MRI Signature of α-Synuclein Pathology in Asymptomatic Stages and a Memory Clinic Population

Laura E M Wisse; Nicola Spotorno; Marcello Rossi; Michel J Grothe; Angela Mammana; Pontus Tideman; Simone Baiardi; Olof Strandberg; Alice Ticca; Danielle van Westen; Niklas Mattsson-Carlgren; Sebastian Palmqvist; Erik Stomrud; Piero Parchi; Oskar Hansson

doi:10.1001/jamaneurol.2024.2713

. 2024 Jul 28;81(10):1051–1059. doi: 10.1001/jamaneurol.2024.2713

MRI Signature of α-Synuclein Pathology in Asymptomatic Stages and a Memory Clinic Population

Laura E M Wisse ^1,^✉, Nicola Spotorno ², Marcello Rossi ³, Michel J Grothe ^4,⁵, Angela Mammana ³, Pontus Tideman ^2,⁶, Simone Baiardi ⁷, Olof Strandberg ², Alice Ticca ⁷, Danielle van Westen ^8,⁹, Niklas Mattsson-Carlgren ^2,^10,¹¹, Sebastian Palmqvist ^2,⁶, Erik Stomrud ^2,⁶, Piero Parchi ^3,⁷, Oskar Hansson ^2,^6,^✉, for the Alzheimer’s Disease Neuroimaging Initiative

¹Department of Clinical Science Lund, Lund University, Lund, Sweden

²Clinical Memory Research Unit, Department of Clinical Sciences Malmö, Lund University, Lund, Sweden

³IRCCS, Istituto delle Scienze Neurologiche di Bologna, Bologna, Italy

⁴Reina Sofia Alzheimer Center, CIEN Foundation, Instituto de Salud Carlos III, Madrid, Spain

⁵Centro de Investigacion Biomédica en Red Sobre Enfermedades Neurodegenerativas, Instituto de Salud Carlos III, Madrid, Spain

⁶Memory Clinic, Skåne University Hospital, Malmö, Sweden

⁷Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy

⁸Department of Diagnostic Radiology, Clinical Sciences, Lund University, Lund, Sweden

⁹Image and Function, Skåne University Hospital, Lund, Sweden

¹⁰Department of Neurology, Skåne University Hospital, Lund, Sweden

¹¹Wallenberg Center for Molecular Medicine, Lund University, Lund, Sweden

Group Information: The Alzheimer’s Disease Neuroimaging Initiative members appear in Supplement 2.

Accepted for Publication: June 15, 2024.

Published Online: July 28, 2024. doi:10.1001/jamaneurol.2024.2713

^✉

Corresponding Authors: Laura E. M. Wisse, PhD, Department of Clinical Sciences Lund, Lund University, Klinikgatan 13B, 22242, Lund, Sweden (laura.wisse@med.lu.se); Oskar Hansson, MD, PhD, Memory Clinic, Skåne University Hospital, SE-205 02, Malmö, Sweden (oskar.hansson@med.lu.se).

Author Contributions: Drs Wisse and Spotorno had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Weiss and Spotorno are considered co–first authors and contributed equally to this work.

Concept and design: Wisse, Spotorno, Hansson.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Wisse, Spotorno, Hansson.

Critical review of the manuscript for important intellectual content: Spotorno, Rossi, Grothe, Mammana, Tideman, Baiardi, Strandberg, Ticca, van Westen, Mattsson-Carlgren, Palmqvist, Stomrud, Parchi.

Statistical analysis: Spotorno, Grothe.

Obtained funding: Wisse, Mattsson-Carlgren, Palmqvist, Parchi, Hansson.

Administrative, technical, or material support: Spotorno, Rossi, Grothe, Mammana, Tideman, Baiardi, Strandberg, Ticca, van Westen, Hansson.

Supervision: Parchi, Hansson.

Conflict of Interest Disclosures: Dr Wisse reported receiving grants from National Institute on Aging, Swedish Research Council, Strategic Research Area MultiPark, and Swedish Alzheimer Foundation during the conduct of the study. Dr Palmqvist reported receiving research support from ki:elements, Alzheimer’s Drug Discovery Foundation, and Avid and consultancy/speaker fees from Bioartic, Biogen, Esai, Lilly, and Roche. Dr Hansson reported having acquired research support (for the institution) from AVID Radiopharmaceuticals, Biogen, C2N Diagnostics, Eli Lilly, Eisai, Fujirebio, GE Healthcare, and Roche; in addition, in the past 2 years, he reported receiving consultancy/speaker fees from AC Immune, Alzpath, BioArctic, Biogen, Bristol Myers Squibb, Cerveau, Eisai, Eli Lilly, Fujirebio, Merck, Novartis, Novo Nordisk, Roche, Sanofi, and Siemens outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by the National Institute of Aging (R01-AG070592, R01AG083740), European Research Council (ADG-101096455), Alzheimer’s Association (ZEN24-1069572, SG-23-1061717), GHR Foundation, Swedish Research Council (2022-00900, 2022-00775, 2021-02219, 2018-02052), ERA PerMed (ERAPERMED2021-184), Knut and Alice Wallenberg Foundation (2022-0231), Strategic Research Area MultiPark (Multidisciplinary Research in Parkinson’s Disease) at Lund University, Swedish Alzheimer Foundation (AF-980907, AF-993465, AF-994229), EU Joint Programme Neurodegenerative Diseases (2019-03401), Wallenberg AI, Autonomous Systems and Software Program and Data-Driven Life Science Joint call for research projects (WASP/DDLS22-066), Swedish Brain Foundation (FO2021-0293, FO2023-0163), Parkinson Foundation of Sweden (1412/22), Cure Alzheimer’s fund, Rönström Family Foundation (FRS-0003), Konung Gustaf V:s och Drottning Victorias Frimurarestiftelse, Crafoord Foundation (20210690), Skåne University Hospital Foundation (2020-O000028), Regionalt Forskningsstöd (2022-1259) and Swedish federal government under the Avtal om Läkarutbildning och Forskning (ALF) agreement on medical education and research (2022-Projekt0080, 2022-Project0107), and grant CP19/00031 from the Miguel Servet Fellowship of the Instituto de Salud Carlos III (Dr Grothe). Data collection at the Laboratory of Neuropathology, IRCCS-ISNB, Bologna Italy, was supported by the grant Ricerca Finalizzata-2021-12374386, funded by the Ministero della Salute.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Group Information: The members of the Alzheimer’s Disease Neuroimaging Initiative appear in Supplement 2.

