Joel S. Owen, Jill Fiedler-Kelly
Introduction to Population Pharmacokinetic / Pharmacodynamic Analysis with Nonlinear Mixed Effects Models
Joel S. Owen, Jill Fiedler-Kelly
Introduction to Population Pharmacokinetic / Pharmacodynamic Analysis with Nonlinear Mixed Effects Models
- Gebundenes Buch
- Merkliste
- Auf die Merkliste
- Bewerten Bewerten
- Teilen
- Produkt teilen
- Produkterinnerung
- Produkterinnerung
Providing a user-friendly, hands-on introduction to the most powerful tool for population PK/PD models, Introduction to Population PK/PD Analysis with Nonlinear Mixed Effects Models introduces the reader to the NONMEM system, a powerful tool for pharmacokinetic / pharmacodynamic analysis. This useful guide helps pharmaceutical scientists and students learn the requisite information needed to perform mixed effect modeling of pharmacologic data using the NONMEM software package. Chapters discuss population model terminology, Bayesian analysis, PK/PD simulation. An associated website hosts…mehr
Andere Kunden interessierten sich auch für
- Michael J. RosenbergThe Agile Approach to Adaptive Research113,99 €
- Philip RoweEssential Statistics for the Pharmaceutical Sciences125,99 €
- Sheila Annie PetersPhysiologically Based Pharmacokinetic (Pbpk) Modeling and Simulations204,99 €
- Jianling WangPredictive Admet180,99 €
- Polypharmacology in Drug Discovery159,99 €
- Jie-Jack LiDrug Discovery151,99 €
- Applications of Ion Chromatography for Pharmaceutical and Biological Products162,99 €
-
-
-
Providing a user-friendly, hands-on introduction to the most powerful tool for population PK/PD models, Introduction to Population PK/PD Analysis with Nonlinear Mixed Effects Models introduces the reader to the NONMEM system, a powerful tool for pharmacokinetic / pharmacodynamic analysis. This useful guide helps pharmaceutical scientists and students learn the requisite information needed to perform mixed effect modeling of pharmacologic data using the NONMEM software package. Chapters discuss population model terminology, Bayesian analysis, PK/PD simulation. An associated website hosts datasets and programming code.
This book provides a user-friendly, hands-on introduction to the Nonlinear Mixed Effects Modeling (NONMEM) system, the most powerful tool for pharmacokinetic / pharmacodynamic analysis.
Introduces requisite background to using Nonlinear Mixed Effects Modeling (NONMEM), covering data requirements, model building and evaluation, and quality control aspects
Provides examples of nonlinear modeling concepts and estimation basics with discussion on the model building process and applications of empirical Bayesian estimates in the drug development environment
Includes detailed chapters on data set structure, developing control streams for modeling and simulation, model applications, interpretation of NONMEM output and results, and quality control
Has datasets, programming code, and practice exercises with solutions, available on a supplementary website
This book provides a user-friendly, hands-on introduction to the Nonlinear Mixed Effects Modeling (NONMEM) system, the most powerful tool for pharmacokinetic / pharmacodynamic analysis.
Introduces requisite background to using Nonlinear Mixed Effects Modeling (NONMEM), covering data requirements, model building and evaluation, and quality control aspects
Provides examples of nonlinear modeling concepts and estimation basics with discussion on the model building process and applications of empirical Bayesian estimates in the drug development environment
Includes detailed chapters on data set structure, developing control streams for modeling and simulation, model applications, interpretation of NONMEM output and results, and quality control
Has datasets, programming code, and practice exercises with solutions, available on a supplementary website
Produktdetails
- Produktdetails
- Verlag: Wiley & Sons
- 1. Auflage
- Seitenzahl: 320
- Erscheinungstermin: 8. Juli 2014
- Englisch
- Abmessung: 240mm x 161mm x 22mm
- Gewicht: 562g
- ISBN-13: 9780470582299
- ISBN-10: 0470582294
- Artikelnr.: 40185970
- Verlag: Wiley & Sons
- 1. Auflage
- Seitenzahl: 320
- Erscheinungstermin: 8. Juli 2014
- Englisch
- Abmessung: 240mm x 161mm x 22mm
- Gewicht: 562g
- ISBN-13: 9780470582299
- ISBN-10: 0470582294
- Artikelnr.: 40185970
Joel S. Owen is Professor of Pharmaceutics at Union University, Jackson, Tennessee and President and Principal Scientist of Joel S. Owen, LLC. He has led workshops on NONMEM and PK/PD modeling concepts and applications and served as Director PK/PD at Cognigen Corporation in Buffalo, New York. He has published 16 articles in research publications. Jill Fiedler-Kelly is Vice President and Chief Scientific Officer of Cognigen Corporation and Adjunct Associate Professor of Pharmaceutical Sciences at the University at Buffalo. She has been teaching workshops and graduate courses on population modeling for over 10 years and has published more than 20 articles and book chapters on pharmacokinetics and pharmacodynamics.
