A new study has uncovered evidence suggesting that a person’s inherited predisposition for higher levels of the tau protein in their blood is associated with an increased likelihood of developing Alzheimer’s disease or its precursor stage. The findings, which also point to potential differences in risk based on sex and age, were published in the journal Neurology.
Alzheimer’s disease is a progressive brain disorder that gradually impairs memory and thinking skills. At the molecular level, it is characterized by the accumulation of two key proteins in the brain: amyloid-beta, which forms plaques between nerve cells, and tau, which forms tangles inside them. While tau protein normally helps stabilize the internal skeleton of brain cells, in Alzheimer’s disease it becomes abnormal and aggregates, disrupting cell function and contributing to neurodegeneration.
Because elevated tau levels in the blood can reflect ongoing damage to brain cells, they are considered an important biomarker for the disease. In the new study, led by geneticist Niki Mourtzi and neurology professor Nikolaos Scarmeas of the National and Kapodistrian University of Athens Medical School, researchers sought to move beyond measuring current tau levels and instead investigate the underlying genetic factors.
“Early detection of Alzheimer’s disease remains challenging, as most biomarkers require invasive procedures or expensive imaging. We aimed to fill this gap by investigating
whether a polygenic risk score for plasma tau, a minimally invasive biomarker, could identify individuals at higher risk for developing Alzheimer’s disease or amnestic mild cognitive impairment,” the researchers told PsyPost.
A polygenic risk score is a tool that estimates an individual’s inherited susceptibility to a specific condition. This single numerical value is calculated by combining the small effects of numerous genetic variants from across a person’s entire genome.
“An important advantage of a polygenic risk score is that it captures inherited genetic variation, allowing us to predict disease risk from birth, decades before amyloid and tau start to accumulate in the brain,” Mourtzi and Scarmeas explained. “Unlike prior studies that focused on cognitive scores, our study evaluated a clinically meaningful outcome over time, providing a more direct link between genetic risk and disease development.
The investigation was conducted in two main phases, using data from two distinct populations. The first part of the study involved the Hellenic Longitudinal Investigation of Aging and Diet (HELIAD), a community-based study in Greece. The researchers analyzed data from 618 participants, who were 65 years or older and did not have Alzheimer’s or amnestic mild cognitive impairment at the beginning of the study. Amnestic mild cognitive impairment is a condition involving memory loss that is often a precursor to Alzheimer’s disease.
For each participant, the team calculated a polygenic risk score for tau based on 21 genetic variations located near the gene that provides the instructions for making the tau protein. The participants were followed for an average of about three years. During this period, 73 individuals were diagnosed with either Alzheimer’s disease or amnestic mild cognitive impairment.
The analysis provided evidence of an association between the genetic score and disease risk. The results showed that for every one standard deviation increase in the polygenic risk score, there was an associated 29% higher risk of developing one of the cognitive conditions. This relationship appeared to be independent of other known risk factors, including age, sex, education, and the presence of the APOE e4 gene, which is the most well-established genetic risk factor for Alzheimer’s.
“Our study is among the first to link a polygenic risk score for plasma tau directly to clinical outcomes rather than cognitive scores,” Mourtzi and Scarmeas said.
When the researchers examined specific subgroups, they observed that the association was not uniform. The link between a higher genetic score and disease risk was stronger in women, who showed a 45% increase in risk for every standard deviation increase in their score. The association also tended to be more pronounced in younger participants (those below the group’s median age of 73), who had an 87% higher risk. In contrast, the associations were not statistically significant for men or for the older participants in the cohort.
“People with a higher genetic predisposition to elevated plasma tau levels face an increased risk of Alzheimer’s disease or its prodromal stage,” Mourtzi and Scarmeas told PsyPost. “In the HELIAD study, those with higher genetic risk had about a 28.5% greater chance of developing Alzheimer’s disease or amnestic mild cognitive impairment. The effect was stronger in women and younger individuals, suggesting that both sex and age influence how genetic risk translates into disease. Early identification of those at higher genetic risk could enable earlier interventions, lifestyle modifications, or monitoring, potentially improving outcomes.”
