Back

Clinicopathological and prognostic correlations of plasma biomarkers in neurodegenerative diseases

Malpetti, M. M.; Chouliaras, L.; Murley, A. G.; Swann, P.; Coyle-Gilchrist, I.; Bevan-Jones, W. R.; Savulich, G.; Perquin, M. N.; Wiggins, J.; Bowns, L.; Su, L.; Hughes, L.; Quaegebeur, A.; Allinson, K.; Rittman, T.; Hong, Y.; Fryer, T.; Roberts, R.; Ashton, N. J.; Blennow, K.; Zetterberg, H.; O'Brien, J. T.; Rowe, J. B.

2025-12-08 neurology
10.64898/2025.12.07.25341788 medRxiv
Show abstract

Plasma biomarkers have emerged as powerful tools to identify Alzheimers disease (AD) pathology and are increasingly used for diagnosis and monitoring. However, their wider differential diagnostic and prognostic value in neurodegenerative diseases other than AD remains unclear. This study tested and compared the diagnostic and prognostic performance of neurofilament light chain (NfL), p-tau217, A{beta}42/40 and GFAP, in a large cohort of participants recruited from memory clinics and parkinsonism services, with survival data and neuropathology confirmation. Participants with AD and non-AD neurodegenerative diseases were recruited from memory and parkinsonism secondary healthcare services (n=646, 94 with mild cognitive impairment, 130 with Alzheimers dementia, 58 with Lewy-body dementia, 58 with behavioural variant frontotemporal dementia, 56 with primary progressive aphasia, 110 with progressive supranuclear palsy, 58 with corticobasal syndrome and 82 with motor neuron disease), and age-/sex-matched healthy volunteers (n=133). Out of 779 participants, 102 patients had amyloid-positivity also assessed by CSF and/or PET, and 48 patients donated their brains for neuropathological assessment. Group differences and differential performance of plasma biomarkers were analysed using non-parametric tests, ROC analyses, and Cox regression for survival prognosis. Plasma p-tau217 showed high accuracy in discriminating patients with AD versus controls (AUC=0.82), and patients with positive vs negative amyloid markers (AUC=0.80). However, plasma p-tau217 and p-tau231 were also elevated in patients with Lewy body dementia (versus controls AUC=0.73) and motor neuron disease (versus controls AUC=0.72). The plasma NfL/p-tau217 ratio showed better performance in differentiating patients with AD versus frontotemporal lobar degeneration (FTLD) pathologies at post-mortem (AUC=0.94, while p-tau217 AUC=0.82). Plasma NfL was the strongest predictor of survival (HR=1.18 [1.01-1.37], p=0.036), over and above other plasma markers and diagnoses. Our findings demonstrate the utility of plasma p-tau217 as diagnostic marker for AD pathology (noting elevation in the clinically distinct MND), and plasma NfL for prognosis across multiple neurodegenerative diseases. The NfL/p-tau217 ratio was the most sensitive and specific marker in differentiating AD from non-AD groups. We propose that the combination of plasma p-tau217 and NfL can enhance diagnostic and prognostic precision by leveraging their complementary strengths.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above