New Search Search History

Holdings Information

Action Navigation

Action Box More Info Navigation

    Action Box

    Controversial statistical issues in clinical trials / Shein-Chung Chow.

    Bibliographic Record Display

    • Title:Controversial statistical issues in clinical trials / Shein-Chung Chow.
    •    
    • Author/Creator:Chow, Shein-Chung, 1955-
    • Published/Created:Boca Raton, FL : CRC Press, ©2011.

    • Holdings

      Holdings Record Display

       
    • Library of Congress Subjects:Drugs--Testing.
      Drug development.
      Clinical trial.
    • Medical Subjects: Clinical Trials as Topic.
      Drug Evaluation, Preclinical.
      Data Interpretation, Statistical.
    • Description:xix, 591 p. : ill ; 25 cm.
    • Series:Chapman & Hall/CRC biostatistics series (Unnumbered)
    • Summary:"Preface In pharmaceutical/clinical development of a test drug or treatment, relevant clinical data are usually collected from subjects with the diseases under study in order to evaluate safety and efficacy of the test drug or treatment under investigation. To provide accurate and reliable assessment, well-controlled clinical trials under valid study design are necessarily conducted. Clinical trial process is a lengthy and costly process, which is necessary to ensure a fair and reliable assessment of the test treatment under investigation. Clinical trial process consists of protocol development, trial conduct, data collection, statistical analysis/interpretation, and reporting. In practice, controversial issues evitably occur regardless the compliance of good statistical practice (GSP) and good clinical practice (GCP). Controversial issues in clinical trials are referred to as debatable issues that are commonly encountered during the conduct of clinical trials. In practice, controversial issues could be raised from, but are not limited to, (1) compromises between theoretical and real/common practices, (2) miscommunication and/or misunderstanding in perception/interpretation among regulatory agencies, clinical scientists, and biostatisticians, and (3) disagreement, inconsistency, miscommunication/misunderstanding, and errors in clinical practice. "--Provided by publisher.
    • Notes:Includes bibliographical references and index.
    • ISBN:9781439849613 (hardback : alk. paper)
      1439849617 (hardback : alk. paper)
    • Contents:Machine generated contents note: 1. Introduction
      1.1. Introduction
      1.2. Pharmaceutical Development
      1.2.1. Nonclinical Development
      1.2.2. Preclinical Development
      1.2.3. Clinical Development
      1.3. Controversial Issues
      1.4. Aim and Structure of the Book
      2. Good Statistical Practices
      2.1. Introduction
      2.2. Statistical Principles
      2.2.1. Bias and Variability
      2.2.2. Confounding and Interaction
      2.2.3. Hypotheses Testing
      2.2.4. Type I Error and Power
      2.2.5. Randomization
      2.2.6. Sample Size Determination/Justification
      2.2.7. Statistical Difference and Scientific Difference
      2.2.8. One-Sided Test versus Two-Sided Test
      2.3. Good Statistical Practices in Europe
      2.4. Implementation of GSP
      2.5. Concluding Remarks
      3. Bench-to-Bedside Translational Research
      3.1. Introduction
      3.2. Biomarker Development
      3.2.1. Optimal Variable Screening
      3.2.2. Model Selection and Validation
      3.2.3. Remarks
      3.3. One-Way/Two-Way Translational Process
      3.3.1. One-Way Translational Process
      3.3.2. Two-Way Translational Process
      3.4. Lost in Translation
      3.5. Animal Model versus Human Model
      3.6. Concluding Remarks
      4. Bioavailability and Bioequivalence
      4.1. Introduction
      4.2. Bioequivalence Assessment
      4.2.1. Study Design
      4.2.2. Statistical Methods
      4.2.3. Remarks
      4.3. Drug Interchangeability
      4.3.1. Drug Prescribability and Drug Switchability
      4.3.2. Population and Individual Bioequivalence
      4.4. Controversial Issues
      4.4.1. Fundamental Bioequivalence Assumption
      4.4.2. One-Fits-All Criterion
      4.4.3. Issues Related to Log Transformation
      4.5. Frequently Asked Questions
      4.5.1. What If We Pass Raw Data Model but Fail Log-Transformed Data Model?
      4.5.2. What If We Pass AUC but Fail Cmax?
      4.5.3. What If We Fail by a Relatively Small Margin?
      4.5.4. Can We Still Assess Bioequivalence If There Is a Significant Sequence Effect?
