Skip to main content
NCBI home page
Human Brain Mapping logoLink to Human Brain Mapping
. 2023 May 1;44(9):3845–3858. doi: 10.1002/hbm.26316

Normalization effect of dopamine replacement therapy on brain functional connectome in Parkinson's disease

Chenqing Wu 1, Haoting Wu 1, Cheng Zhou 1, Xiaojun Guan 1, Tao Guo 1, Zhengye Cao 1, Jingjing Wu 1, Xiaocao Liu 1, Jingwen Chen 1, Jiaqi Wen 1, Jianmei Qin 1, Sijia Tan 1, Xiaojie Duanmu 1, Baorong Zhang 2, Peiyu Huang 1, Xiaojun Xu 1, Minming Zhang 1,✉
PMCID: PMC10203796  PMID: 37126590

Abstract

Dopamine replacement therapy (DRT) represents the standard treatment for Parkinson's disease (PD), however, instant and long‐term medication influence on patients' brain function have not been delineated. Here, a total of 97 drug‐naïve patients, 43 patients under long‐term DRT, and 94 normal control (NC) were, retrospectively, enrolled. Resting‐state functional magnetic resonance imaging data and motor symptom assessments were conducted before and after levodopa challenge test. Whole‐brain functional connectivity (FC) matrices were constructed. Network‐based statistics were performed to assess FC difference between drug‐naïve patients and NC, and these significant FCs were defined as disease‐related connectomes, which were used for further statistical analyses. Patients showed better motor performances after both long‐term DRT and levodopa challenge test. Two disease‐related connectomes were observed with distinct patterns. The FC of the increased connectome, which mainly consisted of the motor, visual, subcortical, and cerebellum networks, was higher in drug‐naïve patients than that in NC and was normalized after long‐term DRT (p‐value <.050). The decreased connectome was mainly composed of the motor, medial frontal, and salience networks and showed significantly lower FC in all patients than NC (p‐value <.050). The global FC of both increased and decreased connectome was significantly enhanced after levodopa challenge test (q‐value <0.050, false discovery rate‐corrected). The global FC of increased connectome in ON‐state was negatively associated with levodopa equivalency dose (r = −.496, q‐value = 0.007). Higher global FC of the decreased connectome was related to better motor performances (r = −.310, q‐value = 0.022). Our findings provided insights into brain functional alterations under dopaminergic medication and its benefit on motor symptoms.

Keywords: dopamine replacement therapy, functional MRI, Parkinson's disease

Two Parkinson's disease‐related connectomes were observed: the increased connectome, mainly consisting of the motor, visual, subcortical, and cerebellum networks, and the decreased connectome, composed of the motor, medial frontal, and salience networks. The functional connectivity (FC) of increased connectome was significantly higher in drug‐naïve patients than in normal control (NC) and was normalized in patients under long‐term dopamine replacement therapy (DRT). The FC of decreased connectome was significantly enhanced after instant treatment, whose global FC was negatively related to motor symptoms.

graphic file with name HBM-44-3845-g005.jpg

1. INTRODUCTION

Parkinson's disease (PD) is the second most common neurodegenerative disease with both motor and nonmotor symptoms (Hayes, 2019; Kalia & Lang, 2015). The core pathology involves dopaminergic neuron degeneration in the nigrostriatal pathway (Kalia & Lang, 2015). Dopamine replacement therapy (DRT) compensates for this deficiency and represents the standard treatment in the last decade (You et al., 2018). Long‐term DRT is beneficial to alleviate motor symptoms and contributes to a better medication response and sustained motor improvement (Cilia et al., 2020; Ferrazzoli et al., 2016). However, DRT failed to provide disease‐modifying effects to slow down the disease progression, and could even cause motor complications such as motor fluctuation and dyskinesia in the late‐stage (Lang & Espay, 2018). Evidence from fundamental research revealed decreased neuron excitability, reduced dopamine accumulation, altered receptor expression, and restored spinal density in the dopaminergic neural circuit of parkinsonism mice under long‐term DRT exposure (Cote & Kuzhikandathil, 2015; Fieblinger et al., 2018; Smith et al., 2014; Suárez et al., 2014). Therefore, long‐term dopaminergic exposure may deeply influence the intrinsic neuronal activity of PD patients, which needs further exploration in vivo.

