Mode
Text Size
Log in / Sign up

Higher GrimAge epigenetic age acceleration is significantly associated with higher dementia riskGrimAge metric shows link to increased risk of dementia

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway
Note that GrimAge is the only metric significantly associated with dementia risk, but overall evidence is insufficient for clinical use.

This meta-analysis evaluated the association between various DNAm-based biological age metrics and neurodegenerative diseases, including Dementia, Parkinson's Disease, Multiple Sclerosis, and Amyotrophic Lateral Sclerosis. The study included a large total sample size of 46365 individuals. The primary objective was to determine if these epigenetic clocks could serve as reliable indicators or biomarkers for neurodegenerative conditions.

The researchers analyzed five specific DNAm-based biological age metrics: GrimAge, Horvath1, Hannum, PhenoAge, and DunedinPACE. These metrics are designed to estimate biological age based on DNA methylation patterns. The study also incorporated Mendelian randomization to investigate potential causal relationships between these metrics and the observed clinical outcomes.

Regarding the primary outcomes, the analysis found that GrimAge epigenetic age acceleration (EAA) was significantly associated with higher dementia risk. Specifically, a higher GrimAge EAA resulted in a hazard ratio (HR) of 1.11 per standard deviation (SD), with a 95% confidence interval (CI) of 1.02 to 1.20 and a p-value of 0.018. In contrast, the other four metrics—Horvath1 EAA, Hannum EAA, PhenoAge EAA, and DunedinPACE—showed no significant association with dementia risk. No specific effect sizes or confidence intervals were reported for these non-significant findings.

Secondary outcomes included the use of Mendelian randomization to establish causality. However, the results from Mendelian randomization provided only limited evidence for causality regarding the relationship between DNAm age metrics and neurodegenerative diseases. Safety and tolerability data were not reported, as these are biological metrics rather than pharmacological interventions.

These findings suggest that while GrimAge shows a statistically significant association with dementia risk, the results are not consistent across all types of DNAm clocks. The evidence for other neurodegenerative diseases, such as Parkinson's Disease, Multiple Sclerosis, and Amyotrophic Lateral Sclerosis, was noted as heterogeneous across different clock generations and study designs. This heterogeneity complicates the ability to draw broad conclusions for all neurodegenerative conditions.

Several methodological limitations were identified. The primary limitations include the inconsistent associations of DNAm age clocks with neurodegenerative diseases and the limited evidence for causality provided by Mendelian randomization. The diversity in study designs and clock generations contributed to the heterogeneity of results for non-dementia conditions.

Clinically, these results indicate that DNAm age clocks currently provide insufficient support for use as reliable disease biomarkers in clinical practice. The findings suggest a need for more specific DNAm clocks and detailed pathway analyses tailored to specific neurodegenerative diseases rather than general biological age metrics. Currently, there is no evidence to support the use of these metrics for routine clinical decision-making or as definitive diagnostic tools. Questions remain regarding the specific biological pathways that GrimAge captures to correlate with dementia risk. Furthermore, more consistent data is needed to determine why GrimAge outperformed other clocks in this specific association. Future research must focus on developing disease-specific DNAm clocks to improve clinical utility.

How this fits prior evidence

How this fits prior evidence: This meta-analysis addresses a gap in identifying specific biomarkers for neurodegenerative diseases. While previous evidence highlighted that social isolation and loneliness are significant determinants of cognitive decline and dementia risk, this study explores biological markers. It does not directly relate to the findings regarding photobiomodulation, KRTCAP2 genetic burden, transposable elements, or dietary habits in Parkinson's disease.

Understanding how our bodies age can help doctors better understand and manage conditions like dementia and other neurological diseases. For many people, the fear of cognitive decline is a significant concern. Researchers are looking for ways to measure biological aging, which is how fast our cells age, rather than just our chronological age in years. This research looked at several different methods for measuring this biological age to see if any of them could help identify risks for dementia.

The researchers conducted a large meta-analysis, which is a study that combines data from many different previous studies to find broader patterns. They looked at data from 46,365 individuals. The researchers tested five different types of DNA-based age clocks. These are tools that look at chemical changes in DNA to estimate how much a person has aged biologically. The five clocks tested were GrimAge, Horvath1, Hannum, PhenoAge, and DunedinPACE.

The results showed that only one of the five tools, called GrimAge, had a significant link to a higher risk of dementia. Specifically, a higher score on the GrimAge scale was associated with an increased risk of developing dementia. The other four clocks tested—Horvath1, Hannum, PhenoAge, and DunedinPACE—did not show a clear or consistent link to dementia risk in this study. This suggests that while some biological clocks are useful for certain health markers, they are not all equally useful for predicting neurological conditions.

It is important to note that this study has several limitations. The evidence for other diseases like Parkinson's or Multiple Sclerosis was inconsistent across the different types of clocks. Additionally, the researchers used a method called Mendelian randomization to see if aging actually caused the disease, but this provided only limited evidence for a direct cause. Because the results were inconsistent across different types of clocks, the findings are not yet strong enough to be used as a definitive medical test.

For patients today, this means that while the GrimAge clock showed a specific link to dementia risk, it is not currently a tool that doctors can use to diagnose or predict the disease in a clinical setting. More research is needed to develop specific tests for different types of neurological conditions. For now, these findings are a step toward understanding how biological aging relates to brain health, but they do not change the current standard of care for patients with dementia or other related conditions.

What this means for you:
One specific biological age clock showed a link to dementia risk, but other clocks did not show a consistent link.

Study Details

Study typeMeta analysis
Sample sizen = 46,365
EvidenceLevel 1
PublishedOct 2026
View Original Abstract ↓
Hallmarks of biological ageing are closely interconnected with neurodegenerative diseases. Deoxyribonucleic acid methylation (DNAm) clocks quantify biological age and may serve as peripheral biomarkers of neurodegenerative diseases. This systematic review and meta-analysis aim to examine associations between DNAm-based biological age metrics and neurodegenerative diseases. Six databases were systematically searched from 1st January 2011 to 15th April 2026 for studies reporting associations between DNAm-based ages and neurodegenerative diseases including all-cause dementia, Parkinson's Disease, Multiple Sclerosis, and Amyotrophic Lateral Sclerosis. Correlated-and-hierarchical-effects meta-analyses were fitted using restricted maximum likelihood with Knapp-Hartung adjustment. Of 42,023 records screened, 37 observational studies (n=46365; 60.5% female; median of mean age 69.4 years [IQR 62.5-73.3)) and seven Mendelian randomization studies met inclusion criteria. In longitudinal meta-analyses, higher GrimAge epigenetic age acceleration (EAA) was significantly associated with higher dementia risk (HR per SD 1.11 [95% CI 1.02-1.20]; I^2 = 0.0%; p = 0.018), whereas Horvath1 EAA, Hannum EAA, PhenoAge EAA, and DunedinPACE were not significantly associated with dementia risk. Evidence for other neurodegenerative diseases were heterogeneous across clock generations and study designs. Mendelian randomization provided limited evidence for causality. Higher GrimAge EAA was the only DNAm age measure consistently predictive of higher all-cause dementia risk. Overall, however, DNAm age clocks show limited and inconsistent associations with neurodegenerative diseases, providing insufficient support for their use as disease biomarkers. These findings highlight the need to develop and independently validate disease-specific DNAm clocks and undertake pathway analyses to identify methylation changes associated with neurodegeneration and the biological processes they reflect.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.