Preface xiii CHAPTER 1 The Practice of Pharmacometrics 1 1.1 Introduction 1
1.2 Applications of Sparse Data Analysis 2 1.3 Impact of Pharmacometrics 4
1.4 Clinical Example 5 CHAPTER 2 Population Model Concepts and Terminology
9 2.1 Introduction 9 2.2 Model Elements 10 2.3 Individual Subject Models 11
2.4 Population Models 12 2.4.1 Fixed-Effect Parameters 13 2.4.2
Random-Effect Parameters 14 2.5 Models of Random Between-Subject
Variability (L1) 17 2.5.1 Additive Variation 17 2.5.2 Constant Coefficient
of Variation 18 2.5.3 Exponential Variation 18 2.5.4 Modeling Sources of
Between-Subject Variation 19 2.6 Models of Random Variability in
Observations (L2) 19 2.6.1 Additive Variation 20 2.6.2 Constant Coefficient
of Variation 21 2.6.3 Additive Plus CCV Model 22 2.6.4 Log-Error Model 24
2.6.5 Relationship Between RV Expressions and Predicted Concentrations 24
2.6.6 Significance of the Magnitude of RV 25 2.7 Estimation Methods 26 2.8
Objective Function 26 2.9 Bayesian Estimation 27 CHAPTER 3 NONMEM Overview
and Writing an NM-TRAN Control Stream 28 3.1 Introduction 28 3.2 Components
of the NONMEM System 28 3.3 General Rules 30 3.4 Required Control Stream
Components 31 3.4.1 $PROBLEM Record 31 3.4.2 The $DATA Record 32 3.4.3 The
$INPUT Record 35 3.5 Specifying the Model in NM-TRAN 35 3.5.1 Calling
PREDPP Subroutines for Specific PK Models 35 3.5.2 Specifying the Model in
the $PK Block 38 3.5.3 Specifying Residual Variability in the $ERROR Block
45 3.5.4 Specifying Models Using the $PRED Block 49 3.6 Specifying Initial
Estimates with $THETA, $OMEGA, and $SIGMA 50 3.7 Requesting Estimation and
Related Options 56 3.8 Requesting Estimates of the Precision of Parameter
Estimates 62 3.9 Controlling the Output 63 CHAPTER 4 Datasets 66 4.1
Introduction 66 4.2 Arrangement of the Dataset 68 4.3 Variables of the
Dataset 71 4.3.1 TIME 71 4.3.2 DATE 71 4.3.3 ID 72 4.3.4 DV 74 4.3.5 MDV 74
4.3.6 CMT 74 4.3.7 EVID 75 4.3.8 AMT 76 4.3.9 RATE 77 4.3.10 ADDL 78 4.3.11
II 79 4.3.12 SS 80 4.4 Constructing Datasets with Flexibility to Apply
Alternate Models 80 4.5 Examples of Event Records 81 4.5.1 Alternatives for
Specifying Time 81 4.5.2 Infusions and Zero-Order Input 81 4.5.3 Using ADDL
82 4.5.4 Steady-State Approach 83 4.5.5 Samples Before and After Achieving
Steady State 83 4.5.6 Unscheduled Doses in a Steady-State Regimen 84 4.5.7
Steady-State Dosing with an Irregular Dosing Interval 84 4.5.8 Multiple
Routes of Administration 85 4.5.9 Modeling Multiple Dependent Variable Data
Types 86 4.5.10 Dataset for $PRED 86 4.6 Beyond Doses and Observations 87
4.6.1 Other Data Items 87 4.6.2 Covariate Changes over Time 88 4.6.3
Inclusion of a Header Row 89 CHAPTER 5 Model Building: Typical Process 90
5.1 Introduction 90 5.2 Analysis Planning 90 5.3 Analysis Dataset Creation
92 5.4 Dataset Quality Control 93 5.5 Exploratory Data Analysis 94 5.5.1
EDA: Population Description 95 5.5.2 EDA: Dose-Related Data 99 5.5.3 EDA:
Concentration-Related Data 99 5.5.4 EDA: Considerations with Large Datasets
111 5.5.5 EDA: Summary 115 5.6 Base Model Development 116 5.6.1 Standard
Model Diagnostic Plots and Interpretation 116 5.6.2 Estimation of Random
Effects 130 5.6.3 Precision of Parameter Estimates (Based on $COV Step) 137
5.7 Covariate Evaluation 138 5.7.1 Covariate Evaluation Methodologies 140
5.7.2 Statistical Basis for Covariate Selection 141 5.7.3 Diagnostic Plots
to Illustrate Parameter-Covariate Relationships 143 5.7.4 Typical
Functional Forms for Covariate-Parameter Relationships 148 5.7.5 Centering
Covariate Effects 156 5.7.6 Forward Selection Process 160 5.7.7 Evaluation
of the Full Multivariable Model 167 5.7.8 Backward Elimination Process 169
5.7.9 Other Covariate Evaluation Approaches 171 5.8 Model Refinement 172
CHAPTER 6 Interpreting the NONMEM Output 178 6.1 Introduction 178 6.2
Description of the Output Files 178 6.3 The NONMEM Report File 179 6.3.1
NONMEM-Related Output 179 6.3.2 PREDPP-Related Output 180 6.3.3 Output from
Monitoring of the Search 180 6.3.4 Minimum Value of the Objective Function
and Final Parameter Estimates 182 6.3.5 Covariance Step Output 186 6.3.6
Additional Output 187 6.4 Error Messages: Interpretation and Resolution 188
6.4.1 NM-TRAN Errors 188 6.4.2 $ESTIMATION Step Failures 189 6.4.3
$COVARIANCE Step Failures 190 6.4.4 PREDPP Errors 191 6.4.5 Other Types of
