Missing data in clinical trials for weight management

Area of Science:

• Pharmacology
• Biostatistics
• Clinical Trial Design

Background:

• The US FDA approved liraglutide for weight management in 2014.
• Statistical review of liraglutide's application highlighted significant challenges in handling missing data.
• The drug's efficacy for weight loss was established, but estimating the intention-to-treat effect was complex.

Purpose of the Study:

• To illustrate challenges and considerations in handling missing data within clinical trials.
• To analyze the impact of missing data on risk-benefit evaluations for weight management drugs.
• To demonstrate how participant adherence to primary endpoint visits influences statistical review.

Main Methods:

• Review of statistical methodologies used in the liraglutide weight management application.
• Analysis of data from subjects who completed primary endpoint visits despite early treatment cessation.
• Examination of applicant's primary efficacy and sensitivity analyses, and FDA's approaches to missing data.

Main Results:

• Missing data handling presented difficulties in obtaining reliable intention-to-treat effect estimates.
• Data from returning participants were crucial for identifying limitations in the applicant's analyses.
• FDA's analyses utilized this data to address missing data concerns.

Conclusions:

• Effective handling of missing data is critical for robust clinical trial outcomes and regulatory decision-making.
• Participant engagement in endpoint visits is vital for comprehensive data collection and analysis.
• The case of liraglutide underscores the importance of rigorous statistical approaches in drug approval processes.