Meeting Presentation: This paper was presented at the Alzheimer’s Association International Conference; July 28, 2024; Philadelphia, Pennsylvania.

Data Sharing Statement: See Supplement 3.

Additional Contributions: We thank all the BioFINDER study team members as well as participants in the study and their family members for their dedication. Beyond usual salary, no financial compensation was received for these contributions.

^✉

Corresponding author.

PMCID: PMC11284633 PMID: 39068668

This cohort study investigates the association between cerebrospinal fluid seed amplification α-synuclein positivity and magnetic resonance imaging (MRI) structural measures across the continuum from clinically unimpaired to cognitively impaired individuals.

Key Points

Question

What is the association of cerebrospinal fluid seed amplification assay (SAA) α-synuclein positivity with magnetic resonance imaging structural measures across the continuum from clinically unimpaired to cognitively impaired individuals?

Findings

In this cohort study including 2961 participants from the BioFINDER-1 study, BioFINDER-2 study, and Alzheimer’s Disease Neuroimaging Initiative, SAA α-synuclein positivity was consistently associated with an atrophy pattern that mostly restricted nucleus basalis of Meynert (NBM) volume, even in asymptomatic stages.

Meaning

The clinical relevance of the restricted SAA α-synuclein–associated NBM atrophy pattern is highlighted by its contribution to α-synuclein–induced attention/executive function impairment but also suggests potential alternative pathways in which α-synuclein pathology leads to cognitive and functional impairments.

Abstract

Importance

The lack of an in vivo measure for α-synuclein (α-syn) pathology until recently has limited thorough characterization of its brain atrophy pattern, especially during early disease stages.

Objective

To assess the association of state-of-the-art cerebrospinal fluid (CSF) seed amplification assays (SAA) α-syn positivity (SAA α-syn+) with magnetic resonance imaging (MRI) structural measures, across the continuum from clinically unimpaired (CU) to cognitively impaired (CI) individuals, in 3 independent cohorts, and separately in CU and CI individuals, the latter reflecting a memory clinic population.

Design, Setting, and Participants

Cross-sectional data were used from the Swedish BioFINDER-2 study (inclusion, 2017-2023) as the discovery cohort and the Swedish BioFINDER-1 study (inclusion, 2007-2015) and Alzheimer’s Disease Neuroimaging Initiative (ADNI; inclusion 2005-2022) as replication cohorts. All cohorts are from multicenter studies, but the BioFINDER cohorts used 1 MRI scanner. CU and CI individuals fulfilling inclusion criteria and without missing data points in relevant metrics were included in the study. All analyses were performed from 2023 to 2024.

Exposures

Presence of α-syn pathology, estimated by baseline CSF SAA α-syn.

Main Outcomes and Measures

The primary outcomes were cross-sectional structural MRI measures either through voxel-based morphometry (VBM) or regions of interest (ROI) including an automated pipeline for cholinergic basal forebrain nuclei CH4/4p (nucleus basalis of Meynert [NBM]) and CH1/2/3. Secondary outcomes were domain-specific cross-sectional cognitive measures. Analyses were adjusted for CSF biomarkers of Alzheimer pathology.

Results

A total of 2961 participants were included in this study: 1388 (mean [SD] age, 71 [10] years; 702 female [51%]) from the BioFINDER-2 study, 752 (mean [SD] age, 72 [6] years; 406 female [54%]) from the BioFINDER-1 study, and 821 (mean [SD] age, 75 [8] years; 449 male [55%]) from ADNI. In the BioFINDER-2 study, VBM analyses in the whole cohort revealed a specific association between SAA α-syn+ and the cholinergic NBM, even when adjusting for Alzheimer copathology. ROI-based analyses in the BioFINDER-2 study focused on regions involved in the cholinergic system and confirmed that SAA α-syn+ was indeed independently associated with smaller NBM (β = −0.271; 95% CI, −0.399 to −0.142; P <.001) and CH1/2/3 volumes (β = −0.227; 95% CI, −0.377 to −0.076; P =.02). SAA α-syn+ was also independently associated with smaller NBM volumes in the separate CU (β = −0.360; 95% CI, −0.603 to −0.117; P =.03) and CI (β = −0.251; 95% CI, −0.408 to −0.095; P =.02) groups. Overall, the association between SAA α-syn+ and NBM volume was replicated in the BioFINDER-1 study and ADNI cohort. In CI individuals, NBM volumes partially mediated the association of SAA α-syn+ with attention/executive impairments in all cohorts (BioFINDER-2, β = −0.017; proportion-mediated effect, 7%; P =.04; BioFINDER-1, β = −0.096; proportion-mediated effect, 19%; P =.04; ADNI, β = −0.061; proportion-mediated effect, 20%; P =.007).

Conclusions and Relevance

In this cohort study, SAA α-syn+ was consistently associated with NBM atrophy already during asymptomatic stages. Further, in memory clinic CI populations, SAA α-syn+ was associated with NBM atrophy, which partially mediated α-syn–induced attention/executive impairment.

Introduction

Lewy body (LB) pathology consists of pathological aggregates of α-synuclein (α-syn) forming Lewy bodies and Lewy neurites.¹ LB disease, also referred to as neuronal α-synuclein disease,² includes Parkinson disease (PD) and dementia with LB (DLB), initially characterized by motor symptoms,³ and by cognitive and behavioral changes,⁴ respectively. LB pathology is associated with different spreading patterns,⁵ which can start in the olfactory bulb, brainstem, or amygdala and with later stages including limbic or neocortical areas. In both PD and DLB, α-syn deposition patterns are varied but frequently involve limbic and neocortical regions,^5,6 whereas α-syn copathology in Alzheimer disease (AD) remains often more restricted to regions such as the olfactory bulb and amygdala.^5,7 Interestingly, cholinergic regions, such as the nucleus basalis of Meynert (NBM), are also often affected by α-syn pathology.⁸ Less is known about α-syn pathology in clinically normal older individuals, although it has been reported in 5% to 12%^2,9,10,11,12 and most commonly in the brainstem, amygdala, and olfactory bulb.

α-Syn pathology (measured postmortem) has not been consistently associated with brain atrophy on magnetic resonance imaging (MRI),^13,14,15,16 but significant associations often include medial temporal lobe (MTL) regions. Nonsignificant findings are potentially due to limited sample sizes, a limited set of brain regions included, and the time difference between the MRI and autopsy. In cases with concomitant AD pathology, α-syn pathology has been associated with MTL and parietal atrophy, albeit not consistently.^14,17,18 Less work has been done on the association with atrophy in clinically unimpaired (CU) individuals, with only 1 study¹⁹ reporting a negative association of α-syn pathology with amygdala volumes.