The resting‐state functional magnetic resonance imaging (rs‐fMRI) detects neuronal activity at a mesoscopic level and is one of the most popular techniques to explore pathophysiological changes in the neural circuits (Heeger & Ress, 2002). Functional connectivity (FC), calculated using blood oxygenation level dependent signal correlation, reflects the temporal coincidence of neuronal activity (Ganzetti & Mantini, 2013). Levodopa challenge test improved FC within sensorimotor network (SMN), between SMN and visual network, and between SMN and attention network (Caspers et al., 2021; Wu et al., 2009). Different from such phasic stimulation based on single‐dose dopaminergic medication, long‐term DRT imposes a tonic dopaminergic stimulation to modulate neural activity (Albin & Leventhal, 2017). Long‐term dopaminergic medication exposure normalized the increased FC in the basal ganglia, thalamus, cerebellum, and supplementary motor area (SMA) (Ballarini et al., 2018; Vo et al., 2017). Yet, other cross‐sectional studies revealed both decreased and increased FC in wide‐spread brain regions including the cerebellum, basal ganglia, and motor region in long‐term DRT‐treated patients (Tinaz, 2021; Vancea et al., 2019; Wu et al., 2009). Filippi et al. (2021) used a data‐driven approach and found longitudinal FC changes varied across patients with coexistence of increased and decreased connectivity. These heterogeneous results should be interpreted with caution because most studies mentioned above compared DRT‐treated patients with normal controls (NCs) instead of drug‐naïve patients, and it is hard to explicitly distinguish the effects of dopaminergic medication and PD pathophysiology on FC alterations. Therefore, it is significantly important to further stratify DRT‐treated and drug‐naïve patients in comparison, which could better clarify the influence of long‐term DRT on neural circuits and motor symptom changes.

In this study, we first identified connectomes in drug‐naïve patients to reveal intrinsic pathological‐related change, and then comprehensively explored the FC differences in NC, drug‐naïve, and DRT‐treated patients, and association among FC, dopaminergic medication and motor symptoms.

2. MATERIALS AND METHODS

2.1. Participant

This research was approved by the Ethics Committee of the Second Affiliated Hospital, Zhejiang University School of Medicine, and informed consent forms were obtained from all participants following the Declaration of Helsinki.

A total of 382 PD patients were initially diagnosed according to the UK PD Society Brain Bank diagnostic criteria and Movement Disorder Society clinical diagnostic criteria (Hughes et al., 1992; Postuma et al., 2015). Patients were excluded due to ① cerebrovascular diseases, including severe cerebral atrophy, stroke, and head injury (N = 43); ② neurological surgical history (N = 3); ③ treatment with anticholinergic medication (N = 4) or antidepressants (N = 2) which might have interactive effects with dopaminergic medication on FC (Wu et al., 2022) (Figure 1). Drug‐naïve patients with levodopa challenge test were stratified into Group 2 while the others were divided into Group 1. Patients under stable DRT for at least 1 month were enrolled in Group 2 as treated patients (Poulopoulos & Waters, 2010; Ruottinen & Rinne, 1996). Meanwhile, 184 NC were recruited from the community and those with cerebrovascular diseases were also excluded (N = 21) (Figure 1). Propensity score matching was performed separately in Groups 1 and 2 to ensure the similarity distribution of age and gender between patients and NC, as well as disease duration between treated and drug‐naïve patients (Austin, 2011). Groups 1 and 2 were independent with no overlapped participants.

FIGURE 1.

FIGURE 1

Open in a new tab

Inclusion and exclusion criteria of NC, drug‐naïve, and treated patients in two independent groups. DRT, dopamine replacement therapy; NC, normal controls; PD, Parkinson's disease

Basic demographic data, including age, gender, years of education, and disease duration were collected. The duration of dopaminergic medication and the kinds of medication were recorded. The levodopa equivalency dose (LED) was calculated according to the previous statement (Tomlinson et al., 2010). Medication therapy was adjusted to a stable level with optimal motor control and minimal side effects. Motor symptoms severity of patients was assessed using Unified PD Rating Scale (UPDRS) part 3 score and Hoehn and Yahr stage (H‐Y). The UPDRS part 3 score was divided into three parts: scores of bradykinesia, rigidity, and tremor according to previous studies (Kang et al., 2005; Stebbins et al., 2013). These evaluations were conducted before (OFF) and after (ON) levodopa challenge test, except for drug‐naïve patients in Group 1 who were only assessed in OFF‐state: ① OFF‐state: following overnight withdrawal (at least 12 h) of antiparkinsonian medications for DRT‐treated patients and ② ON‐state: 1 h after the levodopa challenge test (administration of 200 mg levodopa and 50 mg benserazide) (Saranza & Lang, 2021).

