New USC PK Workshop Feb 1997

Roger Jelliffe jelliffe at
Tue Dec 24 17:39:05 EST 1996

                                                        December 20, 1996

Dear Colleague:

        The Laboratory of Applied Pharmacokinetics of the USC School of
Medicine is putting on a pharmacokinetic conference on Friday and Saturday,
February 21st and 22nd, 1997, on "Parametric and Nonparametric Population
PK/PD Modeling, Modeling Individualized Antibiotic Therapy and Its Effects,
and 'Multiple Model' Individualization of Drug Dosage". The emphasis will
be on Population Modeling, on modeling drug diffusion into simulated
endocardial vegetations, on modeling the Post-Antibiotic effect, and on
models of bacterial growth and their killing by antibiotics.

WHERE?  WHEN?  The conference will be held in the First Floor Conference
Room and the Computer Classroom of the University of Southern California
School of Pharmacy, 1985 Zonal Avenue, Los Angeles California, 90033, on
Friday and Saturday, February 21st and 22nd, 1997.  One must arrange travel,
hotel, and meal accommodations independently.

OBJECTIVES. The conference objectives are to increase the understanding and
hands - on use of Parametric and Nonparametric Population Pharmacokinetic
Modeling and its role in optimal Bayesian Model-Based, Goal-Oriented,
individualized drug therapy. The concepts and software to implement them
(the USC*PACK clinical software and the parametric iterative 2-stage
Bayesian (IT2B) and nonparametric EM (NPEM) programs) are intended for
general use, with special application to the optimal modeling of and therapy
with cardiovascular drugs, antibiotics, therapy of AIDS, cancer, transplant
situations, psychiatric illnesses, and in Phase 1-2 and 2-3 trials, and
concentration-controlled clinical trials. Information obtained in this
course can also be applied on other equivalent software.

WHO IS IT FOR?  The conference is for physicians and pharmacists with a
working knowledge of pharmacokinetics (Vd, Kel, etc) to apply modern parametric
and nonparametric modeling concepts to the study of population models. Methods
to describe diffusion of antibiotics into simulated endocardial vegetations, the
post-antibiotic effect and its results on bacterial killing will be discussed,
and 'Multiple Model' (MM) approaches to optimally precise dosage regimens will
be introduced. Applications to AZT, Trimethoprim,
Aminoglycosides, Vancomycin, Digoxin, and other drugs will also be discussed.

HANDS-ON SESSIONS! The workshop will be especially oriented around hands-on
use of the software, to maximally familiarize the participants with its
practical use.

BRING YOUR NOTEBOOK!  For those who already have or who wish to obtain
relevant clinical software for this workshop, and who wish to bring it
to the workshop in their own laptop or notebook computer, you are warmly
invited to do so.

THE FACULTY are: Roger Jelliffe, M.D., Prof. of Medicine, USC School of
Medicine, Los Angeles, CA, Course Coordinator; Agneta Hurst, Pharm.D., USC
School of Pharmacy, George Jaresko, Pharm.D., USC School of Pharmacy, David
Bayard, Ph.D., NASA Jet Propulsion Laboratory, Pasadena CA, Mark Milman, Ph.D.,
NASA Jet Propulsion Laboratory, Pasadena CA, Alan Schumitzky, Ph.D., Prof. of
Mathematics, USC, and Michael Van Guilder, Ph.D., USC Laboratory of Applied

        Please call (213) 342-1300, or Fax us at (213) 342-1302, or email us at
jelliffe at   if you have any questions.


                                                Roger W. Jelliffe, M.D.
               Professor of Medicine

Preliminary Program: "Parametric and Nonparametric Population PK/PD
Modeling, Modeling Individualized Antibiotic Therapy and Its Effects, and
'Multiple Model' Individualization of Drug Dosage"
Day 1 - Friday, February 21, 1997
8:30 AM - Registration
9:00 AM - Welcome - Dr. Roger Jelliffe
9:05 AM - An Overview of Population Modeling - Dr. Jelliffe
9:15 AM -  Parametric Population Models
                Naive Pooling, Standard Two Stage, NONMEM
9:45 AM - The Parametric Iterative 2-Stage Bayesian (IT2B) Procedure
10:15 AM - Nonparametric (NP) Population Models - Dr. Alan Schumitzky
                NP Maximum Likelihood
                NP Expectation - Maximization
                Their Relationship to the Separation Theorem and to Optimal
                      Drug Regimens

10:45 AM - Break

11:00 AM - Basic Building Blocks in Population Modeling and Optimal Drug
                      Therapy - Dr. Jelliffe
                Evaluating Renal Function
                Determining the Assay Error Pattern and the Weighting
                      Scheme for Measurements
                When to Get Serum Levels: Optimal Sampling Strategies
                Modeling Environmental Errors in Dosage Preparation and
11:45 AM - Modeling Pharmacologic Effects and Drug Binding to Effect Sites
12:00 Noon - Current Bayesian Modeling and Clinical Models of Drug Diffusion
12:15 PM - Modeling Bacterial Growth and Kill, and the Post-Antibiotic effect

12:30 PM - Lunch

1:30 PM - "Multiple Model" Adaptive Control and Optimal Drug Dosing
                      - Dr. David Bayard
2:00 PM - Applying "Multiple Model" Concepts to Therapy with Vancomycin
                      - Dr. Jelliffe
2:15 PM - Maximum Entropy approaches to obtain discrete population parameter
                      densities - Dr. Mark Milman