With the development of a cerebrospinal fluid (CSF) marker for misfolded α-syn pathology using an in vitro seed amplification assay (SAA), it is now feasible to characterize the SAA α-syn–associated atrophy pattern in vivo in large-scale populations,²⁰ including those that are more difficult to identify in postmortem studies. We aimed to investigate the association of a CSF SAA α-syn measure with structural MRI measures using an unbiased voxel-based morphometry (VBM) approach, adjusting for Alzheimer copathology. Guided by the VBM results, a selected set of regions was investigated in the following: (1) the whole cohort, (2) CU individuals, and (3) a memory clinic of cognitively impaired (CI) individuals. The primary analyses were done in the Swedish BioFINDER-2 study cohort, and the results were validated in 2 independent cohorts (that of the Swedish BioFINDER-1 study and the Alzheimer Disease Neuroimaging Initiative [ADNI]).

Methods

This cohort study was approved by the Swedish Ethical Review Authority. All patients included provided written informed consent. Details on the study population, CSF biomarkers, MRI methodology, and analyses are available in the eMethods in Supplement 1. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

Study Population

Study participants were cognitively and neurologically unimpaired at baseline (in the Swedish BioFINDER-1 study,²¹ the Swedish BioFINDER-2 study,²² and the ADNI cohort) or CI in all cohorts. In the BioFINDER-1 and BioFINDER-2 studies, only CU individuals who did not show parkinsonian symptoms within 1 year of follow-up were included. In the CI group, individuals with mild cognitive impairments or dementia were included. For the BioFINDER-1 and BioFINDER-2 studies, all participants provided written informed consent. Ethical approval was given by the Swedish Ethical Review Authority. For ADNI, the study was approved after ethical review of each site’s local review board, and all participants provided informed written consent. Participant race and ethnicity information was not collected per study protocol.

CSF Markers

Concentrations of CSF amyloid-β (Aβ)42,²³ Aβ40, and phosphorylated tau 181²⁴ (p-tau181) were measured with the Roche Elecsys assays.²⁵ Aβ40 was not available in ADNI.

α-Syn real-time quaking-induced conversion analyses were performed using an in vitro SAA generating a binary assessment of the presence of α-syn pathology (SAA α-syn+/−), albeit with different methods for the BioFINDER⁹ and ADNI cohorts.²⁶ Note that both methods had very similar performance in clinical and neuropathologically verified cohorts^26,27,28,29 (eMethods in Supplement 1).

MRI Methodology

VBM was performed on T1-weighted images in the discovery cohort (BioFINDER-2 cohort) using Statistical Parametric Mapping software (SPM). All images were processed using FreeSurfer, version 6.0 (Laboratory for Computational Neuroimaging at the Athinoula A. Martinos Center for Biomedical Imaging), to obtain cortical thickness and subcortical volumes.

Volume measures of the cholinergic basal forebrain were obtained using an automated method developed by Grothe et al.³⁰ The atlas contains the subdivisions CH1/2/3 and CH4/4p, where CH4/4p includes most of the NBM and CH1/2/3 the remaining basal forebrain nuclei.³¹

Cognitive Tests

z Scores of global cognition and of the memory, visuospatial, and attention/executive domains were used for all participants based on the distribution of CU participants who were Aβ−.

Statistical Analysis

All main analyses were performed in the BioFINDER-2 cohort (discovery cohort), and significant results were replicated in the BioFINDER-1 and ADNI cohorts. We selected BioFINDER-2 as the discovery cohort because all participants underwent MRI in the same center on the same scanner using a state-of-the-art protocol. In the whole cohort in BioFINDER-2, VBM analyses were performed with age, sex, intracranial volume, cognitive status, CSF Aβ42/40 level, and CSF p-tau181 level as covariates. Statistical significance was set to the false discovery rate threshold of 0.05, and minimum cluster-extent of 100 voxels.

For the region-of-interest (ROI) analyses, the association between SAA α-syn+ and MRI was assessed in a multiple regression framework. The same covariates were included; intracranial volume was only included when investigating volumetrics, site when analyzing ADNI, and cognitive status when focusing on the whole cohort. We corrected for site when analyzing ADNI data to account for any differences in scanning, which might occur at different sites, including MRI protocol, scanner brand, and hardware. All ROI-based analyses in the discovery cohort were corrected for multiple comparisons using the Bonferroni test, and reported P values were corrected for multiple comparisons.

Mediation analyses were performed with the R package mediation (R Project for Statistical Computing) to assess whether structural measures mediated an association between SAA α-syn+ and cognitive measures.

Statistical significance was set at a 2-tailed P value <.05. All analyses were performed from 2023 to 2024 using R, version 4.2.2, or Python, version 3.9.16 (open source).

Results

Demographics and clinical information per cohort are shown in the Table. A total of 2961 participants were included in this study: 1388 (mean [SD] age, 71 [10] years; 702 female [51%]; 686 male [49%]) from the BioFINDER-2 study, 752 (mean [SD] age, 72 [6] years; 406 female [54%]; 346 male [46%]) from the BioFINDER-1 study, and 821 (mean [SD] age, 75 [8] years; 374 female [45%]; 449 male [55%]) from ADNI. Compared with the BioFINDER-1 and ADNI cohorts, most notable is the slightly younger age of the CU group in BioFINDER-2 (mean [SD] age, 68.4 [10.7] years vs 72.5 [5.5] and 75.6 [6.5], respectively). Compared with the BioFINDER-2 and BioFINDER-1 cohorts, notable is the higher percentage CSF SAA α-syn+ of the CU group in ADNI (No. [%], 44 [18.6%] vs 46 [6.7%] and 45 [8.6%], respectively). A description of the CI group can be found in the eResults in Supplement 1.

Table. Demographics and Clinical Information of the 3 Cohorts.

