Fundamentals of Clinical Research

1. Variability of Biological Phenomena and Measurement Errors 1.1. Phenotypic Variability 1.2. Temporal Variability. 1.3. Measurement - Related Variability 1.3.1. The Measurement 1.3.2. Measurement Errors 1.4. Variability of Diagnostic Tests Summary 2. Distinctive Aspects of a Biomedical Study. Observational and Experimental Studies 2.1. Distinctive Features of Biomedical Studies 2.2. The Study Protocol 2.3. Observational Studies 2.4. Experimental Studies Summary 3. Observational Studies 3.1. Basic Designs of Observational Studies 3.1.1. Prospective or Cohort Studies 3.1.2. Retrospective Studies or Control Cases 3.1.3. Sample Size 3.2. Bias and Confounding 3.2.1. Control of Bias in Epidemiology 3.2.2. Control of the Phenomenon of Confounding | 3.3. Advantages and Disadvantages of the Different Types of Observational Studies Summary 4 Defining the Treatment Effect 4.1. From the Single Measurement to the Signal 4.2. Identification and Quantification of the End-Points (Individual Subject Level) 4.2.1. Methodological Characteristics of the End-Point 4.2.2 Discriminating Between Primary and Secondary End-Points and Between Efficacy and Safety/Tolerability End-Points 4.3. Identification and Quantification of the Signals (Group Level) 4.4. Statistical Considerations 4.5. Practical, Regulatory, Marketing and Pharmaco-Economic Considerations 4.6. Selection and Characterization of the Primary End-Point and Signal: an Example 4.6.1. Stage One: Define the Main Therapeutic Level 4.6.2. Stage Two: Define the Primary End-Point (Individual Patient Level) 4.6.3. Stages Three and Four: Define the Group Indicator, the Signal, and the Threshold of Clinical Relevance (Treatment Group and Study Levels) 4.7. More Than One Question in the Same Study: the Problem of Multiple Statistical Tests 4.8. Validation of Measurement Scales 4.9. Special Types of End-Points Summary 5. Probability, Inference and Decision Making 5.1. Probability 5.1.1 Definitions 5.1.2. Probability Distribution and Probability Density Function 5.1.3. Normal or Gaussian Distribution 5.2. Basic Concepts of Inference 5.2.1. Hypothesis Testing and Statistical Formulation of the Medical Question 5.2.2. Statistical Estimation as the Tool for Evaluation of Clinical Relevance 5.3. Statistical Inference in the Frequentist and the Bayesian Approaches 5.4. Two Digressions: Measures of Variability and Likelihood Function 5.4.1. Measures of Variability 5.4.2. Likelihood Function 5.5. Frequentist (Classical) Analysis of a Clinical Trial 5.5.1. Hypothesis Testing: the Frequentist Solution 5.5.2. Estimation of the Effect: the Frequentist Solution 5.6. Bayesian Analysis of a Clinical Trial 5.6.1. Hypothesis Testing: the Bayesian Solution 5.6.2. Estimation of the Effect: the Bayesian Solution 5.7. Some Additional Considerations on the Frequentist and Bayesian Approaches 5.8. Parametric and Non-Parametric Inference 5.9. Statistical Decision Making in the Medical Field 5.10. Evidence-Based Medicine Summary 6. The Choice of the Sample 6.1. Which Subjects Should Form the Sample? 6.1.1. Characteristics of the Patients to be Enrolled in the Study 6.1.2. Mechanism of Subject Selection 6.2. How Many Subjects Should Form the Sample? 