Design and Analysis of Group Sequential Trials for Repeated Measurements When Pipeline Data Occurs: A Tutorial

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Longitudinal Data Analysis

Background:

• Group sequential trials (GST) offer early stopping for efficacy or futility in clinical research.
• Established GST methods primarily address immediate endpoints, posing challenges for delayed or repeated measures.
• Pipeline subjects with partially available data at interim analyses are common in delayed endpoint trials.

Purpose of the Study:

• To provide guidance on planning and analyzing group sequential trials with repeated measurements and delayed endpoints.
• To enhance statistical power by incorporating early outcome data from pipeline subjects.
• To develop and present methods for valid p-values and confidence intervals in such complex trial designs.

Main Methods:

• Review and expansion of existing statistical methodologies for group sequential trials.
• Development of methods to formally incorporate data from pipeline subjects at interim analyses.
• Inclusion of nonbinding futility stopping rules and detailed computational procedures for statistical inference.

Main Results:

• Demonstration of how accounting for early measurements increases statistical power in GST.
• Methods presented for the formal incorporation of all available data, including late-emerging outcomes.
• Development of computational tools for valid statistical analysis in complex GST designs.

Conclusions:

• Effective planning and analysis of GST with delayed endpoints and repeated measures are crucial for maximizing data utility.
• The proposed methods and R package facilitate robust statistical inference, enhancing the efficiency of clinical trials.
• This work addresses a critical gap in GST methodology, particularly for trials with longitudinal outcomes and delayed assessments.