Dementia Prevention Digest

Non-invasive early-detection tests gaining momentum

Non-invasive early-detection tests gaining momentum

Key Questions

What new blood tests are being developed for early Alzheimer's detection?

Multiple tools including blood circRNAs predict Alzheimer's years before symptoms with high accuracy (AUC 0.95-0.98) and outperform pTau217 in progression prediction. A DNA-RNA blood model (AUC 0.705) has shown cross-ancestry validation, while finger-prick p-tau217 tests are also advancing.

How is the NHS involved in Alzheimer's blood testing?

A major NHS trial is offering blood tests to over 1,000 patients across the UK to detect Alzheimer's disease, moving toward broader clinical implementation of these tools.

Can eye tests detect mild cognitive impairment?

Eye-movement dynamics using devices like the GAP show promise in detecting MCI with AUC 0.897, linking saccadic reaction time to genes like RALYL and cognitive reserve.

What are the limitations of current blood tests like Lumipulse?

Real-world data shows Lumipulse can produce up to 40% false positives, which tempers enthusiasm despite pTau217 remaining a benchmark test now licensed by ALZpath-Abbott.

Are hearing tests useful for early dementia screening?

Hearing tests that measure brain responses rather than just the ear may identify early cognitive decline, with related eye and hearing assessments emerging as low-cost screening options.

What policy changes support biomarker testing for Alzheimer's?

Delaware has passed biomarker coverage legislation, while Oregon is pushing for similar measures in 2027 to improve access to early detection tools.

How do circRNAs compare to pTau217 for predicting Alzheimer's progression?

CircRNAs outperform pTau217 with a higher hazard ratio (2.9 vs 1.8) for progression and can still function after anti-amyloid therapy, offering better timing of symptom onset.

What other non-invasive methods are advancing for Alzheimer's detection?

Advances include retinal AI, plasma miRNAs, acoustic analysis, digital phenotyping, smartphone memory tests, and SECmeres RNA nanoparticles showing early proof-of-concept.

Multiple new tools: blood circRNAs predict Alzheimer's years before symptoms (AUC 0.95, combined 0.98) and outperform pTau217 in progression prediction (HR 2.9 vs 1.8); they also work after anti-amyloid therapy. A major NHS trial is offering blood tests to over 1,000 patients, moving toward clinical implementation. Eye-movement dynamics detect MCI (AUC 0.897); a new Japanese study using a GAP device shows saccadic reaction time and latency linked to RALYL gene and cognitive reserve. Finger-prick p-tau217 (capillary) is promising but raises equity and counseling gaps. SECmeres (RNA nanoparticles) show proof-of-concept. A DNA-RNA blood model (AUC 0.705) with cross-ancestry validation (Korean and U.S. cohorts) could serve as a triage tool. However, real-world false positives with Lumipulse (40% false positives) temper hype. pTau217 remains benchmark (ALZpath-Abbott licensing). Other advances: GPND-AI multi-disease test, retinal AI, B12 deficiency mimic, plasma miRNAs, Sysmex automated platform, acoustic analysis, digital phenotyping, smartphone memory tests. Hearing and eye tests are emerging as low-cost screening signals. Personal story on testing anxiety adds human perspective. Policy: Delaware biomarker coverage bill, Oregon 2027 legislation push.

Sources (21)
Updated Jul 7, 2026