The accurate prediction of Alzheimer’s disease (AD) risk is fundamental for timely interventions, clinical trial enrollments, and care planning. One approach uses a polygenic risk score (PRS), which sums the small effects of many disease-associated genes into one cumulative score. However, PRS only reflects inherited genetic factors and fails to capture how those genes interact with lifestyle and environmental influences.
Recent advances in measuring proteins in blood samples can now provide insight into what happens when genetic and non-genetic factors interact, but plasma biomarkers alone cannot reliably predict AD.
Both PRS and plasma biomarkers are valuable tools, yet each has its own limitations. Dr. Prokopenko and Dr. Moqri believe that combining the two will yield a more accurate proteogenomic risk score for early AD detection. In preliminary experiments, their proteogenomic risk score outperformed PRS alone at predicting which asymptomatic individuals went on to develop AD. For this project, they intend to further validate and refine their score in a new cohort, while also improving its ability to account for sex, race, and APOE genotype.
The project contains two primary aims. In the first, they will measure concentrations of important AD blood biomarkers, such as pTau217 and Aβ42/40, GFAP, and NfL, in a large cohort from the Mass General Brigham (MGB) Biobank with existing genetic data. This will provide them with the data on plasma biomarker levels needed to generate proteogenomic risk scores for these patients. The second aim is to evaluate the performance of the risk scores for these cases. Ultimately, the goal is for the proteogenomic risk score to accurately predict AD before symptom onset. As part of this aim, they will determine how well the risk score performs across different races and sexes, and whether the APOE genotype has a substantial impact on individual plasma markers or on their composite measures.
This study aims to validate and refine a test for early AD detection that uses both plasma biomarkers and patient genetic information to achieve greater predictive power than either test alone. The long-term goal of this work is to develop a practical, scalable tool for early AD assessment, enabling early prevention and intervention.