      4.5.5. What Should We Do When We Have Almost Identical Means but Still Fail to Meet the Bioequivalence Criterion?
      4.5.6. Power and Sample Size Calculation Based on Raw Data Model and Log-Transformed Model Are Different
      4.5.7. Adjustment for Multiplicity
      4.6. Concluding Remarks
      5. Hypotheses for Clinical Evaluation and Significant Digits
      5.1. Introduction
      5.2. Hypotheses for Clinical Evaluation
      5.3. Statistical Methods for Testing Composite Hypotheses of NS
      5.4. Impact on Power and Sample Size Calculation
      5.4.1. Fixed Power Approach
      5.4.2. Fixed Sample Size Approach
      5.4.3. Remarks
      5.5. Significant Digits
      5.5.1. Chow's Proposal
      5.5.2. Statistical Justification
      5.6. Concluding Remarks
      6. Instability of Sample Size Calculation
      6.1. Introduction
      6.2. Sample Size Calculation
      6.3. Instability and Bootstrap-Median Approach
      6.3.1. Instability of Sample Size Calculation
      6.3.2. Bootstrap-Median Approach
      6.4. Simulation Study
      6.4.1. One-Sample Problem
      6.4.2. Two-Sample Problem
      6.5. Example
      6.6. Concluding Remarks
      7. Integrity of Randomization/Blinding
      7.1. Introduction
      7.2. Effect of Mix-Up Randomization
      7.3. Blocking Size in Randomization
      7.3.1. Probability of Correctly Guessing
      7.3.2. Numerical Study
      7.3.3. Remarks
      7.4. Test for Integrity of Blinding
      7.5. Analysis under Breached Blindness
      7.6. Example
      7.7. Concluding Remarks
      8. Clinical Strategy for Endpoint Selection
      8.1. Introduction
      8.2. Clinical Strategy for Endpoint Selection
      8.3. Translations among Clinical Endpoints
      8.4. Comparison of Different Clinical Strategies
      8.4.1. Test Statistics, Power, and Sample Size Determination
      8.4.2. Determination of the Non-Inferiority Margin
      8.5. Numerical Study
      8.5.1. Absolute Difference versus Relative Difference
      8.5.2. Responders' Rate Based on Absolute Difference
      8.5.3. Responders' Rate Based on Relative Difference
      8.6. Concluding Remarks
      9. Protocol Amendments
      9.1. Introduction
      9.2. Moving Target Patient Population
      9.3. Analysis with Covariate Adjustment
      9.3.1. Continuous Study Endpoint
      9.3.2. Binary Response
      9.4. Assessment of Sensitivity Index
      9.4.1. Case Where ε Is Random and C Is Fixed
      9.4.2. Case Where ε Is Fixed and C Is Random
      9.5. Sample Size Adjustment
      9.6. Concluding Remarks
      10. Seamless Adaptive Trial Designs
      10.1. Introduction
      10.2. Controversial Issues
      10.2.1. Flexibility and Efficiency
      10.2.2. Validity and Integrity
      10.2.3. Regulatory Concerns
      10.3. Types of Two-Stage Seamless Adaptive Designs
      10.4. Analysis for Seamless Design with Same Study Objectives/Endpoints
      10.4.1. Theoretical Framework
      10.4.2. Two-Stage Adaptive Design
      10.4.3. Conditional Power
      10.5. Analysis for Seamless Design with Different Endpoints
      10.6. Analysis for Seamless Design with Different Objectives/Endpoints
      10.6.1. Nonadaptive Version
      10.6.2. Adaptive Version
      10.6.3. Example
      10.7. Concluding Remarks
      11. Multiplicity in Clinical Trials
      11.1. General Concept
      11.2. Regulatory Perspective and Controversial Issues
      11.2.1. Regulatory Perspectives
      11.2.2. Controversial Issues
      11.3. Statistical Method for Adjustment of Multiplicity
      11.3.1. Bonferroni's Method
      11.3.2. Tukey's Multiple Range Testing Procedure
      11.3.3. Dunnett's Test
      11.3.4. Closed Testing Procedure
      11.3.5. Other Tests
      11.4. Gatekeeping Procedures
      11.4.1. Multiple Endpoints
      11.4.2. Gatekeeping Testing Procedures
      11.5. Concluding Remarks
      12. Independence of Data Monitoring Committee
      12.1. Introduction
      12.2. Regulatory Requirements
      12.2.1. Determining Need for a DMC
      12.2.2. Confidentiality of Interim Data and Analysis
      12.2.3. Desirability of an Independent DMC
      12.3. DMC Composition and Charter
      12.3.1. DMC Composition and Support Staff
      12.3.2. DMC Charter
      12.4. DMC's Functions and Activities
      12.4.1. Randomization
      12.4.2. Critical Data Flow
      12.4.3. DMC Report and Analysis Plan
      12.4.4. Sensitivity Analysis
      12.4.5. Executive Summary/Report
