Premium
From Trial and Error to Trial Simulation. Part 1: The Importance of Model‐Based Drug Development for Antidepressant Drugs
Author(s) -
Santen G,
Zwet E,
Danhof M,
Della Pasqua O
Publication year - 2009
Publication title -
clinical pharmacology and therapeutics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.941
H-Index - 188
eISSN - 1532-6535
pISSN - 0009-9236
DOI - 10.1038/clpt.2009.105
Subject(s) - clinical trial , dropout (neural networks) , computer science , type i and type ii errors , drug development , clinical pharmacology , medical physics , management science , psychology , econometrics , medicine , statistics , machine learning , drug , psychiatry , mathematics , pharmacology , engineering
Clinical trial simulation (CTS) allows the investigation of the influence of design characteristics on important aspects of clinical trials such as power and type I error. Simulation scenarios may be critical to decision making and prevention of study failure. The analysis and simulation of clinical trials in depression have, however, suffered from a lack of disease/dropout models that describe the individual time course of the clinical scale of interest. We propose a new model with dual random effects (DREM) derived from functional data analysis, which provides unbiased estimates of parameters and is suitable for the purposes of clinical trial simulation. A comparison of model performance is presented, along with standard statistical methods using various goodness‐of‐fit criteria. Our results show that data simulated using the DREM closely match individual patient data, including real‐life dropout scenarios. In addition, parameterization in terms of interindividual variability ensures easier explanation of findings to clinical scientists, who ultimately make the relevant decisions. Clinical Pharmacology & Therapeutics (2009); 86 3, 248–254. doi: 10.1038/clpt.2009.105