Blood Test Using Circular RNA Biomarkers Predicts Alzheimer’s Progression

Blood-based approaches that can stratify risk years before symptom onset are increasingly central to precision neurology. A newly launched platform now reports high diagnostic and prognostic performance using circular RNA signatures measured from blood.

Circular Genomics (San Diego, CA, USA) has introduced CircPATH, a proprietary circular RNA discovery and validation engine, following publication of findings in Nature Medicine showing that blood-based circular RNA biomarkers can predict progression of Alzheimer’s disease. The platform underpins the company’s diagnostic product development across neurodegenerative conditions. The study is described as the first to demonstrate that circular RNAs can predict progression to symptomatic disease and outperform existing biomarker standards.

CircPATH systematically identifies disease-relevant circular RNAs from complex biological samples, validates them across independent cohorts, and translates them into clinically actionable biomarker panels. By leveraging key features of circular RNAs—including stability, tissue specificity, and independence from linear transcripts—the platform detects brain-enriched, brain-derived signals in blood that cross the blood–brain barrier. These signatures capture multi-pathway Alzheimer’s biology, including amyloid and tau pathology, synaptic dysfunction, neuroinflammation, and amyloid precursor protein modulation.

In a large, well-characterized Alzheimer’s Disease Research Center (ADRC) clinical cohort, CircPATH-identified biomarker signatures achieved an area under the curve (AUC) of 0.945 for detecting biomarker-confirmed Alzheimer’s disease, exceeding plasma pTau217 (AUC 0.877). When integrated with pTau217, the combined model reached an AUC of 0.967, approaching near-perfect diagnostic accuracy and rivaling cerebrospinal fluid tests and positron emission tomography. For predicting progression from cognitively normal status to symptomatic mild cognitive impairment/Alzheimer’s disease, circular RNAs yielded a hazard ratio of 2.92 versus 1.81 for pTau217—a 61% improvement—while the integrated model achieved a hazard ratio of 4.83. 

For identifying individuals who would develop symptoms within five years, the circular RNA model reached an AUC of 0.870 compared with 0.676 for pTau217 alone. The 34-circular RNA signature was discovered and validated across three independent cohorts: two Knight Alzheimer’s Disease Research Center cohorts (n=1,221 and n=551) and the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) study cohort (n=676).

Circular Genomics reports a pipeline of circular RNA‑based diagnostic products for Alzheimer’s, Parkinson’s, and other neurodegenerative conditions, and is pursuing collaborations to integrate CircPATH‑derived biomarkers into drug development programs and clinical pathways.

“This study represents the first time we’ve demonstrated that blood-based circular RNAs can predict progression to symptomatic Alzheimer’s disease, and they do so with performance that exceeds our current gold-standard biomarkers. What makes this particularly exciting is that these biomarkers begin changing years before symptoms appear and they capture disease biology that goes beyond amyloid and tau,” said Dr. Carlos Cruchaga, Professor of Psychiatry and Director of the Neurogenomics and Informatics Center at Washington University School of Medicine in St. Louis, and lead author of the Nature Medicine study.

"We’re opening a new chapter in precision neurology—one where we can identify at-risk individuals earlier, monitor disease more comprehensively, and ultimately intervene more effectively. The potential to transform both clinical care and drug development is immense," added Dr. Cruchaga.

Related Links
Circular Genomics