NONMEM Errors 192 6.4.6 FORTRAN Compiler or Other Run-Time Errors 193 6.5
General Suggestions for Diagnosing Problems 193 CHAPTER 7 App lications
Using Parameter Estimates from the Individual 198 7.1 Introduction 198 7.2
Bayes Theorem and Individual Parameter Estimates 200 7.3 Obtaining
Individual Parameter Estimates 202 7.4 Applications of Individual Parameter
Estimates 204 7.4.1 Generating Subject-Specific Exposure Estimates 204
7.4.2 Individual Exposure Estimates for Group Comparisons 210 CHAPTER 8
Introduction to Model Evaluation 212 8.1 Introduction 212 8.2 Internal
Validation 212 8.3 External Validation 213 8.4 Predictive Performance
Assessment 214 8.5 Objective Function Mapping 217 8.6 Leverage Analysis 220
8.7 Bootstrap Procedures 222 8.8 Visual and Numerical Predictive Check
Procedures 223 8.8.1 The VPC Procedure 223 8.8.2 Presentation of VPC
Results 225 8.8.3 The Numerical Predictive Check (NPC) Procedure 229 8.9
Posterior Predictive Check Procedures 229 CHAPTER 9 User-Written Models 232
9.1 Introduction 232 9.2 $MODEL 235 9.3 $SUBROUTINES 236 9.3.1 General
Linear Models (ADVAN5 and ADVAN7) 236 9.3.2 General Nonlinear Models
(ADVAN6, ADVAN8, ADVAN9, and ADVAN13) 238 9.3.3 $DES 238 9.4 A Series of
Examples 240 9.4.1 Defined Fractions Absorbed by Zero- and First-Order
Processes 240 9.4.2 Sequential Absorption with First-Order Rates, without
Defined Fractions 242 9.4.3 Parallel Zero-Order and First-Order Absorption,
without Defined Fractions 243 9.4.4 Parallel First-Order Absorption
Processes, without Defined Fractions 245 9.4.5 Zero-Order Input into the
Depot Compartment 246 9.4.6 Parent and Metabolite Model: Differential
Equations 247 CHAPTER 10 PK/PD Models 250 10.1 Introduction 250 10.2
Implementation of PD Models in NONMEM 251 10.3 $PRED 252 10.3.1
Direct-Effect PK/PD Examples: PK Concentrations in the Dataset 253 10.3.2
Direct-Effect PK/PD Example: PK from Computed Concentrations 255 10.4 $PK
256 10.4.1 Specific ADVANs (ADVAN1-ADVAN4 and ADVAN10-ADVAN12) 256 10.4.2
General ADVANs (ADVAN5-ADVAN9 and ADVAN13) 257 10.4.3 PREDPP: Effect
Compartment Link Model Example (PD in $ERROR) 257 10.4.4 PREDPP: Indirect
Response Model Example: PD in $DES 259 10.5 Odd-Type Data: Analysis of
Noncontinuous Data 261 10.6 PD Model Complexity 262 10.7 Communication of
Results 263 CHAPTER 11 Simulation Basics 265 11.1 Introduction 265 11.2 The
Simulation Plan 265 11.2.1 Simulation Components 266 11.2.2 The
Input-Output Model 266 11.2.3 The Covariate Distribution Model 270 11.2.4
The Trial Execution Model 273 11.2.5 Replication of the Study 274 11.2.6
Analysis of the Simulated Data 275 11.2.7 Decision Making Using Simulations
275 11.3 Miscellaneous Other Simulation-Related Considerations 276 11.3.1
The Seed Value 276 11.3.2 Consideration of Parameter Uncertainty 277 11.3.3
Constraining Random Effects or Responses 278 CHAPTER 12 Quality Control 285
12.1 Introduction 285 12.2 QC of the Data Analysis Plan 285 12.3 Analysis
Dataset Creation 286 12.3.1 Exploratory Data Analysis and Its Role in
Dataset QC 287 12.3.2 QC in Data Collection 287 12.4 QC of Model
Development 288 12.4.1 QC of NM-TRAN Control Streams 289 12.4.2 Model
Diagnostic Plots and Model Evaluation Steps as QC 290 12.5 Documentation of
QC Efforts 290 12.6 Summary 291 References 292 Index 293
1.2 Applications of Sparse Data Analysis 2 1.3 Impact of Pharmacometrics 4
1.4 Clinical Example 5 CHAPTER 2 Population Model Concepts and Terminology
9 2.1 Introduction 9 2.2 Model Elements 10 2.3 Individual Subject Models 11
2.4 Population Models 12 2.4.1 Fixed-Effect Parameters 13 2.4.2
Random-Effect Parameters 14 2.5 Models of Random Between-Subject
Variability (L1) 17 2.5.1 Additive Variation 17 2.5.2 Constant Coefficient
of Variation 18 2.5.3 Exponential Variation 18 2.5.4 Modeling Sources of
Between-Subject Variation 19 2.6 Models of Random Variability in
Observations (L2) 19 2.6.1 Additive Variation 20 2.6.2 Constant Coefficient
of Variation 21 2.6.3 Additive Plus CCV Model 22 2.6.4 Log-Error Model 24
2.6.5 Relationship Between RV Expressions and Predicted Concentrations 24
2.6.6 Significance of the Magnitude of RV 25 2.7 Estimation Methods 26 2.8
Objective Function 26 2.9 Bayesian Estimation 27 CHAPTER 3 NONMEM Overview
and Writing an NM-TRAN Control Stream 28 3.1 Introduction 28 3.2 Components