2.2. MRI acquisition

All imaging data were acquired with a 3.0 Tesla MRI scanner (Discovery MR750; GE Healthcare). Each participant's head was stabilized with foam pads, and earplugs were provided to reduce the noise during scanning. All participants were told to close their eyes and stay awake during scanning. For all patients, rs‐fMRI data were collected before and after levodopa challenge test (OFF‐ and ON‐state). Note that the data acquisition was performed in a fixed‐order scanning in the OFF condition followed by scanning in the ON condition. This order could not be randomized due to the design of the study.

Rs‐fMRI data were acquired using gradient recalled echo ‐ echo planar imaging sequence: echo time = 30 ms; repetition time = 2000 ms; flip angle = 77°; field of view (FOV) = 240 × 240 mm2; matrix = 64 × 64; slice thickness = 4 mm; slice gap = 0 mm; number of slices = 38 (axial); interleaved slice acquisition; time points = 205. Three‐dimensional T1‐weighted images were acquired using a fast‐spoiled gradient recalled sequence: echo time = 3.036 ms; repetition time = 7.336 ms; inversion time = 450 ms; flip angle = 11°; FOV = 260 × 260 mm2; matrix = 256 × 256; slice thickness = 1.2 mm; number of slices = 196 (sagittal). Both FOVs of the sequences covered the whole brain, including the cerebrum, cerebellum, and brain stem.

2.3. Image preprocessing

Rs‐fMRI data were processed using Statistical Parametric Mapping (SPM12, https://www.fil.ion.ucl.ac.uk/spm/) and Data Processing Assistant for Resting‐State fMRI (DPABI_V3.1_180801, http://www.rfmri.org/) (Yan et al., 2016). In the beginning, the first 10 volumes of the functional time series were removed for scanner stabilization and individuals' adaption to the environment. The remaining images underwent slice timing for interval scanning and realignment to the middle volume to correct for interscan head motion. Then, the processed images were spatially normalized to the standard MNI space through T1 images. Next, spatial smoothing with a Gaussian kernel of 6 × 6 × 6 mm full‐width‐at‐half‐maximum, detrending, nuisance covariates regression (Friston 24‐motion parameters, mean signals of white matter and cerebrospinal fluid) and band‐pass temporal filtering (0.01–0.1 Hz) were sequentially applied to the remaining volumes. Motion artifacts were minimized through motion scrubbing (Jenkinson et al., 2002; Parkes et al., 2018). Volumes with mean frame‐wise displacement ≥0.2 mm were removed and the remaining volumes were applied for matrices construction. After scrubbing, patients with data shorter than 4 min (120 volumes) were excluded (N = 22) and all images were visually checked for quality control.

2.4. FC matrices construction

The brain regions, which represented the nodes of networks, were defined using the Shen 268‐node functional atlas which included the cortex, subcortex, and cerebellum (Shen et al., 2013). The correlation matrices were obtained by calculating the Pearson correlation coefficient between the mean time course of each pair of nodes and the resultant correlation coefficients were transformed using Fisher's z‐transformation (Figure 2a). Accordingly, 268 × 268 symmetric matrices were obtained for all individuals, and each value of the matrix represented the FC strength between all pairs of nodes.

FIGURE 2.