6.2.1. Statistical Considerations 6.2.2. Medical and Practical Aspects Summary 7. The Choice of Treatments 7.1. Study Treatments 7.1.1. How Many Treatments 7.1.2. What Treatments 7.1.3. Blinding of the Study Treatments 7.1.4. Packaging and Logistics 7.2. Concomitant Treatments Summary 8. Experimental Design: Fallacy of "Before-After" Comparisons in Uncontrolled Studies 8.1. Experimental Design: Introductory Concepts 8.2. Before-After Comparison in a Single Group of Subjects 8.3. Temporal Variations of the Disease 8.4. Temporal Variations of Staff, Equipment and Environment 8.5 Statistical Regression Toward the Mean 8.5.1. The Basic Principle 8.5.2. Areas of Biomedical Experiments Affected by Regression Toward the Mean 8.5.3. How to Minimize the Effect of Regression Toward the Mean 8.6. Learning Effect 8.7. Psychological Effect 8.8. The Before-After Design Without Control Group in Oncology Summary 9. Experimental Design: the Randomized Blinded Study as an Instrument to Reduce Bias 9.1. Introduction 9.2. Randomization as Antidote Against Selection Bias 9.2.1. Definition and Conceptual Framework 9.2.2. Types of Randomization 9.2.3. Other Methods for Assigning Patients to Treatments 9.3. Blinding of Treatments as an Antidote Against Assessment Bias 9.4. A Priori Definition of the Statistical Methods and Populations as Antidote Against the Analysis Bias 9.4.1. Methods of Statistical Analysis 9.4.2. Analysis Populations 9.5. Comparison Between an Observational and an Experimental Study Summary 10. Experimental Designs 10.1. Introduction 10.2. Parallel Group Design 10.2.1. Characteristics 10.2.2. Advantages and Disadvantages 10.2.3. Conditions of Applicability 10.3. Variants of the Parallel Group Design 10.3.1. Completely Randomized Parallel Group Design 10.3.2. Stratified Parallel Group Design 10.3.3. Parallel Group Randomized Block Design 10.3.4. Balanced Incomplete Block Design 10.4. Other Designs with Comparison Between Subjects: Dose-Escalation and Dose-Titration 10.4.1. Dose-Escalation Design 10.4.2. Dose-Titration Design 10.5. Complete Cross-Over Design 10.5.1. Characteristics 10.5.2. Advantages and Disadvantages 10.5.3. Conditions of Applicability 10.6. Variants of the Cross-Over Design 10.6.1.Variants Based on the Type of Randomization 10.6.2. Incomplete Cross-Over Designs 10.7. Other Designs with Within-Subject Comparisons: Simultaneous Treatments and Single Patient Designs 10.7.1. Simultaneous Treatments Design 10.7.2. Cross-Over Design on a Single Patient (or "N of 1" Design) 10.8. Factorial Designs 10.8.1. Characteristics 10.8.2. Advantages and Disadvantages 10.8.3. Conditions of Applicability 10.9. Split-Plot Design 10.9.1. Characteristics 10.9.2. Conditions of Applicability 10.10. Non-Controlled Designs in Phase II Oncology Studies Summary 11. Study Variants Applicable to More than One Type of Design: Equivalence Studies, Interim Analyses, Adaptive Plans and Repeated Measurements 11.1. Equivalence and Non-Inferiority Studies 11.1.1. Characteristics 11.1.2. The Statistical Analysis of an Equivalence Study 11.1.3. Planning and Implementation Problems 11.1.4. Analysis and Interpretation Problems 11.2. Studies With Interim Analyses and Sequential Designs 11.2.1. Definitions and Classification 11.2.2. Conditions of Applicability 11.2.3. Choice of the End-Points 11.2.4. Data Management Issues 11.2.5. Statistical Issues and Decision Making Criteria 11.2.6. Conflict of Interest and Confidentiality Issues 11.3. Adaptive (Flexible) Designs 11.4. Studies With Repeated Measurements Summary 12.The Drug Development Process and the Phases of Clinical sx Research 12.1. Overview of the Preclinical Development Process 12.2. The Phases of Clinical Development 12.2.1. Introduction 12.2.2. Phase I 12.2.3. Phase II 12.2.4. Phase III 12.2.5. Registration Dossier 12.2.6. Phase IV 12.2.7. Project Management 12.3. The Phases of Clinical Development for Oncology Compounds 12.3.1. Phase I 12.3.2. Phase II 12.3.3. Phase III 12.4. Accelerating Clinical Development Summary Appendix: Areas Under the Curve of the Standard Normal Distribution References Analytical Index