of the NONMEM System 28 3.3 General Rules 30 3.4 Required Control Stream
Components 31 3.4.1 $PROBLEM Record 31 3.4.2 The $DATA Record 32 3.4.3 The
$INPUT Record 35 3.5 Specifying the Model in NM-TRAN 35 3.5.1 Calling
PREDPP Subroutines for Specific PK Models 35 3.5.2 Specifying the Model in
the $PK Block 38 3.5.3 Specifying Residual Variability in the $ERROR Block
45 3.5.4 Specifying Models Using the $PRED Block 49 3.6 Specifying Initial
Estimates with $THETA, $OMEGA, and $SIGMA 50 3.7 Requesting Estimation and
Related Options 56 3.8 Requesting Estimates of the Precision of Parameter
Estimates 62 3.9 Controlling the Output 63 CHAPTER 4 Datasets 66 4.1
Introduction 66 4.2 Arrangement of the Dataset 68 4.3 Variables of the
Dataset 71 4.3.1 TIME 71 4.3.2 DATE 71 4.3.3 ID 72 4.3.4 DV 74 4.3.5 MDV 74
4.3.6 CMT 74 4.3.7 EVID 75 4.3.8 AMT 76 4.3.9 RATE 77 4.3.10 ADDL 78 4.3.11
II 79 4.3.12 SS 80 4.4 Constructing Datasets with Flexibility to Apply
Alternate Models 80 4.5 Examples of Event Records 81 4.5.1 Alternatives for
Specifying Time 81 4.5.2 Infusions and Zero-Order Input 81 4.5.3 Using ADDL
82 4.5.4 Steady-State Approach 83 4.5.5 Samples Before and After Achieving
Steady State 83 4.5.6 Unscheduled Doses in a Steady-State Regimen 84 4.5.7
Steady-State Dosing with an Irregular Dosing Interval 84 4.5.8 Multiple
Routes of Administration 85 4.5.9 Modeling Multiple Dependent Variable Data
Types 86 4.5.10 Dataset for $PRED 86 4.6 Beyond Doses and Observations 87
4.6.1 Other Data Items 87 4.6.2 Covariate Changes over Time 88 4.6.3
Inclusion of a Header Row 89 CHAPTER 5 Model Building: Typical Process 90
5.1 Introduction 90 5.2 Analysis Planning 90 5.3 Analysis Dataset Creation
92 5.4 Dataset Quality Control 93 5.5 Exploratory Data Analysis 94 5.5.1
EDA: Population Description 95 5.5.2 EDA: Dose-Related Data 99 5.5.3 EDA:
Concentration-Related Data 99 5.5.4 EDA: Considerations with Large Datasets
111 5.5.5 EDA: Summary 115 5.6 Base Model Development 116 5.6.1 Standard
Model Diagnostic Plots and Interpretation 116 5.6.2 Estimation of Random
Effects 130 5.6.3 Precision of Parameter Estimates (Based on $COV Step) 137
5.7 Covariate Evaluation 138 5.7.1 Covariate Evaluation Methodologies 140
5.7.2 Statistical Basis for Covariate Selection 141 5.7.3 Diagnostic Plots
to Illustrate Parameter-Covariate Relationships 143 5.7.4 Typical
Functional Forms for Covariate-Parameter Relationships 148 5.7.5 Centering
Covariate Effects 156 5.7.6 Forward Selection Process 160 5.7.7 Evaluation
of the Full Multivariable Model 167 5.7.8 Backward Elimination Process 169
5.7.9 Other Covariate Evaluation Approaches 171 5.8 Model Refinement 172
CHAPTER 6 Interpreting the NONMEM Output 178 6.1 Introduction 178 6.2
Description of the Output Files 178 6.3 The NONMEM Report File 179 6.3.1
NONMEM-Related Output 179 6.3.2 PREDPP-Related Output 180 6.3.3 Output from
Monitoring of the Search 180 6.3.4 Minimum Value of the Objective Function
and Final Parameter Estimates 182 6.3.5 Covariance Step Output 186 6.3.6
Additional Output 187 6.4 Error Messages: Interpretation and Resolution 188
6.4.1 NM-TRAN Errors 188 6.4.2 $ESTIMATION Step Failures 189 6.4.3
$COVARIANCE Step Failures 190 6.4.4 PREDPP Errors 191 6.4.5 Other Types of
NONMEM Errors 192 6.4.6 FORTRAN Compiler or Other Run-Time Errors 193 6.5
General Suggestions for Diagnosing Problems 193 CHAPTER 7 App lications
Using Parameter Estimates from the Individual 198 7.1 Introduction 198 7.2
Bayes Theorem and Individual Parameter Estimates 200 7.3 Obtaining
Individual Parameter Estimates 202 7.4 Applications of Individual Parameter
Estimates 204 7.4.1 Generating Subject-Specific Exposure Estimates 204
7.4.2 Individual Exposure Estimates for Group Comparisons 210 CHAPTER 8
Introduction to Model Evaluation 212 8.1 Introduction 212 8.2 Internal
Validation 212 8.3 External Validation 213 8.4 Predictive Performance
Assessment 214 8.5 Objective Function Mapping 217 8.6 Leverage Analysis 220
8.7 Bootstrap Procedures 222 8.8 Visual and Numerical Predictive Check
Procedures 223 8.8.1 The VPC Procedure 223 8.8.2 Presentation of VPC
Results 225 8.8.3 The Numerical Predictive Check (NPC) Procedure 229 8.9
Posterior Predictive Check Procedures 229 CHAPTER 9 User-Written Models 232
9.1 Introduction 232 9.2 $MODEL 235 9.3 $SUBROUTINES 236 9.3.1 General
Linear Models (ADVAN5 and ADVAN7) 236 9.3.2 General Nonlinear Models
