Gray matter alterations and their atlas-based spatial associations with neurotransmitter distributions in isolated REM sleep behavior disorder.

Gray matter alterations and their atlas-based spatial associations with neurotransmitter distributions in isolated REM sleep behavior disorder

Abstract

Purpose

Multiple neurotransmitter systems are implicated in the pathophysiology of isolated rapid eye movement sleep behavior disorder (iRBD) and its progression to α-synucleinopathy. This study aimed to identify baseline gray matter (GM) alterations in patients with iRBD compared with healthy controls and to examine their spatial correspondence with known neurotransmitter distributions. We further characterized GM patterns associated with future phenoconversion and their neurochemical context.

Methods

A total of 109 patients with iRBD who completed at least one follow-up evaluation and 24 age- and sex-matched healthy controls underwent structural magnetic resonance imaging. Gray matter volume (GMV) alterations were assessed using the CAT12 toolbox. Spatial correlations between group-level GMV difference maps and neurotransmitter atlases were evaluated using the JuSpace toolbox.

Results

At baseline, patients with iRBD showed reduced GMV in the right cerebellar cortex and vermis, along with increased GMV in the left putamen and pallidum. These GMV alterations demonstrated significant spatial correspondence with dopaminergic receptor and transporter maps, including dopamine D2 receptors and the dopamine transporter (DAT). Among patients with iRBD, those who subsequently phenoconverted exhibited lower baseline GMV in the right medial orbital gyrus compared with non-converters. This pattern showed significant spatial correlations with serotonergic (5-HT2A), glutamatergic (mGluR5), GABAergic (GABAa), and dopaminergic (FDOPA) neurotransmitter maps.

Conclusion

Patients with iRBD exhibit distinct patterns of GMV alteration that spatially align with dopaminergic neurotransmitter distribution maps, while phenoconversion is associated with GM changes corresponding to multiple neurotransmitter systems, although the phenoconversion findings are exploratory.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s11325-026-03845-2.

Keywords: REM sleep behavior disorder, Phenoconversion, Magnetic resonance imaging, Voxel-based morphometry, Neurotransmitter

Introduction

Most patients with isolated rapid eye movement (REM) sleep behavior disorder (iRBD) eventually develop neurodegenerative diseases associated with α-synuclein pathology, and iRBD is therefore widely regarded as a prodromal stage of synucleinopathy [1]. Accumulating evidence suggests that iRBD shares common neurodegenerative processes with established α-synucleinopathies, including Parkinson’s disease (PD) and dementia with Lewy bodies [2].

In PD, motor symptoms are primarily linked to degeneration of the nigrostriatal dopaminergic system, whereas non-motor symptoms are associated with broader involvement of cholinergic, serotonergic, glutamatergic, and noradrenergic neurotransmitter systems [3]. As a prodromal form of α-synucleinopathy, dopaminergic loss has been demonstrated in patients with RBD [4]. Furthermore, multimodal imaging studies have shown that RBD is accompanied by alterations across multiple neurotransmitter systems, including cholinergic, noradrenergic, and dopaminergic circuits [5].

Structural neuroimaging studies have reported gray matter (GM) alterations in patients with idiopathic RBD, involving both subcortical structures like the caudate nucleus and various prefrontal and temporal cortical areas [6]. These GM changes are thought to reflect regional neuronal vulnerability that may be associated with specific neurotransmitter systems [7]. Supporting this notion, recent evidence indicates that regional variability in neurotransmitter distributions is intrinsically linked to patterns of structural brain modification [5]. Accordingly, assessing the spatial correspondence between GM alterations and neurotransmitter maps may provide insights into the neurobiological substrates underlying RBD and its progression toward overt neurodegeneration.

In this study, we aimed to evaluate baseline GM alterations in patients with iRBD compared with healthy controls and to examine their spatial correspondence with established neurotransmitter distributions. In addition, we investigated GM patterns associated with phenoconversion and characterized their corresponding neurochemical profiles. To this end, we employed the JuSpace toolbox to assess spatial correlations between GM difference maps and PET/SPECT-derived neurotransmitter atlases [8], a methodology that has been validated in studies of PD [9], as well as other neurodegenerative disorders, including frontotemporal dementia and Alzheimer’s disease [10].

