and 5% healthy subjects. The structural model was a two compartment model with mixed order (between 0 and 1) absorption and first order elimination rates, and inter-occasion variability on clearance. Modeling was performed using NONMEM V.
Results: Typical clearance was 0.428 L/day and distribution volume 4.0 L in a 70 kg subject. Age, gender, creatinine clearance, white blood cells count and concomitant drug treatment such as steroids, amino-salicylic acid and analogs or anti-infectives did not influence the pharmacokinetics of CZP. Anti-CZP antibodies, repeated administration, weight, monocyte count, immunosuppressant intake and ethnicity had a statistically significant effect on the pharmacokinetic model. Simulations from the final model showed that, at steady state, only the presence of anti-CZP antibodies had a more than 30% effect on Cmax and AUCtau. However, these were detected in only 8% of the patients and did not appear to influence the efficacy endpoint (CDAI score). Doubling body weight, the second most influential covariate, was associated with a 25 % and a 20% decrease in Cmax and AUCtau , respectively.
Conclusions: Amongst the numerous covariates tested for their contribution to the pharmacokinetic variability of CZP, none of them seemed to have a clinically relevant impact.
Poster: Applications- CNS
Emma Bostrom Blood-brain barrier transport helps explain discrepancies in in vivo potency between oxycodone and morphine
Emma Boström, Margareta Hammarlund-Udenaes and Ulrika SH SimonssonDivision of Pharmacokinetics and Drug Therapy, Department of Pharmaceutical Biosciences, Uppsala University, Box 591, SE-751 24 Uppsala, Sweden
Objectives: The objective of this study was to evaluate the brain pharmacokinetic-pharmacodynamic (PKPD) relationships of unbound oxycodone and morphine in order to investigate the influence of blood-brain barrier (BBB) transport on differences in potency between these drugs.
Methods: Microdialysis was used to obtain unbound concentrations in brain and blood. The antinociceptive effect of each drug was assessed using the hot water tail-flick method. A population PK model of morphine was developed using NONMEM. BBB transport was described as the rate (CLin) and extent (Kp,uu) of equilibration, where CLin is the influx clearance across the BBB and Kp,uu is the ratio of the unbound concentration in brain to that in blood at steady state. A joint PKPD model of oxycodone and morphine based on unbound brain concentrations was developed and used as a statistical tool to evaluate differences in the PD parameters of the drugs.
Results: A six-fold difference in Kp,uu between oxycodone and morphine implies that, for the same unbound concentration in blood, the concentrations of unbound oxycodone in brain will be six times higher than those of morphine. A direct effect model was applied to the unbound brain PKPD data. A power model using Effect = Baseline + Slope C³ best described the data. Drug-specific Slope and ³ parameters were supported by the data, making the relative potency of the drugs concentration-dependent. Based on unbound brain concentrations, morphine was the more potent drug. However, based on unbound blood concentrations, oxycodone was more potent.
Conclusions: For centrally acting drugs such as opioids, PKPD relationships used for describing the interaction with the receptor are better obtained by correlating the effects to concentrations of unbound drug in the tissue of interest rather than to blood concentrations.
Poster: Applications- CNS
marc-antoine fabre Population pharmacokinetic analysis in children, adolescents and adults with schizophrenia or bipolar disorder
M.A. Fabre (2), E. Fuseau (2), A. Vermeulen (1), A. Thyssen (1).(1) Johnson & Johnson Pharmaceutical Research & Development; (2) EMF Consulting, BP 2, 13545 Aix en Provence, France
Introduction: Risperidone is an effective and well-tolerated atypical antipsychotic. Oral risperidone is licensed for the treatment of schizophrenia and acute mania associated with Bipolar I disorder in adults.
Objectives: To evaluate the pharmacokinetics (PK) of risperidone (RIS) and the active moiety (AM) in paediatric subjects with schizophrenia or bipolar disorder compared to adults using population PK analysis.Objectives: Text regarding objectives.
Methods: The population PK analysis of RIS and AM plasma concentrations was performed on rich and sparse data from three studies in paediatrics and six studies in adult subjects with either schizophrenia or bipolar disorder and included 304 children/adolescents (9-17 years) and 476 adults (18-71 years). A two-compartmental population pharmacokinetic model was developed using nonlinear mixed effects modelling (NONMEM). Age, sex, race, body weight, creatinine clearance (CRCL) and study were incorporated in the models to test as potential covariates. Since the dataset included children, adolescent and adult data, allometric scaling was used as a prior in the structural model.
Results/Conclusion: The AM and RIS concentrations were best described by a two-compartment disposition model with first order absorption and first order elimination. CRCL, age and weight increased the clearance of the active moiety, which is unlikely to be of clinical relevance as simulated plasma concentrations were similar in children, adolescents and adults.
After including allometric scaling as a prior for clearances and volumes, none of the demographic or biochemical characteristics tested were found to have an effect on any of the PK parameters of risperidone. This suggests that the PK of risperidone was similar between children, adolescents and adults after accounting for body weight differences.
Poster: Applications- CNS
Matt Hutmacher Modeling the Dropout from Longitudinal Adverse Event Data: Selecting Optimal Titration Regimens
Bojan Lalovic, Matthew M Hutmacher, Bill Frame, Kaori Ito, Raymond MillerPfizer
Background: Dizziness and somnolence are the predominant adverse events (AEs) reported in the treatment of Generalized Anxiety Disorder (GAD) with pregabalin, representing a major determinant of study withdrawal (dropout). Various ad-hoc administration regimens (titration schemes) have been implemented throughout the development of this compound to reduce the incidence and severity of AEs and dropout rates, based on clinical considerations.
Objectives: To execute a quantitative, model-based titration schedule optimization, we modeled patient withdrawal resulting from reported AEs across all dose escalation regimens. The model is to be used to provide prospective clinical trial simulations of titration schedules to select an optimal scheme to minimize AEs and resulting study withdrawal.
Methods: We developed a parametric, discrete-time model of dropout (study withdrawal) based on individual daily self-reported AE (dizziness and somnolence) information. Adverse events were reported as none, mild, moderate or severe (ordered categorical data) across six 4-6 week studies. The dropout model focused on AEs due to considerable withdrawal during the initial, titration phase week. Graphics of nonparametric estimates of survival (plots of hazard probabilities vs. time and Kaplan Meier plots of observed vs. model predicted survivorship probabilities) were used as a guide in the selection of appropriate probability distribution model. Exponential, Weibull, log-logistic, linear-exponential and Gompertz parametric hazard probability distributions were examined to describe the time course of dropout risk as a function of the maximum adverse event by day. The hazard varied as a linear or exponential function of time for the linear-exponential and Gompertz models, respectively. The maximum adverse event severity (MAXAE) across dizziness or somnolence resulted in a more parsimonious model compared to the use of either measure of AE alone.
Results: The best fit to the data was achieved using the Gompertz model for hazard probabilities. This two term log-linear hazard function was comprised of time-invariant (intercept) and a time-dependent parameter (slope) allowing for a monotonically decreasing, time-dependent changes in discrete conditional dropout (hazard). Additionally, initial hazard was larger with higher severity of dizziness or somnolence (MAXAE>=2), corresponding to an initially short dropout half-life of 3, 8 and 60 days at day 1, 10 and 30, respectively. On the other hand, dropout half-lives were significantly longer (lower conditional risk) for groups reporting lower MAXAE severities, in the order of months and only slowly increasing throughout the study. For the individuals reporting no adverse events at day 1, 10 and 30, half-lives were 211, 250 and 350 respectively. The model also supported disease duration as a predictor of dropout. Predictive check demonstrated good model performance across all dose arms of the trials.
Conclusions: The discrete time parametric dropout model adequately described the time course of withdrawal across all GAD studies. Prospective simulations highlight the impact of differing titration schemes on dropout probabilities.
References: [1] Feltner DE, Crockatt JG, Dubovsky SJ, et al. A randomized, double-blind, placebo-controlled, fixed-dose, multicenter study of pregabalin in patients with generalized anxiety disorder. J Clin Psychopharmacol 2003 Jun; 23 (3): 240-9[2] Pohl RB, Feltner DE, Fieve RR, et al. Efficacy of pregabalin in the treatment of generalized anxiety disorder: double-blind, placebo-controlled comparison of BID versus TID dosing. J Clin Psychopharmacol 2005; 25: 151-8[3] Rickels K, Pollack MH, Feltner DE, et al. Pregabalin for treatment of generalized anxiety disorder: a 4-week, multicenter, double-blind, placebo-controlled trial of pregabalin and alprazolam. Arch Gen Psychiatry 2005; 62: 1022-30[4] Montgomery SA, Tobias K, Zornberg GL, et al. Efficacy and safety of pregabalin in the treatment of generalized anxiety disorder: a 6-week, multicenter, randomized, double-blind,placebo-controlled comparison of pregabalin and venlafaxine. J Clin Psychiatry 2006; 67: 771-82[5] Pande AC, Crockatt JG, Feltner DE, et al. Pregabalin in generalized anxiety disorder: a placebo-controlled trial. Am J Psychiatry2003 Mar; 160 (3): 533-40
Poster: Applications- CNS
Laura Iavarone Population PK/PD of Alprazolam in the Attenuation of ACTH Activation Induced by Cognitive Performance in Metyrapone-treated Healthy Volunteers
L. Iavarone(1), R. Gomeni(1), E. Merlo-Pich(2)(1) CPK&MS, GSK, Verona, Italy; (2) CPDM, GSK, Verona, Italy;
Objectives: The control of ACTH release from the pituitary is under hormonal controls. Stimulating effects are produced by the peptide CRH released from the hypothalamus under stress. Inhibiting effects are produced by circulating cortisol, whose release from the adrenal glands is in ACTH-dependent. Metyrapone, an inhibitor of the cortisol synthesis, attenuate the cortisol negative feedback on ACTH release, resulting in an enhanced sensitivity to the stimulating effects of CRH. In this work we investigated the effects of Alprazolam on ACTH levels over time(4h) following dosing and in response to a cognitive performance test in volunteers receiving metyrapone 8 h before. The objective was to model the time-course of the inhibition produced by Alprazolam.
Methods: The relationship between ACTH and Alprazolam plasma levels was studied using an indirect PD response model. The rate of change of the ACTH response over time with no drug present can be described by dR/dt=kin-kout"R, kin is the zero-order constant for production and kout is the first-order rate constant for loss. The PK/PD model was developed in a stepwise fashion and described the change in ACTH over time and the effect of cognitive test at 3h post-dose. The final model represents inhibitory processes that operate according to the classical inhibitory function, I(t)=1-(Cp/(Cp+IC50) where Cp is the Alprazolam plasma levels and IC50 is the Alprazolam plasma levels producing 50% of maximum inhibition. The rate of change of R can be described by dR/dt= kin"I(t)-kout"R. A Mixed effect modelling (NONMEM) was used to estimate the model parameters. The circadian fluctuation of ACTH in the absence of Metyrapone was described by a cosine function over a 24h period
Results: Increase of ACTH over time was produced by metyrapone, further enhanced by the cognitive test. Alprazolam was able to decrease ACTH level in the experimental settings. The PK/PD relationships between ACTH exposure and Alprazolam plasma levels able to overall inhibit over the 24h the release of ACTH of the 50% (IC50) was estimated to be 6.22ng/mL.
Conclusions:
Metyrapone increased ACTH level over time and cognitive test produced a peak of ACTH. Both effects were attenuated by alprazolam in exposure-dependent manner, with an IC50 estimated a 6.22 ng/mL. This work indicates the possibility to investigate GABAergic compound using endocrinologic endpoints and PKPD modelling.
References: [1] M Hossain et al. Pharmaceutical Research, Vol. 14, No3, 309-315 (1997)
Poster: Applications- CNS
Maria Kjellsson Modelling Sleep Using Markov Mixed Effects Models
Maria C. Kjellsson (1), Daniele Ouellet (2), Raymond Miller (2), Mats O. Karlsson (1)(1) Uppsala University, Uppsala, Sweden (2) Pfizer Global Research & Developement, Ann Arbor, Michigan, USA
Objectives: To characterize the time course of sleep stages and the concentration-effect relationship of Drug X relative to placebo and to an active comparator using Markov models in patients with insomnia.
Methods: Sleep data were obtained in a 4-way crossover study of low and high doses of Drug X, a standard dose of an active control and placebo in 43 patients with primary insomnia. Sleep stages were measured for 8 hrs overnight at screening (baseline) and for 2 nights of dosing following each treatment. Markov models consisting of submodels for baseline, placebo, Drug X and positive control were developed for each transition. All models were merged into a joint sleep model for simulations.The submodels were developed sequentially, starting with baseline, followed by placebo, and in parallel for each drug. For each additional submodel, the parameters of the previous submodel were fixed. To speed up the model-building process, a number of pre-defined standard models for each submodel were tried. These standard models were chosen based on previous experience with similar data [1] and physiological plausibility.The model development was done using a population analysis approach in NONMEM V, assessing both between subject and between occasion variability (BSV, BOV).A posterior predictive check and simulations of 3 alternate study designs were performed.
Results: The baseline model was in most cases best described by a piece-wise linear function (PWL) of both bedtime (0 to 8 hrs) and stage time (duration within a stage). The PLW had two slopes with an internal breakpoint, which was either fixed at the median or estimated. BSV was characterized in most transitions and BOV in about half of the transitions.Placebo effects were found on 4 transitions, all for transitions between awake, stage 1 and REM. A majority of the drug effects of Drug X were best described as a linear model as a function of drug concentration in the effect compartment. The drug effects of the positive control were described with a linear model changing with the predicted concentrations in central compartment.The predictive performance of the joint model, assessed by simulations of the realized study design, was good, with 16 of 18 pre-defined efficacy parameters well described.Simulations with changing the time of dosing from ½ hour to 1 hour prior to bedtime resulted in a 40% reduction in latency to persistent sleep for the higher dose of Drug X.
Conclusions: The proposed reduced model building process resulted in a model that describes the sleep pattern at baseline, and following placebo, low and high dose of Drug X and positive control.
References: [1] Karlsson MO et al. A pharmacodynamic Markov mixed-effect model for the effect of temazepam on sleep. Clin Pharmacol Ther 2000;68(2):175-88Poster: Applications- CNS
Frank Larsen Non-Linear Mixed Effects PK/PD Modelling of Acute Autoinhibitory Feedback Effects of Escitalopram (ESC) on Extracellular Serotonin (5-HT) Levels in Rat Brain
F. Larsen (1), C. Bundgaard (2)(1) Clinical Pharmacology & Pharmacokinetics, (2) Discovery ADME, H. Lundbeck A/S, Copenhagen, Denmark.