(ADVAN6, ADVAN8, ADVAN9, and ADVAN13) 238 9.3.3 $DES 238 9.4 A Series of
Examples 240 9.4.1 Defined Fractions Absorbed by Zero- and First-Order
Processes 240 9.4.2 Sequential Absorption with First-Order Rates, without
Defined Fractions 242 9.4.3 Parallel Zero-Order and First-Order Absorption,
without Defined Fractions 243 9.4.4 Parallel First-Order Absorption
Processes, without Defined Fractions 245 9.4.5 Zero-Order Input into the
Depot Compartment 246 9.4.6 Parent and Metabolite Model: Differential
Equations 247 CHAPTER 10 PK/PD Models 250 10.1 Introduction 250 10.2
Implementation of PD Models in NONMEM 251 10.3 $PRED 252 10.3.1
Direct-Effect PK/PD Examples: PK Concentrations in the Dataset 253 10.3.2
Direct-Effect PK/PD Example: PK from Computed Concentrations 255 10.4 $PK
256 10.4.1 Specific ADVANs (ADVAN1-ADVAN4 and ADVAN10-ADVAN12) 256 10.4.2
General ADVANs (ADVAN5-ADVAN9 and ADVAN13) 257 10.4.3 PREDPP: Effect
Compartment Link Model Example (PD in $ERROR) 257 10.4.4 PREDPP: Indirect
Response Model Example: PD in $DES 259 10.5 Odd-Type Data: Analysis of
Noncontinuous Data 261 10.6 PD Model Complexity 262 10.7 Communication of
Results 263 CHAPTER 11 Simulation Basics 265 11.1 Introduction 265 11.2 The
Simulation Plan 265 11.2.1 Simulation Components 266 11.2.2 The
Input-Output Model 266 11.2.3 The Covariate Distribution Model 270 11.2.4
The Trial Execution Model 273 11.2.5 Replication of the Study 274 11.2.6
Analysis of the Simulated Data 275 11.2.7 Decision Making Using Simulations
275 11.3 Miscellaneous Other Simulation-Related Considerations 276 11.3.1
The Seed Value 276 11.3.2 Consideration of Parameter Uncertainty 277 11.3.3
Constraining Random Effects or Responses 278 CHAPTER 12 Quality Control 285
12.1 Introduction 285 12.2 QC of the Data Analysis Plan 285 12.3 Analysis
Dataset Creation 286 12.3.1 Exploratory Data Analysis and Its Role in
Dataset QC 287 12.3.2 QC in Data Collection 287 12.4 QC of Model
Development 288 12.4.1 QC of NM-TRAN Control Streams 289 12.4.2 Model
Diagnostic Plots and Model Evaluation Steps as QC 290 12.5 Documentation of
QC Efforts 290 12.6 Summary 291 References 292 Index 293
Preface xiii CHAPTER 1 The Practice of Pharmacometrics 1 1.1 Introduction 1
1.2 Applications of Sparse Data Analysis 2 1.3 Impact of Pharmacometrics 4
1.4 Clinical Example 5 CHAPTER 2 Population Model Concepts and Terminology
9 2.1 Introduction 9 2.2 Model Elements 10 2.3 Individual Subject Models 11
2.4 Population Models 12 2.4.1 Fixed-Effect Parameters 13 2.4.2
Random-Effect Parameters 14 2.5 Models of Random Between-Subject
Variability (L1) 17 2.5.1 Additive Variation 17 2.5.2 Constant Coefficient
of Variation 18 2.5.3 Exponential Variation 18 2.5.4 Modeling Sources of
Between-Subject Variation 19 2.6 Models of Random Variability in
Observations (L2) 19 2.6.1 Additive Variation 20 2.6.2 Constant Coefficient
of Variation 21 2.6.3 Additive Plus CCV Model 22 2.6.4 Log-Error Model 24
2.6.5 Relationship Between RV Expressions and Predicted Concentrations 24
2.6.6 Significance of the Magnitude of RV 25 2.7 Estimation Methods 26 2.8
Objective Function 26 2.9 Bayesian Estimation 27 CHAPTER 3 NONMEM Overview
and Writing an NM-TRAN Control Stream 28 3.1 Introduction 28 3.2 Components
of the NONMEM System 28 3.3 General Rules 30 3.4 Required Control Stream
Components 31 3.4.1 $PROBLEM Record 31 3.4.2 The $DATA Record 32 3.4.3 The
$INPUT Record 35 3.5 Specifying the Model in NM-TRAN 35 3.5.1 Calling
PREDPP Subroutines for Specific PK Models 35 3.5.2 Specifying the Model in
the $PK Block 38 3.5.3 Specifying Residual Variability in the $ERROR Block
45 3.5.4 Specifying Models Using the $PRED Block 49 3.6 Specifying Initial
Estimates with $THETA, $OMEGA, and $SIGMA 50 3.7 Requesting Estimation and
Related Options 56 3.8 Requesting Estimates of the Precision of Parameter
Estimates 62 3.9 Controlling the Output 63 CHAPTER 4 Datasets 66 4.1
Introduction 66 4.2 Arrangement of the Dataset 68 4.3 Variables of the
Dataset 71 4.3.1 TIME 71 4.3.2 DATE 71 4.3.3 ID 72 4.3.4 DV 74 4.3.5 MDV 74
4.3.6 CMT 74 4.3.7 EVID 75 4.3.8 AMT 76 4.3.9 RATE 77 4.3.10 ADDL 78 4.3.11
II 79 4.3.12 SS 80 4.4 Constructing Datasets with Flexibility to Apply
Alternate Models 80 4.5 Examples of Event Records 81 4.5.1 Alternatives for
Specifying Time 81 4.5.2 Infusions and Zero-Order Input 81 4.5.3 Using ADDL
