Functional PET for mapping metabolic dynamics in Parkinson’s disease

Functional PET for mapping metabolic dynamics in Parkinson’s disease | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Functional PET for mapping metabolic dynamics in Parkinson’s disease Vanessa Heinecke, Lilly Machholz, Kenan Steidel, Lenna M. Rüsing, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7156182/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 18 Nov, 2025 Read the published version in Scientific Reports → Version 1 posted 16 You are reading this latest preprint version Abstract Static [ 18 F]-FDG PET studies have shown characteristic metabolic patterns for Parkinson’s disease (PD). However static PET scans offer only a snapshot of synaptic activity, preventing insights into dynamics of glucose consumption on individual level. In this study, we apply a novel dynamic constant infusion PET protocol in combination with MRI for mapping metabolic dynamics in PD. Metabolic time series comparison revealed hypometabolism in bilateral clusters in the substantia nigra in PD patients compared to controls (p FWE < 0.001). The temporal structure allowed us to depict metabolic networks covering and the basal ganglia circuits and typical neural resting-state networks on individual level. A measure of time series variation showed a region-specific increase in time series variation in PD, which related to cognitive PD symptoms. The findings pinpoint the methodological capabilities of the acquisition protocol for evaluating metabolic dynamics and metabolic network activity in PD. Our study lays a foundation for the application of molecular imaging with temporal resolution in context of neurodegenerative diseases. Such protocols can capture interregional metabolic changes on patient level, the applicability of which should be evaluated in prodromal stages. Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Health sciences/Neurology Biological sciences/Neuroscience glucose dynamics metabolic variation seed-based metabolic connectivity neurocognitive networks cognitive symptoms non-motor symptoms imaging biomarkers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Parkinson’s disease (PD) is the neurodegenerative disease with the fastest growing prevalence worldwide 1 . PD is primarily considered a movement disorder, although the complex clinical manifestation, including non-motor symptoms, are not solely attributable to dopaminergic depletion in basal ganglia-cortex loops. Instead, network-level dysfunction is assumed to underly PD and specifically, more widespread aberrant interregional neural communication 2 . [ 18 F]-fluorodeoxyglucose ([ 18 F]-FDG) positron emission tomography (PET), as a direct correlate of neural activity indexing synaptic activity, has increasingly been utilised for brain network visualisation and holds significant potential as a network biomarker candidate 3 . A stereotypic PD-related metabolic pattern with hypometabolism in occipito-parietal regions and hypermetabolism in the supplementary motor area and the putamina relative to healthy controls has been repeatedly observed 4 . In particular, it is well established that hypometabolism in the occipital cortex 5 and regions covering the default-mode network relate to cognitive decline in PD 6 . Reliability of former results have been hampered by the static nature of classical PET acquisitions, providing only a snapshot per subject, so that only limited insights into metabolic dynamics and interregional communication at the individual level have been possible to date. Nonetheless, this information is crucial for uncovering network-level metabolic correlates of individual symptoms. Only recently, Villien et al. introduced a functional [ 18 F]-FDG-PET (fPET) protocol leveraging constant tracer infusion and list-mode acquisition for information on glucose dynamics in the individual subject 7 . The technique offers a direct and quantitative measure of neuronal function 8 . In young healthy subjects, Voigt et al. found a strong association between cognition and metabolic network activity in fronto-parietal areas 9 . Irrespective of this, time series variation measures have been identified as key drivers of cognitive performance in aging 10 . However, metabolic networks and glucose dynamics at the subject-level remain underexplored within the context of neurodegenerative disorders despite their potential role for symptom expression. To close this gap, we combined resting-state [ 18 F]-FDG-fPET with fMRI acquisition to study glucose dynamics in PD. We aimed to evaluate between-group differences in glucose metabolism based on time series data compared to healthy subjects, to establish a measure of glucose dynamics at the subject-level, and to develop a seed-based network approach for analysing metabolic time series data (Fig. 1 ). This approach offers new insights into how glucose dynamics contributes to network-level changes related to individual PD symptoms. The metabolic time series data enabled us to identify hypometabolic clusters within the substantia nigra in our small PD sample. Moreover, we report an increase in time series variation associated with cognitive symptoms in PD and seed-based metabolic networks on subject-level. These findings may lay the groundwork for developing novel network-based imaging markers for neurodegenerative diseases using molecular time series information. Results Study participants and clinical characterisation A total of 14 persons with PD (PwPD) (63.43 ± 8.92 years, three females) and 13 healthy controls (59.54 ± 5.13 years, six females) underwent [ 18 F]-FDG-fPET scanning. Cognitive screening tests did not reveal differences in cognitive function between the groups (Montreal Cognitive Assessment (MoCA): W = 118.5, P = 0.19). The average MDS-UPDRS-III score for PwPD without medication was 35.64 ± 17.54 points, indicating mild to moderate severity of overall motor symptoms (refer to supplementary Table 1 for clinical and demographic information). Nigral hypometabolism and hypermetabolic activity in the motor cortex in PD The voxel-wise group comparison of cerebral [ 18 F]-FDG uptake time series using a flexible factorial design revealed subcortical and cortical regions with altered metabolic activity in PwPD compared to controls (P < 0.05 after family-wise error (FWE) rate correction). Regional hypometabolism was observed in the temporal, parietal, occipital and frontal lobes in PwPD (Fig. 2 a). Parietal clusters included the bilateral angular gyri and precuneus (P FWE <0.0001, Supplementary Table 2). Subcortical hypometabolism was present in the basal ganglia: the bilateral substantia nigra pars compacta and reticulata (Fig. 2 a-c), the putamina and the left caudate nucleus (P FWE <0.0001, Supplementary Table 2). The cerebellum, the right amygdala, the left insula and the dorsal part of the thalamus also showed relative hypometabolism in PwPD (P FWE <0.0001, Supplementary Table 2). Significant hypermetabolic regions in PwPD were observed in a large cluster spanning the supplementary motor area, the middle cingulate and the precuneus (Fig. 2 d, P FWE <0.0001, Supplementary Table 2). The bilateral putamina, the orbitofrontal cortex, the medial occipital lobes, the fusiform gyrus and the cerebellum exhibited increased activity in PwPD (P FWE <0.0001). The right nucleus accumbens also showed relative hypermetabolism in PwPD (P FWE <0.0001, Supplementary Table 2). Extracted relative [ 18 F]-FDG uptake values from all clusters revealed contrast-specific patterns of signal time courses within hypometabolic clusters with a lower level of metabolic activity over time in PwPD (Fig. 2 e) and hypermetabolic clusters characterised by a higher activity over time in PwPD (Fig. 2 f). Contrasting the individual [ 18 F]-FDG activity between hypometabolic clusters in the substantia nigra and hypermetabolic clusters in the motor cortex indicated that reduced subcortical activity is accompanied by corresponding increased cortical activity in PwPD (Fig. 2 g). On the other hand, healthy controls’ interregional association was characterised by higher subcortical activity and lower cortical activity. Entering the extracted values into a group classification analysis yielded an area under the curve of 0.995 for a model that included all clusters with differences between groups, and 0.940 for hypometabolic or 0.945 for hypermetabolic clusters respectively (Fig. 2 h). The corresponding comparisons of static mean scans revealed no significant hypometabolism at the applied threshold and small clusters with hypermetabolism in the motor cortex at uncorrected p-level (not shown). Nigral metabolic networks on subject level and group level Seed-based metabolic connectivity analysis of the data-driven obtained substantia nigra clusters at the subject-level revealed the highest connectivity to nearby areas in the midbrain in all subjects for both sides (Fig. 3 b,c, Supplementary Fig. 1), which remained significant at a threshold of P < 0.05 after FWE correction. At a more liberal threshold, additional clusters in the right and left putamen, right caudate nucleus, right pallidum and in the left ventrolateral thalamus, bilateral ventroposterior thalamus, left inferior lateral thalamus, left anterior pulvinar, medial pulvinar were obtained in healthy controls for the left substantia nigra seed (t > 2, uncorrected P-level). In PwPD, the midbrain cluster extended into thalamic regions including mediodorsal thalamus, left medial pulvinar, left inferior lateral thalamus, left ventrolateral thalamus at a more liberal threshold for the left substantia nigra seed (t > 2, uncorrected P-level). In addition, the left caudate, bilateral putamen and left pallidum were connected to the left nigral seed region in PwPD. Subject-level metabolic connectivity maps described consistently for all subjects the highest connectivity to nearby left midbrain regions (Fig. 3 a). In all subjects, the left substantia nigra was connected to the left thalamus, left putamen and for some subjects also to the contralateral putamen or caudate nucleus. Similar results were observed for the right substantia nigra seed (Supplementary Fig. 1a). In a direct group comparison of metabolic connectivity maps of the left substantia nigra, a cluster with reduced metabolic connectivity extending rostral between the substantia nigra to medial pulvinar was observed. Smaller clusters with hypoconnectivity to the left substantia nigra seed were obtained in the right anterior putamen and pallidum, but only at uncorrected P-level. Alterations in region-wise glucose dynamics in PwPD relate to cognitive performance A group comparison of a measure of variation within the time series per subject in the identified hypo- and hypermetabolic clusters revealed a higher within-subject variance coefficient within the cortical cluster covering the supplementary motor area, precuneus, and bilateral post- and precentral gyri (P = 0.02, Fig. 4 a) in PwPD relative to healthy controls. The subjects’ variation coefficient of this region correlated significantly with cognitive performance evaluated by the screening tool MoCA (P = 0.022, r= -0.44) and cognitive z-scores, which included standardised performance scores from all cognitive domains (Fig. 4 b, P = 0.011, r=-0.48). An independent analysis of atlas-based regions representing the key nodes of canonical resting-state networks showed that dynamic signal changes indicated by variation coefficient occurred in the superior sensorimotor cortex in PwPD (P = 0.036). Again, higher variation coefficients in the superior sensorimotor cortex were associated with worse cognitive performance, measured by MoCA and cognitive z-scores (P = 0.031, r=-0.42, Supplementary Fig. 2b). A categorisation into groups according to cognitive performance based on common criteria revealed that five patients and two controls had mild cognitive impairment (see Supplementary Table 5 for detailed test results) and two out of these five exhibited the lowest cognition z-scores and highest variation coefficients. Altered metabolic level and seed-based networks alongside cognitive decline in PD The implemented seed-based approach for metabolic time series data revealed the typical default mode network structure in both independent imaging modalities in control subjects (Fig. 5 a, supplementary Table 6). The obtained maps contained the posterior cingulate cortex, precuneus and inferior parietal cortex, which was only included on the left hemisphere in the fPET modality (Fig. 5 a). In both modalities, less clusters were observed in PwPD and normal cognition and the least in PwPD and MCI (Fig. 5 a). The extracted time series yielded the lowest [ 18 F]-FDG uptake in the precuneal cortex and the highest [ 18 F]-FDG uptake in PwPD and normal cognition and control subjects in between (Fig. 5 b). By utilising a seed in the superior sensorimotor cortex, a typical cortical motor network was observed in all the groups in the fMRI modality (Fig. 5 a). While PwPD with mild cognitive impairment showed a more posteriorly distributed network and loss of lateral clusters, patients with normal cognition exhibited only