Objectives: To characterise the PK/PD relationships including inter-individual variability (IIV) of ESC-induced 5HT response after acute administration to rats. A mechanistic turnover feedback model was assessed using a non-linear mixed effects (NLME) modelling approach.
Methods: Rats (n=17) were infused with 2.5, 5, or 10 mg/kg ESC or vehicle over 60 min. Extracellular 5-HT in hippocampus was monitored using microdialysis. Simultaneously, serial blood was sampled for ESC unbound plasma levels. The structural PK/PD model was a turnover model with drug-induced inhibition of loss of response (kout) and an inhibitory feedback moderator function resembling the acute mechanism of action of ESC[1]. Response acted linearly on the production (ktol) of the moderator, which acted inversely on the production (kin) of response (% of basal level 5-HT normalised to 100%). Additional model parameters were RBAS (baseline 5-HT), n (Hill-factor), Imax (maximal inhibition of loss of response) and IC50 (plasma levels of ESC resulting in 50% inhibition of loss of response). The PK model was fitted and the final parameter estimates were fixed in the combined PK/PD analysis. The PD model was described by differential equations (ADVAN9). IIV was modelled using exponential errors. The residual variability was proportional for the PK and additive for the PD. The final modelwas evaluated using 95% predictive performance plots and bootstrap analysis. NONMEM VI (Globomax) was used for the modelling.
Results: A two-compartment PK model (ADVAN3, TRANS4) adequately described the ESC plasma levels. IIV was identified for CL (17%), V1 (45%), Q (15%). Dose level on clearance was the only significant covariate. 5-HT levels were significantly increased following drug administration. However, at high doses, the mean response-time curves were almost identical. Therefore, a simple intrinsic turnover model was considered inappropriate. The final model included no further covariates and fitted all the response-data well and resulted in parameter estimates with acceptable precision. IIV was identified for RBAS (16%), Imax (14%) and IC50 (82%). The residual variability was 18% for the PK and 30 response units (%) for the PD.
Conclusion: The NLME turnover feedback model was successfully implemented. Variability estimates were low to moderate except for IIV of the in vivo potency (IC50). The model may serve as a tool to compare the PK/PD behaviour of different SSRIs.
Reference:[1] C. Bundgaard, F. Larsen, M. Jørgensen, J. Gabrielsson, 2006. Mechanistic model of acute autoinhibitory feedback action after administration of SSRIs in rats: application to escitalopram-induced effects on brain serotonin levels. Eur. J. Pharm. Sci. 29, 394-404.Poster: Applications- CNS
Mathilde Marchand Supporting the recommended paediatric dosing regimen for rufinamide using clinical trial simulation
Mathilde Marchand (1, 2), Eliane Fuseau (1), David Critchley (3)(1) EMF consulting, Aix en Provence, France; (2) EA3286 Laboratoire de Toxicocinétique et Pharmacocinétique, Marseille, France; (3) EISAI Global Clinical development, UK
Background: Rufinamide marketing application was reviewed recently by the CHMP (Committee for medicinal products for human use), for the treatment of seizures associated with Lennox-Gastaut syndrome (LGS) as adjunctive therapy in patients 4 years and older. Relationships between pharmacokinetics, efficacy and safety parameters have been established. However, only one study had been conducted in patients with Lennox-Gastaut syndrome. To document the exposure in a larger population, simulations of exposure and efficacy under the proposed dosing regimen were performed, so that the main sources of variability could be understood and dosing regimens found to give exposure similar to that shown to be safe and efficacious in larger populations of patients with others types of epilepsy.
Methods: Monte Carlo simulations were used to investigate the effect on rufinamide exposure and efficacy in the patient population of different proposed dosing regimens. Four dosing regimens, varying in term of initial dose, dose increment and maximum daily dose were defined based on patient body weight from 4 to 35 years of age. Since exposure variability appeared to increase in children with body weight less than 30 kg, additional simulations of exposure were carried out in that population.
Results/Conclusions: The simulations of the different dosing regimen resulted in rufinamide exposure similar to that observed during clinical trials. The results of the simulated total seizure frequency per day show a large variability, which was also observed in the LGS clinical study. The concentrations simulated in patients with a body weight less than 30 kg presented a large interindividual variability than other patients. Additional simulations demonstrated that this increased variability was due to greater valproate concentration in some of the children treated with rufinamide. Complementary investigations of maximum daily dose lead to propose maximum daily dose for patients less than 30 kg receiving antiepileptic drugs: rufinamide and valproate. This recommendation would ensure that children (less than 30 kg) treated with rufinamide and valproate concomitantly would not be overexposed to rufinamide.
Poster: Applications- CNS
Gianluca Nucci Population pharmacokinetic modelling of pimozide and its relation to CYP2D6 genotype.
Gianluca Nucci, Keith Muir and Roberto GomeniClinical Pharmacokinetics Modelling & Simulation, GlaxoSmithKline, Verona Italy
Background: Pimozide is a dopamine receptor antagonist used to treat obsessive compulsive disorder associated with Tourette's syndrome. In humans, pimozide is believed to be primarily metabolised by cytochrome P450 enzyme system CYP3A4 [1, 2] although CYP2D6 may also play a role .
Objectives: The objectives of the population pharmacokinetic analysis were to model the pharmacokinetics of pimozide in subjects of different CYP2D6 phenotypes in order to elucidate the role of CYP2D6 in pimozide disposition.
Methods: A two-compartment model with first-order absorption, and lag-time was implemented for Pimozide as basic model. It was fitted to the data of 32 healthy volunteers using NONMEM. Individual covariate used to refine the model estimates were age, weight, sex and CYP2D6 polymorphisms.
Results: Based on CYP2D6 genotype, the subjects had the following predicted metabolizer status: Extensive (EM) N=26, Intermediate (IM) N=4, and Poor (PM) N=2. CYP2D6 genotype was found to highly significantly improve model fitting to the data, with average population clearance of 14, 36 and 55 L/h in PM, IM and EM respectively. The variability of pimozide plasma concentrations was caused to a relevant degree by CYP2D6 (CVb% is decreased from 60% to 30% when CYP2D6 activity was taken into consideration). Individual body weight was found to be significantly linked to pimozide volume of distribution.
Conclusions: These population PK results suggest that pimozide CYP2D6 metaboliser status impacts significantly on the PK of pimozide, and CYP2D6 is at least as important as CYP3A4 in pimozide metabolism.
References: [1] Desta Z, Kerbusch T, Flockhart DA. Effect of clarithromycin on the pharmacokinetics and pharmacodynamics of pimozide in healthy poor and extensive metabolizers of cytochrome P450 2D6 (CYP2D6).Clin Pharmacol Ther. 65(1):10-20, 1999.[2] Desta Z, Kerbusch T, Soukhova N, Richard E, Ko JW, Flockhart DA. Identification and characterization of human cytochrome P450 isoforms interacting with pimozide. J Pharmacol Exp Ther. 285(2):428-37, 1998.[3] Richard E, Soukova N, & Kerbusch T: Metabolism of pimozide by CYP3A and CYP2D6 in human liver microsomes Clin Pharmacol Ther. 61(2):232, 1997
Poster: Applications- CNS
Gijs Santen Comparing treatment effect in depression trials: Mixed Model for Repeated Measures vs Linear Mixed Model
Gijs Santen(1), Meindert Danhof(1), Oscar Della Pasqua(1,2)1) Division of Pharmacology, LACDR, Leiden University, Leiden, the Netherlands. (2) Department of Clinical Pharmacokinetics/Modeling & Simulation, GlaxoSmithKline, Greenford, UK
Objectives: It is a well known fact that depression trials may fail in 50% of the cases even if effective doses of an antidepressant drug are administered. A high placebo effect, large variability between patients and inadequate endpoints are commonly given as reasons for this high failure rate. Therefore, investigations into alternative endpoints, novel study designs and statistical methods using historical data could lead to a reduction in the failure rate in clinical trials with anti-depressant drugs. In previous work we have explored the sensitivity of the Hamilton Depression Rating Scale (HAM-D) to treatment effect. The current work focuses on the impact of standard statistical analysis used to evaluate effect size in clinical trials, the Linear Mixed Model for Repeated Measures (MMRM) and compares it with a Linear Mixed Model (LMM).
Methods: Data from several double blind randomised placebo-controlled trials in Major Depression were extracted from GlaxoSmithKline's clinical database. Basically, the MMRM models repeated measures within a single individual as multivariate data with an unstructured covariance matrix that is assumed to identical across individuals. Treatment-time and baseline-time interactions are modelled as fixed effects. The LMM models HAMD response with the interactions treatment-time and baseline-time as fixed effects, but includes a subject-specific effect. MMRM analysis was performed in SAS using proc mixed, whilst LMM was also fitted in SAS using proc mixed and WinBUGS, with missing data replacement by the posterior predictive distribution for the specific individual.
Results: Re-analysis of study data revealed minor differences for the estimates obtained with either method. However, when diagnostic plots are evaluated, it is clear that the MMRM shows a bias relative to LMM. Model bias was especially evident across the range of responses in the observed vs predicted plots and over the time course of response of an individual patient. In a few occasions, MMRM resulted in incorrect estimates of significance level and consequently wrong conclusions about treatment effect.
Conclusions: The analysis of HAM-D data with an identical unstructured covariance matrix across individuals may not be appropriate. The use of the LMM with subject-specific random effects and missing data replacement based on posterior prediction distributions may offer a better alternative to current methodology for the assessment of treatment effect in depression.
Poster: Applications- CNS
Armel Stockis Dose-response population modeling of the new antiepileptic drug brivaracetam in add-on treatment of partial onset seizures.
Christian Laveille(1), Eric Snoeck(1), Brigitte Lacroix(2), Maria Laura Sargentini-Maier(2), Armel Stockis(2)(1) Exprimo, Lummen, Belgium; (2)UCB Pharma, Braine-l'Alleud, Belgium.
Objectives: To describe the individual change in seizure frequency from baseline after treatment with brivaracetam or placebo, to model the dose-response relationship and to assess the impact of potential covariates.
Methods: Efficacy data were used from two double-blind, placebo-controlled parallel-group phase-IIB trials in 363 patients aged 16-65 years. Brivaracetam dose levels were 0, 5, 20, 50 and 150 mg/day. Individual seizure frequency was modeled with NONMEM (version V) as a Poisson process, expressed as a function of baseline seizures, drug treatment, placebo effect and subject specific-random effects. The Mixture function was used for partitioning the population in two subgroups of patients exhibiting decreased or increased seizure frequency compared to baseline. In the first group, the drug effect was modeled using an Emax dose-response function on top of the placebo effect, whereas the change in seizure frequency in non-improving patients was dose-independent. In a second step, the dose was replaced by individual posterior estimates of AUCtau in the Emax model.
Results: 73% of the patients were classified as improving on brivaracetam, compared to 57% on placebo. In improving patients, the ED50 was predicted to be 21 mg/day and the maximum seizure reduction from baseline was 70%. The mean seizure reduction in patients improving on placebo was 41%. The mean increase in non-improving patients was 11%. Age, bodyweight, gender, carbamazepine, phenytoin, and the number of concomitant AEDs were found to neither affect the percentage of patients who are likely to improve nor the extent of change in seizure frequency, while country and concomitant levetiracetam influenced the effect size. The concentration-response model did not result in any improvement over the dose-response model.
Conclusions: Emax-modeling of Poisson-transformed seizure count data allowed to demonstrate a dose-response relationship for brivaracetam in 73% of the patients with refractory partial seizures. A dose of 20 mg daily is expected to decrease the seizure frequency by 50% of the maximum from baseline.
Poster: Applications- CNS
Nathalie Toublanc Retrospective population pharmacokinetic analysis of seletracetam in epileptic and healthy adults
Marc-Antoine Fabre, Eliane Fuseau(1), Maria-Laura Sargentini-Maier, Nathalie Toublanc(2)(1) EMF consulting, Aix-en-Provence, France ; (2) UCB S.A., Braine-lAlleud, Belgium
Objectives: Characterization of population pharmacokinetics of the novel SV2A ligand seletracetam (ucb 44212) in healthy and epileptic adult populations to identify covariates that may have a clinically significant influence on its pharmacokinetics (PK) and simulation of the PK profiles in patients with the formulation planned for phase IIb-III studies.
Methods: 233 subjects received single or multiple twice daily doses of instant release formulation of seletracetam in 4 clinical pharmacology studies and 3 phase IIa studies. Seletracetam concentration-time data were analyzed using NONMEM. Food, age, gender, race, body weight (BW), body surface area, treatment duration, dose and concomitant antiepileptic drug (AED), health status and creatinine clearance were tested as possible covariates. Absorption parameters for once a day (o.d) formulation were derived under fasted and fed conditions from a pilot study. They were added to the final model to simulate the profiles after o.d administration, using demographic covariates of the phase IIa patients with appropriate replications.
Results: 194 subjects were Caucasians, 109 females, 24 elderly (above 65) and 124 epileptic subjects, with median (range) for BW and age of 74.5 (43-133) kg, and 36 (17-86) years. Seletracetam plasma concentrations were adequately described by a one-compartment model. BW, sex, age and enzyme inducing AEDs were identified as covariates affecting CL/F, resulting in a reduction of inter-individual variability (IIV) from 22% to 15%. The influence of the statistically significant covariates ranged between 16% (inducer AEDs) and 70 % (BW). BW and gender were also identified as covariates on V/F resulting in a reduction of IIV from 14% to 7%. The population mean of V/F was 0.6 and 0.5 L/kg for males and females. Food also had a significant effect on the absorption rate.Based on simulation of the o.d formulation in a population consisting of replication of phase IIa patients, in both fasted and fed conditions, Cmaxss, Cavss, and Cminss increased in females, by 27% to 43%. The influence of food on the o.d formulation was less than 15% on all parameters. The effect of inducing AEDs was relevant only on Cminss (decrease ca. 27%)..
Conclusions: Although some covariates were statistically significant, based on simulations of concentration vs. time profiles in patients of an o.d. formulation, the only ones that may have a relevant effect on exposure are BW, sex, and enzyme inducing AEDs.
Poster: Applications- CNS
Maud Vernaz-Gris Pooled PK analysis of a new CNS drug, in healthy subjects.
