Workshop: Apr 9-10, 2001

The USC Laboratory of Applied Pharmacokinetics presents a Workshop on

Basic Applications of Pharmacokinetics to Optimal Patient Care: How to do it with a laptop computer at the bedside.

April 9-10, 2001, Moscow, Russia.

This course is for physicians and pharmacists with an interest in population pharmacokinetic/dynamic modeling who have a grasp of the basic aspects of such work. Day 1 will introduce and review Basic PK/PD tools and concepts of pharmacokinetic modeling, and will emphasize their application to optimal patient care. Day 2 will discuss intermediate and advanced PK/PD tools, and concepts, including parametric and nonparametric population modeling. Note: if you would like to bring your own laptop computer to obtain and learn the relevant software (not included in the registration fee), you are encouraged to do so.

Note: Participants planning to come from outside Russia should allow at least one month in order to obtain a visa for entry into Russia.

For more information, please contact

Dr. Irina Bondareva
Institute of Physical and Chemical Medicine
City = Moscow
Country = Russia
Email = lidos@aha.ru

Preliminary Program

Faculty:

Roger Jelliffe, M.D., Professor of Medicine, USC School of Medicine, USA (Coordinator)
Irina Bondareva, Researcher, Moscow.

Monday, April 9, 2001- Pharmacokinetics and Optimal Patient Care

8:30 AM - Registration
9:00 AM - Welcome - Dr. Bondareva
9:15 AM - Review of Basic Pharmacokinetic Concepts - Dr. Jelliffe
		Compartmental Models
			Cumulation and Elimination
			T ½, Fraction lost, Doses sustained.
			Changing T ½, changing dose, outcomes.
9:45 AM - Ways of fitting data for patients
		Linear regression of logs of data
			Must wait for steady state
			Must wait for complete distribution after a dose
		Nonlinear regression on the data itself
			No wait for steady state
			No wait for distribution
		Bayesian fitting - the best
		The MAP Bayesian scenario and feedback strategy

10:30 - Break

10:45 - Estimation of Creatinine Clearance without a urine specimen in unstable patients
11:00 - When to obtain serum data
		Not just the trough
		Capture the dynamics
		Some optimal strategies 
11:40 - How to plan and develop individualized dosage regimens for patients
Set a target goal for each patient according to the need for the drug.
		Aminoglycosides 10 and 1, or 20 and 0.5
		Vancomycin trough 10
		Digoxin - really a 2 compartment model
			Clinical effect correlates better with tissue than serum 
                        concentrations
			How to manage this problem clinically
			Serum troughs 0.9
			Peripheral peaks 7.0
			More for patients with atrial fibrillation

12:15 PM - Lunch

1:30 PM - Introduction to software for clinical applications
a)Select model, set goals, plan initial individualized regimen
Aminoglycosides
A patient with a CCr of 100, 50, 5
When to dose in relation to dialysis
What about "once - daily" dossage?
Vancomycin
	A patient with a CCr of 100, 50, 5
Digoxin
	A patient with a CCr of 100, 50, 5
2:15 PM - 	b)Enter and store past doses and levels
		      A gentamicin example
2:30 PM - Case studies in aminoglycoside therapy
Therapeutic drug monitoring
Making the individualized, Bayesian posterior, model
	Analyzing the data
	A patient on dialysis
2:45 PM - Concentration dependent and non-concentration dependent antibiotics
3:00 PM - Case studies in digoxin therapy
			An initial regimen for a patient with atrial fibrillation
			A case history: another patient with atrial fibrillation
			A patient on digoxin and quinidine

3:30 PM - Break

3:45 PM - Analysis of Antiepileptic Drugs - Dr. Irina Bondareva 
4:15 PM - Modeling diffusion into endocardial vegetations
4:30 PM - Modeling the post-antibiotic effect, and drug penetration into
          endocardial vegetations
4:45 PM - Modeling organism growth and kill
5:15 PM - Adjourn

Tuesday, April 10, 2001 -Modeling Tools and Applications

9:00 AM - Cost-effectiveness of optimal aminoglycoside therapy
9:30 AM - Outcomes in Busulfan therapy for bone marrow transplants in children
10:00 AM - More on Basic Tools
		Modeling the assay error
		When to get levels - optimal experimental design

10:30 AM - Break

10:45 AM - Parametric population models
		What "parametric" means here
		The iterative Bayesian (IT2B) modeling approach
			Separating inter - from intra-individual variability (IIV)
			Separating IIV  from assay error
		Demonstration of the approach - an Amikacin data set

11:15 AM - Nonparametric population modeling approaches
		What "nonparametric" means here
		The NPEM approach
		Using IIV, assay error, and stated ranges
	
11:45 AM - Using population modeling approaches optimally
		Get the assay error polynomial
		Use IT2B - get Gamma
Then use NPEM, get the entire joint density, essentially resolving the population 
into up to one model for each subject studied.

12:15 PM - The separation principle: limitations to current dosage methods

12:30 PM - Lunch


1:10 PM - Introduction to multiple model (MM) dosage design
Software for MM dosage regimens

1:30 PM - Making larger and nonlinear PK/PD models
		Using BOXES to make the model
		Big IT2B - preparing the analysis
		Big NPEM - preparing the analysis
		Getting on the web to the San Diego supercomputer,
                or to our Dell cluster		
		Sending the data and instructions
		Doing the analysis
		Downloading the results, seeing them on the PC
2:15 PM - Getting MM Bayesian posterior joint densities
		MM Bayesian posteriors
A new method - IMM - for detecting changing parameter values in patients
2:30 PM - Other models - Dr. Bondareva
3:00 PM - Coming new things in PK/PD
		Clinical MM dosage design
		Time lags in big IT2B and big NPEM models
		A new NPEM improvement (NPAG) by Bob Leary at SDSC
		
4:00 PM - Adjourn.