Materials and methods

Study participants

Patients with video-polysomnography–confirmed iRBD who underwent T1-weighted magnetic resonance imaging (MRI) at Seoul National University Hospital were consecutively recruited. To evaluate the phenoconversion, we included only patients who completed at least one longitudinal follow-up evaluation. The diagnosis of iRBD was established according to the International Classification of Sleep Disorders, Third Edition (ICSD-3) criteria [11].

Participants were excluded if they met any of the following criteria: (1) a history of other pre-existing neurodegenerative diseases, (2) moderate-to-severe obstructive sleep apnea (defined as an apnea-hypopnea index ≥ 15/hr), or (3) significant structural brain lesions identified on neuroimaging. The control group consisted of 24 age- and sex-matched healthy volunteers without sleep-related, neurological (including dementia or parkinsonism), or psychiatric disorders. Controls were examined by a neurologist and screened by interview for RBD symptoms, parkinsonism, and cognitive impairment; none had a MoCA-K score below 25. Controls did not undergo video-polysomnography.

The study protocol was approved by the Institutional Review Board of Seoul National University Hospital (IRB No. 1702-150-835), and written informed consent was obtained from all participants prior to enrollment.

Clinical evaluation

The severity and frequency of RBD symptoms were assessed using the REM Sleep Behavior Disorder Questionnaire–Korean version (RBDQ-KR), a validated 13-item instrument derived from the original Hong Kong version [12]. Global cognitive function was measured using the Montreal Cognitive Assessment (MoCA) [13]. Additionally, depressive symptoms were quantified using the Korean version of the Geriatric Depression Scale (GDS-K) [14].

Follow-up evaluation

The follow-up protocol has been described in detail previously [15]. Briefly, two neurologists (K.Y.J., H.J.K.) examined each patient every 6 to 12 months, blinded to the imaging analyses. Parkinson’s disease, dementia with Lewy bodies (DLB), and multiple system atrophy (MSA) were diagnosed according to the Movement Disorder Society [16], DLB Consortium [17], and second consensus (Gilman) criteria [18], respectively. Patients who developed one of these disorders were classified as phenoconverters, and those who had not by their last follow-up visit as non-converters (censored and still at risk).

Image acquisition and preprocessing

Structural MRI data were acquired on a 3.0-Tesla Siemens Biograph mMR scanner (Siemens Healthcare Sector, Germany) equipped with a 16-channel head coil. T1-weighted images were obtained using a three-dimensional magnetization-prepared rapid gradient-echo (MPRAGE) sequence with the following parameters: repetition time (TR) = 1670 ms, echo time (TE) = 1.89 ms, flip angle = 9°, field of view (FOV) = 250 mm, acquisition matrix = 256 × 256, voxel size = 1.0 × 1.0 × 1.0 mm³, and 208 sagittal slices. All images were visually inspected by an experienced neuroradiologist to ensure image quality and to exclude incidental structural abnormalities or artifacts.

Image preprocessing was performed using the Computational Anatomy Toolbox (CAT12) implemented in Statistical Parametric Mapping software (SPM12; Wellcome Centre for Human Neuroimaging) running on MATLAB R2024b (MathWorks, Natick, MA) [19]. T1-weighted images were segmented into gray matter, white matter, and cerebrospinal fluid components using default CAT12 settings.

GMV analysis

Voxel-based morphometry (VBM) was performed on segmented gray matter (GM) images using the CAT12 toolbox. Individual GM maps were spatially normalized to the Montreal Neurological Institute (MNI) space using affine and nonlinear registration and modulated with Jacobian determinants to preserve absolute GM volume at each voxel. The modulated GM images were smoothed with an isotropic Gaussian kernel of 8-mm full width at half maximum (FWHM) to improve the signal-to-noise ratio and account for inter-individual anatomical variability.

Spatial correlation with neurotransmitter density maps

Spatial correlations between gray matter morphological alterations and neurotransmitter distributions were assessed using the JuSpace toolbox (version 2.0). Unthresholded voxel-wise t-statistic maps derived from group-level GMV comparisons (iRBD vs. HCs, and phenoconverters vs. non-converters) were used as input for the spatial correlation analyses.