M. Vernaz-Gris (1), E. Fuseau (1), L. Del Frari (2), V. Brunner (2), P. Hermann (2)(1) EMF Consulting, France ; (2) IRPF, France
Objectives: The objectives of the pooled data analysis were to describe the PK of a new drug developed in the CNS area, to evaluate variability in a population of healthy subjects and to provide a simulation model for optimising study design in patients.
Methods: The population PK analysis was performed on a pooled database, including 121 healthy subjects in 6 phase I studies. Subjects received single or repeated oral dose as a solution (0.075 to 2.5 mg) or as a capsule (1 to 2.25 mg). 2490 plasma concentrations were available. The structural PK model was chosen through individual PK modelling on a subset of subjects. Thereafter, population PK modelling on the complete database was used to estimate the parameters (FOCE interaction estimation method).
Results and Conclusion: A two compartment disposition model with first order elimination and a proportional residual error model were used. The drug often displayed either a large peak or two main peaks of plasma concentrations following single and repeated oral dose administration. The absorption was best described by two different processes separated by a lag time: one fraction of the dose is absorbed into the central compartment by a zero order process; the remaining fraction of the dose is absorbed through a first order process. The structural model includes various covariates, including the effect of the dose, food, formulation and time on apparent volumes of distribution (V2/F, V3/F), apparent inter-compartmental clearance (Q/F), duration of zero order absorption process (D2). The absorption was rapid although D2 was increased with administration of a capsule and with food intake. Inter individual variability (IIV) was evaluated for apparent clearance (CL/F), V2/F, V3/F, Q/F and D2. The magnitude of IIV was estimated between 30 and 47 CV% with exponential error models. Based on these results, simulations will be performed for optimising study design in patients.
Poster: Applications- CNS
Sandra Visser Modeling the time-course of the antipyretic effects and prostaglandin inhibition in relation the analgesic effects of naproxen: a compound selection strategy
Sandra Visser(1), Elke Krekels(1), Kristina Ängeby Möller(2), Marie Angesjö(1), Ingemo Sjögren(1) and Odd-Geir Berge(2)(1)DMPK and (2)Disease Biology, Local Discovery, AstraZeneca R&D Södertälje, SE-151 85 Södertälje, Sweden
Objectives: This study aimed to characterize the pharmacokinetic-pharmacodynamic (PKPD) relationship between plasma concentrations, inhibition of TXB2 and PGE2 synthesis, and the antipyretic and analgesic effects of naproxen in rats in order to investigate whether analgesic measurements could be replaced by endpoints that are more sensitive for compound selection.
Methods: Analgesic effects: naproxen (0, 7.5 and 30 µmol/kg p.o.) was given 1h after an intra-articular injection of carrageenan. Weight bearing was assessed at 5 time points per rat using the PawPrint method and plasma concentrations were measured at 25h.
Antipyretic effects: fever was induced by a 2 g/kg s.c. injection of brewer's yeast 4h before naproxen (0, 7.5, 30 and 90 µmol/kg p.o.). Body temperature was measured continuously using telemetry up to 25h after dosing and plasma concentrations were measured at 25h.
Prostaglandin synthesis: TXB2 and PGE2 synthesis over time was measured ex vivo in separate animals using ELISA. Satellite animals were used to obtain the complete PK profile. The naproxen concentrations were analyzed using LC-ESI-MS/MS. Nonmem V was used for all PKPD modelling procedures.
Results: A two-compartment PK model described the concentrations of naproxen best. Fever was identified as a covariate on CL. Individual parameter estimates were used to predict the pharmacokinetic profiles to the analgesic and antipyretic effects.
A sigmoidal relationship between the naproxen concentrations and the inhibition of TXB2 and PGE2 synthesis was observed. Population estimates for potency were 5±1 and 13±4 µM, respectively. Inter-individual variation was around 35% whereas the residual variation was 15%.
A linear model was used to describe the relationship between weight bearing on the affected limb and the concentrations. A tolerance pool model was used to describe the concentration dependent reduction of fever and the observed rebound.
Naproxen was equipotent with respect to the antipyretic and analgesic effects. However, the variability in measurements was much larger and dose separation less clear for the analgesic effects.
Conclusions: The time-courses of the naproxen concentration, inhibition of TXB2 and PGE2 synthesis, antipyretic and analgesic effects in rats were quantified and correlated. Endpoints such as antipyretic effects or the inhibition of TXB2/PGE2 could serve as alternatives to analgesic measurements for identifying differences between compounds in lead optimization.
Poster: Applications- CNS
Katarina Vucicevic Population Pharmacokinetic Modelling of Amitriptyline in Depression Patients
K. Vucicevic(1), B. Miljkovic(1), M. Pokrajac (1), I. Grabnar(2)(1) Department of Pharmacokinetics, Faculty of Pharmacy, University of Belgrade, Serbia; (2) Faculty of Pharmacy, University of Ljubljana, Slovenia
Objectives: This study aimed to characterize population pharmacokinetics of amitriptyline (AMT), in order to identify possible influential covariates and to assess the linearity of AMT kinetics.
Methods: In total 428 plasma samples were obtained from 28 patients diagnosed with major depression after single AMT dose of 75mg or 150mg, and in steady-state achieved with t.i.d. doses. Population PK analysis was performed using NONMEM and Visual-NM. Data were fitted with one and two compartment models for both AMT and its active metabolite nortriptyline (NT), while FOCE INTERACTION was used for estimation. The influences of patients weight, age, sex, co-therapy with fluvoxamine or lithium, daily dose of AMT on PK parameters were examined.
Results: The analysis showed that kinetics of AMT followed two-compartment model with first-order absorption and lag-time. No effect of dose on pharmacokinetic parameters was observed and the individual parameters estimated from steady state data were comparable to parameters estimated from single dose data. Mean (s.e.) parameter estimates were: CL=64.6 (1.5) L/h, Vc=896 (95) L ,Vp=641 (134) L, Q=114 (21) L/h, Ka=0.73 (0.13) h-1, Tlag=0.634 (0.032) h. Interindividual variability of AMT parameters was low and was best described by exponential error model, while proportional error model the most adequately characterized residual variability in AMT concentrations. NT kinetics was formation rate limited. Simultaneous fitting of parent drug and metabolite was a further step in modeling process.
Conclusions: The NONMEM analysis showed linear kinetics of AMT with no significant effect of tested covariates on AMT PK parameters.
Poster: Applications- Coagulation
Xavier Delavenne The use of an indirect response model to assess interaction between drugs: acenocoumarol and amoxicillin + clavulanic acid
X Delavenne (1), T Basset (2), P Girard (3) , H Decousus (1,4), P Mismetti (1,4), S Laporte (1)(1) Department of clinical pharmacology, EA3065, Saint-Etienne University Hospital, France;(2) Laboratory of pharmacology and toxicology, Saint-Etienne University Hospital, France; (3) EA3738, Faculté de Médecine de Lyon Sud, Oullins, France;(4) Department of Internal medicine and therapeutics, EA3065, Saint-Etienne University Hospital, France
Objectives: It has been reported in the literature an increase of anticoagulation level, assessed by prothrombin time, when acenocoumarol, an oral antivitamin K, is associated with amoxicillin plus clavulanic acid (antibiotic drug). The aim of present study was to investigate quantitatively the influence of amoxicillin plus clavulanic acid on pharmacokinetic (PK) and pharmacodynamic (PD) of acenocoumarol.
Methods: A single dose of 8 mg of acenocoumarol were orally administered to 8 healthy volunteers on day 1 and 8. From day 3 to 9, the volunteers received 1g of amoxicillin + 250 mg of clavulanic acid. Eleven blood samples were collected at day 1 and at day 8 for each volunteer; plasma concentrations of acenocoumarol and prothrombin time ratio (PTr) were measured. The PK-PD analysis was performed with a non-linear mixed effect model in NONMEM using FOCE INTERACTION method. In a first step, the structural PK model was identified by pooling the present dataset with other individual data from PK acénocoumarol trials. In a second step, an indirect PK-PD model was build conditional on the individual Bayesian PK parameter estimations from first step; PTr was fitted to an indirect action model with inhibition of the response synthesis as described by Dayneka [1]. The model is composed of two functions, one for the clotting factor (CF) and one for PTr: dCF/dt = Kin*(1-(Cp/(Cp+C50))) Kout*CF PTr = PTr0*CF/(¸+CF), CF are the clotting factors, ¸ is the hyperbolic parameter. Demographic covariates, as well as antibiotic treatment effect, were tested on PK and PD parameters. The model was validated by visual predictive check.
Results: Plasma concentrations of acenocoumarol were best fitted with a two compartment and first-order absorption model. Weight was included as covariate on V2 and antibiotic treatment effect on CL, thus reducing unexplained inter-individual variabilities from 10.0% to 0.3% for V2 and from 12.2% to 8.7% for CL. A significant 15% decrease in anecocoumarol clearance was observed when antibiotic was prescribed. PTr levels were well fitted by the indirect response model, but no significant covariate were identified, and especially the co-administration of antibiotic did not induce any significant changes on the prothrombin time.
Conclusions: To our best knowledge, this is the first application of an indirect response model to acenocoumarol data for assessing drug interaction. The only drug interaction with antibiotic that was found in this study was at the PK level, but not at PD level. Despite some case reports of clinical suspicions, amoxicillin plus clavulanic acid do not seem to affect the pharmacodynamic activity of acenocoumarol as assessed by prothrombin time.
Reference: [1] J Pharmacokinet Biopharm 1993; 21: 457-78.
Poster: Applications- Coagulation
Andreas Velsing Groth A Population PK/PD Model Assessing The Pharmacodynamics Of A Rapid-Acting Recombinant FVII Analogue, NN1731, In Healthy Male Subjects
Andreas Groth1, Judi Møss2, Tine Møller3, Steen Ingwersen11Biomodelling, 2Medical and Science, NovoSeven Key Projects, 3Biostatistics, Novo Nordisk, Bagsværd, Denmark
Objectives: NN1731 is a recombinant analogue of activated human coagulation factor FVII (FVIIa). Preclinical studies have indicated increased activity of NN1731 compared to native FVIIa in thrombin generation; a key step in the pathway to blood coagulation [1]. Thrombin generation from prothrombin may be measured by the appearance in the blood of prothrombin fragments 1 & 2 (F1+2). A first human dose (FHD) trial investigated the safety and the PK of NN1731, including measurements of F1+2 in plasma. It was attempted to investigate the feasibility of establishing a population PK/PD model of NN1731 pharmacologic effect to enable a future estimation of a relative potency of NN1731 compared to native recombinant FVIIa (rFVIIa, NovoSeven®).
Methods: The FHD trial was a dose escalation trial with 4 cohorts of patients (5, 10, 20 and 30 mcg NN1731/kg). The population PK/PD analysis was performed using NONMEM based on plasma measurements of PK and F1+2 for PD. A linear 2-compartment model of NN1731 PK was developed to describe adequately an initial rapid distribution phase followed by a less rapid elimination phase. An indirect response model was used to describe the formation of F1+2 induced by NN1731.
Results: The population PK/PD model was able to describe well the NN1731 PK and the appearance of F1+2.
Conclusions: Population PK/PD modelling may be used to provide an estimate of the relative potency of NN1731 compared to NovoSeven® once comparable measurements in subjects exposed to NovoSeven® become available.
Reference: [1] E. Persson et al, PNAS, 96;13583,2001.
Poster: Applications- CVS
Stefanie Albers Population Pharmacokinetics and Dose Simulation of Carvedilol in Pediatric Patients with Congestive Heart Failure
S. Albers (1), B. Meibohm (2), J. Barrett (3), TS. Mir (4), S. Laer (1)(1) Clinical Pharmacy and Pharmacotherapy, University of Duesseldorf, Germany; (2) Department of Pharmaceutical Sciences, University of Tennessee, Memphis, USA, (3) Childrens Hospital of Philadelphia, Philadelphia, USA, (4) Department of Pediatric Cardiology, University of Hamburg, Germany
Objectives: In-silico tools like population pharmacokinetic (Pop-PK) modeling and simulation play a pivotal role in developing safe and effective dosing strategies especially for pediatric patients. Therefore, the aim of our work was to investigate the ontogeny of carvedilol pharmacokinetics in pediatric patients by Pop-PK analysis. Dose simulations were performed to investigate the carvedilol dosing strategy for pediatric patients. The integration of these dosing regimens into clinical studies might increase the probability of success in future randomized controlled clinical studies aiming at efficacy.
Methods: Data were derived from a prospective, open-label study of carvedilol for the therapy of pediatric patients with congestive heart failure. Up to 13 plasma concentrations were determined from each patient during one dosing interval after 0.09 mg/kg QD and 0.35 mg/kg BID, respectively, using a validated HPLC-assay. Total plasma concentrations were analysed using a nonlinear mixed-effects modelling approach (NONMEM, Version V 1.1). The population model was further used for simulations of different daily doses and dosing intervals. Target parameter for the simulations was the area under the plasma concentration time curve (AUC) as a measure of drug exposure.
Results: 480 carvedilol plasma concentrations of 41 patients (0.1 - 19.3 years; median 3.5) were included in the analysis. Carvedilol pharmacokinetics were best described by a two-compartment model with first order absorption and absorption lag. Allometric weight normalization was used for clearances and volume of distribution parameters. Additionally, a significant influence of age on clearance (CL) and central volume of distribution (V2) was found:
CL [L/h] = 38.1*((weight [kg]/13)**0.75)-((age [years]/3.5)**2.7)
V2 [L] = 22.0*(weight [kg]/13)*(1-(0.13*age [years]/3.5))
Dose simulations revealed that for infants (28 days - 23 months), children (2 - 11 years) and adolescents (12 - 15 years) daily doses of 3, 2 and 1 mg/kg, administered in two or three doses were necessary to reach an exposure (AUC) comparable to adults receiving 0.7 mg/kg/day.
Conclusions: Younger patients have to be treated with higher doses of carvedilol to reach the same exposure as adults. These results have to be considered for further randomized controlled trials investigating the efficacy of carvedilol in order to avoid ineffective drug exposures. Poster: Applications- CVS
Kevin Krudys Can Bayes Prevent QTC-interval prolongation? A challenge beyond random effects.