82 4.5.4 Steady-State Approach 83 4.5.5 Samples Before and After Achieving
Steady State 83 4.5.6 Unscheduled Doses in a Steady-State Regimen 84 4.5.7
Steady-State Dosing with an Irregular Dosing Interval 84 4.5.8 Multiple
Routes of Administration 85 4.5.9 Modeling Multiple Dependent Variable Data
Types 86 4.5.10 Dataset for $PRED 86 4.6 Beyond Doses and Observations 87
4.6.1 Other Data Items 87 4.6.2 Covariate Changes over Time 88 4.6.3
Inclusion of a Header Row 89 CHAPTER 5 Model Building: Typical Process 90
5.1 Introduction 90 5.2 Analysis Planning 90 5.3 Analysis Dataset Creation
92 5.4 Dataset Quality Control 93 5.5 Exploratory Data Analysis 94 5.5.1
EDA: Population Description 95 5.5.2 EDA: Dose-Related Data 99 5.5.3 EDA:
Concentration-Related Data 99 5.5.4 EDA: Considerations with Large Datasets
111 5.5.5 EDA: Summary 115 5.6 Base Model Development 116 5.6.1 Standard
Model Diagnostic Plots and Interpretation 116 5.6.2 Estimation of Random
Effects 130 5.6.3 Precision of Parameter Estimates (Based on $COV Step) 137
5.7 Covariate Evaluation 138 5.7.1 Covariate Evaluation Methodologies 140
5.7.2 Statistical Basis for Covariate Selection 141 5.7.3 Diagnostic Plots
to Illustrate Parameter-Covariate Relationships 143 5.7.4 Typical
Functional Forms for Covariate-Parameter Relationships 148 5.7.5 Centering
Covariate Effects 156 5.7.6 Forward Selection Process 160 5.7.7 Evaluation
of the Full Multivariable Model 167 5.7.8 Backward Elimination Process 169
5.7.9 Other Covariate Evaluation Approaches 171 5.8 Model Refinement 172
CHAPTER 6 Interpreting the NONMEM Output 178 6.1 Introduction 178 6.2
Description of the Output Files 178 6.3 The NONMEM Report File 179 6.3.1
NONMEM-Related Output 179 6.3.2 PREDPP-Related Output 180 6.3.3 Output from
Monitoring of the Search 180 6.3.4 Minimum Value of the Objective Function
and Final Parameter Estimates 182 6.3.5 Covariance Step Output 186 6.3.6
Additional Output 187 6.4 Error Messages: Interpretation and Resolution 188
6.4.1 NM-TRAN Errors 188 6.4.2 $ESTIMATION Step Failures 189 6.4.3
$COVARIANCE Step Failures 190 6.4.4 PREDPP Errors 191 6.4.5 Other Types of
NONMEM Errors 192 6.4.6 FORTRAN Compiler or Other Run-Time Errors 193 6.5
General Suggestions for Diagnosing Problems 193 CHAPTER 7 App lications
Using Parameter Estimates from the Individual 198 7.1 Introduction 198 7.2
Bayes Theorem and Individual Parameter Estimates 200 7.3 Obtaining
Individual Parameter Estimates 202 7.4 Applications of Individual Parameter
Estimates 204 7.4.1 Generating Subject-Specific Exposure Estimates 204
7.4.2 Individual Exposure Estimates for Group Comparisons 210 CHAPTER 8
Introduction to Model Evaluation 212 8.1 Introduction 212 8.2 Internal
Validation 212 8.3 External Validation 213 8.4 Predictive Performance
Assessment 214 8.5 Objective Function Mapping 217 8.6 Leverage Analysis 220
8.7 Bootstrap Procedures 222 8.8 Visual and Numerical Predictive Check
Procedures 223 8.8.1 The VPC Procedure 223 8.8.2 Presentation of VPC
Results 225 8.8.3 The Numerical Predictive Check (NPC) Procedure 229 8.9
Posterior Predictive Check Procedures 229 CHAPTER 9 User-Written Models 232
9.1 Introduction 232 9.2 $MODEL 235 9.3 $SUBROUTINES 236 9.3.1 General
Linear Models (ADVAN5 and ADVAN7) 236 9.3.2 General Nonlinear Models
(ADVAN6, ADVAN8, ADVAN9, and ADVAN13) 238 9.3.3 $DES 238 9.4 A Series of
Examples 240 9.4.1 Defined Fractions Absorbed by Zero- and First-Order
Processes 240 9.4.2 Sequential Absorption with First-Order Rates, without
Defined Fractions 242 9.4.3 Parallel Zero-Order and First-Order Absorption,
without Defined Fractions 243 9.4.4 Parallel First-Order Absorption
Processes, without Defined Fractions 245 9.4.5 Zero-Order Input into the
Depot Compartment 246 9.4.6 Parent and Metabolite Model: Differential
Equations 247 CHAPTER 10 PK/PD Models 250 10.1 Introduction 250 10.2
Implementation of PD Models in NONMEM 251 10.3 $PRED 252 10.3.1
Direct-Effect PK/PD Examples: PK Concentrations in the Dataset 253 10.3.2
Direct-Effect PK/PD Example: PK from Computed Concentrations 255 10.4 $PK
256 10.4.1 Specific ADVANs (ADVAN1-ADVAN4 and ADVAN10-ADVAN12) 256 10.4.2
General ADVANs (ADVAN5-ADVAN9 and ADVAN13) 257 10.4.3 PREDPP: Effect
Compartment Link Model Example (PD in $ERROR) 257 10.4.4 PREDPP: Indirect