Variable	BioFINDER-2		BioFINDER-1		ADNI
Variable	CU (n = 691)	CI (n = 697)	CU (n = 522)	CI (n = 230)	CU (n = 236)	CI (n = 585)
Age, mean (SD), y	68.4 (10.7)	72.7 (7.4)^a	72.5 (5.5)	71.6 (5.5)^a	75.6 (6.5)	74.3 (7.9)^a
Sex, No. (%)
Female	384 (56)	318 (46)^a	313 (60)	93 (40)^a	125 (53)	247 (42)^a
Male	307 (44)	379 (54)	209 (40)	137 (60)	111 (47)	338 (58)
Education, mean (SD), y	13 (4)	12 (4)^a	12 (4)	11 (3)^a	16 (3)	16 (3)^a
MMSE, mean (SD), points	29 (1)	24 (4)^a	29 (1)	27 (2)^a	29 (1)	26 (4)^a
Global cognition (z scored), mean (SD)	0.69 (0.51)	−0.85 (0.78)^a	0.40 (0.73)	−1.15 (0.75)^a	0.89 (0.36)	−0.36 (0.95)^a
Memory (z scored), mean (SD)	−0.72 (0.57)	0.77 (0.76)^a	−0.42 (0.74)	0.99 (1.00)^a	−0.90 (0.59)	0.36 (0.90)^a
Executive functions (z scored), mean (SD)	0.56 (0.85)	−0.67 (0.71)^a	0.27 (0.84)	−0.86 (0.98)^a	0.66 (0.64)	−0.27 (1.00)^a
CSF Aβ42/40, mean (SD)	0.094 (0.03)	0.067 (0.03)^a	0.081 (0.03)	0.063 (0.03)^a	NA	NA
CSF Aβ42, pg mL⁻¹, mean (SD)	1797 (788)	1201 (689)^a	1453 (687)	1056 (661)^a	1232.9 (430)	862.4 (429)^a
CSF Aβ42/40 (p-tau/Aβ42 for ADNI) positivity, No. (%)	211 (31)	483 (69)	163 (31)	144 (63)	51 (22)	370 (63)
CSF p-tau181, pg mL⁻¹, mean (SD)	19.2 (8.7)	28.8 (16.9)^a	20.9 (9.8)	25.9 (14.0)^a	22.3 (10.0)	30.1 (15.4)^a
CSF SAA α-syn positivity (LB pathology), No. (%)	46 (6.7)	172 (24.7)^a	45 (8.6)	49 (21.3)^a	44 (18.6)	157 (26.8)^a

Open in a new tab

Abbreviations: Aβ, amyloid-β; α-syn, α-synuclein; ADNI, Alzheimer Disease Neuroimaging Initiative; BioFINDER-1, the Swedish BioFINDER-1 study; BioFINDER-2, the Swedish BioFINDER-2 study; CI, cognitively unimpaired; CSF, cerebrospinal fluid; CU, clinically unimpaired; LB, Lewy body; MMSE, Mini Mental State Examination; mPACC5, modified Preclinical Alzheimer Cognitive Composite 5; NA, not applicable; p-tau, phosphorylated tau; SAA, seed amplification assay.

^{^a}

Statistical significance between CU and CI at P < .05. In ADNI CSF Aβ42/40 was available only is a subset of participants; therefore, CSF Aβ42 was used as a marker of Aβ pathology. Cutoffs for Aβ positivity can be found in the eMethods in Supplement 1. Global cognition: mPACC5; memory: delayed recall task included in the Alzheimer’s Disease Assessment Scale (ADAS) in the BioFINDER-1 and BioFINDER-2 cohorts and the harmonized composite memory function score in ADNI; executive functions: the symbol digits modalities test in the BioFINDER-1 and BioFINDER-2 cohorts and harmonized composite executive function score in ADNI. The cognitive tests were not available for all participants. Number of participants with available scores for the different cognitive domains: BioFINDER-2: global cognition (CU = 676; CI = 547), memory (CU = 684; CI = 643), executive functions (CU = 679; CI = 569); BioFINDER-1: global cognition (CU = 412; CI = 145), memory (CU = 520; CI = 221), executive functions (CU = 405; CI = 126). A description of the cognitively impaired group can be found in the eMethods in Supplement 1.

Association of α-Syn Pathology and Structural MRI Measures

In the whole BioFINDER-2 cohort (CU and CI), VBM analyses revealed a significant cluster that roughly overlapped with the cholinergic NBM nuclei (Figure 1A). Following this result, the ROI analyses were focused on regions involved in the cholinergic system,^31,32 most of which also have reported vulnerability to α-syn pathology.^6,33 ROI-based analyses revealed significantly smaller NBM volumes (β = −0.271; 95% CI, −0.399 to −0.142; family-wise error P <.001) (eTables 1 to 9 in Supplement 1 contain uncorrected P values) and CH1/2/3 (β = −0.227; 95% CI, −0.377 to −0.076; family-wise error P =.02) in individuals who were SAA α-syn+ compared with those who were α-syn− (Figure 1B and eTable 1 in Supplement 1). Only the findings for NBM were replicated in BioFINDER-1 (β = −0.329; 95% CI, −0.534 to −0.123; P =.002) and ADNI (β = −0.216; 95% CI, −0.364 to −0.067; P =.005) (Figure 1C and D and eTables 2-3 in Supplement 1).

Figure 1. — A, VBM analyses in the whole cohort (cognitively unimpaired and cognitively impaired) of the BioFINDER-2 study. VBM results are shown in orange (P < .05, after false discovery rate correction). Regions in purple and blue depict the atlas-based basal forebrain nuclei. B-D, ROI-based analyses of the association of cerebrospinal fluid SAA α-syn+ with MRI structural measures in the BioFINDER-2 study (B) and replicated in the BioFINDER-1 study (C) and the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (D). The violin plot representing the data distribution is overlaid on the box plot showing the IQR and the median of the distribution. The red line indicates a significant difference between SAA α-syn negative (neg) and positive (pos) individuals (P < .05) after Bonferroni correction over all 8 regions. CH1/2/3 reflects the nuclei in the basal forebrain, not including the NBM; NBM, nucleus basalis of Meynert.

When restricting analyses to the CU population, SAA α-syn+ was associated with significantly smaller NBM volumes (BioFINDER-2, β = −0.360; 95% CI, −0.603 to −0.117; family-wise error P =.03) (Figure 2A and eTable 1 in Supplement 1). This association was replicated in BioFINDER-1 (β = −0.286; 95% CI, −0.570 to −0.001; P = .05) but not in ADNI (Figure 2B and C and eTables 2-3 in Supplement 1).