Kevin M. Krudys, Oscar Della-PasquaClinical Pharmacology & Discovery Medicine, GlaxoSmithKline
Objectives: Early in the course of clinical development, it is important to be able to assess the propensity of non-antiarrhythmic drugs to prolong the QT/QTc interval. The current regulatory guidelines suggest using the largest time-matched mean difference between drug and placebo (baseline-adjusted) over the sampling interval[1], thereby neglecting any exposure-effect relationship and underlying nonlinearity in physiological fluctuation of QT interval. Thus far, most modelling attempts used to characterise drug-induced QT interval prolongation do not account for the limitations in clinical data or disregard model parameterisation in terms of drug-specific and system-specific properties. The aim of this study is to use a Bayesian approach to characterise the exposure-effect relationship of three compounds known to prolong the QT/QTc interval. We show the advantage of this approach to avoid false positives/negatives and optimise the design of QT/QTc specific studies.
Methods: The database consisted of 4 studies of moxifloxacin (400 mg), one study of grepafloxacin (600 mg) and one study of d,l-sotalol (160mg) with a total of 453 subjects. Population PK models were built for each compound to simulate individual concentration values at the times of QT measurements. The pharmacodynamic (PD) model describing QT interval comprises three components: an individual correction factor for RR interval (heart rate), an oscillatory component describing the circadian variation and a truncated Emax model to capture drug effect[2]. Model building was performed in WinBUGS version 1.4. The posterior distributions provided by WinBUGS allowed for the simulation of scenarios to investigate the impact of different study designs.
Results: The PD model provided estimates of a heart rate correction factor of mean (95% credible interval) 0.30(0.28 - 0.32) and a 24 hour circadian component with amplitude 3.34(2.32 - 4.39) ms and phase 3.23(3.01 - 3.62) hrs. The estimated QT prolongation due to moxifloxacin, grepafloxacin and d,l-sotalol at the 75th percentile of the observed concentration range was 7.19(6.06 - 8.35), 19.43(15.83 - 23.13) and 13.68(11.52 - 15.94) ms, respectively. Simulations suggest that the use of a PK/PD model can establish the QT liability of a compound using considerably fewer subject and/or samples than the standard approach.
Conclusions: A population model of QT interval was developed for three compounds known to cause QT prolongation. The explicit description of the exposure-effect relationship, incorporating various sources of variability offers advantages compared to the standard regulatory approach in that it yields estimates of the exposure required to reach clinically relevant increase in QT interval without the requirement for parameterisation of maximum effect (Emax). In addition, the posterior (predictive) distribution can be used directly to translate liability to QT interval prolongation across varying dose ranges.
References: [1] Guidance for Industry: E14 Clinical evaluation of QT/QTc interval prolongation and proarrhythmic potential for non-antiarrhythmic drugs. HYPERLINK "http://www.fda.gov/cder/guidance/6922fnl.pdf" http://www.fda.gov/cder/guidance/6922fnl.pdf [2] Bachman WJ, Gillespie WR. Truncated sigmoid Emax models: a reparameterization of the sigmoid Emax model for use with truncated PK/PD data. In American Society for Clinical Pharmacology and Therapeutics (ASCEPT) Meeting 1998.
Poster: Applications- CVS
Céline LAFFONT Population pharmacokinetic analysis of perindoprilat in hypertensive paediatric patients
Laffont CM (1), Mentré F (2) and Foos-Gilbert E (1)(1) Servier Research Group, Courbevoie, France (2) INSERM U738, Paris, France; University Paris 7, Paris, France
Introduction: The use of angiotensin-converting enzyme (ACE) inhibitors in the treatment of children suffering from hypertension or congestive heart failure has been widely recognized as useful. However, little is known about the pharmacokinetics of ACE inhibitors in these patients.
Objective: To develop a population pharmacokinetic model for perindoprilat in paediatric patients using both paediatric and adult data.
Material and Methods: An orodispersible formulation of perindopril (prodrug) was developed for paediatric use. This formulation was administered once per day to 60 hypertensive children and adolescents [median age (range) = 6 (2-15) years)] in a 4-month Phase II study. It was also given once per day to 24 healthy adult volunteers for one week in a Phase I study. Sparse (n=3-5) or extensive (n=12) blood sampling was performed at steady-state in the Phase II and Phase I studies, respectively. Perindoprilat plasma concentrations in paediatric patients (254 observations) were analysed together with adult data (286 observations) in order to better evaluate the impact of body weight on perindoprilat pharmacokinetic parameters. The population pharmacokinetic analysis was performed using NONMEM V.
Results: Perindoprilat plasma concentrations were analysed as the sum of unbound perindoprilat concentrations and concentrations of perindoprilat bound to circulating ACE. Unbound concentrations were best described by a one-compartment model with first-order formation and lag-time. Perindoprilat binding to ACE was modelled using a Michaelis-Menten relationship, assuming a single saturable binding site. Since it was not possible to estimate perindoprilat binding parameters from the present data, those were fixed to parameter estimates found in a previous population analysis in adults, for which adequate blood sampling was performed. Perindoprilat apparent clearance (CL/F) and volume of distribution (V/F) were related to body weight using general allometric equations (power model). Estimation of power model parameters revealed that CL/F and V/F increased proportionally with body weight. Creatinine clearance was estimated from serum creatinine using Schwartz formula. It was found to significantly influence CL/F and V/F, which is consistent with previous findings in adults [1] and the renal elimination of perindoprilat. Altogether, body weight and creatinine clearance explained a large part of the between-subject variability in CL/F and V/F for the paediatric population.
Conclusion: This population PK model provides a rationale for the adjustment of the initial therapeutic dose in paediatric patients (which may be further modified based on clinical evaluation). The initial therapeutic dose should be administered on a µg/kg basis, and should be reduced in case of moderate renal impairment in agreement with the recommendations made in adults.
References: [1] Parker E, Aarons L, Rowland M and Resplandy G (2005) The pharmacokinetics of perindoprilat in normal volunteers and patients: influence of age and disease state. Eur J Pharm Sci 26:104-113
Poster: Applications- CVS
Divya Menon Pharmacokinetics / Pharmacodynamics of Intravenous Bolus Nicardipine in Adults Undergoing Cardiovascular Surgery
Divya Menon (1), John T. Mondick (1), Bhuvana Jayaraman (1), Albert T. Cheung (2), Jeffrey S. Barrett (1)(1) Laboratory for Applied PK/PD, Division of Clinical Pharmacology and Therapeutics, The Childrens Hospital of Philadelphia, PA. (2) Department of Anesthesia, University of Pennsylvania, Philadelphia, PA.
Background: Nicardipine is a dihydropyridine calcium channel antagonist with selectivity for the coronary and peripheral vasculature, and minimal negative inotropic effects. The antihypertensive effect of nicardipine is characterized by a rapid onset of action and a short duration of effect, making it suitable for use in perioperative conditions. IV bolus administration of nicardipine in anesthetized patients has been shown to result in an acute decrease in arterial blood pressure without reflex tachycardia. Previous studies have concluded that nicardipine obeys linear pharmacokinetics.
Purpose: The objectives of this investigation were the following: (1) Construct a population pharmacokinetic / pharmacodynamic (PK/PD) model to describe the effect of an intravenous bolus nicardipine dose on the mean arterial blood pressure in anesthetized patients undergoing cardiovascular surgery. (2) Identify covariates that are predictors of variability. (3) Utilize the population PK/PD model to plan a multiple dose trial from which drug labeling can be defined.
Methods: A population PK/PD model was constructed from data collected from 40 anesthetized patients undergoing cardiovascular surgery, randomized to receive 0.25, 0.5, 1 or 2 mg of nicardipine. Nicardipine was administered directly into the right atrial port of the pulmonary artery catheter. PK samples were collected at predose, 2, 5, 7, 10, 20, 30, 90, 120, 180 and 240 min. Systolic and diastolic blood pressure was measured predose and every 15 seconds for up to 240s after nicardipine administration and every 60 s thereafter for up to 30 minutes. The data was analyzed using nonlinear mixed effects modeling with NONMEM. The predictive performance of the final model was evaluated by the predictive check method.
Results: The pharmacokinetics of nicardipine following IV bolus administration was well described by a two compartment model. The effects of nicardipine on mean arterial pressure were modeled using a slow receptor-binding model. The population mean (%CV) PK parameter estimates for CL, V1, Q and V2 were 23.6 (27.2) L/h, 0.22 (30.8) L, 10.3 (30.7) L/h and 3.08 (23.1) L. A graded increase in clearance was observed with increasing doses.
Conclusions: Contrary to the results reported following administration of an infusion, an effect mediated increase in the clearance of nicardipine was observed. We are exploring design elements in the planned multiple dose trial to aid us in discriminating equally plausible models. All possible models and approaches are discussed.
Reference: Cheung et al, Anesth Analg 1999, 89:1116-23Poster: Applications- Endocrine
Silke Dittberner Determination of the absolute bioavailability of BI 1356, a substance with non-linear pharmacokinetics, using a population pharmacokinetic modelling approach
Dittberner, S. (1), V. Duval (2), A. Staab (2), I. Troconiz (3), U. Graefe-Mody (2), U. Jaehde (1)(1) Dept. Clinical Pharmacy, Institute of Pharmacy, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany; (2) Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach a.d.R., Germany; (3) School of Pharmacy, University of Navarra, Pamplona, Spain
Objectives & Background: BI 1356 is a new dipeptidyl-peptidase IV (DPP IV) inhibitor under clinical development for type 2 diabetes mellitus. BI 1356 exhibits non-linear pharmacokinetics most likely caused by a concentration-dependent protein binding to its target protein DPP IV. The non-linearity makes it difficult to determine the absolute bioavailability using non-compartmental analysis. Therefore the objective of this analysis was to develop a model with a physiologically plausible explanation for the non-linear pharmacokinetics and to use this model to determine the absolute bioavailability. In addition, urine data were used to confirm the model structure. The urinary elimination is assumed to be dependent only on the unbound concentration and therefore provides additional information about the binding. The model is based on healthy volunteer pharmacokinetic data from an intravenous single rising dose study including a crossover arm with oral administration for the estimation of the absolute bioavailability.
Methods: Single dose plasma concentration-time profiles of 28 healthy volunteers (0.5, 2.5 or 10 mg BI 1356 intravenously, or 5 mg intravenously plus 10 mg BI 1356 orally) consisting of 862 BI 1356 plasma and 304 urine concentrations were included in the population PK analysis. The modelling was performed using the FOCE INTERACTION estimation method implemented in NONMEM V.
Results & Discussion: Plasma concentration-time profiles of BI 1356 in healthy volunteers were best described by a three-compartment model with concentration-dependent protein binding in the central and in one peripheral compartment. The absorption of BI 1356 was modelled using a first-order process. Inter-individual variability was estimated on the absorption parameters (Ka, Lag time and F). Using this model it was possible to determine the absolute bioavailability of BI 1356 despite its non-linear pharmacokinetics. The absolute bioavailability was estimated to be around 30 %. It is most likely that the concentration-dependent protein binding in the central and the peripheral compartment represents binding of BI 1356 to DPP IV available in both plasma and tissue (e.g. kidney, liver, lung). The urine data were best modelled assuming a linear urinary elimination of the unbound concentration of BI 1356. The parameter estimates were nearly identical to those obtained using only the plasma data. The urine data from the 0.5 mg intravenous dose group was slightly overpredicted, however no alternative model tested described the data better. These results confirm the model structure and suggest that the urine data could be explained without an additional non-linear process.
Conclusions: A population pharmacokinetic model including physiological knowledge showed that the non-linear plasma and urine pharmacokinetics of BI 1356, a new DPP IV inhibitor, could be explained by concentration dependent protein binding, most likely to its target protein. Using this model it was possible to determine the absolute bioavailability of BI 1356 despite its non-linear pharmacokinetics. The absolute bioavailability was estimated to be around 30 %.
Poster: Applications- Endocrine
Daniel Jonker Pharmacokinetic modelling of the once-daily human glucagon-like peptide-1 analogue, liraglutide, in healthy volunteers and comparison to exenatide
Daniël M. Jonker, Estelle Watson, Anders Dyhr Toft, Peter Kristensen, Lotte Bjerre Knudsen, Steen H. IngwersenNovo Nordisk A/S, Bagsværd, Denmark
Objectives: Glucagon-like peptide-1 (GLP-1) is an incretin hormone that has been shown to stimulate insulin secretion in a glucose-dependent manner, to inhibit glucagon secretion and to delay gastric emptying. These are promising properties for treatment of type 2 diabetes. Liraglutide is a long-acting human GLP-1 analogue that was designed to be eliminated considerably slower than the natural hormone. The aim of this modelling was to develop a pharmacokinetic model on available data on liraglutide and exenatide in man, and thereby compare their pharmacokinetic properties.
Methods: The pharmacokinetics of liraglutide were studied in 72 healthy male subjects using a dose escalation design. Each subject received a single sc injection of 1.25 up to 20 ug/kg liraglutide or vehicle and 22 venous blood samples were collected up to 48 hours after dosing. A population-based modelling approach was taken to describe the time course of liraglutide concentrations and its between-subject variability (BSV) using NONMEM V. During model building, goodness-of-fit was assessed by the objective function value and a range of graphical procedures. The estimates from the final model were used to generate a predicted PK profile for once daily sc dosing. Similarly, a PK profile for twice daily sc dosing of exenatide was generated from literature data.
Results: The absorption of liraglutide following subcutaneous (sc) administration was slow, with peak concentrations occurring at 9-12 hours post-dosing. Liraglutide absorption was adequately described by a dose proportional zero order process and a subsequent first order process. A one-compartment model with a central volume amounting to 0.086 L/kg (rSE 11%, BSV 33%) was estimated, showing that after absorption, liraglutide is mainly confined to the central circulation. Clearance was independent of dose and was estimated to be 0.0060 L/hr kg-1 (rSE 6.8%, BSV 13%). The mean elimination half-life after sc dosing was 13 hr (by noncompartmental analysis). The absolute bioavailability was found to be 51% (rSE 8%, BSV 30%).
Conclusions: The model and data demonstrate suitable pharmacokinetic properties for 24h-coverage with once daily sc administration of liraglutide. In addition, the PK profile of once daily liraglutide was found to have a lower peak to trough variation than that of twice-daily exenatide.
Poster: Applications- Endocrine
Thomas Klitgaard Population Pharmacokinetic Model for Human Growth Hormone in Adult Patients in Chronic Dialysis vs. Healthy Subjects
T Klitgaard , JN Nielsen, MS Fitsios, and M LangeNovo Nordisk A/S Denmark; Novo Nordisk Inc. USA
Objectives: To develop a population pharmacokinetics (popPK) model of human growth hormone (hGH) in patients with end-stage renal disease (ESRD) and healthy volunteers (HVs), after s.c. administration of rhGH, to explore and to support the design of future clinical trials.