small deficits in the precentral gyrus in comparison to controls. Conversely, the fPET motor networks included more clusters in the frontal lobe and subcortical clusters, which were absent in the fMRI modality in all groups (Fig. 5 a). The extracted time series revealed significant between-group differences (H = 9.06, P = 0.01) with the lowest [ 18 F]-FDG uptake in the superior sensorimotor cortex in controls and a significantly higher [ 18 F]-FDG uptake in PwPD with normal cognition in comparison to controls (P < 0.01 after Holm-Bonferroni correction) and the highest [ 18 F]-FDG uptake in patients with MCI (Fig. 5 b). Using the anterior cingulate cortex, the common structure of the salience network was observed in all groups in the fMRI modality (Fig. 5 a). In contrast, fPET networks were more restricted to the seed. The extracted time course revealed the lowest activity over time in the PD group with cognitive impairment and the highest in PwPD with normal cognition (Fig. 5 b). The right dorsolateral prefrontal cortex was connected to the inferior parietal cortex with similarly located clusters in both modalities, smaller clusters in fPET and less parietal clusters in the group of PwPD with MCI. The extracted time course showed a comparable activity in PwPD and healthy controls and lowest activity in patients with MCI over time (Fig. 5 b). Discussion Stereotypic metabolic patterns are promising candidates as circuit biomarkers for Parkinson’s disease. However, the methodological constraints of standard PET acquisition protocols, which provide only a static snapshot of glucose consumption, have hindered their routine application as interregional network markers on a subject-level. This study lays the groundwork for the application of a molecular imaging protocol with time series information in the context of neurodegeneration to overcome this hurdle. The key innovation of this protocol is the potential to derive metabolic time series reflecting glucose dynamics in the individual subject. We identified three PD-related metabolic changes based on this novel data structure. Firstly, we detected a typical pattern, including a bilateral hypometabolic cluster in the substantia nigra and hypermetabolic activity in the motor cortex in a small PD sample, which static mean scans could not capture. Secondly, symptom-related alterations in glucose dynamics at the individual level were identified and associated with distinct PD symptoms. Lastly, the data structure allowed us to depict metabolic connectivity maps on a subject-level, revealing subcortical basal-ganglia networks and typical cortical resting-state networks per subject based on metabolic time series information (Fig. 6 ). The observed clusters with hypometabolism in PwPD in our study align with previously reported patterns, indicating occipito-parietal hypometabolism in PD based on static PET 5 , 11 , 12 . Subcortical hypometabolism in midbrain clusters encompassing parts of the substantia nigra revealed by voxel wise group comparison in a high-resolution static PET data set were first described by some of the authors of this work in 2020 12,13 and also examined longitudinally as progression marker 14 . Another study reported similar findings with a region of interest-based approach in idiopathic PD as well as atypical parkinsonian syndromes and found lower metabolism in entities with worse nigrostriatal pathology 15 . The here reported subcortical clusters were closely confined to the bilateral substantia nigra pars compacta and reticulata as confirmed by the automated anatomical labeling atlas version 3 and the PD25 atlas. Given the small sample size and the absence of similar patterns in mean scans in our pilot study, it is plausible that the temporal resolution of the data with molecular time series information per subject allowed for the identification of more disease-related changes that correspond spatially with known spatial distribution of nigral cell loss. Additional multimodal imaging – such as neuromelanin-sensitive MRI and small animal PET studies in synucleinopathy models – may help corroborate the co-localisation with nigral pathology and dopaminergic cell loss 16 . Confirmation of a mechanistic association with nigral degeneration would underscore the translational potential of this imaging marker as a non-invasive tool for visualising nigral degeneration. There is preliminary evidence for nigral hypometabolism in patients with idiopathic rapid-eye movement sleep behavior disorder compared to matched controls by using an atlas-based approach with the median pons uptake value as the reference 17 . Due to the static character of commonly applied PET protocols, the variation coefficient of glucose dynamics had not been analysed yet. We report changes in the subject-level variation coefficient in the data-driven motor cortex cluster as well as in an atlas-based superior sensorimotor region in PwPD. The level of variation within the time series was associated with cognition in our cohort with patients presenting with MCI showing the highest variation coefficients in the superior sensorimotor cortex. Only a few studies have examined comparable measures of the blood oxygenation level dependent (BOLD) signal variability. One study using a working memory task found age-related increases in BOLD variability in older adults, particularly in the left and right precentral gyrus 10 . In individuals aged ~ 20–66 years, greater variability was associated with poor performance both during and outside the scanner. Our data suggest a similar association between the superior sensorimotor area, including pre-motor areas with cognitive impairment in PD based on a direct measure of neural activity. Another study has reported atrophy in the precentral gyrus and supplementary motor area in PD with MCI, which was accompanied by hypermetabolic changes in PD with dementia in a longitudinal design 18 . Together both findings might be indicative of adaptive processes that result from atrophy in this region. These findings should be analysed in longitudinal studies with larger sample sizes to examine the potential of these metabolic measures for detecting the progression of early cognitive dysfunction. The implementation of seed-based analysis for dynamic fPET data for the nigro-striatal system enabled the description of subcortical metabolic networks on subject-level. The nigral seed’s time course exhibited a plausible correlation with striatal and thalamic regions in all subjects often with continuous clusters to thalamic regions that appear like a continuous cluster between the midbrain and the thalamus. Two interesting aspects should be mentioned at this point. Firstly, a very recent study has provided evidence for a direct dopaminergic connection between the substantia nigra pars compacta and the thalamus in young healthy subjects by using multi-shell high-angular resolution diffusion MRI 19 . Second, the observed thalamic regions were among the most prominent regions within the a disease-specific metabolic network revealed by independent component analysis in this data set 20 : it is unknown how the here reported hypometabolic clusters, located ventrally to the brain stem part of the metabolic motor network, relate to the observed hypersynchronous motor network. If both findings are reflective of thalamic disinhibition in PwPD, is rather speculative and needs to be clarified in additional studies, incorporating larger sample sizes. The implementation of seed-based analysis for dynamic fPET data enabled the identification of canonical resting-state networks based on metabolic time series. At the applied threshold, the highest spatial similarity between fPET and fMRI was observed in the DMN across all groups. The seed-based analyses presented here can be only compared to previous studies that applied static PET to PD cohorts. In accordance with Sala et al., fewer clusters were observed in the prefrontal cortex in PD, with no clusters found in PwPD and MCI 21 . While the motor network identified using our superior sensorimotor cortex seed extended into lateral frontal and subcortical areas, such as the thalamus, in the PET modality, the fMRI network was confined to the pre- and postcentral gyri and supplementary motor area. Although the representation of the networks at the group level is of course still highly dependent on the group size, the results demonstrate that it is possible to use this technique to identify resting networks of comparable spatial extent through seed-based methods based on metabolic time series. Although our study provides important initial insights into metabolic time series, the protocol has some limitations. Firstly, we had a relatively small sample size, which was limited by the Federal Bureau for Radiation Protection and the ethics committee and represents an inherent limitation. This limits our power for detecting group-level networks and may contribute to the rather seed-confined nigral fPET networks in analyses with strict cluster-level FWE-correction. Due to the small sample size, especially the analysis of sub groups with different cognition levels are highly explorative and should be validated in larger samples. Secondly, we applied strict inclusion criteria to ensure that patients tolerate an off-phase longer than 12 h, which limits the generalisability of our findings to other disease stages. In addition, our multimodal acquisition protocol is afflicted with some restrictions. As both imaging procedure were performed consecutively in different scanners at different day times, it may be of question whether both represent the same resting-state of brain function. However, several requirements were undertaken to guarantee comparability: both procedures were performed in rooms with dimmed light under standardised conditions and standardised instructions were utilised like in comparable studies 12 . Our resting-state network analyses revealed common large-scale networks with comparable spatial distribution in both modalities. However, it is important to acknowledge that fPET-based networks are still derived based on a lower temporal resolution compared to fMRI. Nevertheless, it is unlikely that the slow dynamics of neurometabolic coupling can be better resolved in seconds 8 . Due to a low activity in initial frames, only the last 30 frames were retained in the presented analyses and it needs to be considered that preprocessing pipelines for constant infusion fPET are not yet as standardised as fMRI processing pipelines including denoising and intensity normalisation 8 . Finally, mapping small nuclei in subcortical areas is always subject to uncertainty. Therefore, all available atlases, including Parkinson's-specific atlases, were used to ensure that spatial mapping was as accurate as possible. Calculation of ROI-based measures, like the applied measure for time series variation, is dramatically dependent on ROI definition. Therefore, data-driven ROIs as well as atlas-based approaches were performed to validate the findings. additional ROI-based analyses depend on ROIs derived from the Human Connectome Project 22 . As described in Yaeger at al. 23 , it needs to be considered that these ROIs were derived from young adults and the transferability to elderly subjects needs to be handled with caution 23 , 24 . Conclusion In conclusion, the results of this study support the use of the constant infusion fPET protocol in context of neurodegenerative diseases and demonstrate its ability to detect subcortical metabolic alterations in PD, unidentified by corresponding averaged mean scans. The findings are consistent with patterns obtained using static protocols and validate our previous findings regarding midbrain hypometabolism in an independent, non-high-resolution small data set. Our study provides first insights into subject-level glucose dynamics and network connectivity based on metabolic time series information in a neurodegenerative disease. Methods Study participants and data collection The study received approval from the local ethics committee of the medical faculty of the Philipps-University of Marburg (146/19). Authorization for radiation exposure was obtained by the Federal Office for Radiation Protection. The study was carried out in adherence to the principles outlined in the Declaration of Helsinki and participants declared their written informed consent before participating. A total of 14 healthy controls (HC) and 15 Parkinson’s disease (PD) subjects were recruited, of which 13 HC and 14 PD patients provided [ 18 F]-FDG-PET data. PD subjects were recruited through the central study coordination of the Department of Neurology at the University Hospital of Marburg in Germany. HC subjects were recruited through advertisements. Inclusion criteria: german speaking, older than 50 years old, three to eight years of illness, Hoehn & Yahr (H&Y) stage 1-2.5 in motoric OFF-state, no therapeutic changes within three months. Patients needed to be able to endure a medication break of 12 h of non-retard and 72 h of retarded PD medication. Exclusion criteria: structural cerebral damage (e.g. vascular events, tumors), severe depression