“We were somewhat surprised by the pronounced sex- and age-specific effects. This may be influenced by sex-specific genetic mechanisms: for example, X-linked genes such as USP11 are more highly expressed in female brains and can promote tau accumulation, while other X chromosome loci like CHST7 may facilitate tau fibril formation and propagation. We also found that genetic risk was more relevant in younger participants, suggesting that inherited tau- related risk is more influential earlier in life before lifestyle, comorbidities, or other environmental factors become dominant.”
To see if these findings were robust, the researchers sought to replicate them in a much larger and more diverse group of people from the UK Biobank. This second part of the analysis included over 142,000 individuals aged 60 and older who were free of dementia at the start of the study. These participants were followed for an average of nearly 13 years, during which 2,737 developed Alzheimer’s disease.
In this large cohort, a higher polygenic risk score for tau was also associated with an increased risk of an Alzheimer’s diagnosis, which supports the initial findings. The effect size was smaller, with a one standard deviation increase in the score corresponding to about a 5% increase in risk.
The subgroup analyses by sex and age did not produce significant results in this larger sample. However, when the researchers created a smaller UK Biobank subsample that was statistically matched to the Greek cohort based on age, sex, and other characteristics, the results were more aligned. In this matched group, a higher score was linked to a 50% increased risk of developing Alzheimer’s.
“Although the individual effect of the tau PRS is modest, it remained consistent across two large, independent cohorts, reinforcing its potential utility,” Mourtzi and Scarmeas said. “When combined with established risk factors such as APOE genotype, age, sex, or genetic risk for other Alzheimer’s-related biomarkers (e.g., amyloid, hippocampal atrophy, white matter hyperintensities) can help identify people who may be at higher risk for Alzheimer’s disease, potentially years or even decades before symptoms appear.”
It is important to note that polygenic risk scores are predictive tools, not diagnostic certainties. They are based on common genetic variants and do not account for the influence of rare genes, lifestyle choices, or environmental factors, all of which play a part in the development of complex diseases like Alzheimer’s. The score was also developed and tested in populations of European ancestry, meaning its predictive power might not be the same in individuals from other backgrounds.
“It represents only one factor, as other variables like lifestyle, environment, and chance also play a significant role,” the researchers noted. “A high polygenic risk score does not guarantee a person will develop Alzheimer’s disease, and a low polygenic risk score does not exclude the possibility of developing it. However, polygenic risk scores can be seen as an important tool to identify individuals at higher risk and take early preventive actions, such as lifestyle modifications, monitoring, or participation in clinical studies aimed at reducing risk.”
Looking ahead, the research team suggests that this polygenic risk score for tau could be combined with other genetic scores, such as those for amyloid buildup or brain atrophy, to create a more comprehensive risk assessment model. Such a multifaceted approach could improve the ability to stratify individuals by their overall genetic risk, helping to target preventive strategies and guide enrollment in clinical trials for new therapies.
“We aim to integrate tau-related polygenic risk scores with additional genetic and imaging biomarkers to develop comprehensive, multifactorial models for Alzheimer’s disease risk prediction,” Mourtzi and Scarmeas explained. “We have already computed polygenic risk scores for other relevant endophenotypes, including amyloid deposition, hippocampal atrophy, and white matter hyperintensities, and intend to combine these scores into a single composite measure that captures overall genetic and neuroimaging risk. This integrative approach has the potential to enable early, personalized interventions and to refine risk stratification strategies in both research and clinical settings.”
The study, “Longitudinal Association of a Polygenic Risk Score for Plasma T-Tau With Incident Alzheimer Dementia and Mild Cognitive Impairment,” was authored by Niki Mourtzi, Sokratis Charisis, Eva Ntanasi, Alexandros Hatzimanolis, Alfredo Ramirez, Stefanos N. Sampatakakis, Mary Yannakoulia, Mary H. Kosmidis, Efthimios Dardiotis, George Hadjigeorgiou, Paraskevi Sakka, Eirini Mamalaki, Christopher Papandreou, Marios K. Georgakis, and Nikolaos Scarmeas.
