New Multi-Ancestry Genetic Score Improves Alzheimer’s Risk Prediction for Diverse Populations
A new multiancestry polygenic risk score for Alzheimer’s disease successfully predicts cognitive decline and neuropathological markers across diverse populations, addressing a long-standing demographic imbalance in genetic screening tools. Developed by a team of researchers including Dr. Lindsay Farrer and Dr. Xiaoling Zhang at Boston University, the expanded model was tested on tens of thousands of individuals to improve risk assessment for non-white patients.
- A newly developed multi-ancestry polygenic risk score (PRS) significantly improves Alzheimer’s disease risk prediction for African American, Hispanic, and East Asian populations compared to legacy tools built entirely on European ancestry data.
- The study analyzed summarized information from genetic markers across a multi-ancestry group comprising more than 63,000 Alzheimer’s cases and 484,000 controls.
- The validated PRS correlates strongly with real clinical markers, including memory decline, reduced hippocampal volume on brain MRIs, and abnormal cerebrospinal fluid levels of amyloid-beta and phosphorylated Tau (pTau).
The Clinical Gap in Legacy Polygenic Risk Scores
Late-onset Alzheimer’s disease affects an estimated 6.9 million people in the United States, presenting as a progressive neurodegenerative disorder marked by memory impairment and cognitive deterioration. While the APOE epsilon 4 allele remains the most prominent genetic risk factor, genome-wide association studies have identified numerous other variants influencing susceptibility. Historically, these insights relied heavily on data from individuals of European descent, rendering earlier polygenic risk scores inconsistent and often ineffective when applied to genetically diverse patient cohorts.
According to co-corresponding author Dr. Lindsay A. Farrer, chief of biomedical genetics at the Boston University Chobanian & Avedisian School of Medicine, the underrepresentation of diverse genetic ancestries in genome-wide association study datasets has created a critical challenge in clinical risk application. By incorporating genetic data from African American, Hispanic, and East Asian populations, researchers have enhanced the transferability and overall accuracy of predictive modeling.
Methodology and Validation Across Diverse Cohorts
The research team utilized summary statistics derived from a multi-ancestry group comprising more than 63,000 Alzheimer’s disease cases and 484,000 age-matched controls. This expanded framework allowed investigators to construct a robust polygenic risk score capable of crossing ancestral boundaries. The model underwent initial testing in an independent, ancestrally diverse sample of 10,612 Alzheimer’s cases and 16,625 elderly controls. Subsequent validation in a mixed-ancestry cohort of 1,500 cases and 75,500 elderly controls confirmed its enhanced predictive capacity.
When compared directly against older risk scores built solely on European ancestry data, the multi-ancestry polygenic risk score demonstrated superior performance in identifying true clinical diagnoses across all evaluated populations. Investigators evaluated these associations using data from multiple major initiatives, including the Alzheimer’s Disease Sequencing Project, the Framingham Heart Study, the Alzheimer’s Disease Neuroimaging Initiative, and the Korean Brain Aging Study for the Early Diagnosis and Prediction of Alzheimer’s Disease.
Neuropathological Biomarkers and Clinical Trajectories
Beyond clinical diagnosis, the research team analyzed how the new polygenic risk score intersects with tangible brain pathology. The score demonstrated significant associations with poorer performance in memory, executive function, and language domains. Brain magnetic resonance imaging revealed that individuals with higher risk scores exhibited a reduced volume in the hippocampus, the brain structure most vulnerable during the early pathogenesis of Alzheimer’s disease.
Analysis of cerebrospinal fluid identified correlations with abnormal levels of hallmark proteins, including amyloid-beta and phosphorylated Tau, with a notable elevation in pTau deviation observed among female participants. Longitudinal evaluations further established that individuals carrying a very high multi-ancestry polygenic risk score experienced accelerated cognitive decline over time. Funding for this NIH-funded study underscores the critical value of diverse cohorts in refining clinical tools for neurodegenerative diseases.
Future Directions in Precision Neurology
The development of an equitable polygenic risk score marks a pivotal step toward precision neurology, enabling clinicians to identify high-risk individuals long before overt dementia manifests. Enhanced risk stratification allows researchers to match diverse patient populations appropriately to emerging clinical trials and targeted therapeutic interventions.
Disclaimer: The information provided in this article is for educational and scientific communication purposes only and does not constitute medical advice. Always consult with a qualified healthcare provider regarding any medical condition, diagnosis, or treatment plan.