and motor complications, signs of dementia, safety concerns about MRI scanning like pacemaker, artificial heart valves, metal in the body (e.g. total endoprostheses) and claustrophobia, pregnancy and a blood glucose > 180 mg/dl at the time of PET examination. Clinical assessment Motor severity was tested according to part III of the Movement Disorder Society Unified Parkinson`s disease rating scale (MDS-UPDRS-III) in ON and OFF-state 25 . The levodopa equivalent daily dose (LEDD) was determined using established criteria 26 . All subjects underwent a cognitive test battery including Montreal Cognitive Assessment (MoCA), revised Wechsler Memory Scale (WMS-R), Parkinson Neuropsychometric Dementia Asessment (PANDA) 27 , and Regensburg Word Fluency test (RWT). The categorisation of all our subjects into mild cognitive impairment (MCI) and normal cognition (NC) was carried out according to Movement Disorder Society Level II criteria 28 . MCI was diagnosed in PD patients, when a difference of >/= 1.5*standard deviation was observed in relation to age-matched norm means in at least two cognitive test results regardless of domain affiliation. Resting-state [ 18 F]-FDG-PET and fMRI acquisition The dynamic [ 18 F]-FDG-PET scans were acquired on a SIEMENS Biograph 6 Scanner (Siemens, Germany) at the Department of Nuclear Medicine at the University Hospital of Marburg, Germany. Measurements of all subjects were carried out in OFF-state after overnight fasting and testing of blood sugar levels under standardised conditions. An average, 199.3 ± 5.27 MBq of [ 18 F]-FDG were infused via i.v. injection continuously using a perfusor (Braun, Germany) at a rate of 0.01 ml/s. Tomographic images were acquired dynamically for 90 min. MRI scanning was performed on a Trio Tim Syngo 3 Tesla MR-scanner (Siemens, Erlangen) at the Department for Psychiatry and Psychotherapy of the University Hospital of Marburg, Germany. Participants underwent structural MRI with the following parameters: repetition time (TR): 1900 ms, echo time (TE): 2.52 ms, voxel size: 1.0 × 1.0 × 1.0 mm 3 . For fMRI measurements, subjects were instructed to keep their eyes opened and to avoid thinking about anything in particular. The eye area was checked by camera throughout the measurement. The 8-minute lasting multiband echo-planar imaging sequence 29 was characterised by the following parameters: 490 time points, TR 1040 ms, TE 30.0 ms, 3 × 3 × 3 mm 3 voxel size and 48 slices. DICOM files were converted into NifTi files using the dicom2niix tool in MRICroGL. The detailed overview about further preprocessing and analysis of the data is described in Ruppert-Junck et al. 2024 20 and corresponding scripts are available on our Github repository: https://github.com/ruppertm/fPETDynamics.git . Neuroimaging data analysis Voxel-wise group comparison in a subject-and-time design A flexible factorial design in SPM12 was applied to evaluate voxel-wise differences in metabolic activity within the time series between healthy controls and PD patients, considering the last 30 frames of spatially normalised scans per subject. Two main effects were created: the first with factor number 1 (Subject) and the second with factor number 2 (Time). Global normalisation was conducted using ANCOVA with reference to the global mean. Additionally, a grey matter mask in MNI-space was applied to restrict the analysis to grey matter regions (ICBM 2009c non-linear symmetric, FSL). Results were considered statistically significant if p < 0.05 after family-wise error (FWE) rate correction at cluster-level and surpassed a minimum cluster size of 20 voxels. Anatomical labelling done by using the AAL v3 Atlas. Activity values per region were extracted using the MarsBaR toolbox in SPM12 and global mean-normalised values were plotted in R. Mean normalised uptake values per contrast were entered into a logistic regression and utilised to derive receiver operating characteristic curves per contrast or for all clusters. Functional and metabolic connectivity analyses The fMRI data set was preprocessed according to standardised procedures as described recently 20 . Seed-based resting-state functional connectivity analysis was performed with the default weighted general linear model using the CONN toolbox 22 . Anatomical labelling was performed by using the Harvard-Oxford atlas as implemented in CONN. Metabolic connectivity within the striato-nigro-thalamic system Seed-based metabolic connectivity analyses were performed using scripts relying on SPM12 in Matlab v23a which can be found on our Github Repository ( https://github.com/ruppertm/fPETDynamics.git ). Mean uptake time series were extracted from substantia nigra clusters per subject from global-mean corrected scans and utilised as covariate of interest in a voxel-wise regression analysis per subject. The obtained t -maps per subject were transformed into z-maps. All individuals’ first-level contrast images were entered into a one sample t -test for visualisation on group level and into second-level group comparison evaluating voxel-wise differences in metabolic connectivity. A masked analyses was performed focused on the striato-nigro-thalamic network by using as mask created with the corresponding regions from AALv1 and TD atlas. Subject-level contrast images were visualised with effects between 0.2–1.2 and group results presented at P < 0.05 after FWE correction. Calculation of measures of glucose dynamics – variation coefficient Differences between the groups were derived by permutation tests ( https://github.com/ruppertm/fPETDynamics.git ). The association to clinical variables (UPDRS-III, MoCa, cognitive z-scores) was analysed by linear regression and correlation analyses in R. Network ROIs were defined via independent component analysis of Human Connectome Project data (N = 497) (Whitfield-Gabrieli & Nieto-Castanon, 2012). Network seeds are listed with x, y, z coordinates for the centroid of each seed. Statistical Analysis of clinical data Statistical analysis of clinical data were performed on demographic, behavioral and clinical data using R (RRID: SCR_001905) 30 . Group comparison were performed with two-sided Welch’s t-test or ( Mann-Whitney U test ) based on the results of Shapiro-Wilk test of normality. Multiple group comparisons were performed by using Kruskal-Wallis test and pairwise Wilcoxon tests with Bonferroni-Holm correction for multiple comparison. Declarations Acknowledgements We wish to convey our deep gratitude to the individuals who made this research possible: the members of the Core Facility Brain Imaging Marburg, particularly Prof. Andreas Jansen, Dr. Jens Sommer, Mechthild Wallnig and Rita Werner. We would like to extend our appreciation to our colleagues from the Department of Nuclear Medicine at the University Hospital Marburg, as well as Stefanie Spriewald from the study coordination of the Department of Neurology, for their invaluable assistance with data acquisition. Lastly, we would like to express our heartfelt appreciation to all the participants who generously contributed to this current research project. Author contributions Conceptualisation, D.P., V.H., M.R-J., K.S.; Methodology, K.S.,M.R-J,L.R.,F.T.; Software,K.S.,L.R.,F.T.,M.R-J.,V.H; Formal Analysis, L.M., L.R.,M.R-J.,F.T.;Investigation, J.F., V.H., M.R-J.,K.S.,D.L,M.B; Data curation, D.P., M.R-J; Resources, L.T.,D.P.,M.L; Writing – Original draft, V.H., M.R-J., D.P.; Writing – review & Editing, V.H.,D.L,M.R-J.,D.P.,F.T.,L.R.,L.M.,M.L.,M.B,L.T.,M.L; Funding acquisition, D.P; Project administration, D.P.,M.R-J; Visualisation, M.R-J, L.R.,V.H.; Supervision, D.P.,M.R-J Data availability The data supporting the findings of the present study will be made available by the corresponding author upon reasonable request. Competing interests LT reports grants, personal fees, and non-financial support from SAPIENS Steering Brain Stimulation, Medtronic, Boston Scientific, and St. Jude Medical, and has received payments from Bayer Healthcare, UCB Schwarz Pharma, and Archimedes Pharma and also honoraria as a speaker on symposia sponsored by Teva Pharma, Lundbeck Pharma, Bracco, Gianni PR, Medas Pharma, UCB Schwarz Pharma, Desitin Pharma, Boehringer Ingelheim, GSK, Eumecom, Orion Pharma, Medtronic, Boston Scientific, Cephalon, Abbott, GE Medical, Archimedes, and Bayer. DP received honoraria as a speaker at symposia sponsored by Boston Scientific Corp, Medtronic, AbbVie Inc., Zambon, and Esteve Pharmaceuticals GmbH. He received payments as a consultant for Boston Scientific Corp and Bayer, and he received a scientific grant from Boston Scientific Corp for a project entitled: ‘Sensor-based optimisation of Deep Brain Stimulation settings in Parkinson’s disease’ (compareDBS). Finally, DP was reimbursed by Esteve Pharmaceuticals GmbH and Boston Scientific Corp for travel expenses to attend congresses. The remaining authors declare no conflicts of interest. Correspondence and requests for materials should be addressed to Marina Ruppert-Junck. References Dorsey, E. R. & Bloem, B. R. The Parkinson Pandemic—A Call to Action. JAMA Neurol. 75 , 9 (2018). Palop, J. J., Chin, J. & Mucke, L. A network dysfunction perspective on neurodegenerative diseases. Nature 443 , 768–773 (2006). Schindlbeck, K. A. & Eidelberg, D. Network imaging biomarkers: insights and clinical applications in Parkinson’s disease. Lancet Neurol. 17 , 629–640 (2018). Niethammer, M. et al. Gene therapy reduces Parkinson’s disease symptoms by reorganizing functional brain connectivity. Sci Transl Med 10 , (2018). Huang, C. et al. Metabolic brain networks associated with cognitive function in Parkinson’s disease. NeuroImage 34 , 714–723 (2007). Ruppert, M. C. et al. The default mode network and cognition in Parkinson’s disease: A multimodal resting-state network approach. Hum. Brain Mapp. 10.1002/hbm.25393 (2021). Villien, M. et al. Dynamic functional imaging of brain glucose utilization using fPET-FDG. NeuroImage 100, 192–9 (2014). Jamadar, S. D. et al. Metabolic and Hemodynamic Resting-State Connectivity of the Human Brain: A High-Temporal Resolution Simultaneous BOLD-fMRI and FDG-fPET Multimodality Study. Cereb. Cortex N Y N . 1991 10.1093/cercor/bhaa393 (2021). Voigt, K. et al. Metabolic and functional connectivity provide unique and complementary insights into cognition-connectome relationships. Cereb. Cortex N Y N . 1991 10.1093/cercor/bhac150 (2022). Boylan, M. A. et al. Greater BOLD Variability is Associated With Poorer Cognitive Function in an Adult Lifespan Sample. Cereb. Cortex . 31 , 562–574 (2021). Eckert, T. et al. FDG PET in the differential diagnosis of parkinsonian disorders. NeuroImage 26 , 912–921 (2005). Ruppert, M. C. et al. Network degeneration in Parkinson’s disease: multimodal imaging of nigro-striato-cortical dysfunction. Brain J. Neurol. 143 , 944–959 (2020). Glaab, E. et al. Integrative analysis of blood metabolomics and PET brain neuroimaging data for Parkinson’s disease. Neurobiol. Dis. 124 , 555–562 (2019). Steidel, K. et al. Longitudinal trimodal imaging of midbrain-associated network degeneration in Parkinson’s disease. NPJ Park Dis. 8 , 79 (2022). Schröter, N. et al. Nigral glucose metabolism as a diagnostic marker of neurodegenerative parkinsonian syndromes. NPJ Park Dis. 8 , 123 (2022). Jang, D. P. et al. Functional neuroimaging of the 6-OHDA lesion rat model of Parkinson’s disease. Neurosci. Lett. 513 , 187–192 (2012). Diaz-Galvan, P. et al. Brain glucose metabolism and nigrostriatal degeneration in isolated rapid eye movement sleep behaviour disorder. Brain Commun. 5 , fcad021 (2022). González-Redondo, R. et al. Grey matter hypometabolism and atrophy in Parkinson’s disease with cognitive impairment: a two-step process. Brain 137 , 2356–2367 (2014). Cirillo, G. et al. Evidence for direct dopaminergic connections between substantia nigra pars compacta and thalamus in young healthy humans. Front. Neural Circuits . 18 , 1522421 (2025). Ruppert-Junck, M. C. et al. Connectivity based on glucose dynamics reveals exaggerated sensorimotor network coupling on subject-level in Parkinson’s disease. Eur. J. Nucl. Med. Mol. Imaging . 10.1007/s00259-024-06796-6 (2024). Sala, A. et al. Altered brain metabolic connectivity at multiscale level in early Parkinson’s disease. Sci. Rep. 7 , 4256 (2017). Whitfield-Gabrieli, S. & Nieto-Castanon, A. Conn: a functional connectivity toolbox for correlated and anticorrelated brain networks. Brain Connect. 2 , 125–141 (2012). Yeager, B. E., Twedt, H. P., Bruss, J., Schultz, J. & Narayanan, N. S. Cortical and subcortical functional connectivity and cognitive impairment in Parkinson’s disease. NeuroImage Clin. 42 , 103610 (2024). Pereira, J. B. et al. Assessment of cortical degeneration in patients with Parkinson’s disease by voxel-based morphometry, cortical folding, and cortical thickness. Hum. Brain Mapp. 33 , 2521–2534 (2012). Goetz, C. G. et al. Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS‐UPDRS): Scale presentation and clinimetric testing results. Mov. Disord . 