These maps were correlated with PET/SPECT-derived neurotransmitter atlases encompassing serotonergic (5-HT1A, 5-HT1B, and 5-HT2A receptors, serotonin transporter [SERT]), dopaminergic (dopamine D1 and D2 receptors, dopamine transporter [DAT], and [¹⁸F]-FDOPA), GABAergic (GABAa), opioid (µ-opioid receptor), cannabinoid (CB1), noradrenergic (norepinephrine transporter [NAT]), cholinergic (vesicular acetylcholine transporter [VAChT]), and glutamatergic (metabotropic glutamate receptor 5 [mGluR5]) systems.

Statistical analysis

Demographic and clinical variables were compared between groups using independent-sample t tests or Mann–Whitney U tests for continuous variables, and chi-square tests for categorical variables, as appropriate. Continuous data are reported as mean ± standard deviation, as appropriate. All clinical statistical analyses were performed using SPSS software (version 22.0; IBM Corp., Armonk, NY, USA).

Group differences in gray matter volume (GMV) were examined using a general linear model (GLM) implemented in SPM12. Separate models were constructed to compare (1) patients with iRBD versus healthy controls and (2) phenoconverters versus non-converters. To control for potential confounding effects, total intracranial volume (TIV) was included as a nuisance covariate, and age, sex, and years of education were entered as covariates in all models. Statistical significance was defined at a voxel-level threshold of p < 0.001 (uncorrected), followed by cluster-level family-wise error (FWE) correction at p < 0.05. As an exploratory time-to-event analysis, we fitted a Cox proportional-hazards model with the time from MRI to phenoconversion (or last follow-up) as the outcome and the mean gray matter volume of the orbitofrontal cluster as the predictor, adjusted for age, sex, education, and TIV.

For spatial correlation analyses, Spearman’s rank correlation coefficients were calculated between unthresholded voxel-wise t-statistic maps and each neurotransmitter atlas using the JuSpace toolbox (version 2.0). Spatial correlations were computed using the unthresholded t-maps of the iRBD > control and non-converter > converter contrasts, such that positive t values indicate greater gray matter volume in patients with iRBD and in non-converters, respectively. Correlation coefficients were Fisher’s z-transformed to normalize their distribution. The statistical significance was assessed using 10,000 exact spatial permutations, in which the null distribution is generated from surrogate maps that preserve the spatial autocorrelation structure of the maps. To control for multiple comparisons across neurotransmitter maps, false discovery rate (FDR) correction was applied with a significance threshold of p < 0.05.

Results

Patient characteristics

The mean age of the patients with iRBD was 69 years, with 56 participants (51.4%) being male, which was comparable to the HC group. The mean RBDQ-KR score was 48.7 ± 18.7. Patients with iRBD showed higher scores on the Geriatric Depression Scale–Korean version (GDS-K; 9.7 ± 7.6) compared with controls. The mean MoCA-K score in the iRBD group was 26.1 ± 2.7, which was lower than that of the control group (Table 1).

Table 1Demographic and clinical characteristics of patients with isolated REM sleep behavior disorder (iRBD) and healthy controls

Controls iRBD p-value

n = 24 n = 109

Age (years) 67.2 ± 3.6 68.8 ± 5.6 0.173

Sex, male (%) 13 (54.2) 56 (51.4) 0.804

BMI (kg/m2) 22.8 ± 2.4 23.7 ± 2.8 0.125

Education (years) 14.1 ± 2.6 12.7 ± 3.9 0.097

RBDQ-KR 3.9 ± 2.6 48.7 ± 18.7 < 0.001

MoCA-K 27.4 ± 1.2 26.1 ± 2.7 0.020

GDS-K 2.9 ± 3.2 9.7 ± 7.6 < 0.001

TIV 1497.6 ± 123.2 1482.4 ± 127.9 0.590

Abbreviation: iRBD, isolated rapid eye movement sleep behavior disorder; BMI, body mass index; RBDQ-KR, REM sleep behavior disorder questionnaire–Korean version; MoCA-K, Korean version of the Montreal Cognitive Assessment; GDS-K, Korean version of the Geriatric Depression Scale; TIV, total intracranial volume

Participants were followed for a mean duration of 4.5 ± 1.2 years. During this period, 16 patients (14.7%) phenoconverted to a neurodegenerative disorder, including Parkinson’s disease (n = 12), multiple system atrophy (n = 3), and dementia with Lewy bodies (n = 1). The mean interval from baseline to conversion was 3.9 ± 2.0 years. Phenoconverters tended to have a lower body mass index (BMI) and higher GDS-K scores at baseline compared with non-converters. (Supplementary Table 1).