Figure 2. — A-C, ROI-based analyses in CU individuals. D-F, ROI-based analyses in CI individuals. Main analyses are performed in the BioFINDER-2 study and replicated in the BioFINDER-1 and Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohorts. The violin plot representing the data distribution is overlaid on the box plot showing the IQR and the median of the distribution. The red line indicates a significant difference between SAA α-syn negative (neg) and positive (pos) individuals (P < .05) after Bonferroni correction over all 8 regions. CH1/2/3 reflects the nuclei in the basal forebrain, not including the NBM; NBM, nucleus basalis of Meynert.

In CI, a significant association between SAA α-syn+ and smaller NBM volumes was found in BioFINDER-2 (β = −0.251; 95% CI, −0.408 to −0.095; family-wise error P =.02) and replicated in BioFINDER-1 (β = −0.367; 95% CI, −0.677 to −0.057; P = .02) and ADNI (β = −0.242; 95% CI, −0.417 to −0.066; P = .007) (Figure 2E and F and eTables 1-3 in Supplement 1).

Role of Structural MRI Measures in the Association of α-Syn+ With Cognition

In CU, neither SAA α-syn+ nor NBM volumes were significantly associated with cognitive functioning (eTables 4-5 in Supplement 1).

In CI, attention/executive functioning was the only cognitive domain that was associated with both SAA α-syn+ (β = −0.231; 95% CI, −0.367 to −0.096; family-wise error P =.006) and NBM volume (β = 0.107; 95% CI, 0.043-0.170; family-wise error P =.008) in BioFINDER-2 (eTables 4-5 in Supplement 1). This was replicated in BioFINDER-1 (α-syn, β = −0.507; 95% CI, −0.891 to −0.124; P =.01; NBM, β = 0.237; 95% CI, 0.085-0.389; P =.003) and ADNI (α-syn, β = −0.305; 95% CI, −0.469 to −0.141; P < .001; NBM, β = 0.283; 95% CI, 0.207-0.361; P < .001) (eTables 6-7 in Supplement 1). Mediation models showed that NBM volumes partially mediated the association of SAA α-syn+ with attention/executive functioning in all 3 cohorts (BioFINDER-2, β = −0.017; proportion-mediated effect = 7%; P =.04; BioFINDER-1, β = −0.096; proportion-mediated effect = 19%; P =.04; ADNI, β = −0.061; proportion-mediated effect = 20%; P =.007) (Figure 3).

Figure 3. — The symbol digit modalities test is used in BioFINDER-1 and BioFINDER-2 cohorts, and the harmonized composite executive function score is used in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort.

^aMediation models are shown for the BioFINDER-1, BioFINDER-2, and ADNI cohorts.

Sensitivity Analyses

Post hoc power analyses revealed that power for detecting an association of SAA α-syn+ with NBM volumes was limited in ADNI for the CU group and could potentially explain the null finding. Post hoc power analyses for the other regions and groups are also shown in eFigure 1 in Supplement 1.

We performed sensitivity analyses in the BioFINDER-2 cohort using a temporal meta-ROI encompassing regions covering Braak stage I to IV (referred to as Cho I-IV³⁴) standardized uptake value ratio tau positron emission tomography (tau-PET) instead of CSF p-tau as tau-PET tracks disease severity more closely (eMethods in Supplement 1 contains more information on tau-PET). For 7 cases, tau-PET was missing. Similar results were found, although SAA α-syn+ was also associated with MTL regions after Bonferroni correction, whereas these were only trends when correcting for CSF p-tau. More specifically, in the whole cohort, significantly smaller NBM (β = −0.268; 95% CI, −0.395 to −0.142; family-wise error P <.001), CH1/2/3 (β = −0.216; 95% CI, −0.367 to −0.066; family-wise error P =.03), and amygdala volumes (β = −0.148; 95% CI, −0.252 to −0.043; family-wise error P =.04) were found in individuals with SAA α-syn+ compared with those with α-syn− (eTable 8 in Supplement 1). In the CU group, significantly smaller NBM volumes (β = −0.364; 95% CI, −0.605 to −0.124; family-wise error P =.02) and enthorinal cortex thickness (β = −0.276; 95% CI, −0.467 to −0.085; family-wise error P =.03) were found in individuals with SAA α-syn+ compared with those who were α-syn− (eTable 8 in Supplement 1). In the CI group, only NBM volumes (β = −0.244; 95% CI, −0.397 to −0.091; family-wise error P =.01) were associated with SAA α-syn+. We could not replicate these findings in BioFINDER-1 or ADNI because tau-PET was not available in a sufficient number of cases with α-syn SAA testing done in CSF samples collected close in time to the PET examinations.

We also conducted all ROI analyses in individuals negative for Aβ in all 3 cohorts, to negate the effect of AD pathology in a different way (ie, instead of adjusting for AD CSF biomarkers). We only performed these analyses in the whole cohort as we had limited power for further subgroupings. In the BioFINDER-2 cohort, significantly smaller NBM volumes (β = −0.345; 95% CI, −0.545 to −0.145; family-wise error P =.005) (eTable 9 in Supplement 1) were found in individuals with SAA α-syn+ compared with those who were α-syn−, which was replicated in the BioFINDER-1 cohort (β = −0.388; 95% CI, −0.677 to −0.100; P =.008) and ADNI (β = −0.256; 95% CI, = −0.484 to −0.027; P =.03). These results largely replicate the results of the main analyses.

Finally, receiver operating characteristic analyses were performed to assess the potential discriminatory power of NBM volume. The area under the curve (AUC) revealed a limited single-participant diagnostic utility in distinguishing between SAA α-syn+ and α-syn− groups (AUCs in all cohorts below 0.75) (eFigure 2 in Supplement 1).

Discussion

Using state-of-the art in vivo biomarkers of α-syn+, we found a significant and consistent SAA α-syn–associated atrophy pattern localized to the NBM in both CU and CI individuals. α-Syn pathology was, in fact, first described in the NBM by Friedrich Lewy³⁵ and occurs already in asymptomatic cases in the NBM.³³ A recent imaging-pathologic association study in patients with AD found LB pathology to be a primary pathological correlate of in vivo NBM atrophy on MRI.³⁶ Additionally, NBM atrophy has consistently been reported in patients with clinically defined PD and DLB.^37,38,39 The present results suggest that this occurs already during asymptomatic disease stages, independently from AD copathology.