23 , 2129–2170 (2008). Jost, S. T. et al. Levodopa Dose Equivalency in Parkinson’s Disease: Updated Systematic Review and Proposals. Mov. Disord Off J. Mov. Disord Soc. 38 , 1236–1252 (2023). Kalbe, E. et al. Screening for cognitive deficits in Parkinson’s disease with the Parkinson neuropsychometric dementia assessment (PANDA) instrument. Parkinsonism Relat. Disord . 14 , 93–101 (2008). Litvan, I. et al. Diagnostic criteria for mild cognitive impairment in Parkinson’s disease: Movement Disorder Society Task Force guidelines. Mov. Disord Off J. Mov. Disord Soc. 27 , 349–356 (2012). Auerbach, E. J., Xu, J., Yacoub, E., Moeller, S. & Uğurbil, K. Multiband accelerated spin-echo echo planar imaging with reduced peak RF power using time‐shifted RF pulses. Magn. Reson. Med. 69 , 1261–1267 (2013). R Core Team. R: A Language and Environment for Statistical Computing. R Found. Stat. Comput (2018). Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterial.pdf Cite Share Download PDF Status: Published Journal Publication published 18 Nov, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 10 Sep, 2025 Reviews received at journal 04 Sep, 2025 Reviews received at journal 02 Sep, 2025 Reviews received at journal 31 Aug, 2025 Reviews received at journal 21 Aug, 2025 Reviewers agreed at journal 17 Aug, 2025 Reviewers agreed at journal 16 Aug, 2025 Reviewers agreed at journal 15 Aug, 2025 Reviews received at journal 15 Aug, 2025 Reviewers agreed at journal 14 Aug, 2025 Reviewers agreed at journal 14 Aug, 2025 Reviewers invited by journal 13 Aug, 2025 Editor assigned by journal 31 Jul, 2025 Editor invited by journal 29 Jul, 2025 Submission checks completed at journal 25 Jul, 2025 First submitted to journal 25 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7156182","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":502154200,"identity":"374a2936-d872-4186-8f2f-0ae618445fe8","order_by":0,"name":"Vanessa Heinecke","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Vanessa","middleName":"","lastName":"Heinecke","suffix":""},{"id":502154201,"identity":"b37d16d0-5677-407f-b605-4ddb91135c2e","order_by":1,"name":"Lilly Machholz","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Lilly","middleName":"","lastName":"Machholz","suffix":""},{"id":502154202,"identity":"9f84befc-86c1-4949-8bfe-d923ed6b0504","order_by":2,"name":"Kenan Steidel","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Kenan","middleName":"","lastName":"Steidel","suffix":""},{"id":502154203,"identity":"c2976d1a-46ed-4e29-9a91-cd73bdecb6f9","order_by":3,"name":"Lenna M. Rüsing","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Lenna","middleName":"M.","lastName":"Rüsing","suffix":""},{"id":502154204,"identity":"5aa9a99e-ee8e-4e0d-9898-ab5da03c96d6","order_by":4,"name":"Falk K. Thiemig","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Falk","middleName":"K.","lastName":"Thiemig","suffix":""},{"id":502154205,"identity":"55ebbaed-e019-45b8-ae90-76560a03a834","order_by":5,"name":"Damiano Librizzi","email":"","orcid":"","institution":"Philipps University of Marburg","correspondingAuthor":false,"prefix":"","firstName":"Damiano","middleName":"","lastName":"Librizzi","suffix":""},{"id":502154206,"identity":"42c4153e-ef55-4628-9f23-97be919ec155","order_by":6,"name":"Maya Beckersjürgen","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Maya","middleName":"","lastName":"Beckersjürgen","suffix":""},{"id":502154207,"identity":"3c3d7d88-6d84-421e-833a-db4ed782e130","order_by":7,"name":"Jennifer Fuchs","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Jennifer","middleName":"","lastName":"Fuchs","suffix":""},{"id":502154208,"identity":"9c078412-27ce-4ff4-b345-b3a3b5d1e6bc","order_by":8,"name":"Markus Luster","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Markus","middleName":"","lastName":"Luster","suffix":""},{"id":502154209,"identity":"9f136414-f3dc-4b2e-ac1e-db1827f06ced","order_by":9,"name":"Lars Timmermann","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"Lars","middleName":"","lastName":"Timmermann","suffix":""},{"id":502154210,"identity":"ffea677f-8c7e-4866-b69a-8849bb8f47d0","order_by":10,"name":"David Pedrosa","email":"","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Pedrosa","suffix":""},{"id":502154211,"identity":"97522dba-44a7-44e0-b5e3-abea3615ac7e","order_by":11,"name":"Marina C. Ruppert-Junck","email":"data:image/png;base64,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","orcid":"","institution":"Philipps-University Marburg","correspondingAuthor":true,"prefix":"","firstName":"Marina","middleName":"C.","lastName":"Ruppert-Junck","suffix":""}],"badges":[],"createdAt":"2025-07-18 09:38:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7156182/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7156182/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-28456-x","type":"published","date":"2025-11-18T15:57:58+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89592591,"identity":"e057eb03-4152-401b-a373-486e5f6fb251","added_by":"auto","created_at":"2025-08-21 16:12:42","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2261211,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic overview of the analysis pipeline enabled by the dynamic constant infusion fPET protocol\u003c/strong\u003e. The temporal structure of our data provides several novel avenues in context of studying metabolic changes in neurodegenerative diseases: metabolic activity within the time series can be analysed in a subject x time design, providing detailed molecular information per subject (a), within-subject variation in metabolic activity can be studied and related to group level and behavioral variables (b), seed-based connectivity can be studied on individual level based on metabolic time series, directly compared to fMRI connectivity maps and compared between groups (c). VC = variation coefficient, fPET = functional positron emission tomography, HC=healthy controls, PD = Parkinson’s disease\u003c/p\u003e","description":"","filename":"Fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/9a2e3a5ae734ea468379b3f4.jpg"},{"id":89593396,"identity":"e8492cc4-d0a3-4058-b368-4ae049106e89","added_by":"auto","created_at":"2025-08-21 16:20:42","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1586758,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHypometabolic and hypermetabolic clusters in PwPD relative to healthy controls based on metabolic time series comparison.\u003c/strong\u003e SPM result of HC\u0026gt;PD visualised on MNI standard template. Sagittal, coronal and axial view of regions of hypometabolic (a-c) or hypermetabolic (d) activity obtained by voxel-wise group comparison of [\u003csup\u003e18\u003c/sup\u003eF]-FDG fPET scans from 13 healthy controls and 14 PD patients (P \u0026lt; 0.05 after FWE cluster level correction, extent threshold \u0026gt; 20 voxels). (e-f) Extracted individual and group mean time courses per cluster (MNI peak coordinates). (g) Regression plots for subcortical hypometabolism vs. cortical hypermetabolism (filled dots represent mean uptake per region). (h) Receiver operating characteristic curve illustrating the trade-off between sensitivity and specificity for group classification based on all clusters, hypometabolic or hypermetabolic clusters respectively. All images are shown in neurological display. PD = Parkinson’s disease, HC = healthy controls, SN = substantia nigra\u003c/p\u003e","description":"","filename":"Fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/cf45e6bf31f83093f7bd0c34.jpg"},{"id":89594369,"identity":"3d0e0c45-e8bc-4842-8d92-021edc3d1606","added_by":"auto","created_at":"2025-08-21 16:28:42","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1966932,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSeed-based nigral networks in PwPD and healthy controls on subject- and group-level.\u003c/strong\u003e SPM result of first-level single-subject regression analysis using the left substantia nigra as seed visualised on MNI standard template. Sagittal, coronal and axial view of regions of (a) subject-level contrast images or (b) group-level \u003cem\u003et\u003c/em\u003e-maps obtained by voxel-wise analysis of [\u003csup\u003e18\u003c/sup\u003eF]-FDG PET contrast images from 13 healthy controls and 14 PD patients (P \u0026lt; 0.05 after FWE cluster level correction). Color bar represents subject-level effect in contrast images (a) or \u003cem\u003et\u003c/em\u003e-values of a voxel wise one-sample \u003cem\u003et\u003c/em\u003e-test of contrast images (b). Displayed slices in (a) are positioned at axial z = -14, +8, coronal y= -14 and sagittal x= -9. All images are shown in neurological display.\u003c/p\u003e","description":"","filename":"Fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/b02a044145b6f9a64df48f9e.jpg"},{"id":89594370,"identity":"04687c05-4d90-465e-8ae5-df1dc17d9107","added_by":"auto","created_at":"2025-08-21 16:28:42","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":587504,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMetabolic time series variation in PwPD relative to healthy controls.\u003c/strong\u003e Subject-level variation coefficients from extracted metabolic time series in (\u003cstrong\u003ea)\u003c/strong\u003e regions with metabolic alterations in PwPD in the present cohort. Marked dots represent outliers with values \u0026gt; 1.5 times the inter-quartile distance. \u003cstrong\u003eb\u003c/strong\u003e shows a scatterplot relating regional variation coefficients to cognitive performance measured by cognition z-scores. The subjects marked in rectanglular shape represent the subjects classified as MCI. All images are shown in neurological display. Abbreviations: HC = control subjects, PwPD = persons with Parkinson’s disease, SN = substantia nigra, OC = occipital cortex, AP = anterior putamen, MFC = mid frontal cortex, AG = angular gyrus, PP = posterior putamen, MC = motor cortex, VC = variation coefficient\u003c/p\u003e","description":"","filename":"Fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/e5c7fb1cee1c1ad6023f13a5.jpg"},{"id":89592606,"identity":"faea5856-21bd-4b0b-9bb1-94ae16f50233","added_by":"auto","created_at":"2025-08-21 16:12:42","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1785573,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSeed-based metabolic and hemodynamic networks in PwPD with and without cognitive impairment and healthy controls.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003eGroup-level metabolic (fPET) and hemodynamic (fMRI) seed-based networks obtained by voxel-wise one-sample \u003cem\u003et\u003c/em\u003e-tests with respective subject-level contrast images for n=11 HC, n=9 PD-NC (fMRI:n=7) and n=5 PD-MCI. \u003cstrong\u003eb\u003c/strong\u003eExtracted fPET time course of the corresponding seed region with subject-level time courses (dotted line) and mean time course per group (filled line). Colorbar indicates t-values between 3 and 11 for comparability of the modalities. All images are shown in neurological display. Abbreviations: HC = control subjects, PD = Parkinson’s disease, ACC = anterior cingulate cortex, LPFC-R = right lateral prefrontal cortex, PCC = precuneus cortex, SMS = superior sensorimotor cortex, MCI = mild cognitive impairment, NC = normal cognition, ROI = region of interest\u003c/p\u003e","description":"","filename":"Fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/a453a56a84de00c18060296d.jpg"},{"id":89592601,"identity":"7dae3e72-f60a-44de-b7ba-f33b117f5b00","added_by":"auto","created_at":"2025-08-21 16:12:42","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1066349,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic overview of metabolic features captured by the constant infusion fPET protocol in PD.\u003c/strong\u003e Our study pinpoints three main effects on characterising PD-related metabolic changes based on this novel data structure. Firstly, we detected a typical pattern, including a bilateral hypometabolic cluster in the substantia nigra and hyperactivity in the motor cortex in a small PD sample. Second, symptom-related alterations in variations of glucose consumption on individual level could be captured. Lastly, the data structure enabled us to depict seed-based metabolic connectivity maps on subject-level, which revealed a metabolic network covering the basal ganglia for nigral seed regions.