Gray matter volume alterations

VBM analysis identified distinct patterns of structural brain changes in patients with iRBD. Compared with healthy controls, patients with iRBD exhibited significant reductions in gray matter volume (GMV) in the cerebellar exterior and vermis, along with increased GMV in the left pallidum and putamen. Within the iRBD cohort, patients who subsequently phenoconverted showed significantly lower baseline GMV in the right medial orbital gyrus compared with non-converters. (Table 2; Fig. 1) In an exploratory Cox proportional-hazards analysis, lower baseline orbitofrontal gray matter volume was associated with a higher hazard of phenoconversion (hazard ratio = 2.9 per 1-SD decrease; 95% CI 1.6–5.2; p < 0.001), which persisted after adjustment for age, sex, education, and TIV (HR ≈ 3.1) and for GDS-K (HR ≈ 2.7).

Table 2Brain regions showing significant gray matter volume differences between groups

Cluster k Peak cluster p T Z Peak region MNI coordinates

(voxels) (FWE-corrected) x y z

Controls > iRBD 2686 < 0.0001 5.11 4.86 Cbll vermis, exterior 20 -90 -21

Controls < iRBD 1049 0.019 4.39 4.22 Lt. putamen/pallidum -22 -12 2

Non-converters> Converters 870 0.048 4.63 4.40 Rt. Medial orbital gyrus 10 18 -24

Abbreviation: GMV, gray matter volume; MNI, Montreal Neurological Institute; FWE, family-wise error

Atlas-based spatial associations of GMV alterations with neurotransmitter distributions

Spatial correlation analyses revealed significant correlations between GMV alterations in patients with iRBD and dopaminergic neurotransmitter maps, including the dopamine D2 receptor (ρ = 0.37, FDR-corrected p = 0.001) and the DAT (ρ = 0.31, FDR-corrected p = 0.004). The spatial pattern of GMV differences between phenoconverters and non-converters demonstrated significant correlations with serotonergic (5-HT2a; ρ = 0.32, FDR-corrected p = 0.003), glutamatergic (mGluR5; ρ = 0.45, FDR-corrected p = 0.004), and GABAergic (GABAa; ρ = 0.29, FDR-corrected p = 0.005) neurotransmitter systems. A weaker association was observed for [18F]-FDOPA (ρ = −0.22, FDR-corrected p = 0.046), whereas the dopamine D1 and D2 receptor maps showed nominal associations that did not survive FDR correction (ρ = 0.21 and 0.20; FDR-corrected p = 0.056 and 0.060, respectively). (Table 3; Fig. 1)

Table 3Spatial correlations between group-level gray matter volume changes and neurotransmitter distribution maps

Neurotransmitter iRBD > Control Non-converter > Converter

Spearman’s rho FDR-correctedp-value Spearman’s rho FDR-correctedp-value

5HT1a 0.086 0.8424 0.231 0.1104

5HT2a 0.013 0.8867 0.321 0.0028

5HT1b −0.130 0.3912 0.009 0.9226

SERT 0.182 0.1702 −0.089 0.5196

D1 0.121 0.3912 0.208 0.0557

D2 0.373 0.0014 0.199 0.0602

DAT 0.312 0.0042 0.083 0.5196

FDOPA −0.056 0.8424 −0.219 0.0458

GABAa −0.015 0.8867 0.293 0.0051

Mu −0.065 0.8867 0.140 0.6267

CB1 −0.025 0.8867 −0.028 0.9012

NAT 0.175 0.1702 0.069 0.5781

mGluR5 0.073 0.8787 0.446 0.0042

VAChT 0.218 0.0709 0.096 0.5069

Abbreviations: 5-HT, serotonin; SERT, serotonin transporter; DAT, dopamine transporter; FDOPA, 6-[¹⁸F]-fluoro-L-DOPA; GABAa, gamma-aminobutyric acid type A receptor; CB1, cannabinoid receptor type 1; NAT, norepinephrine transporter; mGluR5, metabotropic glutamate receptor 5; VAChT, vesicular acetylcholine transporter; FDR, false discovery rate

