Potential of F-18-PI-2620 Tau PET to Improve the Imaging-Based Diagnosis of Progressive Supranuclear Palsy (PSP)

Ziel/Aim PSP is a 4-repeat tauopathy movement disorder which can be imaged by the second-generation tau PET tracer F-18-PI-2620. The in vivo diagnosis of PSP is currently established on clinical grounds and supported by midbrain (MB) atrophy estimation on structural MRI. This evaluation investigated the diagnostic potential of F-18-PI-2620 compared to that of the MB atrophy MRI.

Methodik/Methods Dynamic (0-60min p.i.) F-18-PI-2620 PET and structural MRI data of 36 patients with clinical diagnosis of PSP and those of 10 healthy controls (HCs) were analyzed. Distribution volume ratios (DVRs, MRTM2) for typical tau-in-PSP target regions were determined in PMOD. In the MR images, the MB diameter, MB area, MB/pons area ratio, and MR Parkinsonian Index were measured. Furthermore, the PET and MRI images were visually assessed by three blinded experts for typical PSP signs.

Ergebnisse/Results Following a ROC analysis, the MB/pons area ratio discriminated best between the PSP patients and the HCs for the MRI parameters (area under the curve[AUC]=0.93; sensitivity/specificity [sens/spec] 83 %/90 %). For the PET parameters the globus pallidus DVRs were the best discriminators (AUC = 0.91; sens/spec 81 %/100 %).The combination of both parameters (≥1 abnormal parameter defined abnormality) resulted in sens/spec 97 %/90 %. The visual assessments lead to sens/spec in the separation between PSPs and HCs of 75 %/80 % for PET and of 53 %/100 % for MRI (at least one/three signs positive → whole brain positive) and of 81 %/80 % for the combination of both. All of the above semi-quantitative PET and MRI measures apart from the MR Parkinsonian Index correlated with age (p = 0.017 to <0.009) but only the MRI measures correlated with the PSP Rating Scale scores (p = 0.001 to <0.001).

Schlussfolgerungen/Conclusions These results provide evidence for an improved imaging-based PSP diagnosis by F-18-PI-2620 tau PET when added to structural MRI.

Publication History

Article published online:
08 April 2021