Statistical methods for handling missing data to align with treatment policy strategy

Area of Science:

• Clinical Trials Methodology
• Statistical Analysis in Pharmaceuticals
• Pharmacometrics

Background:

• The International Council for Harmonization (ICH) E9(R1) addendum emphasizes pre-trial estimand selection based on study objectives.
• Intercurrent events, such as treatment modifications, are critical considerations in defining estimands for clinical trials.
• The treatment policy strategy is commonly used to maintain the integrity of the planned treatment regimen analysis.

Purpose of the Study:

• To explain methods for handling missing data following intercurrent events in antihyperglycemic product development.
• To evaluate five distinct statistical imputation methods within the treatment policy framework.
• To demonstrate the application of these methods in real-world antihyperglycemic drug development.

Main Methods:

• Application of five statistical imputation methods for missing data after intercurrent events.
• Utilizing the treatment policy strategy for data analysis.
• Comparison of imputation methods through Markov Chain Monte Carlo (MCMC) simulations.

Main Results:

• Five statistical methods were applied to impute missing data under the treatment policy strategy.
• Markov Chain Monte Carlo simulations were employed to compare the performance of these methods.
• Three of the evaluated methods were successfully applied to estimate treatment effects for marketed antihyperglycemic agents.

Conclusions:

• The treatment policy strategy provides a robust framework for handling intercurrent events in clinical trials.
• The choice of imputation method can significantly impact the estimation of treatment effects.
• The study validates the practical application of specific imputation techniques in antihyperglycemic drug development.