DESIGN AND ANALYSIS OF BIOAVAILABILITY AND BIOEQUIVALENCE STUDIES

Transcription

1 DESIGN AND ANALYSIS OF BIOAVAILABILITY AND BIOEQUIVALENCE STUDIES SHEIN-CHUNG CHOW Biostatistics Department Bristol-Myers Squibb Company Plainsboro, New Jersey JEN-PEI LIU Biostatistics Department Berlex Laboratories, Inc. Wayne, New Jersey Marcel Dekker, Inc. New York Basel Hong Kong

2 Preface iii 1. Introduction History of Bioavailability Studies Formulations and Routes of Administration Pharmacokinetic Parameters Clinically Important Differences Assessment of Bioequivalence Decision Rules and Regulatory Aspects Statistical Considerations Aims and Structure of the Book Designs of Bioavailability Studies Introduction The Study Objective Basic Design Considerations The Parallel Design The Crossover Design The Balanced Incomplete Block Design The Selection of Design 46 vii

3 3. Statistical Inferences for Effects from a Standard 2 x 2 Crossover Design Introduction 3.2 The Carry-Over Effects The Direct Drug Effect The Period Effect The Analysis of Variance An Example Statistical Methods for Average Bioavailability Introduction The Confidence Interval Approach The Methods of Interval Hypotheses Testing Bayesian Methods Nonparametric Methods Discussion and Other Alternatives Power and Sample Size Determination Introduction Hypotheses and Type I and Type II Errors Power and Relative Efficiency Sample Size Determination Transformation and Analysis of Individual Subject Ratios Introduction Multiplicative (or Log-Transformed) Model Bioequivalence Measures Maximum Likelihood Estimator Minimum Variance Unbiased Estimator Mean of Individual Subject Ratios Ratio of Formulation Means Comparison of MLE, MVUE, MIR, and RM An Example Discussion The Assessment of Inter- and Intra-Subject Variabilities Introduction Variability and Decision-Making Point and Interval Estimates

4 ix 7.4 Tests for Equality of Variabilities Equivalence in Variability of Bioavailability CV Assessment Discussion Assumptions and Outliers Detection Introduction Tests for Assumptions The Definition of Outlying Observations Detection of Outlying Subjects Detection of Outlying Observations Discussion Optimal Crossover Designs for Two Formulations Introduction Balaam's Design The Two-Sequence Dual Design Optimal Four-Period Designs Transformation and Individual Subject Ratios Discussion Assessment of Bioequivalence for More than Two Formulations Introduction Assessment of Average Bioavailability with More than Two Formulations Analysis for Williams Designs Analysis for Balanced Incomplete Block Design Discussion Assessment of Bioequivalence for Drugs with Negligible Plasma Levels Introduction Design and Clinical Endpoints Statistical Considerations Weighted Least Squares Method Log-Linear Models Generalized Estimating Equations (GEEs) Analysis of Time to Onset of a Therapeutic Response Discussion 341

5 12. Some Related Problems in Bioavailability Studies References Introduction Drug Interaction Study Dose Proportionality Study Steady State Analysis Population Pharmacokinetics Appendix A: Statistical Tables Appendix A. 1 Appendix A.2 Appendix A.3 Appendix A.4 Appendix A.5 Appendix B: SAS Programs Index Appendix B.I Appendix B.2 Appendix B.3 Appendix B.4 Areas of Upper Tail of the Standard Normal Distribution Upper Quantiles of a x 2 Distribution Upper Quantiles of a Central t Distribution Upper Quantiles of an F Distribution Upper Quantiles of the Distribution of Wilcoxon-Mann-Whitney Statistic Procedures for Average Bioavailability Bayesian Method by Grieve MVUE, MLE, RI, MIR for a 2 X 2 Crossover Design Parametric and Nonparametric Versions of Pitman-Morgan Test