Recursive partitioning of resistant mutations for longitudinal markers based on a U-type score

Area of Science:

• Virology
• Biostatistics
• Genetics

Background:

• Antiretroviral treatment failure in human immunodeficiency virus (HIV) infections is often caused by the development of drug resistance mutations.
• Analyzing high-dimensional viral sequence data alongside longitudinal patient outcomes presents a significant statistical challenge.

Purpose of the Study:

• To develop and evaluate a novel recursive partitioning method for correlating high-dimensional viral genetic sequences with repeatedly measured clinical outcomes.
• To assess the utility of this method in the context of antiretroviral therapy, specifically for the drug efavirenz.

Main Methods:

• A recursive partitioning approach was developed, utilizing U-type score statistics as the splitting criterion.
• The method is designed to handle high-dimensional data and longitudinal outcomes.
• Simulation studies were conducted to evaluate the method's finite-sample properties.

Main Results:

• The proposed recursive partitioning method demonstrated flexibility in analyzing longitudinal data.
• The method was successfully applied to real-world data from three phase II clinical trials involving efavirenz.
• The approach effectively correlated viral sequence variations with treatment outcomes.

Conclusions:

• The developed recursive partitioning method offers a robust tool for analyzing complex viral sequence data in relation to treatment outcomes.
• This statistical approach can enhance the understanding of antiretroviral treatment failure and guide therapeutic strategies.
• The method's applicability extends to various problems involving longitudinal data analysis in clinical research.