Methods: The analysis was performed using NONMEM, based on samples from 11 patients (pts.) with ESRD and 10 healthy volunteers (HVs). Subjects received 7 (HVs) or 8 (ESRD pts.) daily doses of 50 µg/kg rhGh. Samples for PK were drawn every 30 minutes for 24 h, following dosing on days 0, 7 and 8 (ESRD pts. only). On day 9, ESRD pts. underwent 3-4 h of dialysis. Various compartment models with both first order and Michaelïs-Menten (MM) type of absorption and elimination were explored. Influence of covariates subject group (ESRD/HV) , gender, weight, and dialysis flow rate on key model parameters was examined. Finally, a visual posterior predictive check of the model was performed, to assess its predictive performance.
Results: The final model was one-compartmental with MM-absorption and MM-elimination. S.c. volume of distribution was 0.45 L/kg, baseline hGH was 0.52 µg/L, maximum absorption rate was 11.3 µg/kg/h, and the amount corresponding to half-maximum elimination rate was 18.9 µg/kg. Only the covariate ESRD/HV was significant (p8 for septic neonates.
Poster: Applications- Other topics
Anthe Zandvliet Dose individualization of indisulam to reduce the risk of severe myelosuppression
Anthe S. Zandvliet (1), Jan H.M. Schellens (2,3), William Copalu (4), Jantien Wanders (4), Jos H. Beijnen (1,2), Alwin D.R. Huitema (1)1 Department of Pharmacy & Pharmacology, The Netherlands Cancer Institute/Slotervaart Hospital, Amsterdam, The Netherlands, 2 Department of Biomedical Analysis, Section of Drug Toxicology, Utrecht University, Utrecht, The Netherlands, 3 Department of Clinical Pharmacology, Division of Medical Oncology, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands and 4 Eisai Ltd., London, UK
Objectives: Chemotherapy with indisulam often causes hematological toxicity. The objective of this study was to evaluate the influence of patient characteristics on the pharmacokinetics and pharmacodynamics of indisulam and to identify patients at risk for developing severe neutropenia or thrombocytopenia after treatment with indisulam.
Methods: Semi-physiological models of indisulam pharmacokinetics and hematological toxicity were used for the covariate analysis.[1,2] Concentrations of indisulam, cell counts of neutrophils and thrombocytes and patient characteristics (demographics, physical condition, prior medication, prior radiotherapy, concomitant medication, CYP2C genotype and biochemistry) of 13 clinical studies including 412 patients were used. Relationships between patient characteristics and pharmacokinetic and pharmacodynamic parameters were evaluated with the population approach using NONMEM. A simulation study was performed to determine the relative risk of dose limiting myelosuppression for the 2.5 and 97.5 percentiles of each patient characteristic. A relative risk of less than 0.9 or more than 1.1 was considered clinically relevant. A dosing algorithm was developed to assist in dose individualization.
Results: Body surface area (BSA), race and CYP2C genotype had a significant impact on indisulam elimination.(pu?uVu÷í÷â÷Ú÷í÷â÷Ú÷í÷ÚÏÚÁ¶«¶¶Úâ÷Ú÷í÷â÷Ú÷í÷â÷Ú÷í÷â÷Ú÷í÷â÷Ú÷í÷âhu®0J6]mH sH hghu®0JmH sH hghu®mH sH jhghu®Uhghu®6]mH sH hu®5\mH sH hu®mH sH hu®6]mH sH hu®0JmH sH jhu®U9VuWuuuÑuÒuäuåuvv^v_v`vyvzv²v³vwww w!wYwZw¤w¥w¦wºw»wówôw6x7x8xMxNxxxâxãxäxåxèxòxóxôxyy
yy*y+y÷ï÷å÷Ú÷ï÷åÍ÷Ú÷ï÷åÍ÷Ú÷ï÷åÍ÷Ú÷ï÷åÍ÷Ú÷ï÷åÍ÷Úï½ï÷ï÷ªh
MmHnHuh²:'h
M0JmHnHu$jh²:'h
M0JUmHnHuh
Mhu®5CJ$aJ$mH sH hu®0J6]mH sH hu®6]mH sH hu®0JmH sH hu®mH sH jhu®U3+y,y-yLyMyNyhyiyjylymynyoypyqyyyyy yÔyÕyÖyðyñyòyôyæÓź«º«º«ÓÓÅÅwÓdTº«ºC«º jê$h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jm$h
Mh
M>*B*UmHnHphÿuh
MmHnHu jð#h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHu$jh²:'h
M0JUmHnHu2js#h
Mh
M>*B*UmHnHphÿuóxoy÷yåzv{]|}â}I~Õ~½X/â°U
Í:Â
'ùóóóóóóóóóóóùóóóùóóóóùóóóùóó
Æd#
Æd#
ôyõyöy÷yøyùyzzzz'zÂzÃzÄzÞzßzàzâzãzäzåzæzçz{{{{{S{T{U{o{ðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2ja&h
Mh
M>*B*UmHnHphÿu jä%h
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jg%h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuo{p{q{s{t{u{v{w{x{{{{{¨{:|;|*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jÞ&h
MUmHnHu||}|~||é|ê|ë|}}} }
}}}
}}*}+},}-}9}¿}À}Á}Û}Ü}Ý}æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jÌ)h
MUmHnHu2jO)h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jÒ(h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jU(h
Mh
M>*B*UmHnHphÿuÝ}ß}à}á}â}ã}ä}~~~~~&~'~(~B~C~D~F~G~H~I~J~K~g~h~i~j~w~²~³~´~Î~õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jC+h
Mh
M>*B*UmHnHphÿu jÆ*h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jI*h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu Î~Ï~Ð~Ò~Ó~Ô~Õ~Ö~×~ó~ô~õ~ö~¶·¸º»¼½¾¿ÛÜïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jº,h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j=,h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jÀ+h
MUmHnHuÜÝÞì567QRSUVWXYZvwxyòóôæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C j®.h
MUmHnHu2j1.h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j´-h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j7-h
Mh
M>*B*UmHnHphÿu3456fgh§¨©ª·
(õæÓÊӼʼ¢ÓõæõæõæÓÊӼʼgÓTõæõ%h²:'h
M0J6]mHnHsH u2j%0h
Mh
M>*B*UmHnHphÿu j¨/h
MUmHnHuh²:'h
M0JmHnHsH u2j+/h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu()*,-./01MNOPa¿ÀÁÛÜÝßàáâãäïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j1h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j1h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j¢0h
MUmHnHu|}~ ¡½¾¿ÀÞßàúûüþæÓÀ°¥¥
¥Ó|Ón|nTÓ°¥¥C¥ j3h
MUmHnHu2j3h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j2h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j2h
Mh
M>*B*UmHnHphÿuþÿ !"0©ª«®¯°±²ÎÏÐÑß2
3
4
N
ðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2j5h
Mh
M>*B*UmHnHphÿu j4h
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j
4h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuN
O
P
R
S
T
U
V
W
s
t
u
v
÷
ø
ù
89ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j~6h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j6h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j5h
MUmHnHu9:;Kª«¬ÆÇÈÊËÌÍÎÏëìíî3457æÓÀ°¥¥
¥Ó|Ón|nTÓn¥¥C¥ jr8h
MUmHnHu2jõ7h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jx7h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jû6h
Mh
M>*B*UmHnHphÿu789:;h
Mh
M>*B*UmHnHphÿu jN>h
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jÑ=h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuïðñóôõö÷ø !; ¡¢£¤ÀÁìÜÑÂѱÂÑÂmìÜÑÂÑ\ÂÑ j*Dh
MUmHnHu2jCh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHu j0Ch
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u ÁÂÃÐ%&'ABCEFGHIJfghisÁÂÃÝÞßæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jFh
MUmHnHu2j¡Eh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j$Eh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j§Dh
Mh
M>*B*UmHnHphÿußáâãäåæ\]^xyz|}~ ¬>?@ZõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jGh
Mh
M>*B*UmHnHphÿu jGh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jFh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu Z[\^_`abc¬®ÈÉÊÌÍÎÏÐÑíîïðþïàÕ๫¹«Â«ÕàÕàÕ๫¹«fÂS%h²:'h
M0J6]mHnHsH u2jIh
Mh
M>*B*UmHnHphÿu jIh
MUmHnHu2jHh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jHh
MUmHnHuþ¢£¤¦§¨©ª«ÇÈÉÊÚNOPjklnopqrsïäÕäÄÕäÕ±¨±¨±mïäÕä\ÕäÕ±¨±¨ jKh
MUmHnHu%h²:'h
M0J6]mHnHsH u2jJh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHu jJh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u¦óôõ4567B¡¢£½¾¿æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jôLh
MUmHnHu2jwLh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu júKh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j}Kh
Mh
M>*B*UmHnHphÿu¿ÁÂÃÄÅÆâãäåù¢£¤¦§¨©ª«ÇÈÉÊØ
%õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jkNh
Mh
M>*B*UmHnHphÿu jîMh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jqMh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu %&')*+,-.JKLMfgh§¨©ª·ïàÕ๫¹«Â«ÕàÕàÕ๫¹«fÂS%h²:'h
M0J6]mHnHsH u2j_Ph
Mh
M>*B*UmHnHphÿu jâOh
MUmHnHu2jeOh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jèNh
MUmHnHu·"#$>?@BCDEFGcdefyçèé
()ïäÕäÄÕäÕ±¨±¨±mïäÕä\ÕäÕ±¨±¨ jÖQh
MUmHnHu%h²:'h
M0J6]mHnHsH u2jYQh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHu jÜPh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u)*+:µ¶·¹º»¼½¾ÚÛÜÝìz{|æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jÊSh
MUmHnHu2jMSh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jÐRh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jSRh
Mh
M>*B*UmHnHphÿu»¼½¾Îÿ !"#$@ABCPíîï õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jAUh
Mh
M>*B*UmHnHphÿu jÄTh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jGTh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu
./01C¦§¨ÂÃÄÆÇÈÉÊËçèïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j¸Vh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j;Vh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j¾Uh
MUmHnHuèéêùNOPjklnopqrs"#$æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C j¬Xh
MUmHnHu2j/Xh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j²Wh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j5Wh
Mh
M>*B*UmHnHphÿu$&'()*+GHIJXÓÔÕïðñóôõö÷ø)¦õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2j#Zh
Mh
M>*B*UmHnHphÿu j¦Yh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j)Yh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ¦§¨ª«¬®¯ËÌÍÎß5679:;Z[ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j[h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j[h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j Zh
MUmHnHuö£¤£¤N¥¦s¦f§K¨©¢©£ª««¬Ã¬-®ð®¯.°ô°ùùùùóùùóùùùùóùùùùùóùùùùùùùùù
Æd#
Æd#
[\]k
$ % & ( ) * + , - I J K L _ « ¬ Ç È É æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C j]h
MUmHnHu2j]h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j\h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j\h
Mh
M>*B*UmHnHphÿuÉ Ë Ì Í Î Ï Ð ì í î ï ¡¡¡,¡-¡.¡0¡1¡2¡3¡4¡5¡Q¡R¡S¡T¡e¡ß¡à¡á¡û¡õæÓÊӼʼ¢Ó¼õæõæõæÓÊӼʼwÓdTõæõh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j_h
Mh
M>*B*UmHnHphÿu j^h
MUmHnHu2j^h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHuû¡ü¡ý¡ÿ¡¢¢¢¢¢ ¢!¢"¢#¢9¢¾¢¿¢À¢Ú¢Û¢Ü¢Þ¢ß¢à¢á¢â¢ã¢ÿ¢£ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j|`h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jÿ_h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j_h