      12.4.6. DMC Meetings
      12.4.7. DMC Documents and Information Dissemination
      12.4.8. DMC Recommendations
      12.4.9. DMC Organizational Flow
      12.5. Independence of DMC
      12.5.1. Some Observations
      12.5.2. Controversial Issues
      12.6. Concluding Remarks
      13. Two-Way ANOVA versus One-Way ANOVA with Repeated Measures
      13.1. Introduction
      13.2. One-Way ANOVA with Repeated Measures
      13.3. Two-Way ANOVA
      13.4. Statistical Evaluation
      13.5. Simulation Study
      13.6. Example
      13.7. Discussion
      14. Validation of QOL Instruments
      14.1. Introduction
      14.2. QOL Assessment
      14.3. Performance Characteristics
      14.3.1. Validity
      14.3.2. Reliability
      14.3.3. Reproducibility
      14.4. Responsiveness and Sensitivity
      14.4.1. Statistical Model
      14.4.2. Precision Index
      14.4.3. Power Index
      14.4.4. Sample Size Determination
      14.5. Utility Analysis and Calibration
      14.5.1. Utility Analysis
      14.5.2. Calibration
      14.6. Analysis of Parallel Questionnaire
      14.7. Example
      14.8. Concluding Remarks
      15. Missing Data Imputation
      15.1. Introduction
      15.2. Last Observation Carry Forward
      15.2.1. Bias-Variance Trade-Off
      15.2.2. Hypothesis Testing
      15.3. Mean/Median Imputation
      15.4. Regression Imputation
      15.5. Marginal/Conditional Imputation for Contingency
      15.5.1. Simple Random Sampling
      15.5.2. Goodness-of-Fit Test
      15.6. Testing for Independence
      15.6.1. Results under Stratified Simple Random Sampling
      15.6.2. When Number of Strata Is Large
      15.7. Controversial Issues
      15.8. Recent Development
      15.9. Concluding Remarks
      16. Center Grouping
      16.1. Introduction
      16.2. Selection of the Number of Centers
      16.3. Impact of Treatment Imbalance on Power
      16.4. Center Grouping
      16.5. Procedure for Center Grouping
      16.6. Example
      17. Non-Inferiority Margin
      17.1. Introduction
      17.2. Non-Inferiority Margin
      17.3. Statistical Test Based on Treatment Difference
      17.3.1. Tests Based on Historical Data under Constancy Condition
      17.3.2. Constancy Condition
      17.3.3. Tests without Historical Data
      17.3.4. Example
      17.4. Statistical Tests Based on Relative Risk
      17.4.1. Hypotheses for Non-Inferiority Margin
      17.4.2. Tests Based on Historical Data under Constancy Condition
      17.4.3. Tests without Historical Data
      17.4.4. Example
      17.5. Mixed Non-Inferiority Margin
      17.5.1. Hypotheses for Mixed Non-Inferiority Margin
      17.5.2. Non-Inferiority Tests
      17.5.3. Example
      17.6. Recent Developments
      17.6.1. Special Issue of the Journal of Biopharmaceutical Statistics
      17.6.2. FDA Draft Guidance
      17.7. Concluding Remarks
      18. QT Studies with Recording Replicates
      18.1. Introduction
      18.2. Study Designs and Models
      18.3. Power and Sample Size Calculation
      18.3.1. Parallel-Group Design
      18.3.2. Crossover Design
      18.3.3. Remarks
      18.4. Adjustment for Covariates
      18.4.1. Parallel-Group Design
      18.4.2. Crossover Design
      18.5. Optimization for Sample Size Allocation
      18.6. Test for QT/QTc Prolongation
      18.6.1. Parallel-Group Design
      Contents note continued: 18.6.2. Crossover Design
      18.6.3. Numerical Study
      18.7. Recent Developments
      18.8. Concluding Remarks
      19. Multiregional Clinical Trials
      19.1. Introduction
      19.2. Multiregional (Multinational), Multicenter Trials
      19.2.1. Multicenter Trials
      19.2.2. Multiregional (Multinational), Multicenter Trials
      19.3. Selection of the Number of Sites
      19.3.1. Two-Stage Sampling
      19.3.2. Testing Procedure
      19.3.3. Optimal Selection
      19.3.4. Example
      19.4. Sample Size Calculation and Allocation
      19.4.1. Some Background
      19.4.2. Proposals of Statistical Guidance
      -Asian Perspective
      19.5. Statistical Methods for Bridging Studies
      19.5.1. Test for Consistency
      19.5.2. Test for Reproducibility and Generalizability
      19.5.3. Test for Similarity
      19.6. Concluding Remarks
      20. Dose Escalation Trials
      20.1. Introduction
      20.2. Traditional Escalation Rule
      20.3. Continual Reassessment Method
      20.3.1. Implementation of CRM
      20.3.2. CRM in Conjunction with Bayesian Approach
