TY - JOUR
T1 - MRI Clustering Reveals Three ALS Subtypes With Unique Neurodegeneration Patterns
AU - Tan, Harold H G
AU - Westeneng, Henk-Jan
AU - Nitert, Abram D
AU - van Veenhuijzen, Kevin
AU - Meier, Jil M
AU - van der Burgh, Hannelore K
AU - van Zandvoort, Martine J E
AU - van Es, Michael A
AU - Veldink, Jan H
AU - van den Berg, Leonard H
N1 - Funding Information:
The authors would like to thank all participants who volunteered for this study. Furthermore, we would like to thank Simone M.A. Vugts for her efforts in coordinating the study and for her help with data acquisition. This study was funded by the Netherlands ALS Foundation (Stichting ALS Nederland). This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement no 772376—EScORIAL).
Publisher Copyright:
© 2022 The Authors. Annals of Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association.
PY - 2022/12
Y1 - 2022/12
N2 - Objective: The purpose of this study was to identify subtypes of amyotrophic lateral sclerosis (ALS) by comparing patterns of neurodegeneration using brain magnetic resonance imaging (MRI) and explore their phenotypes. Methods: We performed T1-weighted and diffusion tensor imaging in 488 clinically well-characterized patients with ALS and 338 control subjects. Measurements of whole-brain cortical thickness and white matter connectome fractional anisotropy were adjusted for disease-unrelated variation. A probabilistic network-based clustering algorithm was used to divide patients into subgroups of similar neurodegeneration patterns. Clinical characteristics and cognitive profiles were assessed for each subgroup. In total, 512 follow-up scans were used to validate clustering results longitudinally. Results: The clustering algorithm divided patients with ALS into 3 subgroups of 187, 163, and 138 patients. All subgroups displayed involvement of the precentral gyrus and are characterized, respectively, by (1) pure motor involvement (pure motor cluster [PM]), (2) orbitofrontal and temporal involvement (frontotemporal cluster [FT]), and (3) involvement of the posterior cingulate cortex, parietal white matter, temporal operculum, and cerebellum (cingulate-parietal–temporal cluster [CPT]). These subgroups had significantly distinct clinical profiles regarding male-to-female ratio, age at symptom onset, and frequency of bulbar symptom onset. FT and CPT revealed higher rates of cognitive impairment on the Edinburgh cognitive and behavioral ALS screen (ECAS). Longitudinally, clustering remained stable: at 90.4% of their follow-up visits, patients clustered in the same subgroup as their baseline visit. Interpretation: ALS can manifest itself in 3 main patterns of cerebral neurodegeneration, each associated with distinct clinical characteristics and cognitive profiles. Besides the pure motor and frontotemporal dementia (FTD)-like variants of ALS, a new neuroimaging phenotype has emerged, characterized by posterior cingulate, parietal, temporal, and cerebellar involvement. ANN NEUROL 2022;92:1030–1045.
AB - Objective: The purpose of this study was to identify subtypes of amyotrophic lateral sclerosis (ALS) by comparing patterns of neurodegeneration using brain magnetic resonance imaging (MRI) and explore their phenotypes. Methods: We performed T1-weighted and diffusion tensor imaging in 488 clinically well-characterized patients with ALS and 338 control subjects. Measurements of whole-brain cortical thickness and white matter connectome fractional anisotropy were adjusted for disease-unrelated variation. A probabilistic network-based clustering algorithm was used to divide patients into subgroups of similar neurodegeneration patterns. Clinical characteristics and cognitive profiles were assessed for each subgroup. In total, 512 follow-up scans were used to validate clustering results longitudinally. Results: The clustering algorithm divided patients with ALS into 3 subgroups of 187, 163, and 138 patients. All subgroups displayed involvement of the precentral gyrus and are characterized, respectively, by (1) pure motor involvement (pure motor cluster [PM]), (2) orbitofrontal and temporal involvement (frontotemporal cluster [FT]), and (3) involvement of the posterior cingulate cortex, parietal white matter, temporal operculum, and cerebellum (cingulate-parietal–temporal cluster [CPT]). These subgroups had significantly distinct clinical profiles regarding male-to-female ratio, age at symptom onset, and frequency of bulbar symptom onset. FT and CPT revealed higher rates of cognitive impairment on the Edinburgh cognitive and behavioral ALS screen (ECAS). Longitudinally, clustering remained stable: at 90.4% of their follow-up visits, patients clustered in the same subgroup as their baseline visit. Interpretation: ALS can manifest itself in 3 main patterns of cerebral neurodegeneration, each associated with distinct clinical characteristics and cognitive profiles. Besides the pure motor and frontotemporal dementia (FTD)-like variants of ALS, a new neuroimaging phenotype has emerged, characterized by posterior cingulate, parietal, temporal, and cerebellar involvement. ANN NEUROL 2022;92:1030–1045.
UR - http://www.scopus.com/inward/record.url?scp=85138335156&partnerID=8YFLogxK
U2 - 10.1002/ana.26488
DO - 10.1002/ana.26488
M3 - Article
C2 - 36054734
SN - 0364-5134
VL - 92
SP - 1030
EP - 1045
JO - Annals of Neurology
JF - Annals of Neurology
IS - 6
ER -