MUmHnHu££££££7£8£9£;££?£@£\£]£^£_£o£à£á£â£ü£ý£þ£¤æÓź«º«º«ÓÓÅÅwÓdTº«ºC«º jpbh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jóah
Mh
M>*B*UmHnHphÿuh
MmHnHu jvah
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHu$jh²:'h
M0JUmHnHu2jù`h
Mh
M>*B*UmHnHphÿu¤¤¤¤¤¤!¤"¤#¤$¤1¤¤¤¤¤¤¤ ¤¡¤¢¤£¤¤¤¥¤Á¤Â¤Ã¤Ä¤Ó¤+¥,¥-¥G¥ðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2jçch
Mh
M>*B*UmHnHphÿu jjch
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jíbh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuG¥H¥I¥K¥L¥M¥N¥O¥P¥l¥m¥n¥o¥{¥ë¥ì¥í¥¦¦ ¦¦
¦¦¦¦¦-¦.¦ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j^eh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jádh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jddh
MUmHnHu.¦/¦0¦O¦P¦Q¦k¦l¦m¦p¦q¦r¦s¦t¦u¦¦¦¦¦¥¦B§C§D§^§_§`§c§æÓź«º«º«ÓÓÅÅwÓdTº«ºC«º jRgh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jÕfh
Mh
M>*B*UmHnHphÿuh
MmHnHu jXfh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHu$jh²:'h
M0JUmHnHu2jÛeh
Mh
M>*B*UmHnHphÿuc§d§e§f§g§h§§
§§§§'¨(¨)¨C¨D¨E¨H¨I¨J¨K¨L¨M¨i¨j¨k¨l¨}¨ñ¨ò¨ó¨
©ðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2jÉhh
Mh
M>*B*UmHnHphÿu jLhh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jÏgh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHu
©©©©©©©©©3©4©5©6©J©~©©©©©©© ©¡©¢©£©¤©À©Á©ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j@jh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jÃih
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jFih
MUmHnHuÁ©Â©Ã©Ï©ªªªªªª ª¡ª¢ª£ª¤ª¥ªÁªÂªÃªÄªâªãªäªþªÿª««æÓÀ°¥¥
¥Ó|Ón|nTÓn¥¥C¥ j4lh
MUmHnHu2j·kh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j:kh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j½jh
Mh
M>*B*UmHnHphÿu««««««$«%«&«'«5«b«c«d«~«««««
««««¤«¥«¦«§«µ«ñ«ò«ó«
¬ðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2j«mh
Mh
M>*B*UmHnHphÿu j.mh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j±lh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHu
¬¬¬¬¬¬¬¬¬3¬4¬5¬6¬F¬¬ ¬¡¬»¬¼¬½¬À¬Á¬Â¬Ã¬Ä¬Å¬á¬â¬ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j"oh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j¥nh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j(nh
MUmHnHuâ¬ã¬ä¬ó¬fgh¨©ª«¹ ®
®®%®&®'®æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jqh
MUmHnHu2jph
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jph
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2joh
Mh
M>*B*UmHnHphÿu'®*®+®,®-®.®/®K®L®M®N®[®Ì®Í®Î®è®é®ê®í®î®ï®ð®ñ®ò®¯¯¯¯"¯c¯d¯e¯¯õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jrh
Mh
M>*B*UmHnHphÿu jrh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jqh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ¯¯¯¯
¯¯¯¯¯¥¯¦¯§¯¨¯½¯
°°°&°'°(°+°,°-°.°/°0°L°M°ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jth
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jsh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j
sh
MUmHnHuM°N°O°d°Ð°Ñ°Ò°ì°í°î°ñ°ò°ó°ô°õ°ö°±±±±&±±®±¯±É±Ê±Ë±æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jøuh
MUmHnHu2j{uh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jþth
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jth
Mh
M>*B*UmHnHphÿu˱αϱбѱұӱï±ð±ñ±ò±ÿ±²²²¹²º²»²¾²¿²À²Á²Â²Ã²ß²à²á²â²³³³ ³õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓ¼õæõ2jowh
Mh
M>*B*UmHnHphÿu jòvh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2juvh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHuô°Ñ±Á²(³Ç³¦´¤µ|¶Y·Õ·¸Y¹Hº»Ö»|¼à¼j½,¾¿¿^ÀÁkÁKÂ/Ã%Ä)ÅÆùùóùùùùùùùùùùùùóùùùùùùóùùùùù
Æd#
Æd#
³!³"³%³&³'³(³)³*³F³G³H³I³U³£³¤³¥³¿³À³Á³Ä³Å³Æ³Ç³È³É³å³æ³ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jæxh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jixh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jìwh
MUmHnHuæ³ç³è³ù³´´´´´ ´£´¤´¥´¦´§´¨´Ä´Å´Æ´Ç´Ö´µµµµµµæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jÚzh
MUmHnHu2j]zh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jàyh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jcyh
Mh
M>*B*UmHnHphÿuµ¡µ¢µ£µ¤µ¥µ¦µÂµÃµÄµÅµÒµX¶Y¶Z¶t¶u¶v¶y¶z¶{¶|¶}¶~¶¶¶¶¶¶5·6·7·Q·õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jQ|h
Mh
M>*B*UmHnHphÿu jÔ{h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jW{h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu Q·R·S·V·W·X·Y·Z·[·w·x·y·z··±·²·³·Í·Î·Ï·Ò·Ó·Ô·Õ·Ö·×·ó·ô·ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jÈ}h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jK}h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jÎ|h
MUmHnHuô·õ·ö·¸{¸|¸}¸¸¸¸¸¸¸¸ ¸¡¸½¸¾¸¿¸À¸Î¸5¹6¹7¹Q¹R¹S¹æÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C j¼h
MUmHnHu2j?h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jÂ~h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jE~h
Mh
M>*B*UmHnHphÿuS¹V¹W¹X¹Y¹Z¹[¹w¹x¹y¹z¹¹$º%º&º@ºAºBºEºFºGºHºIºJºfºgºhºiºvºäºåºæº»õæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2j3h
Mh
M>*B*UmHnHphÿu j¶h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j9h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu »»»»»»» »
»&»'»(»)»G»²»³»´»Î»Ï»Ð»Ó»Ô»Õ»Ö»×»Ø»ô»õ»ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jªh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j-h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j°h
MUmHnHuõ»ö»÷»¼X¼Y¼Z¼t¼u¼v¼y¼z¼{¼|¼}¼~¼¼¼¼¼¼¼½¼¾¼Ø¼Ù¼Ú¼Ý¼æÓÀ°¥¥
¥Ó|Ón|nTÓn¥¥C¥ jh
MUmHnHu2j!h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j¤h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j'h
Mh
M>*B*UmHnHphÿuݼ޼߼à¼á¼â¼þ¼ÿ¼½½½F½G½H½b½c½d½g½h½i½j½k½l½½½½½£½¾ ¾
¾$¾ðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2jh
Mh
M>*B*UmHnHphÿu j
h
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j
h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHu$¾%¾&¾)¾*¾+¾,¾-¾.¾J¾K¾L¾M¾X¾â¾ã¾ä¾þ¾ÿ¾¿¿¿¿¿¿¿$¿%¿ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jh
MUmHnHu%¿&¿'¿4¿_¿`¿a¿{¿|¿}¿¿¿¿¿¿
¿¡¿¢¿£¿¤¿°¿:À;À*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j h
Mh
M>*B*UmHnHphÿuXÀ[À\À]À^À_À`À|À}À~ÀÀÀçÀèÀéÀÁÁÁÁ Á
ÁÁÁ
Á)Á*Á+Á,ÁGÁHÁIÁcÁõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓ¼õæõ2j÷h
Mh
M>*B*UmHnHphÿu jzh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jýh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHucÁdÁeÁhÁiÁjÁkÁlÁmÁÁÁÁÁÁ'Â(Â)ÂCÂDÂEÂHÂIÂJÂKÂLÂMÂiÂjÂïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jnh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jñh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jth
MUmHnHujÂkÂlÂ}ÂÃÃ
Ã'Ã(Ã)Ã,Ã-Ã.Ã/Ã0Ã1ÃMÃNÃOÃPÃ`ÃÄÄÄÄÄÄæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jbh
MUmHnHu2jåh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jhh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jëh
Mh
M>*B*UmHnHphÿuÄ"Ä#Ä$Ä%Ä&Ä'ÄCÄDÄEÄFÄYÄÅÅÅ!Å"Å#Å&Å'Å(Å)Å*Å+ÅGÅHÅIÅJÅ^ÅøÅùÅúÅÆõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jÙh
Mh
M>*B*UmHnHphÿu j\h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jßh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ÆÆÆÆÆÆÆÆÆ:Æ;Æ*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jVh
MUmHnHuƯÆ?ÇéÇÈÉËÉ^ÊèÊRËüË»ÌRÍÎøΡÏUÐ
ÑÒ³ÒbÓLÔ²Ô
Õ7Ö×É×oØIÙùùùùùùùùóùùùùùùùùùùùùóùùùùùù
Æd#
Æd#
ÎÆÏÆÐÆÛÆÇÇÇ7Ç8Ç9ÇÇ?Ç@ÇAÇ]Ç^Ç_Ç`ÇrÇÅÇÆÇÇÇáÇâÇãÇæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jDh
MUmHnHu2jÇh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jJh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jÍh
Mh
M>*B*UmHnHphÿuãÇæÇçÇèÇéÇêÇëÇÈÈ È
ÈÈeÈfÈgÈÈÈÈÈÈÈÈÈȧȨȩȪȷÈùÈúÈûÈÉõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2j»h
Mh
M>*B*UmHnHphÿu j>h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jÁh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ÉÉÉÉÉÉÉÉÉ;ÉÉLɧɨɩÉÃÉÄÉÅÉÈÉÉÉÊÉËÉÌÉÍÉéÉêÉïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j2h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jµh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j8h
MUmHnHuêÉëÉìÉüÉ:Ê;Ê*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j,h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j¯h
Mh
M>*B*UmHnHphÿuâÊåÊæÊçÊèÊéÊêÊËËË Ë.Ë/Ë0ËJËKËLËOËPËQËRËSËTËpËqËrËsËËØËÙËÚËôËõæÓÊӼʼ¢Ó¼õæõæõæÓÊӼʼwÓdTõæõh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jh
Mh
M>*B*UmHnHphÿu j h
MUmHnHu2j£h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHuôËõËöËùËúËûËüËýËþËÌÌÌÌ+ÌÌÌ̴̵̸̳̹̺̻̼̽ÌÙÌÚÌïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jh
MUmHnHuÚÌÛÌÜÌîÌ.Í/Í0ÍJÍKÍLÍOÍPÍQÍRÍSÍTÍpÍqÍrÍsÍÍõÍöÍ÷ÍÎÎÎæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jh
MUmHnHu2jh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jh
Mh
M>*B*UmHnHphÿuÎÎÎÎÎÎÎ7Î8Î9Î:ÎRÎÔÎÕÎÖÎðÎñÎòÎõÎöÎ÷ÎøÎùÎúÎÏÏÏÏ*Ï}Ï~ÏÏÏõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jh
Mh
M>*B*UmHnHphÿu jh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j
h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ÏÏÏÏÏ Ï¡Ï¢Ï£Ï¿ÏÀÏÁÏÂÏÑÏ1Ð2Ð3ÐMÐNÐOÐRÐSÐTÐUÐVÐWÐsÐtÐïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jöh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jyh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jüh
MUmHnHutÐuÐvÐÐéÐêÐëÐÑÑÑ
ÑÑÑ
ÑÑÑ+Ñ,Ñ-Ñ.Ñ*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jð h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2js h
Mh
M>*B*UmHnHphÿuüÑÿÑÒÒÒÒÒ Ò!Ò"Ò#Ò0ÒÒÒÒ«Ò¬ÒÒ°Ò±Ò²Ò³Ò´ÒµÒÑÒÒÒÓÒÔÒãÒ>Ó?Ó@ÓZÓõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2ja£h
Mh
M>*B*UmHnHphÿu jä¢h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jg¢h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ZÓ[Ó\Ó_Ó`ÓaÓbÓcÓdÓÓÓÓÓÓ(Ô)Ô*ÔDÔEÔFÔIÔJÔKÔLÔMÔNÔjÔkÔïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jؤh
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j[¤h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jÞ£h
MUmHnHukÔlÔmÔÔÔÔªÔ«Ô¬Ô¯Ô°Ô±Ô²Ô³Ô´ÔÐÔÑÔÒÔÓÔáÔaÕbÕcÕ}Õ~ÕÕÕæÓź«º«º«ÓÓÅÅwÓdTº«ºC«º j̦h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jO¦h
Mh
M>*B*UmHnHphÿuh
MmHnHu jÒ¥h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHu$jh²:'h
M0JUmHnHu2jU¥h
Mh
M>*B*UmHnHphÿuÕÕÕ
ÕÕÕ£Õ¤Õ¥Õ¦Õ¸ÕÖÖÖ/Ö0Ö1Ö4Ö5Ö6Ö7Ö8Ö9ÖUÖVÖWÖXÖdÖáÖâÖãÖýÖðÝÔÝÆÔƬÝ~ð~mð~ðÝÔÝÆÔÆSÝ~ð~2jC¨h
Mh
M>*B*UmHnHphÿu jƧh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jI§h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuýÖþÖÿÖ××××××#×$×%×&×5ץצק×Á×Â×Ã×Æ×Ç×È×É×Ê×Ë×ç×è×ïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« jº©h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j=©h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jˬh
MUmHnHuè×é×ê×û×KØLØMØgØhØiØlØmØnØoØpØqØØØØØØ%Ù&Ù'ÙAÙBÙCÙæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C j®«h
MUmHnHu2j1«h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j´ªh