Response Model Example: PD in $DES 259 10.5 Odd-Type Data: Analysis of
Noncontinuous Data 261 10.6 PD Model Complexity 262 10.7 Communication of
Results 263 CHAPTER 11 Simulation Basics 265 11.1 Introduction 265 11.2 The
Simulation Plan 265 11.2.1 Simulation Components 266 11.2.2 The
Input-Output Model 266 11.2.3 The Covariate Distribution Model 270 11.2.4
The Trial Execution Model 273 11.2.5 Replication of the Study 274 11.2.6
Analysis of the Simulated Data 275 11.2.7 Decision Making Using Simulations
275 11.3 Miscellaneous Other Simulation-Related Considerations 276 11.3.1
The Seed Value 276 11.3.2 Consideration of Parameter Uncertainty 277 11.3.3
Constraining Random Effects or Responses 278 CHAPTER 12 Quality Control 285
12.1 Introduction 285 12.2 QC of the Data Analysis Plan 285 12.3 Analysis
Dataset Creation 286 12.3.1 Exploratory Data Analysis and Its Role in
Dataset QC 287 12.3.2 QC in Data Collection 287 12.4 QC of Model
Development 288 12.4.1 QC of NM-TRAN Control Streams 289 12.4.2 Model
Diagnostic Plots and Model Evaluation Steps as QC 290 12.5 Documentation of
QC Efforts 290 12.6 Summary 291 References 292 Index 293
1.2 Applications of Sparse Data Analysis 2 1.3 Impact of Pharmacometrics 4
1.4 Clinical Example 5 CHAPTER 2 Population Model Concepts and Terminology
9 2.1 Introduction 9 2.2 Model Elements 10 2.3 Individual Subject Models 11
2.4 Population Models 12 2.4.1 Fixed-Effect Parameters 13 2.4.2
Random-Effect Parameters 14 2.5 Models of Random Between-Subject
Variability (L1) 17 2.5.1 Additive Variation 17 2.5.2 Constant Coefficient
of Variation 18 2.5.3 Exponential Variation 18 2.5.4 Modeling Sources of
Between-Subject Variation 19 2.6 Models of Random Variability in
Observations (L2) 19 2.6.1 Additive Variation 20 2.6.2 Constant Coefficient
of Variation 21 2.6.3 Additive Plus CCV Model 22 2.6.4 Log-Error Model 24
2.6.5 Relationship Between RV Expressions and Predicted Concentrations 24
2.6.6 Significance of the Magnitude of RV 25 2.7 Estimation Methods 26 2.8
Objective Function 26 2.9 Bayesian Estimation 27 CHAPTER 3 NONMEM Overview
and Writing an NM-TRAN Control Stream 28 3.1 Introduction 28 3.2 Components
of the NONMEM System 28 3.3 General Rules 30 3.4 Required Control Stream
Components 31 3.4.1 $PROBLEM Record 31 3.4.2 The $DATA Record 32 3.4.3 The
$INPUT Record 35 3.5 Specifying the Model in NM-TRAN 35 3.5.1 Calling
PREDPP Subroutines for Specific PK Models 35 3.5.2 Specifying the Model in
the $PK Block 38 3.5.3 Specifying Residual Variability in the $ERROR Block
45 3.5.4 Specifying Models Using the $PRED Block 49 3.6 Specifying Initial
Estimates with $THETA, $OMEGA, and $SIGMA 50 3.7 Requesting Estimation and
Related Options 56 3.8 Requesting Estimates of the Precision of Parameter
Estimates 62 3.9 Controlling the Output 63 CHAPTER 4 Datasets 66 4.1
Introduction 66 4.2 Arrangement of the Dataset 68 4.3 Variables of the
Dataset 71 4.3.1 TIME 71 4.3.2 DATE 71 4.3.3 ID 72 4.3.4 DV 74 4.3.5 MDV 74
4.3.6 CMT 74 4.3.7 EVID 75 4.3.8 AMT 76 4.3.9 RATE 77 4.3.10 ADDL 78 4.3.11
II 79 4.3.12 SS 80 4.4 Constructing Datasets with Flexibility to Apply
Alternate Models 80 4.5 Examples of Event Records 81 4.5.1 Alternatives for
Specifying Time 81 4.5.2 Infusions and Zero-Order Input 81 4.5.3 Using ADDL
82 4.5.4 Steady-State Approach 83 4.5.5 Samples Before and After Achieving
Steady State 83 4.5.6 Unscheduled Doses in a Steady-State Regimen 84 4.5.7
Steady-State Dosing with an Irregular Dosing Interval 84 4.5.8 Multiple
Routes of Administration 85 4.5.9 Modeling Multiple Dependent Variable Data
Types 86 4.5.10 Dataset for $PRED 86 4.6 Beyond Doses and Observations 87
4.6.1 Other Data Items 87 4.6.2 Covariate Changes over Time 88 4.6.3
Inclusion of a Header Row 89 CHAPTER 5 Model Building: Typical Process 90
5.1 Introduction 90 5.2 Analysis Planning 90 5.3 Analysis Dataset Creation
92 5.4 Dataset Quality Control 93 5.5 Exploratory Data Analysis 94 5.5.1
EDA: Population Description 95 5.5.2 EDA: Dose-Related Data 99 5.5.3 EDA:
Concentration-Related Data 99 5.5.4 EDA: Considerations with Large Datasets
111 5.5.5 EDA: Summary 115 5.6 Base Model Development 116 5.6.1 Standard
Model Diagnostic Plots and Interpretation 116 5.6.2 Estimation of Random
Effects 130 5.6.3 Precision of Parameter Estimates (Based on $COV Step) 137
5.7 Covariate Evaluation 138 5.7.1 Covariate Evaluation Methodologies 140
