Background: Alzheimer's disease (AD) diagnosis often relies on invasive cerebrospinal fluid (CSF) sampling or expensive cerebral positron emission tomography (PET) imaging. Noninvasive tests could facilitate early detection. Plasma phosphorylated-tau (pTau) isoforms are promising AD biomarkers that correlate with brain amyloid-β (Aβ) deposition. Objective: To explore the diagnostic accuracy of pTau-based plasma biomarkers to forecast CSF amyloid-positive status in a real-world consecutive population of subjects with cognitive impairment and complete plasma and CSF biomarker profiles. Methods: We retrospectively studied 138 consecutive patients with cognitive impairment. Plasma biomarkers (pTau217, pTau181, Aβ42, Aβ40) were measured by automated immunoassay. Diagnostic accuracy for CSF amyloid status was assessed using area-under-the curve (AUC) values from receiver-operating characteristic (ROC) curve analysis. A dual-threshold strategy was used to define low- and high-risk groups and estimate how many lumbar punctures could be avoided. Results: Among 138 patients (37% CSF amyloid-positive), pTau217/Aβ42 ratio had the highest AUC (0.920) for predicting amyloid positivity, followed by pTau217 (AUC 0.904). These values exceeded the accuracy of pTau181-based markers. Using dual cut-offs, a 90% sensitivity-90% specificity strategy could avoid about 93.5% of lumbar punctures using pTau217/Aβ42 (10.9% misclassified), and 84.1% when using pTau217 (11.2% misclassified). Stricter thresholds (95%–95% and 97.5%–97.5%) further reduced misclassification rates but at the expense of fewer avoidable invasive procedures. Conclusions: Plasma pTau217 (alone or combined with Aβ42) shows high accuracy for detecting AD pathology and could serve as a scalable, noninvasive diagnostic tool. This approach may triage patients for confirmatory testing and substantially reduce the need for invasive CSF examination.
Plasma pTau217/Aβ42 and pTau217 outperform pTau181/Aβ42 and pTau181 in predicting cerebrospinal fluid amyloid positivity: A real-world retrospective study
Calcagno, Andrea;
2025-01-01
Abstract
Background: Alzheimer's disease (AD) diagnosis often relies on invasive cerebrospinal fluid (CSF) sampling or expensive cerebral positron emission tomography (PET) imaging. Noninvasive tests could facilitate early detection. Plasma phosphorylated-tau (pTau) isoforms are promising AD biomarkers that correlate with brain amyloid-β (Aβ) deposition. Objective: To explore the diagnostic accuracy of pTau-based plasma biomarkers to forecast CSF amyloid-positive status in a real-world consecutive population of subjects with cognitive impairment and complete plasma and CSF biomarker profiles. Methods: We retrospectively studied 138 consecutive patients with cognitive impairment. Plasma biomarkers (pTau217, pTau181, Aβ42, Aβ40) were measured by automated immunoassay. Diagnostic accuracy for CSF amyloid status was assessed using area-under-the curve (AUC) values from receiver-operating characteristic (ROC) curve analysis. A dual-threshold strategy was used to define low- and high-risk groups and estimate how many lumbar punctures could be avoided. Results: Among 138 patients (37% CSF amyloid-positive), pTau217/Aβ42 ratio had the highest AUC (0.920) for predicting amyloid positivity, followed by pTau217 (AUC 0.904). These values exceeded the accuracy of pTau181-based markers. Using dual cut-offs, a 90% sensitivity-90% specificity strategy could avoid about 93.5% of lumbar punctures using pTau217/Aβ42 (10.9% misclassified), and 84.1% when using pTau217 (11.2% misclassified). Stricter thresholds (95%–95% and 97.5%–97.5%) further reduced misclassification rates but at the expense of fewer avoidable invasive procedures. Conclusions: Plasma pTau217 (alone or combined with Aβ42) shows high accuracy for detecting AD pathology and could serve as a scalable, noninvasive diagnostic tool. This approach may triage patients for confirmatory testing and substantially reduce the need for invasive CSF examination.| File | Dimensione | Formato | |
|---|---|---|---|
|
Plasma pTau217 CSF amyloid positivity Imperiale 2025.pdf
file disponibile agli utenti autorizzati
Licenza:
Copyright dell'editore
Dimensione
950.3 kB
Formato
Adobe PDF
|
950.3 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


