Friday, November 29, 2002
Location -
The lectures will be held in the UMR 5558, 5rd Floor, Room of the UFR,
Buidling Gregor Mendel - 741, University Claude Bernard - Lyon I, 43, boulevard
du 11 novembre 1918, 69622 Villeurbanne Cidex, France
This course is intended for biomedical scientists, physicians and pharmacists with an interest in optimal methods of population pharmacokinetic / pharmacodynamic modeling, and also for those interested in therapeutic drug monitoring and optimally precise individualization of drug therapy for patient care.
Prior experience in compartmental modeling or clinical pharmacokinetics will be an advantage. Participants will be introduced to the USC*PACK software, which can be used both for therapeutic drug monitoring and optimal individualization of drug dosage regimens, as well as for parametric and nonparametric population PK/PD and physiological modeling.
The course will examine the concepts and review and compare the performance of current and new methods of parametric and nonparametric population PK/PD modeling, with special emphasis on the FOCE parametric Iterative 2 Stage Bayesian (IT2B) and the nonparametric adaptive grid (NPAG) programs.
This course will also introduce the new MM-USC*PACK software for "Multiple Model" design of dosage regimens that achieve target therapeutic goals with maximal precision. This method is based on nonparametric population models. It also obtains a patient's Bayesian posterior nonparametric individual model, and, if needed, to detect and quantify the interoccasional variability in each patient's individual model. This permits detection of unsuspected changes in parameter values such as take place with the volume of distribution (and other parameters), in aminoglycoside antibiotics, for example, with changes in the patient's status. This sequential Bayesian "Interacting Multiple Model" approach to interoccasional intra-individual variability comes from the aerospace community, where it is used to track evasive targets. It is new, to our knowledge, in the pharmacokinetic community. It is designed to track the behavior of drugs, especially in unstable patients, to detect unsuspected changes in a patient's parameter values during the period of the data analysis, and to permit achievement of target therapeutic goals with maximum precision. Such changes take place in patients not only when they are acutely ill and highly unstable, but also in children as they grow and mature, when therapy must be guided during all this time. In addition, changes also take place with varying compliance, and a changing apparent volume of distribution may well be a clue to good or poor compliance in a patient.
In addition, new work on stochastic analysis of patient data in the Surgical Intensive Care Unit will be presented. New methods of analyzing such data, predicting patient outcome, and suggesting optimal therapy will be presented.
For more information and Registration, please contact
Pascal Maire, Pharm.D., Ph.D.
UMR CNRS 5558 / ADCAPT / Service Pharmaceutique
Htpital Antoine Charial, Hospices Civils de Lyon,
40, Avenue de la Table de Pierre,
69340 Francheville, France
Phone (33) 04 72 32 34 87
Fax (33) 04 72 32 39 08
Email adcapt@cismsun.univ-lyon1.fr
Morning session Basic Pharmacokinetics, Introduction to Population Modeling
08:30 AM - Registration 09:00 AM - Welcome - Drs. Flandrois, Carret, Maire, Jelliffe. 09:15 AM - Introduction to basic concepts in pharmacokinetics, including Review of basic pharmacokinetic behavior - Dr. Maire Drug elimination and renal function - Dr. Jelliffe 09:30 AM - Evaluating renal function Dr. Jelliffe 09:45 AM - Bayes' theorem and the MAP Bayesian scenario of planning, monitoring, and adjusting drug dosage for patients - Dr. Maire 10:00 AM - Introduction to population modeling - Dr. Jelliffe Types of PK models Linear regression, NLLS, MAP Bayesian 10:30 AM BREAK 10:45 AM - Demo - the IT2B program. Modeling Amikacin - Dr.Jelliffe A typical patient data file Running the program. Getting gamma, ranges, evaluating the results 11:15 AM - Demo - NPEM: modeling Amikacin further - Dr. Jelliffe Evaluating the results - The log-likelihood function The 2 and 3-D plots of the marginal and joint PDF's 11:45 AM - Comparing parametric and nonparametric approaches - IT2B, NPEM, and NPAG. Their statistical consistency and efficiency. Their results - Dr. Jelliffe 12:15 PM - LUNCH
Afternoon Session I Multiple Model Dosage Design
02:00 PM - Multiple Model (MM) dosage design for maximum precision regimens - Dr. Jelliffe 02:30 PM - Getting MM Bayesian posterior individual parameter distributions. The Interacting MM (IMM) approach - Dr. Jelliffe 03:00 PM - Introduction to the new MM - USC*PACK Clinical Programs to achieve target goals with maximum precision - Dr. Jelliffe Demo - 1 compartment model: Planning the initial regimen 03:30 PM - BREAK
Afternoon Session II Advanced Pop Modeling - Large and Nonlinear Models
03:45 PM - Making large and nonlinear population models - Dr. Maire
Demo - Using BOXES making a Michaelis-Menten model of Piperacillin - Dr. Jelliffe
04:15 AM - Demo - setting up Big IT2B Modeling Piperacillin - Dr. Jelliffe
A typical subject data file
Setting up the model, the data, the instructions, sending it, analyzing it.
Evaluating the results
04:45 AM - Big NPEM Modeling Piperacillin
Setting up the model, the data, sending it, analyzing it, evaluating the results
05:15 PM - Group Discussion - All Together.
06:00 PM - Adjourn.