\u003c/p\u003e","description":"","filename":"Fig6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/55247a6a8b344d474d2da245.jpg"},{"id":96650143,"identity":"6e19e196-f7d8-44da-be97-e8c45f68b5df","added_by":"auto","created_at":"2025-11-24 16:08:41","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10344223,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/a7e3bb94-f9cf-43d1-93fa-59441139d2e3.pdf"},{"id":89592590,"identity":"e4c34822-172d-4eed-b354-4bfdfee85a18","added_by":"auto","created_at":"2025-08-21 16:12:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1488054,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7156182/v1/89954c33f45b85cc68a288b0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Functional PET for mapping metabolic dynamics in Parkinson’s disease","fulltext":[{"header":"Introduction","content":"\u003cp\u003eParkinson\u0026rsquo;s disease (PD) is the neurodegenerative disease with the fastest growing prevalence worldwide\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. PD is primarily considered a movement disorder, although the complex clinical manifestation, including non-motor symptoms, are not solely attributable to dopaminergic depletion in basal ganglia-cortex loops. Instead, network-level dysfunction is assumed to underly PD and specifically, more widespread aberrant interregional neural communication\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-fluorodeoxyglucose ([\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG) positron emission tomography (PET), as a direct correlate of neural activity indexing synaptic activity, has increasingly been utilised for brain network visualisation and holds significant potential as a network biomarker candidate\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eA stereotypic PD-related metabolic pattern with hypometabolism in occipito-parietal regions and hypermetabolism in the supplementary motor area and the putamina relative to healthy controls has been repeatedly observed \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. In particular, it is well established that hypometabolism in the occipital cortex\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e and regions covering the default-mode network relate to cognitive decline in PD\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Reliability of former results have been hampered by the static nature of classical PET acquisitions, providing only a snapshot per subject, so that only limited insights into metabolic dynamics and interregional communication at the individual level have been possible to date. Nonetheless, this information is crucial for uncovering network-level metabolic correlates of individual symptoms. Only recently, Villien et al. introduced a functional [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG-PET (fPET) protocol leveraging constant tracer infusion and list-mode acquisition for information on glucose dynamics in the individual subject\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. The technique offers a direct and quantitative measure of neuronal function\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. In young healthy subjects, Voigt et al. found a strong association between cognition and metabolic network activity in fronto-parietal areas\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Irrespective of this, time series variation measures have been identified as key drivers of cognitive performance in aging\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. However, metabolic networks and glucose dynamics at the subject-level remain underexplored within the context of neurodegenerative disorders despite their potential role for symptom expression.\u003c/p\u003e\u003cp\u003eTo close this gap, we combined resting-state [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG-fPET with fMRI acquisition to study glucose dynamics in PD. We aimed to evaluate between-group differences in glucose metabolism based on time series data compared to healthy subjects, to establish a measure of glucose dynamics at the subject-level, and to develop a seed-based network approach for analysing metabolic time series data (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This approach offers new insights into how glucose dynamics contributes to network-level changes related to individual PD symptoms. The metabolic time series data enabled us to identify hypometabolic clusters within the substantia nigra in our small PD sample. Moreover, we report an increase in time series variation associated with cognitive symptoms in PD and seed-based metabolic networks on subject-level. These findings may lay the groundwork for developing novel network-based imaging markers for neurodegenerative diseases using molecular time series information.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eStudy participants and clinical characterisation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eA total of 14 persons with PD (PwPD) (63.43\u0026thinsp;\u0026plusmn;\u0026thinsp;8.92 years, three females) and 13 healthy controls (59.54\u0026thinsp;\u0026plusmn;\u0026thinsp;5.13 years, six females) underwent [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG-fPET scanning. Cognitive screening tests did not reveal differences in cognitive function between the groups (Montreal Cognitive Assessment (MoCA): W\u0026thinsp;=\u0026thinsp;118.5, P\u0026thinsp;=\u0026thinsp;0.19). The average MDS-UPDRS-III score for PwPD without medication was 35.64\u0026thinsp;\u0026plusmn;\u0026thinsp;17.54 points, indicating mild to moderate severity of overall motor symptoms (refer to supplementary Table\u0026nbsp;1 for clinical and demographic information).\u003c/p\u003e\u003cp\u003e\u003cb\u003eNigral hypometabolism and hypermetabolic activity in the motor cortex in PD\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe voxel-wise group comparison of cerebral [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake time series using a flexible factorial design revealed subcortical and cortical regions with altered metabolic activity in PwPD compared to controls (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 after family-wise error (FWE) rate correction). Regional hypometabolism was observed in the temporal, parietal, occipital and frontal lobes in PwPD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Parietal clusters included the bilateral angular gyri and precuneus (P\u003csub\u003eFWE\u003c/sub\u003e\u0026lt;0.0001, Supplementary Table\u0026nbsp;2). Subcortical hypometabolism was present in the basal ganglia: the bilateral substantia nigra pars compacta and reticulata (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea-c), the putamina and the left caudate nucleus (P\u003csub\u003eFWE\u003c/sub\u003e\u0026lt;0.0001, Supplementary Table\u0026nbsp;2). The cerebellum, the right amygdala, the left insula and the dorsal part of the thalamus also showed relative hypometabolism in PwPD (P\u003csub\u003eFWE\u003c/sub\u003e\u0026lt;0.0001, Supplementary Table\u0026nbsp;2). Significant hypermetabolic regions in PwPD were observed in a large cluster spanning the supplementary motor area, the middle cingulate and the precuneus (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed, P\u003csub\u003eFWE\u003c/sub\u003e\u0026lt;0.0001, Supplementary Table\u0026nbsp;2). The bilateral putamina, the orbitofrontal cortex, the medial occipital lobes, the fusiform gyrus and the cerebellum exhibited increased activity in PwPD (P\u003csub\u003eFWE\u003c/sub\u003e\u0026lt;0.0001). The right nucleus accumbens also showed relative hypermetabolism in PwPD (P\u003csub\u003eFWE\u003c/sub\u003e\u0026lt;0.0001, Supplementary Table\u0026nbsp;2).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eExtracted relative [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake values from all clusters revealed contrast-specific patterns of signal time courses within hypometabolic clusters with a lower level of metabolic activity over time in PwPD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee) and hypermetabolic clusters characterised by a higher activity over time in PwPD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef). Contrasting the individual [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG activity between hypometabolic clusters in the substantia nigra and hypermetabolic clusters in the motor cortex indicated that reduced subcortical activity is accompanied by corresponding increased cortical activity in PwPD (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg). On the other hand, healthy controls\u0026rsquo; interregional association was characterised by higher subcortical activity and lower cortical activity. Entering the extracted values into a group classification analysis yielded an area under the curve of 0.995 for a model that included all clusters with differences between groups, and 0.940 for hypometabolic or 0.945 for hypermetabolic clusters respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eh). The corresponding comparisons of static mean scans revealed no significant hypometabolism at the applied threshold and small clusters with hypermetabolism in the motor cortex at uncorrected p-level (not shown).\u003c/p\u003e\u003cp\u003e\u003cb\u003eNigral metabolic networks on subject level and group level\u003c/b\u003e\u003c/p\u003e\u003cp\u003eSeed-based metabolic connectivity analysis of the data-driven obtained substantia nigra clusters at the subject-level revealed the highest connectivity to nearby areas in the midbrain in all subjects for both sides (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb,c, Supplementary Fig.\u0026nbsp;1), which remained significant at a threshold of P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 after FWE correction. At a more liberal threshold, additional clusters in the right and left putamen, right caudate nucleus, right pallidum and in the left ventrolateral thalamus, bilateral ventroposterior thalamus, left inferior lateral thalamus, left anterior pulvinar, medial pulvinar were obtained in healthy controls for the left substantia nigra seed (t\u0026thinsp;\u0026gt;\u0026thinsp;2, uncorrected P-level). In PwPD, the midbrain cluster extended into thalamic regions including mediodorsal thalamus, left medial pulvinar, left inferior lateral thalamus, left ventrolateral thalamus at a more liberal threshold for the left substantia nigra seed (t\u0026thinsp;\u0026gt;\u0026thinsp;2, uncorrected P-level). In addition, the left caudate, bilateral putamen and left pallidum were connected to the left nigral seed region in PwPD. Subject-level metabolic connectivity maps described consistently for all subjects the highest connectivity to nearby left midbrain regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). In all subjects, the left substantia nigra was connected to the left thalamus, left putamen and for some subjects also to the contralateral putamen or caudate nucleus. Similar results were observed for the right substantia nigra seed (Supplementary Fig.\u0026nbsp;1a). In a direct group comparison of metabolic connectivity maps of the left substantia nigra, a cluster with reduced metabolic connectivity extending rostral between the substantia nigra to medial pulvinar was observed. Smaller clusters with hypoconnectivity to the left substantia nigra seed were obtained in the right anterior putamen and pallidum, but only at uncorrected P-level.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eAlterations in region-wise glucose dynamics in PwPD relate to cognitive performance\u003c/b\u003e\u003c/p\u003e\u003cp\u003eA group comparison of a measure of variation within the time series per subject in the identified hypo- and hypermetabolic clusters revealed a higher within-subject variance coefficient within the cortical cluster covering the supplementary motor area, precuneus, and bilateral post- and precentral gyri (P\u0026thinsp;=\u0026thinsp;0.02, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea) in PwPD relative to healthy controls. The subjects\u0026rsquo; variation coefficient of this region correlated significantly with cognitive performance evaluated by the screening tool MoCA (P\u0026thinsp;=\u0026thinsp;0.022, r= -0.44) and cognitive z-scores, which included standardised performance scores from all cognitive domains (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb, P\u0026thinsp;=\u0026thinsp;0.011, r=-0.48). An independent analysis of atlas-based regions representing the key nodes of canonical resting-state networks showed that dynamic signal changes indicated by variation coefficient occurred in the superior sensorimotor cortex in PwPD (P\u0026thinsp;=\u0026thinsp;0.036). Again, higher variation coefficients in the superior sensorimotor cortex were associated with worse cognitive performance, measured by MoCA and cognitive z-scores (P\u0026thinsp;=\u0026thinsp;0.031, r=-0.42, Supplementary Fig.\u0026nbsp;2b). A categorisation into groups according to cognitive performance based on common criteria revealed that five patients and two controls had mild cognitive impairment (see Supplementary Table\u0026nbsp;5 for detailed test results) and two out of these five exhibited the lowest cognition z-scores and highest variation coefficients.