A new Alzheimer’s blood test may show when symptoms are getting close

A research-stage blood test may help distinguish Alzheimer’s disease that is biologically present from disease that is approaching noticeable symptoms. Scientists identified a pattern of circular RNA molecules that predicted progression better than a leading protein biomarker in the study population. The finding could eventually help clinical trials enroll people near symptom onset, but the assay is not yet a routine diagnostic test.

The signal comes from loops of genetic material

Circular RNAs are RNA molecules whose ends join to form closed loops. Unlike the linear messenger RNA used to carry instructions for protein production, many circular RNAs appear to help regulate cellular activity. Their stability in blood and connection to recent biological processes make them attractive candidates for biomarkers.

An NIH-funded analysis released July 1 examined blood data from more than 1,200 participants across multiple independent cohorts. Researchers identified 34 circular RNAs associated with Alzheimer’s disease. Higher levels of certain molecules were linked to nearly triple the risk of progressing to symptoms, according to the agency’s summary.

Amyloid can appear long before cognitive changes

Several current blood tests measure proteins associated with amyloid plaques, including phosphorylated tau markers such as pTau217. Those tests can provide useful evidence of Alzheimer’s pathology, but a positive result may occur years or even decades before cognitive impairment. Detecting disease biology and estimating when symptoms may emerge are different clinical questions.

The circular-RNA model performed similarly to a pTau217 model when identifying Alzheimer’s pathology. Its stronger result appeared in the forward-looking analysis: the 34-RNA signature better predicted which participants would progress to symptomatic disease. Additional experiments suggested that levels began separating from normal approximately two to four years before symptom onset.

Independent cohorts strengthen the result

Biomarker studies can produce misleading signals when a model fits quirks in one dataset rather than a general biological pattern. Testing across independent groups helps reduce that risk. The researchers reported similar findings in samples from two separate cohorts, supporting the idea that the signature was not limited to a single participant pool.

The peer-reviewed study in Nature Medicine also builds on earlier work linking circular RNAs in brain tissue to dementia and neuropathological severity. Moving the measurement into blood matters because a blood draw is more practical and repeatable than sampling brain tissue or cerebrospinal fluid. Even so, replication across broader populations and clinical settings remains necessary.

Timing could change how prevention trials are designed

A treatment intended to delay cognitive decline may work differently depending on how close a participant is to developing symptoms. Trials that enroll people across a very wide presymptomatic window can struggle to detect an effect because many participants might not decline during the study period. A marker of near-term progression could create a more informative group.

Researchers also see potential for monitoring treatment response. Newer therapies can change amyloid-related markers, yet those changes may not capture every part of the disease process. Circular RNAs could provide a complementary view of recent biology rather than replacing established biomarkers. That possibility must be tested prospectively before the signature can guide treatment decisions.

A promising model is not yet a clinic-ready test

The study established an association and predictive model, not a commercially available assay with regulatory clearance. Laboratories need standardized methods for collecting blood, measuring the RNAs and converting results into an interpretable score. Developers must also determine thresholds, error rates and how performance changes across age groups, ancestries, medical conditions and stages of disease.

False reassurance and false alarms both carry consequences. A test that estimates symptom timing would require careful counseling because risk is not certainty and cognitive changes can have many causes. Clinical utility depends not only on statistical accuracy but also on whether the result leads to better decisions, more effective trial enrollment or improved patient outcomes.

Age-related illness, medications and other neurological conditions can complicate interpretation. A useful assay would need prespecified rules for combining the RNA signature with medical history, cognitive testing and established biomarkers. Researchers would also need to show that repeated measurements are stable enough to distinguish a biological change from ordinary laboratory variation.

Researchers are now working toward a usable assay

The team is collaborating with commercial partners to translate the circular-RNA findings into clinical assays. That process includes analytical validation, larger prospective studies and comparison with existing blood, imaging and cognitive measures. The most useful future tool may combine several kinds of evidence rather than rely on one marker.

The new work narrows an important gap in Alzheimer’s testing. Blood biomarkers increasingly show whether characteristic pathology is present, while the 34-RNA signature may add information about when that pathology is likely to become symptomatic. The distinction remains provisional, but it gives researchers a concrete path toward tests that address disease timing as well as detection.

For now, the result belongs to research rather than routine screening. It supports continued development and prospective validation, not a promise that an individual blood sample can predict an exact date of cognitive decline.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview