Browse by author
Lookup NU author(s): Professor Nicola PaveseORCiD
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
© The Author(s) 2025.Two Parkinson’s disease subtypes—“Brain-First” and “Body-First”—have been proposed based on putative sites of onset. We examined whether “Body-First” markers relate to more symmetric striatal [123I]-FP-CIT uptake and whether imaging could discriminate the subtypes. In a retrospective cohort of 158 de novo PD patients imaged at diagnosis and followed for six years, patients were classified as “Body-First” if baseline REM sleep behavior disorder, constipation, or neurogenic orthostatic hypotension was present. DaTQUANT provided semiquantitative metrics; a radiomics-based classifier was also trained on DAT-SPECT images. Neither asymmetry indices nor other DaTQUANT measures differed between groups (all p > 0.05). Radiomics showed poor discrimination (AUC ≈ 0.46). Clinically, “Body-First” patients displayed a more adverse course, with higher MCI prevalence and greater MMSE decline and neuropsychiatric burden, whereas motor severity and complications were comparable between groups. These data suggest DAT-SPECT—conventional or radiomics-enhanced—does not separate proposed subtypes at diagnosis, although “Body-First” features forecast worse non-motor progression.
Author(s): Palermo G, Aghakhanyan G, Bellini G, Giannoni S, Vadi G, Francischello R, Nonne G, Tedeschi MG, Morganti R, Frosini D, Pavese N, Volterrani D, Ceravolo R
Publication type: Article
Publication status: Published
Journal: npj Parkinson's Disease
Year: 2025
Volume: 11
Issue: 1
Online publication date: 20/11/2025
Acceptance date: 25/09/2025
Date deposited: 01/12/2025
ISSN (electronic): 2373-8057
Publisher: Nature Research
URL: https://doi.org/10.1038/s41531-025-01164-z
DOI: 10.1038/s41531-025-01164-z
Data Access Statement: The datasets generated and/or analyzed during the current study contain sensitive patient information and are not publicly available due to privacy and ethical restrictions. De-identified derived data (e.g., DaTQUANT outputs, radiomics feature matrices, and summary statistics) can be shared upon reasonable request to the corresponding author and following approval by the local ethics committee(s) and execution of a data-sharing agreement. Analyses used standard, publicly available software (DaTQUANT®, PyRadiomics, XGBoost, R, and SPSS). Custom scripts used to preprocess data, extract features, and run the nested cross-validation workflow are available from the corresponding author upon reasonable request. Analyses used standard, publicly available software (DaTQUANT®, PyRadiomics, XGBoost, R, and SPSS). Custom scripts used to preprocess data, extract features, and run the nested cross-validation workflow are available from the corresponding author upon reasonable request.
Altmetrics provided by Altmetric