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eAltered metabolic level and seed-based networks alongside cognitive decline in PD\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe implemented seed-based approach for metabolic time series data revealed the typical default mode network structure in both independent imaging modalities in control subjects (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea, supplementary Table\u0026nbsp;6). The obtained maps contained the posterior cingulate cortex, precuneus and inferior parietal cortex, which was only included on the left hemisphere in the fPET modality (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). In both modalities, less clusters were observed in PwPD and normal cognition and the least in PwPD and MCI (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). The extracted time series yielded the lowest [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake in the precuneal cortex and the highest [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake in PwPD and normal cognition and control subjects in between (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). By utilising a seed in the superior sensorimotor cortex, a typical cortical motor network was observed in all the groups in the fMRI modality (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). While PwPD with mild cognitive impairment showed a more posteriorly distributed network and loss of lateral clusters, patients with normal cognition exhibited only small deficits in the precentral gyrus in comparison to controls. Conversely, the fPET motor networks included more clusters in the frontal lobe and subcortical clusters, which were absent in the fMRI modality in all groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). The extracted time series revealed significant between-group differences (H\u0026thinsp;=\u0026thinsp;9.06, P\u0026thinsp;=\u0026thinsp;0.01) with the lowest [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake in the superior sensorimotor cortex in controls and a significantly higher [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake in PwPD with normal cognition in comparison to controls (P\u0026thinsp;\u0026lt;\u0026thinsp;0.01 after Holm-Bonferroni correction) and the highest [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG uptake in patients with MCI (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). Using the anterior cingulate cortex, the common structure of the salience network was observed in all groups in the fMRI modality (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). In contrast, fPET networks were more restricted to the seed. The extracted time course revealed the lowest activity over time in the PD group with cognitive impairment and the highest in PwPD with normal cognition (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). The right dorsolateral prefrontal cortex was connected to the inferior parietal cortex with similarly located clusters in both modalities, smaller clusters in fPET and less parietal clusters in the group of PwPD with MCI. The extracted time course showed a comparable activity in PwPD and healthy controls and lowest activity in patients with MCI over time (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eStereotypic metabolic patterns are promising candidates as circuit biomarkers for Parkinson\u0026rsquo;s disease. However, the methodological constraints of standard PET acquisition protocols, which provide only a static snapshot of glucose consumption, have hindered their routine application as interregional network markers on a subject-level. This study lays the groundwork for the application of a molecular imaging protocol with time series information in the context of neurodegeneration to overcome this hurdle. The key innovation of this protocol is the potential to derive metabolic time series reflecting glucose dynamics in the individual subject. We identified three PD-related metabolic changes based on this novel data structure. Firstly, we detected a typical pattern, including a bilateral hypometabolic cluster in the substantia nigra and hypermetabolic activity in the motor cortex in a small PD sample, which static mean scans could not capture. Secondly, symptom-related alterations in glucose dynamics at the individual level were identified and associated with distinct PD symptoms. Lastly, the data structure allowed us to depict metabolic connectivity maps on a subject-level, revealing subcortical basal-ganglia networks and typical cortical resting-state networks per subject based on metabolic time series information (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe observed clusters with hypometabolism in PwPD in our study align with previously reported patterns, indicating occipito-parietal hypometabolism in PD based on static PET\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Subcortical hypometabolism in midbrain clusters encompassing parts of the substantia nigra revealed by voxel wise group comparison in a high-resolution static PET data set were first described by some of the authors of this work in 2020\u003csup\u003e12,13\u003c/sup\u003e and also examined longitudinally as progression marker\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Another study reported similar findings with a region of interest-based approach in idiopathic PD as well as atypical parkinsonian syndromes and found lower metabolism in entities with worse nigrostriatal pathology\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. The here reported subcortical clusters were closely confined to the bilateral substantia nigra pars compacta and reticulata as confirmed by the automated anatomical labeling atlas version 3 and the PD25 atlas. Given the small sample size and the absence of similar patterns in mean scans in our pilot study, it is plausible that the temporal resolution of the data with molecular time series information per subject allowed for the identification of more disease-related changes that correspond spatially with known spatial distribution of nigral cell loss. Additional multimodal imaging \u0026ndash; such as neuromelanin-sensitive MRI and small animal PET studies in synucleinopathy models \u0026ndash; may help corroborate the co-localisation with nigral pathology and dopaminergic cell loss\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Confirmation of a mechanistic association with nigral degeneration would underscore the translational potential of this imaging marker as a non-invasive tool for visualising nigral degeneration. There is preliminary evidence for nigral hypometabolism in patients with idiopathic rapid-eye movement sleep behavior disorder compared to matched controls by using an atlas-based approach with the median pons uptake value as the reference\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eDue to the static character of commonly applied PET protocols, the variation coefficient of glucose dynamics had not been analysed yet. We report changes in the subject-level variation coefficient in the data-driven motor cortex cluster as well as in an atlas-based superior sensorimotor region in PwPD. The level of variation within the time series was associated with cognition in our cohort with patients presenting with MCI showing the highest variation coefficients in the superior sensorimotor cortex. Only a few studies have examined comparable measures of the blood oxygenation level dependent (BOLD) signal variability. One study using a working memory task found age-related increases in BOLD variability in older adults, particularly in the left and right precentral gyrus\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. In individuals aged\u0026thinsp;~\u0026thinsp;20\u0026ndash;66 years, greater variability was associated with poor performance both during and outside the scanner. Our data suggest a similar association between the superior sensorimotor area, including pre-motor areas with cognitive impairment in PD based on a direct measure of neural activity. Another study has reported atrophy in the precentral gyrus and supplementary motor area in PD with MCI, which was accompanied by hypermetabolic changes in PD with dementia in a longitudinal design\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Together both findings might be indicative of adaptive processes that result from atrophy in this region. These findings should be analysed in longitudinal studies with larger sample sizes to examine the potential of these metabolic measures for detecting the progression of early cognitive dysfunction.\u003c/p\u003e\u003cp\u003eThe implementation of seed-based analysis for dynamic fPET data for the nigro-striatal system enabled the description of subcortical metabolic networks on subject-level. The nigral seed\u0026rsquo;s time course exhibited a plausible correlation with striatal and thalamic regions in all subjects often with continuous clusters to thalamic regions that appear like a continuous cluster between the midbrain and the thalamus. Two interesting aspects should be mentioned at this point. Firstly, a very recent study has provided evidence for a direct dopaminergic connection between the substantia nigra pars compacta and the thalamus in young healthy subjects by using multi-shell high-angular resolution diffusion MRI\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Second, the observed thalamic regions were among the most prominent regions within the a disease-specific metabolic network revealed by independent component analysis in this data set\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e: it is unknown how the here reported hypometabolic clusters, located ventrally to the brain stem part of the metabolic motor network, relate to the observed hypersynchronous motor network. If both findings are reflective of thalamic disinhibition in PwPD, is rather speculative and needs to be clarified in additional studies, incorporating larger sample sizes.\u003c/p\u003e\u003cp\u003eThe implementation of seed-based analysis for dynamic fPET data enabled the identification of canonical resting-state networks based on metabolic time series. At the applied threshold, the highest spatial similarity between fPET and fMRI was observed in the DMN across all groups. The seed-based analyses presented here can be only compared to previous studies that applied static PET to PD cohorts. In accordance with Sala et al., fewer clusters were observed in the prefrontal cortex in PD, with no clusters found in PwPD and MCI\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. While the motor network identified using our superior sensorimotor cortex seed extended into lateral frontal and subcortical areas, such as the thalamus, in the PET modality, the fMRI network was confined to the pre- and postcentral gyri and supplementary motor area. Although the representation of the networks at the group level is of course still highly dependent on the group size, the results demonstrate that it is possible to use this technique to identify resting networks of comparable spatial extent through seed-based methods based on metabolic time series.\u003c/p\u003e\u003cp\u003eAlthough our study provides important initial insights into metabolic time series, the protocol has some limitations. Firstly, we had a relatively small sample size, which was limited by the Federal Bureau for Radiation Protection and the ethics committee and represents an inherent limitation. This limits our power for detecting group-level networks and may contribute to the rather seed-confined nigral fPET networks in analyses with strict cluster-level FWE-correction. Due to the small sample size, especially the analysis of sub groups with different cognition levels are highly explorative and should be validated in larger samples. Secondly, we applied strict inclusion criteria to ensure that patients tolerate an off-phase longer than 12 h, which limits the generalisability of our findings to other disease stages.\u003c/p\u003e\u003cp\u003eIn addition, our multimodal acquisition protocol is afflicted with some restrictions. As both imaging procedure were performed consecutively in different scanners at different day times, it may be of question whether both represent the same resting-state of brain function. However, several requirements were undertaken to guarantee comparability: both procedures were performed in rooms with dimmed light under standardised conditions and standardised instructions were utilised like in comparable studies\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Our resting-state network analyses revealed common large-scale networks with comparable spatial distribution in both modalities. However, it is important to acknowledge that fPET-based networks are still derived based on a lower temporal resolution compared to fMRI. Nevertheless, it is unlikely that the slow dynamics of neurometabolic coupling can be better resolved in seconds\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Due to a low activity in initial frames, only the last 30 frames were retained in the presented analyses and it needs to be considered that preprocessing pipelines for constant infusion fPET are not yet as standardised as fMRI processing pipelines including denoising and intensity normalisation\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eFinally, mapping small nuclei in subcortical areas is always subject to uncertainty. Therefore, all available atlases, including Parkinson's-specific atlases, were used to ensure that spatial mapping was as accurate as possible. Calculation of ROI-based measures, like the applied measure for time series variation, is dramatically dependent on ROI definition. Therefore, data-driven ROIs as well as atlas-based approaches were performed to validate the findings. additional ROI-based analyses depend on ROIs derived from the Human Connectome Project \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. As described in Yaeger at al.