The association between SAA α-syn+ and smaller NBM volumes was consistently found in the CI group in all 3 cohorts, whereas SAA α-syn+ did not appear to contribute to the more widespread atrophy often observed in CI memory clinic populations. Interestingly, the observed smaller NBM volumes still contributed to the SAA α-syn–associated attention/executive impairments in this group. Indeed, both α-syn pathology^40,41 and the NBM, or cholinergic system in general,⁴² have been associated with attention/executive impairments. However, it should be noted that NBM volume only explained a small portion of the association of SAA α-syn+ with attention/executive functioning. Although this could partly be explained by the known imprecision of NBM volumetrics and cognitive tests, it is likely that NBM volume is not the sole mediating factor but rather that downstream microstructural and functional changes in the NBM and associated networks might explain a larger portion of the α-syn–induced pathway to attention/executive impairments. Other imaging techniques or sequences might, therefore, be better suited to identify additional downstream neurodegenerative effects of α-syn pathology in vivo.

Although the use of a discovery cohort and 2 replication cohorts, which differ in terms of demographics, dementia diagnoses, and SAA methods, can be considered a strength and speaks to the replicability of our reported association of SAA α-syn + with NBM atrophy, it can also be considered a limitation. We cannot preclude that a different SAA α-syn-related atrophy pattern would be found if one would select a different discovery cohort, and the current study cannot claim that the NBM is the only location of α-syn–associated atrophy. The same holds for our findings regarding attention/executive function impairment. To investigate this further, we performed exploratory VBM analyses in BioFINDER-1 and ADNI (eFigure 3 in Supplement 1), which replicated our findings for the NBM (although this did not survive correction for multiple comparisons for BioFINDER-1 likely due to low statistical power in that cohort) but also showed more widespread results also including significant confluent clusters in the MTL in ADNI.

Previous neuropathology studies on α-syn–associated atrophy patterns indeed often included the MTL^13,14,15,16 but have been inconsistent. An SAA α-syn–associated MTL atrophy pattern was not replicated here in the main analyses. However, a significant association of SAA α-syn+ with entorhinal cortex thickness before false discovery rate correction was found in the CU group, and several MTL subregions trended toward significance in the whole cohort in BioFINDER-2. Moreover, when correcting for tau-PET instead of CSF–p-tau, SAA α-syn+ was also associated with amygdala volumes in the whole cohort and with entorhinal cortex thickness in the CU group in the BioFINDER-2 study. Finally, exploratory VBM analyses also showed an association between SAA α-syn+ and several MTL regions, including the amygdala and entorhinal cortex in the whole cohort in ADNI. Together these findings suggest that the α-syn-related atrophy pattern, although still restricted, potentially exceeds the NBM and also includes the MTL. It is unclear why the effects were relatively stronger in the CU group, but this finding warrants further research in different study populations.

Most interestingly, the SAA α-syn–associated atrophy pattern was restricted, which fits with the lack of a clear and consistent atrophy pattern reported in previous literature^{9,13,14,15,16} and the fact that NBM was generally not included in previous studies. A similarly selective atrophy of the NBM was also reported in an early VBM study of patients with clinically defined DLB.⁴³

Limitations

This study had some limitations. It should be noted that most extant studies (including the current) only investigated a set of selected regions and did not study smaller brainstem regions which have often been reported to accumulate α-syn pathology early on.^13,14,15,16 The restricted atrophy pattern also fits with the suggestion that α-syn pathology is more associated with synaptic than neuron loss.⁴⁴ Other imaging techniques or sequences might, therefore, be better suited to identify additional downstream neurodegenerative effects of α-syn pathology in vivo and they may provide metrics with higher single-subject diagnostic utility, which appeared to be limited for the NBM volume.

Conclusions

In summary, this cohort study was the first study, to our knowledge, of an in vivo–identified atrophy pattern found to be associated with SAA α-syn+, which already occurs in asymptomatic disease stages. Although SAA α-syn–associated atrophy remains largely restricted to the NBM in CI individuals, its clinical relevance is highlighted by its contribution to α-syn–induced attention/executive function impairment. Moreover, these findings can help with the interpretation of atrophy patterns identified on MRI in the context of a memory clinic population.

Supplement 1.

eMethods.

eResults. Description of Cognitively Impaired Groups

eTable 1. Region of Interest Analyses Associating SAA α-syn+ With MRI Structural Measures in the Whole Cohort, CU and CI Subgroups in BioFINDER-2

eTable 2. Region of Interest Analyses Associating SAA α-syn+ With MRI Structural Measures in the Whole Cohort, CU and CI Subgroups in BioFINDER-1

eTable 3. Region of Interest Analyses Associating SAA α-syn+ With MRI Structural Measures in the Whole Cohort, CU and CI Subgroups in ADNI

eTable 4. Association Between SAA α-syn+ and Cognitive Performance in the CU and CI Groups in BioFINDER-2

eTable 5. Association Between NBM Volume and Cognitive Performance in the CU and CI Groups in BioFINDER-2

eTable 6. Association Between SAA α-syn+ or NBM Volume and Cognitive Performance in the CI Group in BioFINDER-1 (Only Replication of Statistically Significant Results in BioFINDER-2)

eTable 7. Association Between SAA α-syn+ or NBM Volume and Cognitive Performance in the CI Group in ADNI (Only Replication of Statistically Significant Results in BioFINDER-2)

eTable 8. Region of Interest Analyses Associating SAA α-syn+ With MRI Structural Measures in the Whole Cohort, CU and CI Subgroups in BioFINDER-2, While Correcting for Tau-PET Cho I-IV SUVR Instead of CSF p-Tau

eTable 9. Region of Interest Analyses Associating SAA α-syn+ With MRI Structural Measures in the Whole Cohort, CU and CI Subgroups in BioFINDER-2, BioFINDER-1 and ADNI

eFigure 1. Post Hoc Power Analyses for NBM (A) and CH1/2/3 (B) for the Whole Cohorts, for NBM for Clinically Unimpaired (C) and NBM for Cognitively Impaired Individuals (D) to Test to What Extent the Two Replication Cohorts Were Suited for Detecting the Same Effects Found in the BioFINDER-2 Cohort

eFigure 2. Receiver Operating Characteristic (ROC) Analyses to Investigate the Diagnostic Utility of NBM Volume in Distinguishing Between α-syn+ and α-syn- Individuals

eFigure 3. VBM Analyses in the Whole Cohort (CU and CI) of BioFINDER-1 (A) and ADNI (B)

eReferences

jamaneurol-e242713-s001.pdf^{(1.2MB, pdf)}

Supplement 2.

Nonauthor Collaborators. Alzheimer’s Disease Neuroimaging Initiative.

jamaneurol-e242713-s002.pdf^{(217.5KB, pdf)}

Supplement 3.

