Estimation and testing in targeted goup sequential covariate-adjusted randomized clinical trials.

Area of Science:

• Biostatistics
• Clinical Trials Methodology
• Statistical Inference

Background:

• Adaptive clinical trials allow for pre-planned modifications during the trial.
• Group sequential methods enable interim analyses and early stopping.
• Targeted Maximum Likelihood Estimation (TMLE) offers a robust statistical framework.

Purpose of the Study:

• To develop and analyze adaptive group sequential covariate-adjusted randomized clinical trials.
• To integrate TMLE with adaptive sampling schemes for robust inference.
• To extend existing TMLE results to non-i.i.d. settings and incorporate group-sequential testing.

Main Methods:

• Construction of an adaptive sampling design that targets a user-specified optimal design.
• Application of semiparametric TMLE for statistical inference under adaptive sampling.
• Integration of group-sequential testing procedures with the TMLE framework.

Main Results:

• Demonstration of a robust procedure for adaptive group sequential trials, consistent even with model misspecification.
• Extension of TMLE inference to adaptive sampling schemes, moving beyond independent and identically distributed (i.i.d.) settings.
• Validation of theoretical results through simulation studies, supporting efficiency conjectures.

Conclusions:

• The developed methodology provides a robust and potentially efficient approach for adaptive group sequential clinical trials.
• The integration of TMLE with adaptive designs enhances statistical inference capabilities.
• This work extends the application of TMLE in complex clinical trial settings.