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j7ªh
Mh
M>*B*UmHnHphÿuCÙFÙGÙHÙIÙJÙKÙgÙhÙiÙjÙxÙÙÙٵٶٷٺٻټٽپٿÙÛÙÜÙÝÙÞÙðÙ-Ú.Ú/ÚIÚõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2j%h
Mh
M>*B*UmHnHphÿu j¨¬h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j+¬h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu IÙ½ÙQÚíÚÛÜÉÜaÝðÝÞmßàuáAâðâäëäÂå æãæçPè«èé·é¹éùùùùùùùùùùùùùùùùùóùùùóùèãed¿=æI
$a$ed¿=æIgd
M
Æd#
Æd#
IÚJÚKÚNÚOÚPÚQÚRÚSÚoÚpÚqÚrÚÚÉÚÊÚËÚåÚæÚçÚêÚëÚìÚíÚîÚïÚÛÛïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j®h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j®h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j¢h
MUmHnHuÛ
ÛÛÛ\Û]Û^ÛxÛyÛzÛ}Û~ÛÛÛÛÛÛÛ Û¡ÛÛâÛãÛäÛþÛÿÛÜæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C j°h
MUmHnHu2j°h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu j¯h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2j¯h
Mh
M>*B*UmHnHphÿuÜÜÜÜÜÜÜ$Ü%Ü&Ü'Ü8ܥܦܧÜÁÜÂÜÃÜÆÜÇÜÈÜÉÜÊÜËÜçÜèÜéÜêÜõÜ=Ý>Ý?ÝYÝõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2j²h
Mh
M>*B*UmHnHphÿu j±h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j
±h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu YÝZÝ[Ý^Ý_Ý`ÝaÝbÝcÝÝÝÝÝÝÌÝÍÝÎÝèÝéÝêÝíÝîÝïÝðÝñÝòÝÞÞïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j~³h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2j³h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu j²h
MUmHnHuÞÞÞ ÞoÞpÞqÞÞÞÞÞÞÞÞÞÞ±Þ²Þ³Þ´ÞÃÞIßJßKßeßfßgßæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jrµh
MUmHnHu2jõ´h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jx´h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jû³h
Mh
M>*B*UmHnHphÿugßjßkßlßmßnßoßßßßßßãßäßåßÿßààààààà à%à&à'à(à5àQáRáSámáõæÓÊӼʼ¢ÓõæõnæõæÓÊӼʼTÓõæõ2jé¶h
Mh
M>*B*UmHnHphÿu jl¶h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jïµh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu mánáoárásátáuáváwááááá£áâââ9â:â;â>â?â@âAâBâCâ_â`âïàÕ๫¹«Â~nÕàÕ]àÕ๫¹« j`¸h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jã·h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jf·h
MUmHnHu`âaâbârâÌâÍâÎâèâéâêâíâîâïâðâñâòâãããã!ãøãùãúãäääæÓÀ°¥¥
¥Ó|Ón|nTÓÀ°¥¥C jTºh
MUmHnHu2j×¹h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu jZ¹h
MUmHnHujh
MUmHnHuh
MmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u$jh²:'h
M0JUmHnHu2jݸh
Mh
M>*B*UmHnHphÿuäääääää:ä;ä*B*UmHnHphÿu jN»h
MUmHnHuh²:'h
M0JmHnHsH u%h²:'h
M0J6]mHnHsH u2jѺh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHujh
MUmHnHuh
MmHnHu ºå»å¼å¿åÀåÁåÂåÃåÄåàåáåâåãåüåýåþåææææææ æ!æ"æ>æ?æ@æAæVæïàÕ๫¹«Â«ÕàÕàÕ๫¹«fÂS%h²:'h
M0J6]mHnHsH u2j¿½h
Mh
M>*B*UmHnHphÿu jB½h
MUmHnHu2jżh
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHuh
MmHnHujh
MUmHnHu jH¼h
MUmHnHuVæ¿æÀæÁæÛæÜæÝæàæáæâæãæäæåæççççç^ç_ç`çzç{ç|ççççççç ç¡çïäÕäÄÕäÕ±¨±¨±mïäÕä\ÕäÕ±¨±¨ j6¿h
MUmHnHu%h²:'h
M0J6]mHnHsH u2j¹¾h
Mh
M>*B*UmHnHphÿuh²:'h
M0JmHnHuh
MmHnHu$jh²:'h
M0JUmHnHu j*mH sH hýdõh
M0JmH sH h
MH*hýdõh
MH*mH sH jÙÃh
MUhýdõh
M0JmH sH j=Ãh
MUh
Mhýdõh
M6]mH sH hýdõh
MmH sH =Ù
Ü
?BJMX[ª
üþTVáâ?@\^äñó !2!ÆÇÒÝåIU¸!Â!Ã!"w"x"óèóèóèóèóèóèóèóèÛèÛèÛèÛèÓèÓÆÓèÛèÛè¼´¦è¦èÛèÛèÛèÛèÛèÂjÅh
MUhýdõh
M5\mH sH hýdõh
M6]mH sH juÄh
MU
h
M0Jh
Mhýdõh
M0JmH sH jh
MUhýdõh
M0JmH sH hýdõh
MmH sH hýdõh
MH*mH sH 8ñú
þ!ÅÇÝI¸!V"w"y"å"#ð%È)É+6-W-Y-ì-ä.ðëëãÞÖÎëëëëëãÞÖÎëëëëãÞÖÎ$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI
&F¤d¤d[$\$ed¿=æIx"y""(####ð%ø%È)Ð)É+Õ+7-W-X-Y-e-..ã.ä.ï.ï/07£779W9X9Y9e9999«9ª@@A£A¤A¥A²AAB%C&C3CãDëDZEeE÷EÿEXKpK»RÇRñUüUZ(Z½Z[[üîãîãÖãÖãÖãÖãüÎüîãîãÖãÖãÖãüÆüîãîãÖãÖãÖãÖãü¾üîãîãÖã±ãÖãÖãÖãÖãÖãÖãü©jÇh
MUhýdõh
M0JmH sH jåÆh
MUjIÆh
MUjÅh
MUhýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M6]mH sH h
MBä.ï/3769W9Y9}99ª@A£A¥AîA&CZE÷EüGXKQ»RñUZðZúúúúòíåÝúúúúòíåÝúúúúúúúúú$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æIðZ[[¸[T\Ý^*aÄcÝek©k«kllMmr6tjuWvxvzvôvxOyYy{÷òêâÝÝÝÝÝ÷òêâÝÝÝÝÝ÷òêâÝÝÝed¿=æI$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æI[["[ÿ[S\T\_\|]]Ý^å^T_[_j`k`*a2abb@bCbÄcÐcÝeèeg-gAgCgFgGg_h`hÄiÅiAk©kªk«k¹kSlll§lMmUmwnynpp¬qq
rrrrrÑrÓrs¡s6tBtjuuuvuFvüîãîãÖãÊãÖãÊãÖãÖã½ã½ãÖãÖã½ãÊãÖãÖãÖãüµüîãîãÖãÖãÊãÊãÊã½ÊãÖãÊãÊãÖãÖ§ãhýdõh
M5\mH sH jÈh
MUhýdõh
M0JmH sH hýdõh
MH*mH sH hýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M6]mH sH h
MBFvxvyvzv
vwxx+xOyXyYycy{"{#{'|-|ø~¼ÉâãäñG{+,:ê,±È)YZ[Ò$bm¬ ¡¡¡¡Ã¡Ë¢Ì¢Ö¢\¤i¤°°÷µÿµøº»½£½üôüæÛæÛÎÛÎÛ¾Û¾±Û¾ÛÎÛÎÛ©üæÛüæÛÎÛÎÛ±Û±ÛæÛÎÛÎÛÎÛü¡üæÛæÛÎÛÎÛÎÛÎÛÎÛÎjñÉh
MUjUÉh
MUhýdõh
M0JmH sH hýdõh
M0J5\mH sH hýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M6]mH sH j¹Èh
MUh
M?{'|}'~ø~E|Ë*Uͼ°âäG,êÖ×±ÉzúúúúúúúòúúúòúúúçâÚòúúúúúú$a$ed¿=æIed¿=æI
$a$ed¿=æIgd
M$a$ed¿=æIed¿=æIz)ZbÏ ¡¡s¡Ì¢\¤N¨¬°÷µt·Ï¹øº½÷Â)Ã+Ã¥ÃJÄËúúúúúúòíåÝúúúúúúúúúúòíåÝú$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI£½ÈÂ)Ã*Ã+Ã9ÃËÃIÄJÄWÄÆÆÖÇØÇüÇþÇÈÈÈÈÒÈÔÈÉÉ=ÊBÊË'Ë(ËOËÅËÈËÎÎÎÎ ÎèΦϪÏTÒYÒZÒ
Ò¢Ò¤ÒÔÔ*Õ,ÕîÕðÕ&×6×8×>×@×׬׮×úÙüÙ`ÚbÚÜÜ ÜõñéñÛõÛõÎõñõñõñÂõñÂõñõ¶õ¶õÎõ©õÂõñõñõ©õ©õ©õ©õ©õñõñõñõÎõ©õ©õñõñõñõ©õhýdõh
M0JmH sH hýdõh
MH*mH sH hýdõh
MH*mH sH hýdõh
M0JmH sH hýdõh
M6]mH sH jÊh
MUh
Mhýdõh
MmH sH BË ÎTÒ&×ÜÝáæÁèáèãèOéÒë´î§ï]ò3ôSôUô·ô3õ°öH÷AúküúúúúúúúúïêâÚúúúúÒêâÚúúúú$a$ed¿=æI$a$ed¿=æI$a$ed¿=æIed¿=æI
$a$ed¿=æIgd
Med¿=æI Ü(ÜÝËÝááääääÐäÒäæææuèvèèè¼è½èáèâèãèñèÁéÑëÒëßë´î¿î§ï¯ï]òeò4ôSôTôUôcôåô2õ3õ>õ°öµöH÷Q÷AúJúküwü"þ-þ.þZþþ þ¢þïþóèóèÛè×è×è×èÛÉèÁèÁ´Áè¬×èèÛèÛèÛèÛè××èèÛèÛèÛèÛèÛèÛÉèóèÛèjÅËh
MUhýdõh
M6]mH sH j)Ëh
MUhýdõh
M0JmH sH jh
MUhýdõh
M5\mH sH h
Mhýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M0JmH sH ;kü"þí
ܾM
r)_s"B%õ%;'f'h'à';(+;-©0á1ê455¦56y9°;ôïçßÚÚÚÚÚÒïçßÚÚÚÚÚÒïçßÚÚ$a$ed¿=æIed¿=æI$a$ed¿=æI$a$ed¿=æIed¿=æI
$a$ed¿=æIgd
M¹F¤NQ_g ¡C D s"{"B%N%õ%&&&& &"&G&d&e&h&z&&&&&Ï&ò&ó&õ&'''('*'f'g'h''î':(;(F(+'+;-C-©0´0÷óåÚåÚÍÚÁÚÍÚÁÚÁÚÍÚÍÚͳڦÚÍÚ¦ÚÍÚó óó¦ÚÍÚ¦ÚÍÚóåÚåÚÍÚÍÚÍÚÍj5Îh
MU
h
M0J
h
M0Jhýdõh
M0JmH sH hýdõh
M5\mH sH hýdõh
MH*mH sH hýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M6]mH sH h
MjÍh
MU9´0á1ì1í1Ç4555,57666§6y99°;¸;f@~@°B»BåBlCmCnC{CÎCÛCÜCæC9DBDOOgQsQóQþQÿQ¡RºRÔRîRS&S'ScSdSSSS@TTöUV^V_V`VnVõèÚõÖÎÖÀõÀõèõèõèõèõèõÖ¸ÖÀõÀõèõèõèõèõèÚõ«õ«õÖ£Ö££Öõ«õ«õÖÀj Ðh
MU
h
M0Jjh
MUhýdõh
M0JmH sH jmÏh
MUhýdõh
M6]mH sH jÑÎh
MUh
Mhýdõh
M5\mH sH hýdõh
M0JmH sH hýdõh
MmH sH :°;f@°B8ClCnC¾CÜC9DOgQóQ*V^V`V§V´WÙZ(]^'^)_9``úúúïêâÚúúúúúÒêâÚúúúúúúú$a$ed¿=æI$a$ed¿=æI$a$ed¿=æIed¿=æI
$a$ed¿=æIgd
Med¿=æInVÙV³W´WÁW(]3]^%^``Ô`laÜbîbÇcÈcÉcÖcvdddd!f)fgg\ihij¡j,kFkHkll#l$l%l0lklxlyll
l½nÏnÐnín÷nQo[oÁoÔoq0q1qTr|rsDsÛstttõçõÚõÚõÚõÖõÖõÚõÎÖçõçõÚõÚõÚõÚõÚõÁÚõ¹Ö±ÖçõçõÚ£õÚ£õÚõÚõÚõÚ£õÚõÚõÚõÚhýdõh
M5\mH sH jAÑh
MUh
MmH sH hýdõh
M0JmH sH j¥Ðh
MUh
Mhýdõh
M0JmH sH hýdõh
M6]mH sH hýdõh
MmH sH @``Ô`í` a&a;aPa]ahalaÜbcÇcÉcUdd!fg\ijl#lððððððððððëëãÞÖÎëëëëëÃ
$a$ed¿=æIgd
M$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI
&F¤d¤d[$\$ed¿=æI#l%l`lyl½nínQoÁopqTrsÛstvYxZxxx÷ìáÙÙÇÇÇÙÙµµµÙÙ°¥ ed¿=æI
$a$ed¿=æIgd
Med¿=æI
&F
¤d¤d[$\$ed¿=æIgd
M
&F ¤d¤d[$\$ed¿=æIgd
Med¿=æIgd
M
$a$ed¿=æIgd
M
$a$ed¿=æIgd
Med¿=æIgd
Mttvvv[xxxxxØyu{v{{Ã|Ë|fqr
-
.
J
K
s
t
u
ú
º»Æ3;ùú
KW2XYZg5YZe>>>>$If»kd ù$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI.Û/Û8Û?ÛFÛMÛTÛD>>>>>$If»kdú$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æITÛUÛbÛiÛpÛwÛ~ÛD>>>>>$If»kdðú$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI~ÛÛÛ
ÛÛÛÛD>>>>>$If»kdØû$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIÛÛÛÛ¢Û©Û±ÛD>>>>>$If»kdÀü$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI±Û²Û»ÛÂÛÉÛÐÛØÛD>>>>>$If»kd¨ý$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIØÛÙÛæÛíÛôÛûÛÜD>>>>>$If»kdþ$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIÜÜÜ
ÜÜÜÜD>>>>>$If»kdxÿ$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIÜÜÜ Ü'Ü.Ü6ÜD>>>>>$If»kd`$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI6Ü7Ü@ÜGÜNÜUÜ]ÜD>>>>>$If»kdH$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI]Ü^ÜkÜrÜyÜÜÜD>>>>>$If»kd0$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIÜܱÜD>$If»kd$$Ifed¿=æIÖÖrñÿ2
!A
ììqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI±Ü²ÜºÜÃÜËÜÓÜÜÜ$Ifgkd$$Ifed¿=æIÖÖñÿ!*!Ö0ÿÿÿÿÿÿö6ÖÿÖÿÖÿÖÿ3Ö4ÖaöbÖit¿=æI±Ü²ÜÜÜÝÜÝÝ6Ý7ÝCÝ6àAàâÚâÛâÜâëâãã®ã¹ãâäêäõä÷äååVæXæTç\çÛéçéëë¦ìÉìÊìËìÙìíïíðíûíAîIîÊïÒïø±ø`úúúñíåíåíåØÍØÍíÅí·Í·ÍØÍØÍ«Í«Í«ÍØÍØÍØÍí£í·Í·ÍØÍÍØÍØÍíh
Mh
MmH sH hýdõh
M0JmH sH jh
MUhýdõh
MH*mH sH hýdõh
M6]mH sH jh
MUhýdõh
MmH sH hýdõh
M0JmH sH h
MCJaJh
Mhýdõh
MCJaJmH sH 3ÜÜÝÜæÜîÜöÜþÜÝD>>>>>$If»kd$$Ifed¿=æIÖÖrñÿ2
°!A
ìëqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIÝÝÝÝ%Ý-Ý6ÝD>>>>>$If»kd$$Ifed¿=æIÖÖrñÿ2
°!A
ìëqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æI6Ý7Ý6à¶âÚâÜâD??72ed¿=æI$a$ed¿=æIed¿=æI»kd$$Ifed¿=æIÖÖrñÿ2
°!A
ìëqÖ0ÿÿÿÿÿÿö6ÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿÖÿÿÿÿÿ3Ö4ÖaöbÖit¿=æIÜâbã®ãâäTçÛéë¥ìÉìËìAíðíÊïÆòyõø^úú
ú0û©ûþÈÕ÷ïêêêêêâÝ÷ïêêêêêÕÝ÷ïêêêêed¿=æIgd
Med¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æI$a$ed¿=æIúú
úúPû¨û©û³ûþ
þÈÏÕß¡þëìú)!åîÔà¬&¸&*¡*¢*º*ô-8.9.:.X.7/n0o0z011®4¶427>788N8(9Ñ:Ò:Ý:%û;Â@C³DF.F0F«FGoIJNPRRãRýS«UÜXD[C\O^v^x^û^úúúúòíåÝúúúúÕíåÝúúúúúòíåed¿=æIgd
M$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æIû^²`]aöcÄeÍfKg»iâiäij|lDnto9pqÉq±r&s(sOsQs¶s÷st½y÷òòòòòòêåÝ÷òòòòòòòòêåÝ÷òò$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIhaöcþcÄeÌeÍfÙfKgVgiâiãiäiüi?k{l|llDnLn®nÅnÇnànånþno*oÉqÑq±r½r)sOsPsQsbsÇsös÷stt£t$u'u½yÅyá{í{~~¬~ú~û~ü~«ÕÖäöþÊÑõèõèõèõèõäÜäÎõÎõèõèõÂõÂõÂõÂõèõèõäºäÎõÎõèõèõ®õèõèõèõä¦äÎõÎõèõèõèõèjª h
MUhýdõh
MH*mH sH j h
MUhýdõh
M>*mH sH hýdõh
M6]mH sH jrh
MUh
Mhýdõh
M0JmH sH hýdõh
MmH sH A½yá{~Ó~ú~ü~lÖöÊNýÿsDô¦qòÌÊúúúòíåÝúúúúÕíåÝúúúúòíåÝúed¿=æIgd
M$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æINZ×üýþÿùCDTÌÔôü¦®^aq}!$ó)`ÉÊÕ ,-DLg¹ºäå @ A õèõäÙÑäÃõÃõèõ¶õèõèõªõèõªõä¢äÃõÃõèõèõõäääääh
MCJaJh
M5\
h
M0Jhýdõh
MH*mH sH jâ!h
MUhýdõh
MH*mH sH hýdõh
M0JmH sH hýdõh
M6]mH sH jF!h
MUh
Mh
MmH sH h
Mhýdõh
M0JmH sH hýdõh
MmH sH 2DMgnsvz~
¦¬±¹º¼ÁÆËÑ×úúúôôôôôôôôïôôôôôôôôêôôôôôôFfM&Ffj#$Ifed¿=æI×Üäåêïôùÿ
# ( . 4 9 @ A Q U W \ ùùôùùùùùùùùïùùùùùùùùêåÜÜÜ $$Ifa$ed¿=æIFfö.Ff,Ff0)$IfA Q c d µ Ç È É ã ä ¡¡Y¡Z¡¡¡¡Ï¢Ñ¢Û¢Ý¢ê¢ë¢Ñ£Ü£Ý£0¤L¤þ¤¥G¥H¥I¥V¥Á¥¦¦¦¨¨ªª{««*6u®®Õ¯û¯ü¯ý¯°üõíüíâõíüõíüíüíüíÕâÉâÉâÉâÕ»â®â¨ü üââÕâÕâÉâÕâÕâÕâüüj}@h
MUhýdõh
M6]mH sH já?h
MU
h
M0Jhýdõh
M0JmH sH hýdõh
M5\mH sH hýdõh
MH*mH sH hýdõh
M0JmH sH hýdõh
MmH sH h
MCJaJh
M5\h
M6\ c d m v öNöööö¨kdí0$$Ifed¿=æIiÖÖ\zÿÌd·ââââÖ0ÿÿÿÿÿÿö=6ÖÿÿÿÿÖÿÿÿÿÖÿÿÿÿÖÿÿÿÿ3Ö4ÖiaöbÖit¿=æI $$Ifa$ ´ µ ¹ » À Ç WRLCCCC $$Ifa$$Ifed¿=æI¨kd³1$$Ifed¿=æIiÖÖ\zÿÌd·ââââÖ0ÿÿÿÿÿÿö=6ÖÿÿÿÿÖÿÿÿÿÖÿÿÿÿÖÿÿÿÿ3Ö4ÖiaöbÖit¿=æIÇ È É Ð × Ù Bþ}
|ÆÏEFGU õèõàÜÎõÎõèõèõèõèõÜÆÜÎõÎõèõèõèõèõèõèõܾÜÎõÎõèõèõèõèõ±õè£õÜÜÎõjÁDh
MUhýdõh
M5\mH sH hýdõh
M0JmH sH j%Dh
MUjCh
MUhýdõh
M6]mH sH h
MjíBh
MUhýdõh
M0JmH sH hýdõh
MmH sH =î?ñVõ«öÎöÐö2÷¥÷7þ}|"EGÝ^ ¤.