Data Sharing Statement.

jamaneurol-e242713-s003.pdf^{(17.2KB, pdf)}

References

1.Braak H, Del Tredici K, Rüb U, de Vos RAI, Jansen Steur ENH, Braak E. Staging of brain pathology related to sporadic Parkinson disease. Neurobiol Aging. 2003;24(2):197-211. doi: 10.1016/S0197-4580(02)00065-9 [DOI] [PubMed] [Google Scholar]
2.Simuni T, Chahine LM, Poston K, et al. A biological definition of neuronal α-synuclein disease: toward an integrated staging system for research. Lancet Neurol. 2024;23(2):178-190. doi: 10.1016/S1474-4422(23)00405-2 [DOI] [PubMed] [Google Scholar]
3.Taylor JP, McKeith IG, Burn DJ, et al. New evidence on the management of Lewy body dementia. Lancet Neurol. 2020;19(2):157-169. doi: 10.1016/S1474-4422(19)30153-X [DOI] [PMC free article] [PubMed] [Google Scholar]
4.McKeith IG, Dickson DW, Lowe J, et al. ; Consortium on DLB . Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65(12):1863-1872. doi: 10.1212/01.wnl.0000187889.17253.b1 [DOI] [PubMed] [Google Scholar]
5.Attems J, Toledo JB, Walker L, et al. Neuropathological consensus criteria for the evaluation of Lewy pathology in post-mortem brains: a multicenter study. Acta Neuropathol. 2021;141(2):159-172. doi: 10.1007/s00401-020-02255-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Beach TG, Adler CH, Lue L, et al. ; Arizona Parkinson’s Disease Consortium . Unified staging system for Lewy body disorders: correlation with nigrostriatal degeneration, cognitive impairment and motor dysfunction. Acta Neuropathol. 2009;117(6):613-634. doi: 10.1007/s00401-009-0538-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Toledo JB, Gopal P, Raible K, et al. Pathological α-synuclein distribution in subjects with coincident Alzheimer and Lewy body pathology. Acta Neuropathol. 2016;131(3):393-409. doi: 10.1007/s00401-015-1526-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Pasquini J, Brooks DJ, Pavese N. The cholinergic brain in Parkinson disease. Mov Disord Clin Pract. 2021;8(7):1012-1026. doi: 10.1002/mdc3.13319 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Palmqvist S, Rossi M, Hall S, et al. Cognitive effects of Lewy body pathology in clinically unimpaired individuals. Nat Med. 2023;29(8):1971-1978. doi: 10.1038/s41591-023-02450-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Klos KJ, Ahlskog JE, Josephs KA, et al. α-Synuclein pathology in the spinal cords of neurologically asymptomatic aged individuals. Neurology. 2006;66(7):1100-1102. doi: 10.1212/01.wnl.0000204179.88955.fa [DOI] [PubMed] [Google Scholar]
11.Parkkinen L, Soininen H, Alafuzoff I. Regional distribution of α-synuclein pathology in unimpaired aging and Alzheimer disease. J Neuropathol Exp Neurol. 2003;62(4):363-367. doi: 10.1093/jnen/62.4.363 [DOI] [PubMed] [Google Scholar]
12.Gibb WR, Lees AJ. The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson disease. J Neurol Neurosurg Psychiatry. 1988;51(6):745-752. doi: 10.1136/jnnp.51.6.745 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Bejanin A, Murray ME, Martin P, et al. Antemortem volume loss mirrors TDP-43 staging in older adults with non-frontotemporal lobar degeneration. Brain. 2019;142(11):3621-3635. doi: 10.1093/brain/awz277 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.de Flores R, Wisse LEM, Das SR, et al. Contribution of mixed pathology to medial temporal lobe atrophy in Alzheimer disease. Alzheimers Dement. 2020;16(6):843-852. doi: 10.1002/alz.12079 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Harding AJ, Broe GA, Halliday GM. Visual hallucinations in Lewy body disease relate to Lewy bodies in the temporal lobe. Brain. 2002;125(pt 2):391-403. doi: 10.1093/brain/awf033 [DOI] [PubMed] [Google Scholar]
16.Woodworth DC, Sheikh-Bahaei N, Scambray KA, et al. ; Alzheimer’s Disease Neuroimaging Initiative . Dementia is associated with medial temporal atrophy even after accounting for neuropathologies. Brain Commun. 2022;4(2):fcac052. doi: 10.1093/braincomms/fcac052 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Toledo JB, Cairns NJ, Da X, et al. ; Alzheimer’s Disease neuroimaging Initiative (ADNI) . Clinical and multimodal biomarker correlates of ADNI neuropathological findings. Acta Neuropathol Commun. 2013;1(1):65. doi: 10.1186/2051-5960-1-65 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Dawe RJ, Bennett DA, Schneider JA, Arfanakis K. Neuropathologic correlates of hippocampal atrophy in the elderly: a clinical, pathologic, postmortem MRI study. PLoS One. 2011;6(10):e26286. doi: 10.1371/journal.pone.0026286 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Wennberg AM, Whitwell JL, Tosakulwong N, et al. The influence of tau, amyloid, alpha-synuclein, TDP-43, and vascular pathology in clinically normal elderly individuals. Neurobiol Aging. 2019;77:26-36. doi: 10.1016/j.neurobiolaging.2019.01.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Candelise N, Baiardi S, Franceschini A, Rossi M, Parchi P. Towards an improved early diagnosis of neurodegenerative diseases: the emerging role of in vitro conversion assays for protein amyloids. Acta Neuropathol Commun. 2020;8(1):117. doi: 10.1186/s40478-020-00990-x [DOI] [PMC free article] [PubMed] [Google Scholar]
21.The Swedish BioFINDER study. ClinicalTrials.gov identifier: NCT01208675. Updated November 3, 2023. Accessed November 24, 2023. https://clinicaltrials.gov/study/NCT01208675
22.The Swedish BioFINDER 2 study (BioFINDER2). ClinicalTrials.gov identifier: NCT03174938. Updated November 7, 2023. Accessed November 24, 2023. https://clinicaltrials.gov/study/NCT03174938
23.Bittner T, Zetterberg H, Teunissen CE, et al. Technical performance of a novel, fully automated electrochemiluminescence immunoassay for the quantitation of β-amyloid (1-42) in human cerebrospinal fluid. Alzheimers Dement. 2016;12(5):517-526. doi: 10.1016/j.jalz.2015.09.009 [DOI] [PubMed] [Google Scholar]
