Plasma pTau217 and Neurofilament Light Chain as Differentiative Markers for Progressive Supranuclear Palsy From...

Progressive supranuclear palsy (PSP) is a 4-repeat tauopathy neurodegenerative disorder that presents with parkinsonism features and is often indistinguishable from Parkinson disease (PD) in the early stages. The potential of plasma levels of phosphorylated tau (pTau) to differentiate PSP from PD remains unclear. The aim of this study was to identify noninvasive, cost-effective alternatives to tau PET imaging for differentiating PSP from PD.

We measured plasma biomarkers, including pTau species (pTau217, pTau181, and pTau231), neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP), in 2 cohorts recruited from multiple centers across Taiwan. The exploratory cohort included patients with PSP and corticobasal degeneration (CBS) who underwent ¹⁸F-florzolotau PET to confirm underlying tau pathology. All participants with PSP and CBS in this cohort were confirmed negative for amyloid pathology based on either amyloid PET imaging or the plasma Aβ42/Aβ40 ratio. An independent validation cohort, consisting of clinically diagnosed PSP patients without ¹⁸F-florzolotau PET imaging, was also recruited.

A total of 275 participants were included in this study: 195 participants in the exploratory cohort (46 controls, 64 patients with PD and normal cognition, 18 patients with CBS, and 67 patients with PSP) and 80 patients with PSP in the validation cohort. Patients with PSP and CBS had higher levels of pTau217, pTau181, pTau231, and NfL compared with the other groups (all p < 0.05). In the PSP exploratory cohort, plasma levels of pTau217 (r = -0.270, p = 0.031), NfL (r = -0.363, p = 0.003), and GFAP (r = -0.264, p = 0.035) negatively correlated with cognitive performance, while only NfL levels were associated with PSP Rating Scale scores (r = 0.303, p = 0.023). Plasma pTau217 levels positively correlated with cortical tau burden in CBS but negatively correlated with tau tracer uptake in the subthalamic nucleus on ¹⁸F-florzolotau PET in PSP. The combination of pTau217 and NfL with the midbrain-to-pons ratio on brain MRI showed satisfactory performance in differentiating PSP from PD (area under the curve [AUC] 0.946, p < 0.0001), which was comparable to ¹⁸F-florzolotau PET (AUC 0.870, p = 0.001) and confirmed in the validation cohort (AUC 0.825, p < 0.0001).

The combination of plasma pTau217 and NfL with brain MRI represents a noninvasive and clinically feasible strategy for distinguishing PSP from PD.