FIGURE 2

Open in a new tab

Workflow of image processing and statistical analyses. (a) FC matrices construction: The correlation matrices for each individual were obtained by calculating the Pearson correlation coefficient between the mean time course of each pair of regions (nodes) defined by the Shen atlas and the resultant correlation coefficients were transformed using Fisher's z‐transformation. (b) Disease‐related connectome identification in Group 1: NBS was performed to assess FC differences between drug‐naïve patients and NC which were stratified into increased (higher FC in drug‐naïve patients than NC) and decreased (lower FC in drug‐naïve patients than NC) disease‐related connectomes. Two connectomes were binarized for FC calculation in the next steps. (c) Global, within, and between‐network FC calculation in Group 2: FC was extracted based on increased and decreased connectomes identified in step B. Global and network‐level FC were calculated according to 10 canonical networks. (d) Statistical analyses in Group 2: General linear model, paired t test, repeated measure ANOVA, and partial correlation were performed. ANOVA, analysis of variance; FC, functional connectivity; NBS, network‐based statistics; NC, normal controls; PD, Parkinson's disease.

2.5. Disease‐related connectomes identification

Group 1 (51 NC and 51 drug‐naïve patients) was used for identifying disease‐related connectomes. Network‐based statistics (NBS) were performed to assess FC differences between OFF‐state drug‐naïve patients and NC using GRETNA toolkit (version 2.0.0) (Figure 2b) (De Micco et al., 2021; Wang et al., 2015; Zalesky et al., 2010). This method controlled for the huge number of multiple comparisons involved in testing for statistical differences at each connection of the matrix. To be specific, the two‐sample t test was computed between OFF‐state drug‐naïve patients and NC for each functional connection with age and gender as covariates, obtaining the corresponding p‐value of each connection. The connections with a p‐value <.0001 in the t test were selected to construct multiple connected components. The number of functional connections within each component, or their size, was stored. Subsequently, the permutation test was used to ascribe a p‐value to each connected component based on its size. A total of 1000 random permutations were generated independently, where the group (drug‐naïve patient and NC) to which each subject belonged was randomly exchanged. The two‐sample t‐test was recalculated and the same threshold was applied to define sets of suprathreshold connections. The maximal component size derived from each of the 1000 permutations was then determined and stored to yield an empirical estimate of the null distribution of maximal component size. Finally, the P‐value of each observed connected component was determined by the percentage of permutations with the maximal component size greater than the observed size in all 1000 permutations. Components with p‐value <.05 after permutation tests were considered significant.

These significant connected components were defined as disease‐related connectomes and categorized into the increased connectome (higher FC in drug‐naïve patients than NC) and decreased connectome (lower FC in drug‐naïve patients than NC), which were then binarized into two masks. To determine medication effects specifically on disease‐related alterations, further analyses were restricted to these two masks.

2.6. Global, within, and between networks FC calculation

To avoid the pitfall of circularity (Kriegeskorte et al., 2009), further analyses were only carried out in Group 2 (Figure 2c,d). The 268 brain regions (nodes) were classified into 10 canonical networks: medial frontal; frontoparietal; default mode; motor; visual (I, II, and association); subcortical; and cerebellum networks (Noble et al., 2017; Suo et al., 2022). The top five internetwork and intranetwork pairs with the most connections within the increased (or decreased) binary mask were selected for statistical analyses. To be specific, functional connections were extracted from each individual's FC matrix using these binary masks and the average FC value of extracted connections was calculated on global, within‐network, and between‐network levels for each individual according to the previous studies (De Micco et al., 2021; Tinaz, 2021; Wang et al., 2021).

All calculations mentioned above were performed in MATLAB (R2021a for windows, MathWorks) with in‐house scripts.

2.7. Statistical analyses

Statistical analyses were performed using IBM SPSS Statistics software (version 26.0). The one‐sample Kolmogorov–Smirnov test was used to verify whether a continuous variable comes from a normal distribution. Continuous variables with normal distribution were presented as mean and standard deviation (SD) and compared using the two‐sample t test or one‐way analysis of variance (ANOVA). Continuous variables with nonparametric distribution were reported as median and interquartile ranges and compared using Wilcoxon rank‐sum test or Kruskal–Wallis test. Chi‐squared test was used for categorical variable analyses. A two‐tailed p‐value <.05 was considered statistically significant.