\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, it needs to be considered that these ROIs were derived from young adults and the transferability to elderly subjects needs to be handled with caution\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, the results of this study support the use of the constant infusion fPET protocol in context of neurodegenerative diseases and demonstrate its ability to detect subcortical metabolic alterations in PD, unidentified by corresponding averaged mean scans. The findings are consistent with patterns obtained using static protocols and validate our previous findings regarding midbrain hypometabolism in an independent, non-high-resolution small data set. Our study provides first insights into subject-level glucose dynamics and network connectivity based on metabolic time series information in a neurodegenerative disease.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cb\u003eStudy participants and data collection\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe study received approval from the local ethics committee of the medical faculty of the Philipps-University of Marburg (146/19). Authorization for radiation exposure was obtained by the Federal Office for Radiation Protection. The study was carried out in adherence to the principles outlined in the Declaration of Helsinki and participants declared their written informed consent before participating.\u003c/p\u003e\u003cp\u003eA total of 14 healthy controls (HC) and 15 Parkinson’s disease (PD) subjects were recruited, of which 13 HC and 14 PD patients provided [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG-PET data. PD subjects were recruited through the central study coordination of the Department of Neurology at the University Hospital of Marburg in Germany. HC subjects were recruited through advertisements. Inclusion criteria: german speaking, older than 50 years old, three to eight years of illness, Hoehn \u0026amp; Yahr (H\u0026amp;Y) stage 1-2.5 in motoric OFF-state, no therapeutic changes within three months. Patients needed to be able to endure a medication break of 12 h of non-retard and 72 h of retarded PD medication. Exclusion criteria: structural cerebral damage (e.g. vascular events, tumors), severe depression and motor complications, signs of dementia, safety concerns about MRI scanning like pacemaker, artificial heart valves, metal in the body (e.g. total endoprostheses) and claustrophobia, pregnancy and a blood glucose \u0026gt; 180 mg/dl at the time of PET examination.\u003c/p\u003e\u003cp\u003e\u003cb\u003eClinical assessment\u003c/b\u003e\u003c/p\u003e\u003cp\u003eMotor severity was tested according to part III of the Movement Disorder Society Unified Parkinson`s disease rating scale (MDS-UPDRS-III) in ON and OFF-state\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. The levodopa equivalent daily dose (LEDD) was determined using established criteria\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. All subjects underwent a cognitive test battery including Montreal Cognitive Assessment (MoCA), revised Wechsler Memory Scale (WMS-R), Parkinson Neuropsychometric Dementia Asessment (PANDA)\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, and Regensburg Word Fluency test (RWT).\u003c/p\u003e\u003cp\u003eThe categorisation of all our subjects into mild cognitive impairment (MCI) and normal cognition (NC) was carried out according to Movement Disorder Society Level II criteria\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. MCI was diagnosed in PD patients, when a difference of \u0026gt;/= 1.5*standard deviation was observed in relation to age-matched norm means in at least two cognitive test results regardless of domain affiliation.\u003c/p\u003e\u003cp\u003e\u003cb\u003eResting-state [\u003c/b\u003e\u003csup\u003e\u003cb\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/b\u003e\u003c/sup\u003e\u003cb\u003eF]-FDG-PET and fMRI acquisition\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe dynamic [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG-PET scans were acquired on a SIEMENS Biograph 6 Scanner (Siemens, Germany) at the Department of Nuclear Medicine at the University Hospital of Marburg, Germany. Measurements of all subjects were carried out in OFF-state after overnight fasting and testing of blood sugar levels under standardised conditions. An average, 199.3 ± 5.27 MBq of [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG were infused via i.v. injection continuously using a perfusor (Braun, Germany) at a rate of 0.01 ml/s. Tomographic images were acquired dynamically for 90 min.\u003c/p\u003e\u003cp\u003eMRI scanning was performed on a Trio Tim Syngo 3 Tesla MR-scanner (Siemens, Erlangen) at the Department for Psychiatry and Psychotherapy of the University Hospital of Marburg, Germany. Participants underwent structural MRI with the following parameters: repetition time (TR): 1900 ms, echo time (TE): 2.52 ms, voxel size: 1.0 × 1.0 × 1.0 mm\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. For fMRI measurements, subjects were instructed to keep their eyes opened and to avoid thinking about anything in particular. The eye area was checked by camera throughout the measurement. The 8-minute lasting multiband echo-planar imaging sequence\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e was characterised by the following parameters: 490 time points, TR 1040 ms, TE 30.0 ms, 3 × 3 × 3 mm\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e voxel size and 48 slices. DICOM files were converted into NifTi files using the dicom2niix tool in MRICroGL.\u003c/p\u003e\u003cp\u003eThe detailed overview about further preprocessing and analysis of the data is described in Ruppert-Junck et al. 2024\u003csup\u003e20\u003c/sup\u003e and corresponding scripts are available on our Github repository: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/ruppertm/fPETDynamics.git\u003c/span\u003e\u003cspan address=\"https://github.com/ruppertm/fPETDynamics.git\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cb\u003eNeuroimaging data analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eVoxel-wise group comparison in a subject-and-time design\u003c/b\u003e\u003c/p\u003e\u003cp\u003eA flexible factorial design in SPM12 was applied to evaluate voxel-wise differences in metabolic activity within the time series between healthy controls and PD patients, considering the last 30 frames of spatially normalised scans per subject. Two main effects were created: the first with factor number 1 (Subject) and the second with factor number 2 (Time). Global normalisation was conducted using ANCOVA with reference to the global mean. Additionally, a grey matter mask in MNI-space was applied to restrict the analysis to grey matter regions (ICBM 2009c non-linear symmetric, FSL). Results were considered statistically significant if p \u0026lt; 0.05 after family-wise error (FWE) rate correction at cluster-level and surpassed a minimum cluster size of 20 voxels. Anatomical labelling done by using the AAL v3 Atlas. Activity values per region were extracted using the MarsBaR toolbox in SPM12 and global mean-normalised values were plotted in R. Mean normalised uptake values per contrast were entered into a logistic regression and utilised to derive receiver operating characteristic curves per contrast or for all clusters.\u003c/p\u003e\u003cp\u003e\u003cb\u003eFunctional and metabolic connectivity analyses\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe fMRI data set was preprocessed according to standardised procedures as described recently\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Seed-based resting-state functional connectivity analysis was performed with the default weighted general linear model using the CONN toolbox\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Anatomical labelling was performed by using the Harvard-Oxford atlas as implemented in CONN.\u003c/p\u003e\u003cp\u003e\u003cb\u003eMetabolic connectivity within the striato-nigro-thalamic system\u003c/b\u003e\u003c/p\u003e\u003cp\u003eSeed-based metabolic connectivity analyses were performed using scripts relying on SPM12 in Matlab v23a which can be found on our Github Repository (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/ruppertm/fPETDynamics.git\u003c/span\u003e\u003cspan address=\"https://github.com/ruppertm/fPETDynamics.git\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Mean uptake time series were extracted from substantia nigra clusters per subject from global-mean corrected scans and utilised as covariate of interest in a voxel-wise regression analysis per subject. The obtained \u003cem\u003et\u003c/em\u003e-maps per subject were transformed into z-maps. All individuals’ first-level contrast images were entered into a one sample \u003cem\u003et\u003c/em\u003e-test for visualisation on group level and into second-level group comparison evaluating voxel-wise differences in metabolic connectivity. A masked analyses was performed focused on the striato-nigro-thalamic network by using as mask created with the corresponding regions from AALv1 and TD atlas. Subject-level contrast images were visualised with effects between 0.2–1.2 and group results presented at P \u0026lt; 0.05 after FWE correction.\u003c/p\u003e\u003cp\u003e\u003cb\u003eCalculation of measures of glucose dynamics – variation coefficient\u003c/b\u003e\u003c/p\u003e\u003cp\u003eDifferences between the groups were derived by permutation tests (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/ruppertm/fPETDynamics.git\u003c/span\u003e\u003cspan address=\"https://github.com/ruppertm/fPETDynamics.git\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The association to clinical variables (UPDRS-III, MoCa, cognitive z-scores) was analysed by linear regression and correlation analyses in R.\u003c/p\u003e\u003cp\u003eNetwork ROIs were defined via independent component analysis of Human Connectome Project data (N = 497) (Whitfield-Gabrieli \u0026amp; Nieto-Castanon, 2012). Network seeds are listed with x, y, z coordinates for the centroid of each seed.\u003c/p\u003e\u003cp\u003e\u003cb\u003eStatistical Analysis of clinical data\u003c/b\u003e\u003c/p\u003e\u003cp\u003eStatistical analysis of clinical data were performed on demographic, behavioral and clinical data using R (RRID: SCR_001905)\u003csup\u003e\u003cb\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/b\u003e\u003c/sup\u003e. Group comparison were performed with two-sided Welch’s t-test or (\u003cem\u003eMann-Whitney U test\u003c/em\u003e) based on the results of Shapiro-Wilk test of normality. Multiple group comparisons were performed by using Kruskal-Wallis test and pairwise Wilcoxon tests with Bonferroni-Holm correction for multiple comparison.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe wish to convey our deep gratitude to the individuals who made this research possible: the members of the Core Facility Brain Imaging Marburg, particularly Prof. Andreas Jansen, Dr. Jens Sommer, Mechthild Wallnig and Rita Werner. We would like to extend our appreciation to our colleagues from the Department of Nuclear Medicine at the University Hospital Marburg, as well as Stefanie Spriewald from the study coordination of the Department of Neurology, for their invaluable assistance with data acquisition. Lastly, we would like to express our heartfelt appreciation to all the participants who generously contributed to this current research project.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualisation, D.P., V.H., M.R-J., K.S.; Methodology, K.S.,M.R-J,L.R.,F.T.; Software,K.S.,L.R.,F.T.,M.R-J.,V.H; Formal Analysis, L.M., L.R.,M.R-J.,F.T.;Investigation, J.F., V.H., M.R-J.,K.S.,D.L,M.B; Data curation, D.P., M.R-J; Resources, L.T.,D.P.,M.L; Writing \u0026ndash; Original draft, V.H., M.R-J., D.P.; Writing \u0026ndash; review \u0026amp; Editing, V.H.,D.L,M.R-J.,D.P.,F.T.,L.R.,L.M.,M.L.,M.B,L.T.,M.L; Funding acquisition, D.P; Project administration, D.P.,M.R-J; Visualisation, M.R-J, L.R.,V.H.; Supervision, D.P.,M.R-J\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data supporting the findings of the present study will be made available by the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLT reports grants, personal fees, and non-financial support from SAPIENS Steering Brain Stimulation, Medtronic, Boston Scientific, and St. Jude Medical, and has received payments from Bayer Healthcare, UCB Schwarz Pharma, and Archimedes Pharma and also honoraria as a speaker on symposia sponsored by Teva Pharma, Lundbeck Pharma, Bracco, Gianni PR, Medas Pharma, UCB Schwarz Pharma, Desitin Pharma, Boehringer Ingelheim, GSK, Eumecom, Orion Pharma, Medtronic, Boston Scientific, Cephalon, Abbott, GE Medical, Archimedes, and Bayer. DP received honoraria as a speaker at symposia sponsored by Boston Scientific Corp, Medtronic, AbbVie Inc., Zambon, and Esteve Pharmaceuticals GmbH. He received payments as a consultant for Boston Scientific Corp and Bayer, and he received a scientific grant from Boston Scientific Corp for a project entitled: \u0026lsquo;Sensor-based optimisation of Deep Brain Stimulation settings in Parkinson\u0026rsquo;s disease\u0026rsquo; (compareDBS). Finally, DP was reimbursed by Esteve Pharmaceuticals GmbH and Boston Scientific Corp for travel expenses to attend congresses. The remaining authors declare no conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrespondence\u003c/strong\u003e and requests for materials should be addressed to Marina Ruppert-Junck.