Discussion

This study identified distinct patterns of GMV alterations associated with iRBD and its future phenoconversion, and examined the spatial correspondence between these structural changes and established neurotransmitter distributions. Compared with healthy controls, patients with iRBD exhibited reduced GMV in the cerebellar exterior and vermis, along with increased GMV in the left putamen and pallidum, which showed spatial alignment with dopaminergic receptor and transporter distributions. In addition, patients who subsequently phenoconverted demonstrated reduced baseline GMV in the right medial orbital gyrus, a region whose spatial pattern corresponded predominantly to serotonergic (5-HT2A), glutamatergic (mGluR5), and GABAergic (GABAa) receptor distributions.

Structural alterations involving the cerebellum have been consistently reported in iRBD. Previous neuroimaging studies have described both cerebellar GMV increases [6] and decreases [20] in patients with iRBD. In the present study, we observed GMV reduction predominantly involving the right posterior cerebellum, which anatomically overlaps with the region corresponding to lobule VIIB reported in the previous study [20]. Our results further support a potential role of the cerebellum in the pathogenesis of iRBD. Increased GMV in the putamen has also been reported in iRBD [21], although findings across studies have been heterogeneous [22]. These regional GMV increases may reflect adaptive or compensatory change [23], but alternative explanations—including tissue-segmentation effects related to the signal properties of the basal ganglia, iron-related influences on T1 contrast, glial or inflammatory changes, vascular or metabolic factors, and registration bias—cannot be excluded, and a compensatory mechanism cannot be confirmed without longitudinal, diffusion, susceptibility, or molecular imaging.

The spatial pattern of GMV alterations observed in iRBD showed significant correspondence with dopaminergic maps, particularly D2 receptor and DAT atlases, consistent with prior imaging studies [4]. Given the high density of D2 receptors and the dopamine transporter in the striatum and the striatal location of one of the clusters, this correspondence should be interpreted as an atlas-based spatial association rather than evidence of a specific dopaminergic mechanism. Cortical thickness alterations in iRBD have previously been shown to spatially align with DAT density maps [24], and presynaptic dopaminergic imaging studies have demonstrated substantial nigrostriatal dysfunction in iRBD, with more pronounced tracer reductions in patients exhibiting subtle motor features [25, 26]. Meta-analytic evidence further supports reduced nigrostriatal dopaminergic function in idiopathic RBD, particularly in the putamen [27].

Within the iRBD cohort, reduced baseline GMV in the right medial orbital gyrus was associated with subsequent phenoconversion. This finding is in line with previous reports identifying frontal cortical atrophy, including the inferior frontal gyrus, as a structural feature associated with progression to α-synucleinopathies [28]. The medial orbital region is also considered part of an olfactory-related cortical network [29], dysfunction of which represents a hallmark prodromal feature of synucleinopathies [30]. Consistent with this notion, olfactory impairment in iRBD has been associated with an increased risk of phenoconversion over relatively short follow-up periods [31, 32].

Among the neurotransmitter systems examined, mGluR5 showed the largest effect size (ρ = 0.45), whereas the 5-HT2A atlas showed the lowest FDR-corrected p value. This finding aligns with evidence that elevated 5-HT2A receptor binding in the frontal cortex of patients with Parkinson’s disease is associated with increased α-synuclein burden [33]. In the context of RBD, serotonergic pathways play an important role in the regulation of REM sleep without atonia [34]. Although pharmacological modulation of 5-HT2A receptors has not resulted in consistent symptomatic improvement in RBD [35], the present findings suggest that serotonergic receptor architecture may characterize cortical regions that are structurally vulnerable during phenoconversion and may provide neurochemical context for the underlying neurodegenerative trajectory.

Glutamatergic signaling, as captured by the mGluR5 receptor map, also showed a robust spatial correspondence with GMV alterations associated with phenoconversion. mGluR5 is a G protein–coupled receptor that modulates N-methyl-D-aspartate (NMDA) receptor activity and has been implicated in neurodegenerative cascades. Experimental studies have demonstrated that mGluR5 activation can exacerbate excitotoxic neuronal injury [36], whereas pharmacological inhibition of mGluR5 attenuates axonal degeneration in animal models of PD [37]. In line with these findings, a recent PET study reported elevated mGluR5 binding in patients with PD and comorbid RBD, which correlated with the severity of REM sleep without atonia [38].