2:45 PM - Break

3:00 PM - Making a Population Model of Amikacin - Dr. Jelliffe
3:35 PM - The Front part - an Iterative Bayesian Population (IT2B) Modeler
                Evaluating Bioavailability
                Parameterizing the Model: Choosing Volume, 1/V, Clearance,
                      or Kel
                Entering the Assay Error Pattern Polynomial
                Selecting the Convergence Criterion and iterations
                Defining the first Bayesian prior
                Setting the Initial estimates for the parameter ranges
                Defining the ranges to pass to the nonparametric back part
                Seeing the output and the individual estimates
                Checking the results obtained
4:15 PM - The Back part - the Nonparametric Population Modeler itself
                Using the program the standard way - picking up data and
                      ranges from the front part
                Selecting the number of grid (support) points for the initial
                      joint probability density function (PDF)
                Choosing plot and print options
                Evaluating the Nonparametric Output
                        The log-likelihood function
                        The difference between a cycle likelihood and the max
                        The means, modes, skewnesses, kurtoses, and
                                the percentiles
                        Other single-point Descriptors of Dispersion - the
                                DF-50 and DF-95
                        The 2-D plots of the Marginal PDF's
                        The 3-D plots of pairs of Joint Marginal PDF's
                                Comparing these with covariances and correlations
                        Linking Nonparametric models to Multiple Model
                                Adaptive Control
                        Evaluating the convergence - Absolute and normalized
                                convergence plots
                        Getting individual Bayesian Posterior patient
                                parameter densities

5:30 PM - Adjourn

An evening get - together dinner at the USC Faculty Center

No - host cocktails --------------- 6:00 PM

Dinner ------------------------------- 7:00 PM

Day 2 - Saturday, February 22, 1997

9:00 AM - A Review of Yesterday's Population Modeling - Dr. Jelliffe
9:15 AM - A Tougher problem: Making a Population Model of Trimethoprim
                Drs. Hurst, Jaresko, Jelliffe, and Van Guilder

10:30 AM - Break

10:45 AM - Clinical Applications: Digoxin - Dr. Jelliffe
                Planning the initial regimen. Controlling the
                                Peripheral Compartment
                An interesting patient with Atrial Fibrillation
                A patient with very high levels - why?
11:30 AM - Clinical Applications: Trimethoprim
                Setting the goals for PCP pneumonia
11:45 AM - Clinical Applications: AZT

12:30 PM - Lunch

1:30 PM - Clinical Applications: Vancomycin - Dr. Jelliffe
                Setting the trough goals. Planning the initial regimen
2:15 PM - Clinical Applications: Theophylline
                Setting the goals and planning the regimen: regular
                                and sustained release forms
3:00 PM - Clinical Applications: Analyzing Bacterial Growth and Kill

3:30 PM - Break

3:45 PM - Clinical Applications: Gentamicin - Dr. Jelliffe
                Setting the goals: once daily or what?
                Modeling diffusion into vegetations or abscesses
                Effect models - occupation of toxic binding sites
                Modeling bacterial growth and kill, and post
                                - antibiotic effect.
4:50 PM - Once Daily Aminoglycoside Therapy - Pros and Cons
                Evaluating 80 q 8 versus 240 q 24 - 2 cases, 3 values of CCr
                Serum AUC, and AUC in a peripheral effect compartment
                Evaluating the relative benefit of nonlinear uptake on
                                toxic binding sites
                General Discussion

5:30 PM - Adjourn

Registration Form (You may use this sheet)

        I wish to register to attend the USC Workshop on "Parametric and
Nonparametric Population PK/PD Modeling, Modeling Individualized Antibiotic
Therapy and Its Effects, and 'Multiple Model' Individualization of Drug Dosage"
on February 21 and 22, 1997.

                Registration ($295.00) ________________________________________

                Get - Together Dinner Friday night ($25.00)____________________

                                                          Total:   $___________

                (Sorry - we cannot handle credit cards or purchase orders.  We
                can handle checks and foreign travelers checks in US dollars.)

                Please make checks payable to the:

                        Laboratory of Applied Pharmacokinetics

                Name of Registrant ______________________________       _


                Address__________________________________________         _

                _______________________________________________ __

                City, State, Zip_______________________________      _____


                Phone (____)________________Fax (____)_________________

                Your Email address, please! ____________________________

                I am           am not            a member of ASM.
                I am______ am not_______ a member of AMIA.
                I am______ am not_______ a member of IMIA.

                Please return this form, with your check, directly to:

                Roger W. Jelliffe, M.D.
                Laboratory of Applied Pharmacokinetics
                USC School of Medicine
                CSC 134-B
                2250 Alcazar Street
                Los Angeles, CA 90033

        Our phone is (213) 342-1300, and our Fax is (213) 342-1302.

        Note:  Please register early. First come are first served.
        Registration is limited to the available spaces available.
        Cancellations must be made at least 10 days before the
        workshop in order to obtain a refund.

        I would _____ would not _____ also like to receive information
        about the USC software.
        I heard about this conference from______mailing  _____word of
        mouth _____electronic bulletin board_____other
        (please specify)____________________

More information about the Bionews mailing list