Multi-threshold polygenic risk improves hippocampal-based cognitive decline prediction

Area of Science:

• Neuroscience
• Genetics
• Medical Imaging

Background:

• Predicting cognitive decline using brain Magnetic Resonance Imaging (MRI) is crucial in neuroscience.
• Hippocampal volume (HV) is a key biomarker for cognitive function.
• Normative models can be enhanced with multimodal information for better prediction.

Purpose of the Study:

• To augment normative models with genetic information for improved cognitive decline prediction.
• To enhance the accuracy of prognostication models for neurodegenerative diseases.

Main Methods:

• Gaussian Process Regression (GPR) was used to integrate multi-threshold polygenic scores (PGS) with demographic and imaging data.
• Models were trained on 23,997 participants from UK Biobank (UKBB).
• Validation was performed on 3,000 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and European Prevention of Alzheimer's Disease (EPAD) cohorts.

Main Results:

• Genetically-informed models significantly strengthened associations across six experimental designs.
• Prediction accuracy was enhanced for 13 key neurocognitive measures, including Mini-Mental State Examination (MMSE), Clinical Dementia Rating (CDR), and Alzheimer's Disease Assessment Scale (ADAS).
• The models showed improved prediction of future cognitive decline.

Conclusions:

• Integrating multi-threshold PGS with neuroimaging-based predictive models shows promise.
• This approach can improve prognostication for neurodegenerative diseases.
• Findings support enhanced early intervention strategies for conditions like Alzheimer's disease.