The spatial correspondence with GABAA distributions may provide an indirect neurochemical context for the structural changes in iRBD, although GABAergic function was not measured in this study. Animal models have demonstrated that impaired GABAa-mediated inhibition is associated with RBD-like behaviors [39]. Moreover, studies in early-stage Parkinson’s disease have reported divergent alterations in basal ganglia GABA levels depending on the presence of RBD, despite comparable nigral iron and neuromelanin burdens [40]. Together, these findings support a role for GABAergic circuitry in shaping vulnerability patterns observed in prodromal synucleinopathies.

The negative correlation with FDOPA, although weak (ρ = −0.22, FDR-corrected p = 0.046), may suggest that brain regions with lower dopaminergic synthesis capacity are more susceptible to GMV reduction associated with phenoconversion in iRBD. Previous imaging studies have reported reduced AADC activity in the putamen [5, 41] and thalamus [41] in patients with iRBD compared with healthy controls. Moreover, reduced AADC activity has been linked to increased microglial activation, which may contribute to structural brain changes [41]. Given the weak and borderline correlation, however, this interpretation remains speculative.

Several limitations of this study should be acknowledged. First, both the healthy control group (n = 24) and the subset of patients who phenoconverted (n = 16) were relatively small. Although statistical significance for voxel-based morphometry analyses was determined using a cluster-level FWE correction following an initial voxel-level threshold, the limited sample size may still affect the stability and generalizability of the findings. Second, prodromal assessments such as MDS-UPDRS-III and olfactory testing were not performed in all patients and did not differ between groups in those assessed. We therefore cannot exclude that the orbitofrontal finding partly identifies patients with more advanced prodromal disease at baseline. Patients with iRBD had poorer cognition and higher depressive-symptom scores than controls; these differences may partly contribute to the structural findings. The phenoconversion-related findings did not survive adjustment for GDS-K; because depressive symptoms are themselves a prodromal feature of synucleinopathy, this may represent partial over-control, and these findings should therefore be regarded as exploratory. Controls did not undergo video-polysomnography, so unrecognized REM sleep without atonia cannot be excluded. Third, JuSpace analyses rely on neurotransmitter atlases derived from healthy individuals and therefore provide only indirect information regarding neurochemical alterations in patients. Consequently, causal inferences regarding specific neurotransmitter systems cannot be drawn from the present findings alone and should be addressed in future studies using subject-level molecular imaging. Finally, because spatial correlation analyses were performed using group-level t-statistic maps, we were unable to assess direct associations between individual-level structural brain measures and specific clinical variables. Moreover, whole-brain VBM with 8-mm smoothing has limited sensitivity for the small pontine and medullary structures central to REM atonia; the absence of brainstem findings therefore does not exclude brainstem involvement, and our results should not be regarded as a complete account of iRBD pathophysiology. Because the outcomes were heterogeneous (Parkinson’s disease, MSA, and DLB), these findings reflect prodromal synucleinopathy in general and cannot be attributed to a single disorder. Therefore this study does not provide an individual-level prediction model, internal or external validation, and the results represent group-level differences rather than a validated prognostic biomarker.

Conclusion

GMV alterations in patients with iRBD demonstrate significant spatial alignment with the distribution of dopaminergic receptors and transporters. In addition, structural atrophy associated with future phenoconversion shows spatial correspondence with serotonergic (5-HT2A), glutamatergic (mGluR5), and GABAergic (GABAA) distribution maps. Although based on indirect spatial correlations, these findings provide a neurochemical context for regional brain vulnerability in iRBD; they represent group-level differences rather than a validated prognostic biomarker.

Supplementary Information

Below is the link to the electronic supplementary material.

Acknowledgements

None.

Author contributions

JIB: Conceptualization, Formal analysis, Funding acquisition, Writing – original draft; HS: Conceptualization, Methodology, Formal analysis, Writing – review & editing; JHS: Data curation, Investigation, Methodology; HJK: Conceptualization, Methodology, Supervision, Writing – review & editing; WCS: Conceptualization, Supervision, Writing – review & editing; CHS: Conceptualization, Supervision, Writing – review & editing; KYJ: Conceptualization, Supervision, Writing – review & editing.

