Sequential Parallel Comparison Design for Assessing Induction, Maintenance, Long-Term, and Other Treatment Effects on...

Area of Science:

• Clinical Trial Design
• Biostatistics
• Pharmacoeconomics

Background:

• Assessing long-term treatment efficacy in chronic diseases requires evaluating both induction and maintenance effects.
• Existing clinical trial designs, such as lead-in, re-randomized withdrawal, and treat-through, have limitations.
• Sequential parallel comparison designs (SPCD) are valuable, especially when placebo effects are substantial.

Purpose of the Study:

• To propose and evaluate a novel two-stage sequential parallel comparison design (SPCD) for clinical trials.
• To assess induction, maintenance, and long-term treatment effects using a binary endpoint.
• To offer a hybrid design that simplifies operations and improves data utilization compared to existing methods.

Main Methods:

• A two-stage sequential parallel comparison design (SPCD) is employed.
• A weighted combination test is utilized to integrate data from both stages for overall treatment effect evaluation.
• Multiple imputation is applied to handle missing data, specifically missing not at random (MNAR).

Main Results:

• The proposed SPCD simplifies trial operations by avoiding re-randomization.
• The design provides controlled data for robust long-term efficacy and safety analyses.
• Simulations demonstrate the performance of the proposed statistical methods.

Conclusions:

• The novel SPCD offers an efficient and effective approach for evaluating long-term treatment effects in chronic disease clinical trials.
• This design enhances data utilization and simplifies operational aspects compared to traditional methods.
• The statistical methods, including weighted combination tests and multiple imputation, are suitable for analyzing data from this SPCD.