P×úü,JúúúòíåÝúúúúúòíåÝúúúúòíåÝú$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI ] ^ l ¤¯.
6
P[Øúûü
:IJV&.PrstHITèë;>(+,7äìZ]ÃË>"J"¸#Ã#$£$¤$¦$8%`%a%c%&B&j&k&l&òçÚçÚçÚçÚçÖÎÖòçòçÚçÚçÚçÖÆÖòçòçÚçºçºçºçÚçÚçºçÚçÚçÚççÚççÚç§ÖÖjFh
MU
h
M0Jhýdõh
M0JmH sH hýdõh
MH*mH sH jùEh
MUj]Eh
MUh
Mhýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M6]mH sH >¸&OrtÄI,äÃ>"¸#G&j&l&Ä&¯'/,â/y1!7D7F7úúúúòíåÝúúúúúúúòíåÝúúúúòí$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æIl&y&ú&®'¯'¼'/,7,â/ú/y11
12F2H2J2í233!3È3Þ3Ô45+6P6Q6S6à677D7E7F7R7Ø78i=====/>f>g>t>R@S@^@óDûDDEGE^EaE~EEyHòçòçÚçÚçÚçÚÌç¿çÚç¿çÚç¿ç¿ç¿çÚç¹³¯§¯òçò篯òçòçÚçÌÚçÚççççhýdõh
MH*mH sH jÍGh
MUj1Gh
MUh
M
h
M0J
h
M0Jhýdõh
M0JmH sH hýdõh
M5\mH sH hýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M6]mH sH :F778h===Ï=g>{H?JþMCPEPhPjPéPQÙTaUV?Y¿YZ_ZZ÷ïêâÝ÷ïêêêêêâÝ÷ïêêêêêÎÎÎ
&F¤d¤d[$\$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æI$a$ed¿=æIyHzH{HH@JMJþM N
NNN?N@NbNcNhNiN¨N©NÚNÛNàNáNOO:O;O*B*mH phÿsH hýdõh
M0JmH sH jh
MUhýdõh
M5\mH sH hýdõh
MmH sH hýdõh
M0JmH sH 1WW?YGY×ZöZ[X]c]d]^^@^A^o^p^q^8_Z_[_\_g_»_¿_À_Ì_ccg
gi)iØjãjäjúkSlTlUlblmnn#n$o(oppqóèÛè×ÛèÛÉ迳¿¦¿³è××èèÛèÛèÛèÛèÛÉè××èèÛè{èóèhýdõh
M0JmH sH j¡Ih
MUhýdõh
M6]mH sH jIh
MUhýdõh
M0JmH sH hýdõh
M>*mH sH jh
M>*Uhýdõh
M5\mH sH h
Mhýdõh
M0JmH sH hýdõh
MmH sH hýdõh
MH*mH sH 0Z×ZöZX]7_Z_\_°_À_c¦egiØj0lSlUl ln)ray3~K4ððëëãÞÖÎëëëëëëãÞÖÎëëëëëã$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI
&F¤d¤d[$\$ed¿=æIqq)r2r²rµrvvwvayvyyyyÀyÁyzzzz{¢{3~H~s~t~
~~~~~~12KWXþ±»*mH sH hýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M0JmH sH ;46Ý
÷ 89;Ñ+Jì{þ! # y ¬ y¤¨£ªúòêåååååÝúòêååååååÝúòêååå$a$ed¿=æIed¿=æI$a$ed¿=æI$a$ed¿=æIed¿=æI®÷ÿ ¬8Cñ9:;J
stmnJVìõ{«ÿ! " # 1 « ¬ · ¢¢w¤x¤y¤¤¨¨£ª¯ªS«^«¬X¬Y¬Z¬g¬ÕN®O®Z®°óèÛèÛèÛè×Ï×ÁèÁèÛèµèµèµèóèóèÛèÛèÛèÛè××ÁèÁèÛèóèÛèÛèÛèÛèÛè×¥×ÁèÁèÛèjLh
MUjuKh
MUhýdõh
MH*mH sH hýdõh
M6]mH sH jÙJh
MUh
Mhýdõh
M0JmH sH hýdõh
MmH sH hýdõh
M0JmH sH A£ªS«5¬X¬Z¬O®°C±1¶¸þ¸!¹#¹«¹Hº¼¬¼"ÁìÃ'ÄJÄLÄ·ÄìÅÝÆúúòíåÝúúúúúòíåÝúúúúúòíåÝú$a$ed¿=æI$a$ed¿=æIed¿=æI$a$ed¿=æIed¿=æI° °C±J±1¶I¶¸¸¸3¸n¸à¸!¹"¹#¹0¹Ï¹GºHºSº¼¼¬¼³¼"Á5ÁìÃ÷ÃøÃJÄKÄLÄ\ÄÅëÅìÅøÅ Æ
Æ4Æ7ÆÝÆæÆGÉJÉLÊOÊìÊïÊËËÎËÐËÙËÌÌØÌ(Î3Î4ÎϳϴÏóèóèóèóÚèÖèÖÎÖÀèÀèóèóèóèóèóÚè¸ÖÀèÀèóèóè¬èóè¬è è¬è¬èóèóèóÚèÖjåMh
MUhýdõh
MH*mH sH hýdõh
MH*mH sH jIMh
MUhýdõh
M6]mH sH jLh
MUh
Mhýdõh
M5\mH sH hýdõh
MmH sH hýdõh
M0JmH sH *mH sH hýdõh
M0JmH sH h
M6]hýdõh
MmH sH hýdõh
M6]mH sH h
Messes. PK Sim (, version 3.0 software (Bayer technology service) was used for the simulations.
Results/Conclusions: The preclinical PBPK in addition to physicochemical properties and in vitro data allowed a retrospective prediction of the pharmacokinetics in human. Factors influencing human pharmacokinetics were determined. Then PBPK simulations and allometric scaling were compared for the prediction of human clearance. The simulated profiles obtained after several doses in a population were compared to clinical data collected in healthy subjects. Since in vitro and in vivo preclinical data as well as physicochemical properties are available early in the drug development, the use of PBPK simulation should be applied at this early stage to help building a rationale for the selection of the first dose to administer to humans.
Poster: Methodology- PBPK
Gianluca Nucci A Bayesian approach for the integration of preclinical information into a PBPK model for predicting human pharmacokinetics
Bizzotto R, Nucci G, Poggesi I, Gomeni RClinical Pharmacokinetics/Modeling&Simulation, GlaxoSmithKline, Verona, Italy
Objectives: To evaluate the performance of a Bayesian approach for integrating preclinical in vitro and in vivo information into a PBPK model to predict human pharmacokinetics.
Methods: A basic, generic whole body physiologically based pharmacokinetic model was implemented in rats, dogs and humans [1]. The predictive performance of the basic model in these species was initially evaluated using a dataset of 23, 21 and six compounds for rats, dogs and humans, respectively. Since in vivo pharmacokinetic data in animals are always generated before the first study in humans, our aim was to take advantage of this information to improve the predictive performance of the basic PBPK. Our proposal is to use a Bayesian approach to estimate few critical parameters of the PBPK model, to provide a better adherence to the observed pharmacokinetics in rats and dogs, and to eventually use these parameters for predicting the PK in humans. For the application of this proposal the Bayesian parameters identification as implemented in SAAM II was used.
Results: The basic model, applied to the preclinical species showed reliability similar or better than that reported in the literature. Average fold-errors for the main pharmacokinetic parameters were lower than 2, with 91% of the compound parameters predicted within a 3-fold error. The proposed approach increased the reliability of the prediction of the pharmacokinetic data in humans. For the six compounds, for which human oral pharmacokinetic data were available, the average fold-errors for the systemic exposure markedly decreased from 3.6 (basic) to 1.9 (new approach).
Conclusions: During the drug development process, incremental in silico, in vitro and in vivo information is gained before a candidate drug is effectively given to humans. The basic PBPK model proposed by Poulin [1] may be based on too simple assumptions and limited input data for providing predictions in human with the desired level of reliability. In this work the PBPK model prediction were refined applying a Bayesian approach to fine tune few critical PBPK model parameters based on in vivo animal data. If these promising results will be confirmed on a more extensive dataset of compounds, this approach will strongly improve the design and safety of first time in human studies.
References: [1] Poulin P, Theil FP. J Pharm Sci 2002, 91:1358-1370.
Software demonstration
Roger Jelliffe The USC*PACK collection of BigWinPops software for nonparametric adaptive grid (NPAG) population PK/PD modeling, and the MM-USCPACK clinical software
R Jelliffe, A Schumitzky, D Bayard, R Leary, M Van Guilder, A Gandhi, M Neely, and A Bustad.Laboratory of Applied Pharmacokinetics, USC Keck School of Medicine, Los Angeles CA, USA.
The BigWinPops maximum likelihood nonparametric population adaptive grid (NPAG) modeling software runs in XP. The user defines a PK/PD model using the BOXES program to make the structural model. This is compiled and linked transparently. The subject data files are entered, and instructions. Routines for checking data files and for viewing results are provided. Likelihoods are exact. Behavior is statistically consistent, so studying more subjects gives estimates progressively closer to the true values. Stochastic convergence is as good as theory predicts. Parameter estimates are precise [1]. The software is available by license from the University for a nominal donation.
The MM-USCPACK clinical software [2] uses NPAG population models, currently for a 3 compartment linear system, and computes the dosage regimen to hit desired targets with minimum expected weighted squared error, thus providing, for the first time, maximal precision in dosage regimen design, a feature not seen with other currently known clinical software. Models for planning, monitoring, and adjusting therapy with aminoglycosides, vancomycin (including continuous IV vancomycin), digoxin, carbamazepine, and valproate are available.
The interactive multiple model (IMM) Bayesian fitting option [3] now allows parameter values to change if needed during the period of data analysis, and provides more precise tracking of the changing behavior of drugs in clinically unstable patients
In all the software, creatinine clearance is estimated based on one or two either stable or unstable serum creatinines, age, gender, height, and weight [4].
References: [1] Bustad A, Terziivanov D, Leary R, Port R, Schumitzky A, and Jelliffe R: Parametric and Nonparametric Population Methods: Their Comparative Performance in Analysing a Clinical Data Set and Two Monte Carlo Simulation Studies. Clin. Pharmacokinet., 45: 365-383,2006.[2] Jelliffe R, Schumitzky A, Bayard D, Milman M, Van Guilder M, Wang X, Jiang F, Barbaut X, and Maire P: Model-Based, Goal-Oriented, Individualized Drug Therapy: Linkage of Population Modeling, New "Multiple Model" Dosage Design, Bayesian Feedback, and Individualized Target Goals. Clin. Pharmacokinet. 34: 57-77, 1998.[3]. Bayard D, and Jelliffe R: A Bayesian Approach to Tracking Patients having Changing Pharmacokinetic Parameters. J. Pharmacokin. Pharmacodyn. 31 (1): 75-107, 2004.[4]. Jelliffe R: Estimation of Creatinine Clearance in Patients with Unstable Renal Function, without a Urine Specimen. Am. J. Nephrology, 22: 3200-324, 2002.
PAGE
PAGE 91