Researchers have identified a specific signature of circular RNAs (circRNAs) in human blood that can predict Alzheimer’s disease (AD) and its progression with high accuracy. According to a study published in Nature Medicine, this blood-based biomarker panel outperforms existing tests like plasma pTau217 in forecasting the transition from cognitively unimpaired status to symptomatic Alzheimer’s, potentially offering a non-invasive tool for early clinical detection.
How Circular RNAs Function as Biomarkers
Circular RNAs are stable, closed-loop molecules that are highly expressed in the brain. Unlike linear messenger RNA (mRNA), their circular structure makes them resistant to degradation by common cellular enzymes, which allows them to persist in the bloodstream. Because they are capable of crossing the blood-brain barrier, they serve as unique biological messengers reflecting neurological health.
In the study, researchers analyzed RNA-sequencing data from 1,221 participants, including both cognitively unimpaired individuals and those with Alzheimer’s disease, sourced from the Knight Alzheimer Disease Research Center (Knight-ADRC). Using bioinformatics tools—specifically DCC, CIRI2, and CIRI3—the team identified 34 specific circRNAs that were significantly associated with Alzheimer’s pathology. These molecules showed high levels of expression in brain tissue, confirming their neurological origin.
Diagnostic Accuracy Compared to Existing Tests
The researchers developed a predictive model based on these 34 circRNAs and benchmarked it against established diagnostic standards, including CSF biomarkers, amyloid-PET imaging, and plasma pTau217.
- Biomarker-Confirmed Status: The circRNA model achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.945, surpassing the 0.877 AUC observed with plasma pTau217.
- Combined Power: When the circRNA model was integrated with plasma pTau217 levels, the predictive accuracy rose to an AUC of 0.967, suggesting that combining these methods could provide a more robust assessment than any single blood-based test.
Predicting Progression to Symptomatic Alzheimer’s
One of the most significant findings involves the model’s ability to predict which cognitively healthy individuals will progress to symptomatic Alzheimer’s disease. In survival analyses, the circRNA model showed a hazard ratio (HR) of 2.92 for disease progression, significantly higher than the HR of 1.81 associated with pTau217 alone.
Data suggests these changes are detectable in the blood approximately two to four years before the onset of clinical symptoms. By tracking these molecular shifts, clinicians may eventually be able to identify high-risk patients during the presymptomatic phase, a period when therapeutic interventions are most likely to be effective.
Specificity Across Diverse Populations and Dementias
To ensure the model’s reliability, the research team conducted sensitivity analyses across different demographics and neurodegenerative conditions.
- Ancestry: The model demonstrated consistent predictive power across European, African, and mixed-ancestry cohorts, with AUC values remaining high (above 0.90 in several groups).
- Disease Specificity: The circRNA panel appears highly specific to Alzheimer’s disease. When tested against other neurodegenerative conditions, including Parkinson’s disease, frontotemporal dementia (FTD), and dementia with Lewy bodies (DLB), the model showed low predictive power. This indicates that the identified circRNAs capture changes unique to the Alzheimer’s disease process rather than general neurodegeneration.
Key Takeaways
- Early Detection: The 34-circRNA blood test can identify individuals at risk for Alzheimer’s disease up to four years before clinical symptoms manifest.
- Improved Accuracy: Integrating this circRNA panel with current plasma pTau217 testing provides greater diagnostic precision than either method used in isolation.
- Broad Applicability: The model maintains high accuracy across diverse genetic ancestries and is specific to Alzheimer’s, showing little overlap with other forms of dementia like Parkinson’s disease.
- Validation: The findings were successfully replicated in the independent A4 (Anti-Amyloid Treatment in Asymptomatic Alzheimer’s) cohort, reinforcing the potential for clinical utility.
While these results are promising, the researchers emphasize that this model is currently a research tool. Future clinical validation will be necessary before such testing becomes a standard component of medical practice for early Alzheimer’s screening.
Related reading