Biomarkers.

Area of Science:

• Neuroscience
• Clinical Pharmacology
• Biostatistics

Background:

• Alzheimer's disease (AD) and mild cognitive impairment (MCI) management is challenged by heterogeneous patient progression and treatment responses.
• Precision medicine requires identifying patient subgroups who benefit from specific therapies like Donepezil.
• Targeted treatment strategies are crucial for improving therapeutic outcomes in AD and MCI.

Purpose of the Study:

• To develop and apply a novel framework (ITree-MMRM) for identifying patient subgroups with differential long-term treatment effects.
• To pinpoint specific subgroups of MCI patients who exhibit significant long-term benefits from Donepezil treatment.
• To enhance personalized medicine approaches in managing cognitive decline.

Main Methods:

• The ITree-MMRM framework integrates the Interaction Tree (ITree) method with Mixed Model for Repeated Measures (MMRM).
• Applied to a 3-year clinical trial of Donepezil in 769 MCI patients, using Clinical Dementia Rating Scale Sum of Boxes (CDR-SOB) scores.
• Bootstrap pruning was used to refine subgroup identification and mitigate overoptimism.

Main Results:

• An MCI subgroup with lower baseline cognitive scores (MMSE ≤24) showed significant long-term benefits from Donepezil at 36 months.
• Patients with higher baseline cognitive scores demonstrated less benefit from the treatment.
• Simulation studies confirmed the ITree-MMRM framework's robustness and superiority over existing methods.

Conclusions:

• The ITree-MMRM framework successfully identified MCI subgroups with differential long-term responses to Donepezil.
• Findings support personalized medicine strategies for optimizing Donepezil treatment in MCI and AD.
• Further validation in larger clinical trials is warranted to confirm these targeted treatment insights.