5.7.2 Statistical Basis for Covariate Selection 141 5.7.3 Diagnostic Plots
to Illustrate Parameter-Covariate Relationships 143 5.7.4 Typical
Functional Forms for Covariate-Parameter Relationships 148 5.7.5 Centering
Covariate Effects 156 5.7.6 Forward Selection Process 160 5.7.7 Evaluation
of the Full Multivariable Model 167 5.7.8 Backward Elimination Process 169
5.7.9 Other Covariate Evaluation Approaches 171 5.8 Model Refinement 172
CHAPTER 6 Interpreting the NONMEM Output 178 6.1 Introduction 178 6.2
Description of the Output Files 178 6.3 The NONMEM Report File 179 6.3.1
NONMEM-Related Output 179 6.3.2 PREDPP-Related Output 180 6.3.3 Output from
Monitoring of the Search 180 6.3.4 Minimum Value of the Objective Function
and Final Parameter Estimates 182 6.3.5 Covariance Step Output 186 6.3.6
Additional Output 187 6.4 Error Messages: Interpretation and Resolution 188
6.4.1 NM-TRAN Errors 188 6.4.2 $ESTIMATION Step Failures 189 6.4.3
$COVARIANCE Step Failures 190 6.4.4 PREDPP Errors 191 6.4.5 Other Types of
NONMEM Errors 192 6.4.6 FORTRAN Compiler or Other Run-Time Errors 193 6.5
General Suggestions for Diagnosing Problems 193 CHAPTER 7 App lications
Using Parameter Estimates from the Individual 198 7.1 Introduction 198 7.2
Bayes Theorem and Individual Parameter Estimates 200 7.3 Obtaining
Individual Parameter Estimates 202 7.4 Applications of Individual Parameter
Estimates 204 7.4.1 Generating Subject-Specific Exposure Estimates 204
7.4.2 Individual Exposure Estimates for Group Comparisons 210 CHAPTER 8
Introduction to Model Evaluation 212 8.1 Introduction 212 8.2 Internal
Validation 212 8.3 External Validation 213 8.4 Predictive Performance
Assessment 214 8.5 Objective Function Mapping 217 8.6 Leverage Analysis 220
8.7 Bootstrap Procedures 222 8.8 Visual and Numerical Predictive Check
Procedures 223 8.8.1 The VPC Procedure 223 8.8.2 Presentation of VPC
Results 225 8.8.3 The Numerical Predictive Check (NPC) Procedure 229 8.9
Posterior Predictive Check Procedures 229 CHAPTER 9 User-Written Models 232
9.1 Introduction 232 9.2 $MODEL 235 9.3 $SUBROUTINES 236 9.3.1 General
Linear Models (ADVAN5 and ADVAN7) 236 9.3.2 General Nonlinear Models
(ADVAN6, ADVAN8, ADVAN9, and ADVAN13) 238 9.3.3 $DES 238 9.4 A Series of
Examples 240 9.4.1 Defined Fractions Absorbed by Zero- and First-Order
Processes 240 9.4.2 Sequential Absorption with First-Order Rates, without
Defined Fractions 242 9.4.3 Parallel Zero-Order and First-Order Absorption,
without Defined Fractions 243 9.4.4 Parallel First-Order Absorption
Processes, without Defined Fractions 245 9.4.5 Zero-Order Input into the
Depot Compartment 246 9.4.6 Parent and Metabolite Model: Differential
Equations 247 CHAPTER 10 PK/PD Models 250 10.1 Introduction 250 10.2
Implementation of PD Models in NONMEM 251 10.3 $PRED 252 10.3.1
Direct-Effect PK/PD Examples: PK Concentrations in the Dataset 253 10.3.2
Direct-Effect PK/PD Example: PK from Computed Concentrations 255 10.4 $PK
256 10.4.1 Specific ADVANs (ADVAN1-ADVAN4 and ADVAN10-ADVAN12) 256 10.4.2
General ADVANs (ADVAN5-ADVAN9 and ADVAN13) 257 10.4.3 PREDPP: Effect
Compartment Link Model Example (PD in $ERROR) 257 10.4.4 PREDPP: Indirect
Response Model Example: PD in $DES 259 10.5 Odd-Type Data: Analysis of
Noncontinuous Data 261 10.6 PD Model Complexity 262 10.7 Communication of
Results 263 CHAPTER 11 Simulation Basics 265 11.1 Introduction 265 11.2 The
Simulation Plan 265 11.2.1 Simulation Components 266 11.2.2 The
Input-Output Model 266 11.2.3 The Covariate Distribution Model 270 11.2.4
The Trial Execution Model 273 11.2.5 Replication of the Study 274 11.2.6
Analysis of the Simulated Data 275 11.2.7 Decision Making Using Simulations
275 11.3 Miscellaneous Other Simulation-Related Considerations 276 11.3.1
The Seed Value 276 11.3.2 Consideration of Parameter Uncertainty 277 11.3.3
Constraining Random Effects or Responses 278 CHAPTER 12 Quality Control 285
12.1 Introduction 285 12.2 QC of the Data Analysis Plan 285 12.3 Analysis
Dataset Creation 286 12.3.1 Exploratory Data Analysis and Its Role in
Dataset QC 287 12.3.2 QC in Data Collection 287 12.4 QC of Model
Development 288 12.4.1 QC of NM-TRAN Control Streams 289 12.4.2 Model
Diagnostic Plots and Model Evaluation Steps as QC 290 12.5 Documentation of
QC Efforts 290 12.6 Summary 291 References 292 Index 293