24.Lifke V, Kollmorgen G, Manuilova E, et al. Elecsys total-tau and phospho-tau (181P) CSF assays: analytical performance of the novel, fully automated immunoassays for quantification of tau proteins in human cerebrospinal fluid. Clin Biochem. 2019;72:30-38. doi: 10.1016/j.clinbiochem.2019.05.005 [DOI] [PubMed] [Google Scholar]
25.Hansson O, Seibyl J, Stomrud E, et al. ; Swedish BioFINDER study group; Alzheimer’s Disease Neuroimaging Initiative . CSF biomarkers of Alzheimer’s disease concord with amyloid-β PET and predict clinical progression: A study of fully automated immunoassays in BioFINDER and ADNI cohorts. Alzheimers Dement. 2018;14(11):1470-1481. doi: 10.1016/j.jalz.2018.01.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Arnold MR, Coughlin DG, Brumbach BH, et al. α-Synuclein seed amplification in CSF and brain from patients with different brain distributions of pathological α-synuclein in the context of copathology and non-LBD diagnoses. Ann Neurol. 2022;92(4):650-662. doi: 10.1002/ana.26453 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Hall S, Orrù CD, Serrano GE, et al. Performance of α-synuclein RT-QuIC in relation to neuropathological staging of Lewy body disease. Acta Neuropathol Commun. 2022;10(1):90. doi: 10.1186/s40478-022-01388-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Russo MJ, Orru CD, Concha-Marambio L, et al. High diagnostic performance of independent alpha-synuclein seed amplification assays for detection of early Parkinson disease. Acta Neuropathol Commun. 2021;9(1):179. doi: 10.1186/s40478-021-01282-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Bentivenga GM, Mammana A, Baiardi S, et al. Performance of a seed amplification assay for misfolded alpha-synuclein in cerebrospinal fluid and brain tissue in relation to Lewy body disease stage and pathology burden. Acta Neuropathol. 2024;147(1):18. doi: 10.1007/s00401-023-02663-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Grothe MJ, Kilimann I, Grinberg L, Heinsen H, Teipel S. In vivo volumetry of the cholinergic basal forebrain. In: Perneczky R. Biomarkers for Preclinical Alzheimer’s Disease. Vol 137. Humana Press; 2018:213-231. doi: 10.1007/978-1-4939-7674-4_15 [DOI] [Google Scholar]
31.Fritz HJ, Ray N, Dyrba M, Sorg C, Teipel S, Grothe MJ. The corticotopic organization of the human basal forebrain as revealed by regionally selective functional connectivity profiles. Hum Brain Mapp. 2019;40(3):868-878. doi: 10.1002/hbm.24417 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Mesulam MM, Geula C. Nucleus basalis (Ch4) and cortical cholinergic innervation in the human brain: observations based on the distribution of acetylcholinesterase and choline acetyltransferase. J Comp Neurol. 1988;275(2):216-240. doi: 10.1002/cne.902750205 [DOI] [PubMed] [Google Scholar]
33.Markesbery WR, Jicha GA, Liu H, Schmitt FA. Lewy body pathology in normal elderly subjects. J Neuropathol Exp Neurol. 2009;68(7):816-822. doi: 10.1097/NEN.0b013e3181ac10a7 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Cho H, Choi JY, Hwang MS, et al. In vivo cortical spreading pattern of tau and amyloid in the Alzheimer disease spectrum. Ann Neurol. 2016;80(2):247-258. doi: 10.1002/ana.24711 [DOI] [PubMed] [Google Scholar]
35.Liu AKL, Chang RCC, Pearce RKB, Gentleman SM. Nucleus basalis of Meynert revisited: anatomy, history, and differential involvement in Alzheimer and Parkinson disease. Acta Neuropathol. 2015;129(4):527-540. doi: 10.1007/s00401-015-1392-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Teipel SJ, Fritz HC, Grothe MJ; Alzheimer’s Disease Neuroimaging Initiative . Neuropathologic features associated with basal forebrain atrophy in Alzheimer disease. Neurology. 2020;95(10):e1301-e1311. doi: 10.1212/WNL.0000000000010192 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Grothe MJ, Schuster C, Bauer F, Heinsen H, Prudlo J, Teipel SJ. Atrophy of the cholinergic basal forebrain in dementia with Lewy bodies and Alzheimer disease dementia. J Neurol. 2014;261(10):1939-1948. doi: 10.1007/s00415-014-7439-z [DOI] [PubMed] [Google Scholar]
38.Labrador-Espinosa MA, Silva-Rodríguez J, Reina-Castillo MI, Mir P, Grothe MJ. Basal forebrain atrophy, cortical thinning, and amyloid-β status in Parkinson disease-related cognitive decline. Mov Disord. 2023;38(10):1871-1880. doi: 10.1002/mds.29564 [DOI] [PubMed] [Google Scholar]
39.Pereira JB, Hall S, Jalakas M, et al. Longitudinal degeneration of the basal forebrain predicts subsequent dementia in Parkinson disease. Neurobiol Dis. 2020;139:104831. doi: 10.1016/j.nbd.2020.104831 [DOI] [PubMed] [Google Scholar]
40.Quadalti C, Palmqvist S, Hall S, et al. Clinical effects of Lewy body pathology in cognitively impaired individuals. Nat Med. 2023;29(8):1964-1970. doi: 10.1038/s41591-023-02449-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Brenowitz WD, Hubbard RA, Keene CD, et al. Mixed neuropathologies and associations with domain-specific cognitive decline. Neurology. 2017;89(17):1773-1781. doi: 10.1212/WNL.0000000000004567 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Bohnen NI, Kaufer DI, Hendrickson R, et al. Cognitive correlates of cortical cholinergic denervation in Parkinson disease and parkinsonian dementia. J Neurol. 2006;253(2):242-247. doi: 10.1007/s00415-005-0971-0 [DOI] [PubMed] [Google Scholar]
43.Whitwell JL, Weigand SD, Shiung MM, et al. Focal atrophy in dementia with Lewy bodies on MRI: a distinct pattern from Alzheimer disease. Brain. 2007;130(Pt 3):708-719. doi: 10.1093/brain/awl388 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Yousaf T, Dervenoulas G, Valkimadi PE, Politis M. Neuroimaging in Lewy body dementia. J Neurol. 2019;266(1):1-26. doi: 10.1007/s00415-018-8892-x [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials