Shared blood pattern linked to later neurological disease
Nine blood measures ranked among the top 30 predictors for at least seven of eight neurological disease endpoints in UK Biobank data.
Our read
Biomarkers- Studied:
- Cohort study of 274,241 adults aged 40–69 years, median 14.9 years
- What changed:
- Across seven of eight endpoints, nine metabolites ranked in the top 30 baseline predictors of future incident neurological disease
- The catch:
- The signature may reflect unadjusted renal function, body composition, or fasting status
Editor of record: Boris Djordjevic · AgingResearch.News
An observational analysis used published summary data from 274,241 UK Biobank participants recruited at ages 40–69. Researchers compared blood measurements taken before diagnosis with later neurological diagnoses over a median follow-up of 14.9 years. The eight disease endpoints included Alzheimer’s disease, Parkinson’s disease and epilepsy.
Nine of 57 priority blood measures ranked among the top 30 predictors for at least seven endpoints. These measures included metabolites, small molecules produced or used by the body, and ratios between them. The shared pattern exceeded chance expectations. Researchers also used Mendelian randomization, which uses genetic differences to examine possible causal links. None of 45 comparisons met the statistical threshold after accounting for multiple tests. The authors interpreted the pattern as a possible marker or signal of disease developing before diagnosis, rather than a proven cause.
Why it matters
Blood chemistry offers a way to investigate whether age-related neurological illnesses share changes outside the brain before diagnosis.
Caveats
This summary-level observational analysis cannot establish cause, and the lack of a quantitative comparison with non-neurological diseases prevents claims of neurological specificity. Some disease definitions overlapped; frontotemporal and Lewy body dementia were excluded, and independent validation is still needed.
From the paper
- Result
Three pairs showed nominal associations (p < 0.05) that did not survive multiple testing correction: higher genetically instrumented glucose levels were associated with a lower ALS risk (IVW OR = 0.841 [95% CI: 0.717–0.986], p = 0.033) …
- Absolute difference
The full disease-agnostic weights were first reconstructed and matched the deposited disease-agnostic weights to numerical precision (maximum absolute difference = 1.39 × 10^−16).
- Limitation
First, the UK Biobank cohort, while providing exceptional scale, is predominantly of European ancestry (>94%), limiting the generalizability of our findings to other populations.
Quoted word for word from the full text on PubMed Central.
- Neurological disease
- Incident neurological diseases
- Shared pre-diagnostic neurological metabolic signature
- Humans
The paper
Chinese People's Liberation Army; Mayo Clinic in Florida
Biomedicines, Aug 6, 2026, Open access
Comments
Be the first to comment.