Diagnostic accuracy of CSF and plasma pTau-181, pTau-217, and pTau-231 for Alzheimer's disease: a diagnostic meta-analysis.

Abstract

Background: Alzheimer's disease (AD) remains the most common cause of dementia worldwide, yet achieving an accurate early diagnosis continues to pose a major challenge. Recent advances highlight phosphorylated tau (pTau) isoforms-pTau-181, pTau-217, and pTau-231-detected in cerebrospinal fluid (CSF) and plasma as promising biomarkers capable of distinguishing AD from non-Alzheimer's dementias and cognitively healthy individuals (HC). This study aimed to comprehensively assess the diagnostic accuracy of these pTau isoforms through a systematic review and diagnostic meta-analysis.

Methods: Following PRISMA 2020 guidelines and PROSPERO registration (CRD42025632278), we performed a systematic search of PubMed, Scopus, and Web of Science databases through September 2025. Eligible studies evaluated the diagnostic performance of pTau-181, pTau-217, or pTau-231 in CSF or plasma using established AD reference standards. Pooled sensitivity, specificity, diagnostic odds ratio (DOR), and area under the curve (AUC) were calculated using random-effects models. Clinical applicability was assessed via Fagan's nomogram, and publication bias was examined using Deeks' funnel plot asymmetry test.

Results: Forty-eight studies encompassing diverse international cohorts met inclusion criteria. In differentiating AD from HC, CSF pTau-217 demonstrated the highest diagnostic precision (sensitivity = 0.95; specificity = 0.91), outperforming pTau-181 and pTau-231 (AUC = 0.90 each). Plasma biomarkers also exhibited strong diagnostic value: plasma pTau-181 achieved sensitivity of 0.88, specificity of 0.81, and AUC of 0.91, while plasma pTau-231 displayed sensitivity of 0.84, specificity of 0.87, and AUC of 0.86, with low interstudy heterogeneity. Comparative analyses confirmed significantly higher pTau concentrations in AD relative to both non-AD dementias and healthy controls. The overall certainty of evidence ranged from moderate to high, supporting the robustness of these findings.

Conclusions: This meta-analysis supports the strong diagnostic performance of pTau isoforms, particularly CSF pTau-217, in distinguishing Alzheimer's disease from healthy controls and non-Alzheimer's dementias. However, evidence for CSF pTau-217 is limited and largely derived from limited number of studies. Although plasma pTau-181 and pTau-231 show promising diagnostic accuracy, further real-world validation, assay standardization, and assessment within established diagnostic pathways are needed.

Clinical trial number: Not applicable.