Funding

Open Access funding enabled and organized by Seoul National University Hospital. This work was supported by the National Research Foundation of Korea (NRF) (No. NRF-2020R1C1C1013160) and KHU Medical Device Convergence Research Project from Kyung Hee University in 2026(KHU-20263311).

Data availability

All relevant data are provided within the manuscript.

Declarations

Ethics approval

This study was conducted in line with the principles of the Declaration of Helsinki and were approved by the institutional review board of the Seoul National University Hospital (IRB No. 1702-150-835).

Consent to participate

All participants provided written informed consent.

Conflict of interest

JIB is supported by the National Research Foundation of Korea (NRF) (No. 2020R1C1C1013160). HS, JHS, HJK, WCS, CHS and KYJ declare no conflicts of interest in the context of this paper.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Jung-Ick Byun and Huijin Song contributed equally to this work.

Contributor Information

Chul-Ho Sohn, Email: neurorad63@snu.ac.kr.

Ki-Young Jung, Email: jungky@snu.ac.kr.

References

1.Postuma RB et al (2019) Risk and predictors of dementia and parkinsonism in idiopathic REM sleep behaviour disorder: a multicentre study. Brain 142(3):744–759

2.Stefani A et al (2025) From mechanisms to future therapy: a synopsis of isolated REM sleep behavior disorder as early synuclein-related disease. Mol Neurodegener 20(1):19

3.Barone P (2010) Neurotransmission in Parkinson’s disease: beyond dopamine. Eur J Neurol 17(3):364–376

4.Jiménez-Jiménez FJ et al (2021) Neurochemical features of rem sleep behaviour disorder. J personalized Med 11(9):880

5.Knudsen K et al (2018) In-vivo staging of pathology in REM sleep behaviour disorder: a multimodality imaging case-control study. Lancet Neurol 17(7):618–628

6.Wang X et al (2025) Gray matter structural alterations in idiopathic rapid eye movement sleep behavior disorder: A voxel-based meta-analysis. Sleep Med 126:114–121

7.Manninen S et al (2021) Cerebral grey matter density is associated with neuroreceptor and neurotransporter availability: A combined PET and MRI study. NeuroImage 235:117968

8.Dukart J et al (2021) JuSpace: A tool for spatial correlation analyses of magnetic resonance imaging data with nuclear imaging derived neurotransmitter maps. Hum Brain Mapp 42(3):555–566

9.Pang H et al (2024) Disentangling gray matter atrophy and its neurotransmitter architecture in drug-naïve Parkinson’s disease: an atlas-based correlation analysis. Cereb Cortex 34(10):bhae420

10.Morais RF et al (2025) Differential involvement of neurotransmitter pathways in AD, bvFTD and MCI: Whole-brain MRI analysis. Neurobiol Dis 209:106897

11.American Academy of Sleep Medicine (2014) The international classification of sleep disorders, 3rd edn. American Academy of Sleep Medicine, Darien, IL

12.You S et al (2017) The REM sleep behavior disorder screening questionnaire: validation study of the Korean version (RBDQ-KR). J Clin Sleep Med 13(12):1429–1433

13.Nasreddine ZS et al (2005) The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 53(4):695–699

14.Bae JN, Cho MJ (2004) Development of the Korean version of the Geriatric Depression Scale and its short form among elderly psychiatric patients. J Psychosom Res 57(3):297–305

15.Byun J-I et al (2025) Clinical characteristics and phenoconversion in isolated REM sleep behavior disorder: a prospective single-center study in Korea, compared with Montreal cohort. J Clin Sleep Med 21(1):81–88

16.Postuma RB et al (2015) MDS clinical diagnostic criteria for Parkinson’s disease. Mov Disord 30(12):1591–1601

17.McKeith IG et al (2017) Diagnosis and management of dementia with Lewy bodies: Fourth consensus report of the DLB Consortium. Neurology 89(1):88–100

18.Gilman S et al (2008) Second consensus statement on the diagnosis of multiple system atrophy. Neurology 71(9):670–676

19.Gaser C et al (2024) CAT: a computational anatomy toolbox for the analysis of structural MRI data. Gigascience 13:giae049

