Written by Alison Proffitt
Researchers from Washington University School of Medicine, St. Louis, have shown that elevated levels of a group of circular RNAs (circRNAs) in the blood nearly triples patients’ risk of developing symptoms of Alzheimer’s disease (AD), suggesting that these molecules are more sensitive to the onset of symptoms than traditional biomarkers of AD. Their work was published last week in Natural medicine (doi: 10.1038/s41591-026-04485-5).
Biomarker-confirmed diagnosis of Alzheimer’s disease has traditionally used biomarkers that measure amyloid beta (Aβ), tau, and phosphorylated tau (pTau) in cerebrospinal fluid (CSF) or by amyloid positron emission tomography (PET), the researchers wrote. But collecting cerebrospinal fluid through a lumbar puncture is painful, and positron emission tomography is expensive.
Therefore, blood biomarkers are crucial, especially those that can capture the overall disease state and symptoms independently of amyloid plaque. “Plasma pTau217 It is widely recognized by the authors as the leading plasma-based biomarker for preclinical Alzheimer’s disease, and has already been shown to be a marker of amyloid pathology that changes 15 to 20 years before disease onset. However, pTau217 can be unreliable as a biomarker in patients treated with anti-amyloid therapies, because it may return to normal as Aβ plaque is removed from the brain without corresponding changes in cognitive improvement.
“Patients who are treated with new therapies to remove Aβ can become pTau negative but still develop Alzheimer’s disease. These circRNAs may give us a more complete perspective on someone’s overall disease biology,” Krushaja, the study’s corresponding author, said in a press release.
“Aβ and Tau pathology-independent biomarkers are essential for monitoring general neurodegeneration in this therapeutic area,” the researchers concluded.
Enter RNA loops
Unlike amyloid plaques, which slowly accumulate in the brain, circRNAs are more dynamic, reflecting more recent brain activity. Research published in 2022 reported a small number of circRNAs in the blood that were associated with Alzheimer’s disease in a data set of 40 people (doi: 10.1186/s40364-022-00405-0). Next, Kruchaga and his colleagues searched for a targeted list of circRNAs that could distinguish between individuals with Alzheimer’s disease symptoms and individuals without cognitive impairment.
The team generated RNA-seq and longitudinal clinical data from 816 cognitively normal individuals and 405 participants with Alzheimer’s disease to identify circRNAs associated with the clinical state of Alzheimer’s disease, Aβ and Tau stages, and progression to symptomatic Alzheimer’s disease. Any circRNAs that survived multiple assays were considered significant.
The team identified a group of 34 circRNAs associated with Alzheimer’s disease, three of which were nominally important in all neurodegenerative diseases. Interestingly, the authors highlight that 2 of the 34 circRNA biomarkers, “are derived from the gene Beckalma genome-wide AD risk gene associated with Aβ, Tau, and Aβ APOE4 Function, synaptic dysfunction and microglia-derived neuroinflammation.
They built predictive models for this set of circRNAs that successfully identified individuals with Alzheimer’s disease, and performed similarly to models trained on pTau217 protein data.
Based on participants’ longitudinal clinical data, the team found that 78 participants progressed from non-disabled to symptomatic Alzheimer’s disease after blood collection. The researchers were able to perform survival analyzes to determine whether 34 circRNAs in the blood could predict progression to symptoms of Alzheimer’s disease. When they analyzed progression to symptomatic Alzheimer’s disease over 5 years, the circRNA model showed a significantly higher AUC than pTau217 alone.
They explained that the researchers replicated their findings in independent samples from the Knight Alzheimer’s Disease Research Center (ADRC) and anti-amyloid therapy in asymptomatic Alzheimer’s disease populations (A4). Also in these datasets, the circRNA model was better than plasma pTau217 when predicting progression to symptomatic Alzheimer’s disease. “These findings point to the potential of blood circRNAs as a non-invasive, high-precision tool for diagnosing early Alzheimer’s disease and monitoring progress in clinical practice,” the authors wrote.
The test may not be sufficient to stratify patients in clinical trials or treatment, the authors acknowledged, but they found that combining the pTau217 and circRNA models improved the predictive power of which patients would progress to Alzheimer’s disease symptoms in both the discovery dataset and replication datasets.
The researchers, together with commercial partners, are currently working to develop translatable clinical assays for blood-based circRNAs.
“It’s good to have good science and models, but ultimately we’re doing it to help people,” Kruchaga said.