      20.3.3. Extended CRM Trial Design
      20.4. Design Selection and Sample Size
      20.4.1. Criteria for Design Selection
      20.4.2. Sample Size Justification
      20.5. Concluding Remarks
      21. Enrichment Process in Target Clinical Trials
      21.1. Introduction
      21.2. Identification of Differentially Expressed Genes
      21.3. Optimal Representation of in Vitro Diagnostic Multivariate Index Assays
      21.4. Validation of in Vitro Diagnostic Multivariate Index Assays
      21.5. Enrichment Process
      21.6. Study Designs of Target Clinical Trials
      21.7. Analysis of Target Clinical Trials
      21.8. Discussion
      22. Clinical Trial Simulation
      22.1. Introduction
      22.2. Process for Clinical Trial Simulation
      22.2.1. Model and Assumptions
      22.2.2. Performance Characteristics
      22.2.3. Example
      22.2.4. Remarks
      22.3. EM Algorithm
      22.3.1. General Description
      22.3.2. Example
      22.4. Resampling Method: Bootstrapping
      22.4.1. General Description
      22.4.2. Types of Bootstrap Scheme
      22.4.3. Methods for Bootstrap Confidence Intervals
      22.5. Clinical Applications
      22.5.1. Target Clinical Trials with Enrichment Designs
      22.5.2. Dose Escalation Trials
      22.6. Concluding Remarks
      23. Traditional Chinese Medicine
      23.1. Introduction
      23.2. Fundamental Differences
      23.2.1. Medical Theory/Mechanism and Practice
      23.2.2. Medical Practice
      23.2.3. Techniques of Diagnosis
      23.2.4. Treatment
      23.2.5. Remarks
      23.3. Basic Considerations
      23.3.1. Study Design
      23.3.2. Validation of Quantitative Instrument
      23.3.3. Clinical Endpoint
      23.3.4. Matching Placebo
      23.3.5. Sample Size Calculation
      23.4. Controversial Issues
      23.4.1. Test for Consistency
      23.4.2. Animal Studies
      23.4.3. Stability Analysis
      23.4.4. Regulatory Requirements
      23.4.5. Indication and Label
      23.5. Recent Development
      23.5.1. Statistical Quality Control Method for Assessing Consistency
      23.5.2. Stability Analysis for TCM
      23.5.3. Calibration of Study Endpoints
      23.6. Concluding Remarks
      24. Assessment of Follow-On Biologic Products
      24.1. Introduction
      24.2. Regulatory Requirements
      24.3. Criteria for Biosimilarity
      24.3.1. Absolute Change versus Relative Change
      24.3.2. Aggregated versus Disaggregated
      24.3.3. Moment-Based Criteria versus Probability-Based Criteria
      24.3.4. Similarity Factor for Dissolution Profile Comparison
      24.3.5. Consistency in Manufacturing Process/Quality Control
      24.4. Scientific Issues
      24.4.1. Biosimilarity in Biological Activity
      24.4.2. Similarity in Size and Structure
      24.4.3. Problem of Immunogenicity
      24.4.4. Manufacturing Process
      24.4.5. Statistical Considerations
      24.5. Assessing Similarity Using Genomic Data
      24.6. Concluding Remarks
      25. Generalizability/Reproducibility Probability
      25.1. Introduction
      25.2. Estimated Power Approach
      25.2.1. Two Samples with Equal Variances
      25.2.2. Two Samples with Unequal Variances
      25.2.3. Parallel-Group Designs
      25.3. Confidence Bound Approach
      25.4. Bayesian Approach
      25.5. Applications
      25.5.1. Substantial Evidence with a Single Trial
      25.5.2. Sample Size Adjustments
      25.5.3. Generalizability between Patient Populations
      25.6. Concluding Remarks
      26. Good Review Practices
      26.1. Introduction
      26.2. Regulatory Process and Requirements
      26.2.1. Investigational New Drug Application
      26.2.2. New Drug Application
      26.3. Good Review Practices
      26.3.1. Fundamental Values
      26.3.2. Implementation of GRP
      26.3.3. Remarks
      26.4. Obstacles and Challenges
      26.4.1. No Gold Standards for Evaluation of Clinical Data
      26.4.2. One-Fits-All Criterion for Bioequivalence Trials
      26.4.3. Bayesian Statistics in Drug Evaluation
      26.4.4. Adaptive Design Methods in Clinical Trials
      26.5. Concluding Remarks
      27. Probability of Success
      27.1. Introduction
      27.2. Go/No-Go Decision in Development Process
      27.2.1. Simple Approach for Decision Making
      27.2.2. Decision-Tree Approach
      27.2.3. Example
      27.3. POS Assessment
      27.4. Concluding Remarks.
    Session Timeout
    New Session