20.Lee DA, Lee H-J, Park KM (2023) Cerebellar volume reduction in patients with isolated rem sleep behavior disorder: evidence of a potential role of the cerebellum. Eur Neurol 86(5):341–347

21.Chen M et al (2022) Structural and functional brain alterations in patients with idiopathic rapid eye movement sleep behavior disorder. J Neuroradiol 49(1):66–72

22.Holtbernd F et al (2021) Convergent patterns of structural brain changes in rapid eye movement sleep behavior disorder and Parkinson’s disease on behalf of the German rapid eye movement sleep behavior disorder study group. Sleep 44(3):zsaa199

23.Baril AA et al (2017) Gray Matter Hypertrophy and Thickening with Obstructive Sleep Apnea in Middle-aged and Older Adults. Am J Respir Crit Care Med 195(11):1509–1518

24.Rahayel S et al (2023) Mitochondrial function-associated genes underlie cortical atrophy in prodromal synucleinopathies. Brain 146(8):3301–3318

25.Rolinski M et al (2016) Basal ganglia dysfunction in idiopathic REM sleep behaviour disorder parallels that in early Parkinson’s disease. Brain 139(8):2224–2234

26.Yamada G et al (2019) Nigrostriatal dopaminergic dysfunction and altered functional connectivity in REM sleep behavior disorder with mild motor impairment. Front Neurol 10:802

27.Bauckneht M et al (2018) Presynaptic dopaminergic neuroimaging in REM sleep behavior disorder: a systematic review and meta-analysis. Sleep Med Rev 41:266–274

28.de Natale ER, Wilson H, Politis M (2022) Predictors of RBD progression and conversion to synucleinopathies. Curr Neurol Neurosci Rep 22(2):93–104

29.Fjældstad A et al (2017) Brain fingerprints of olfaction: a novel structural method for assessing olfactory cortical networks in health and disease. Sci Rep 7(1):42534

30.Ielo A et al (2025) Structural and functional connectomics of the olfactory system in Parkinson’s disease: a systematic review. Parkinsonism Relat Disord 132:107230

31.Postuma RB et al (2011) Olfaction and color vision identify impending neurodegeneration in rapid eye movement sleep behavior disorder. Ann Neurol 69(5):811–818

32.Dall’Antonia I et al (2025) Refining α-Synucleinopathy Risk in isolated REM sleep behavior disorder Patients Using Repeated Olfactory Testing. Sleep Med p. 106812

33.Rasmussen NB et al (2016) 5-HT2A Receptor Binding in the Frontal Cortex of Parkinson’s Disease Patients and Alpha-Synuclein Overexpressing Mice: A Postmortem Study. Parkinsons Dis 2016:3682936

34.Iranzo A (2018) The REM sleep circuit and how its impairment leads to REM sleep behavior disorder. Cell Tissue Res 373(1):245–266

35.Stefani A et al (2021) Nelotanserin as symptomatic treatment for rapid eye movement sleep behavior disorder: a double-blind randomized study using video analysis in patients with dementia with Lewy bodies or Parkinson’s disease dementia. Sleep Med 81:180–187

36.Conn PJ et al (2005) Metabotropic glutamate receptors in the basal ganglia motor circuit. Nat Rev Neurosci 6(10):787–798

37.Zhang J-N et al (2021) Blockade of metabotropic glutamate receptor 5 attenuates axonal degeneration in 6-hydroxydopamine-induced model of Parkinson’s disease. Mol Cell Neurosci 110:103572

38.Doppler CE et al (2024) Glutamate signaling in patients with Parkinson disease with REM sleep behavior disorder. Neurology 102(9):e209271

39.Brooks PL, Peever JH (2011) Impaired GABA and glycine transmission triggers cardinal features of rapid eye movement sleep behavior disorder in mice. J Neurosci 31(19):7111–7121

40.Zhang Y et al (2025) GABA outperforms iron and neuromelanin in detecting nigrostriatal alterations in early-stage Parkinson’s disease with RBD. npj Parkinson’s Disease 11(1):229

41.Stokholm MG et al (2018) Extrastriatal monoaminergic dysfunction and enhanced microglial activation in idiopathic rapid eye movement sleep behaviour disorder. Neurobiol Dis 115:9–16

Associated Data

Supplementary Materials

Data Availability Statement

All relevant data are provided within the manuscript.