Big Data Mining and Adverse Event Pattern Analysis in Clinical Drug Trials

Area of Science:

• Pharmacovigilance
• Clinical Informatics
• Drug Discovery

Background:

• Drug adverse events (AEs) pose significant risks to patient safety and hinder drug development.
• Existing AE databases primarily focus on U.S. Food and Drug Administration (FDA)-approved drugs.
• ClinicalTrials.gov provides a rich source of AE data from global clinical studies.

Purpose of the Study:

• To create a comprehensive database of drug-AE relationships by extracting data from ClinicalTrials.gov.
• To include both FDA-approved and experimental compounds, expanding beyond current AE databases.
• To facilitate drug development, repositioning, and repurposing through data mining and pattern analysis.

Main Methods:

• Utilized Python scripts to extract drug and AE information from ClinicalTrials.gov.
• Employed regular expressions and a drug dictionary for data processing and structuring.
• Developed a relational database to store and analyze drug-AE relationships.
• Conducted data mining and pattern analysis on the compiled dataset.

Main Results:

• Compiled a database encompassing 8,161 clinical trials, 3,102,675 patients, and 713,103 reported AEs.
• The database includes data on both FDA-approved and experimental drug compounds.
• Identified drug-AE relationships through data mining and pattern analysis.

Conclusions:

• The developed database serves as a valuable tool for researchers.
• It aids in the discovery of novel drug-AE associations.
• Supports the advancement of drug development, repositioning, and repurposing strategies.