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDorsey, E. R. \u0026amp; Bloem, B. R. The Parkinson Pandemic\u0026mdash;A Call to Action. \u003cem\u003eJAMA Neurol.\u003c/em\u003e \u003cb\u003e75\u003c/b\u003e, 9 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePalop, J. J., Chin, J. \u0026amp; Mucke, L. A network dysfunction perspective on neurodegenerative diseases. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e443\u003c/b\u003e, 768\u0026ndash;773 (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchindlbeck, K. A. \u0026amp; Eidelberg, D. Network imaging biomarkers: insights and clinical applications in Parkinson\u0026rsquo;s disease. \u003cem\u003eLancet Neurol.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 629\u0026ndash;640 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNiethammer, M. et al. Gene therapy reduces Parkinson\u0026rsquo;s disease symptoms by reorganizing functional brain connectivity. \u003cem\u003eSci Transl Med\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang, C. et al. Metabolic brain networks associated with cognitive function in Parkinson\u0026rsquo;s disease. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 714\u0026ndash;723 (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRuppert, M. C. et al. The default mode network and cognition in Parkinson\u0026rsquo;s disease: A multimodal resting-state network approach. \u003cem\u003eHum. Brain Mapp.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/hbm.25393\u003c/span\u003e\u003cspan address=\"10.1002/hbm.25393\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVillien, M. et al. Dynamic functional imaging of brain glucose utilization using fPET-FDG. \u003cem\u003eNeuroImage\u003c/em\u003e 100, 192\u0026ndash;9 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJamadar, S. D. et al. Metabolic and Hemodynamic Resting-State Connectivity of the Human Brain: A High-Temporal Resolution Simultaneous BOLD-fMRI and FDG-fPET Multimodality Study. \u003cem\u003eCereb. Cortex N Y N\u003c/em\u003e. \u003cb\u003e1991\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/cercor/bhaa393\u003c/span\u003e\u003cspan address=\"10.1093/cercor/bhaa393\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVoigt, K. et al. Metabolic and functional connectivity provide unique and complementary insights into cognition-connectome relationships. \u003cem\u003eCereb. Cortex N Y N\u003c/em\u003e. \u003cb\u003e1991\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/cercor/bhac150\u003c/span\u003e\u003cspan address=\"10.1093/cercor/bhac150\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoylan, M. A. et al. Greater BOLD Variability is Associated With Poorer Cognitive Function in an Adult Lifespan Sample. \u003cem\u003eCereb. Cortex\u003c/em\u003e. \u003cb\u003e31\u003c/b\u003e, 562\u0026ndash;574 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEckert, T. et al. FDG PET in the differential diagnosis of parkinsonian disorders. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 912\u0026ndash;921 (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRuppert, M. C. et al. Network degeneration in Parkinson\u0026rsquo;s disease: multimodal imaging of nigro-striato-cortical dysfunction. \u003cem\u003eBrain J. Neurol.\u003c/em\u003e \u003cb\u003e143\u003c/b\u003e, 944\u0026ndash;959 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGlaab, E. et al. Integrative analysis of blood metabolomics and PET brain neuroimaging data for Parkinson\u0026rsquo;s disease. \u003cem\u003eNeurobiol. Dis.\u003c/em\u003e \u003cb\u003e124\u003c/b\u003e, 555\u0026ndash;562 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSteidel, K. et al. Longitudinal trimodal imaging of midbrain-associated network degeneration in Parkinson\u0026rsquo;s disease. \u003cem\u003eNPJ Park Dis.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 79 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchr\u0026ouml;ter, N. et al. Nigral glucose metabolism as a diagnostic marker of neurodegenerative parkinsonian syndromes. \u003cem\u003eNPJ Park Dis.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 123 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJang, D. P. et al. Functional neuroimaging of the 6-OHDA lesion rat model of Parkinson\u0026rsquo;s disease. \u003cem\u003eNeurosci. Lett.\u003c/em\u003e \u003cb\u003e513\u003c/b\u003e, 187\u0026ndash;192 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiaz-Galvan, P. et al. Brain glucose metabolism and nigrostriatal degeneration in isolated rapid eye movement sleep behaviour disorder. \u003cem\u003eBrain Commun.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, fcad021 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez-Redondo, R. et al. Grey matter hypometabolism and atrophy in Parkinson\u0026rsquo;s disease with cognitive impairment: a two-step process. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e137\u003c/b\u003e, 2356\u0026ndash;2367 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCirillo, G. et al. Evidence for direct dopaminergic connections between substantia nigra pars compacta and thalamus in young healthy humans. \u003cem\u003eFront. Neural Circuits\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e, 1522421 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRuppert-Junck, M. C. et al. Connectivity based on glucose dynamics reveals exaggerated sensorimotor network coupling on subject-level in Parkinson\u0026rsquo;s disease. \u003cem\u003eEur. J. Nucl. Med. Mol. Imaging\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00259-024-06796-6\u003c/span\u003e\u003cspan address=\"10.1007/s00259-024-06796-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSala, A. et al. Altered brain metabolic connectivity at multiscale level in early Parkinson\u0026rsquo;s disease. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 4256 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWhitfield-Gabrieli, S. \u0026amp; Nieto-Castanon, A. Conn: a functional connectivity toolbox for correlated and anticorrelated brain networks. \u003cem\u003eBrain Connect.\u003c/em\u003e \u003cb\u003e2\u003c/b\u003e, 125\u0026ndash;141 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYeager, B. E., Twedt, H. P., Bruss, J., Schultz, J. \u0026amp; Narayanan, N. S. Cortical and subcortical functional connectivity and cognitive impairment in Parkinson\u0026rsquo;s disease. \u003cem\u003eNeuroImage Clin.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e, 103610 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePereira, J. B. et al. Assessment of cortical degeneration in patients with Parkinson\u0026rsquo;s disease by voxel-based morphometry, cortical folding, and cortical thickness. \u003cem\u003eHum. Brain Mapp.\u003c/em\u003e \u003cb\u003e33\u003c/b\u003e, 2521\u0026ndash;2534 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoetz, C. G. et al. Movement Disorder Society-sponsored revision of the Unified Parkinson\u0026rsquo;s Disease Rating Scale (MDS‐UPDRS): Scale presentation and clinimetric testing results. \u003cem\u003eMov. Disord\u003c/em\u003e. \u003cb\u003e23\u003c/b\u003e, 2129\u0026ndash;2170 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJost, S. T. et al. Levodopa Dose Equivalency in Parkinson\u0026rsquo;s Disease: Updated Systematic Review and Proposals. \u003cem\u003eMov. Disord Off J. Mov. Disord Soc.\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e, 1236\u0026ndash;1252 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKalbe, E. et al. Screening for cognitive deficits in Parkinson\u0026rsquo;s disease with the Parkinson neuropsychometric dementia assessment (PANDA) instrument. \u003cem\u003eParkinsonism Relat. Disord\u003c/em\u003e. \u003cb\u003e14\u003c/b\u003e, 93\u0026ndash;101 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLitvan, I. et al. Diagnostic criteria for mild cognitive impairment in Parkinson\u0026rsquo;s disease: Movement Disorder Society Task Force guidelines. \u003cem\u003eMov. Disord Off J. Mov. Disord Soc.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 349\u0026ndash;356 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAuerbach, E. J., Xu, J., Yacoub, E., Moeller, S. \u0026amp; Uğurbil, K. Multiband accelerated spin-echo echo planar imaging with reduced peak RF power using time‐shifted RF pulses. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cb\u003e69\u003c/b\u003e, 1261\u0026ndash;1267 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eR Core Team. R: A Language and Environment for Statistical Computing. \u003cem\u003eR Found. Stat. Comput\u003c/em\u003e (2018).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"glucose dynamics, metabolic variation, seed-based metabolic connectivity, neurocognitive networks, cognitive symptoms, non-motor symptoms, imaging biomarkers","lastPublishedDoi":"10.21203/rs.3.rs-7156182/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7156182/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eStatic [\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003eF]-FDG PET studies have shown characteristic metabolic patterns for Parkinson\u0026rsquo;s disease (PD). However static PET scans offer only a snapshot of synaptic activity, preventing insights into dynamics of glucose consumption on individual level. In this study, we apply a novel dynamic constant infusion PET protocol in combination with MRI for mapping metabolic dynamics in PD. Metabolic time series comparison revealed hypometabolism in bilateral clusters in the substantia nigra in PD patients compared to controls (p\u003csub\u003eFWE\u003c/sub\u003e \u0026lt; 0.001). The temporal structure allowed us to depict metabolic networks covering and the basal ganglia circuits and typical neural resting-state networks on individual level. A measure of time series variation showed a region-specific increase in time series variation in PD, which related to cognitive PD symptoms. The findings pinpoint the methodological capabilities of the acquisition protocol for evaluating metabolic dynamics and metabolic network activity in PD. Our study lays a foundation for the application of molecular imaging with temporal resolution in context of neurodegenerative diseases. Such protocols can capture interregional metabolic changes on patient level, the applicability of which should be evaluated in prodromal stages.\u003c/p\u003e","manuscriptTitle":"Functional PET for mapping metabolic dynamics in Parkinson’s disease","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-21 16:12:37","doi":"10.21203/rs.3.rs-7156182/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-10T10:02:25+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-04T17:35:38+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-03T01:18:21+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-01T03:13:43+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-21T13:26:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"329936249132203603188379116927929166603","date":"2025-08-17T11:10:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"123369899332518494416674675616684143541","date":"2025-08-16T21:05:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"6314000571053506722866847426751158310","date":"2025-08-16T03:01:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-15T13:35:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"197609474857318267164866138102111012851","date":"2025-08-14T14:23:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"258951000402756659300765790619091058231","date":"2025-08-14T06:06:23+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-14T01:48:20+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-31T11:04:45+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-29T07:11:52+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-25T14:36:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-25T10:51:23+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"869d89c2-e030-4d27-af00-b8c6a784dccc","owner":[],"postedDate":"August 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":53337041,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":53337042,"name":"Health sciences/Diseases"},{"id":53337043,"name":"Health sciences/Neurology"},{"id":53337044,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2025-11-24T16:02:03+00:00","versionOfRecord":{"articleIdentity":"rs-7156182","link":"https://doi.org/10.1038/s41598-025-28456-x","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-11-18 15:57:58","publishedOnDateReadable":"November 18th, 2025"},"versionCreatedAt":"2025-08-21 16:12:37","video":"","vorDoi":"10.1038/s41598-025-28456-x","vorDoiUrl":"https://doi.org/10.1038/s41598-025-28456-x","workflowStages":[]},"version":"v1","identity":"rs-7156182","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7156182","identity":"rs-7156182","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}