All the following statistical analyses were carried out within 12 metrics (global and top 5 network‐level FC of two connectomes) in Group 2. FC differences among NC, OFF‐state drug‐naïve, and treated patients were analyzed using the general linear model (GLM) with age and gender as covariates (Figure 2d). To explicitly interpret the effect of DRT on FC, the “normalization” effect was defined as a pattern in any of the 12 metrics where there is a statistically significant difference of FC between drug‐naïve patients and NC (i.e., FC of drug‐naïve patients > FC of NC for the increased connectome, and vice versa for the decreased connectome) and no significant difference of FC between treated patients and NC at the same time. The instant medication influence on FC was first investigated using the paired T‐test between OFF‐ and ON‐state in all patients. To further compare the difference in FC changes after the levodopa challenge test between drug‐naïve and treated patients, repeated measure ANOVA was carried out with the levodopa challenge state (ON‐ and OFF‐state) as the within‐subject factor and medication group (drug‐naïve and treated) as between‐subject factor. False discovery rate (FDR) was calculated using the MATLAB's mafdr function (https://www.mathworks.cn/help/bioinfo/ref/mafdr.html) and q‐values were presented after multiple comparison corrections (Benjamini & Hochberg, 1995; Benjamini & Yekutieli, 2001). A q‐value of 5% was considered significant after correction.

Partial correlation analyses were performed to examine the association between imaging metrics and motor symptoms (UPDRS Part 3 total score, bradykinesia, rigidity, and tremor scores) in all patients in Group 2, controlling for age, gender, disease duration, and LED. Besides, partial correlation analyses were conducted between imaging metrics and medication usage (medication duration and LED) in treated patients to explore long‐term medication effects on FC, controlling for age, gender, and disease duration. Q‐values were summarized.

FDR correction was applied independently to GLM, paired t test, and partial correlation analyses. All statistical analyses were saved as SPSS syntax and shared in Supplemental Material 2.

2.8. Additional analyses

Considering the arbitrariness of choosing a particular atlas to parcellate brain regions, alternative methods for regional timecousrse calculation and brain parcellation were used to repeat all the network analyses. Specifically, principal component analysis (PCA) was used for timecourse calculation instead of average timecourses, and the first principal component‐based timecourses were extracted. Besides, an alternative atlas, Brainnetome atlas (Fan et al., 2016), was used for brain parcellation. Repeated analyses were, respectively, carried out using PCA‐generated timecourses and Brainnetome atlas.

3. RESULTS

3.1. Demographic and clinical data

A total of 51 NC and 51 drug‐naïve patients (in Group 1), and 43 NC, 46 drug‐naïve, and 43 treated patients (in Group 2) were enrolled, and their characteristics are detailed in Table 1. The information about dopaminergic medication of treated patients was summarized in Supplemental Table 1. No significant difference was found in age, gender, and years of education between NC and patients in Group 1 and Group 2. In Group 2, drug‐naïve and treated patients showed no differences in UPDRS Part 3, tremor, rigidity, or bradykinesia scores in the OFF‐state, however, treated patients had significantly lower UPDRS Part 3 (p = .034) and bradykinesia (p = .038) scores in the ON‐state than drug‐naïve patients.

TABLE 1.

Demographic and clinical data of NC, drug‐naïve, and treated patients

Group 1 Group 2
NC Drug‐naïve PD p‐Value NC Drug‐naïve PD Treated PD p‐Value
(N = 51) (N = 51) (N = 43) (N = 46) (N = 43)
Demographics and clinical data
Age 59.54 ± 7.18 58.55 ± 10.66 .583 59.37 ± 7.17 57.78 ± 9.99 60.03 ± 8.27 .447
Gender (female/male) 31/20 25/26 .233 24/19 18/28 17/26 .203
Year of education 10.00 (8.00–13.00) 9.00 (5.00–12.00) .063 9.00 (6.00–12.00) 9.00 (6.00–12.00) 9.00 (6.00–12.00) .502
Disease duration / 1.82 (0.67–2.94) / / 1.90 (1.13–3.06) 2.85 (1.64–3.66) .111
HY / 2.00 (2.00–2.50) / / 2.00 (1.50–2.50) 2.00 (1.50–2.50) .738
LED / 0 / / 0 375.00 (237.50–475.00) <.001
Medication duration / 0 / / 0 1.79 (0.87–2.83) <.001
OFF state
UPDRS part 3 / 22.00 (13.00–36.00) / / 19.50 (13.75–29.25) 16.00 (10.00–26.00) .136
Tremor / 2.00 (1.00–8.00) / / 4.00 (2.00–5.25) 3.00 (1.00–5.00) .192
Rigidity / 5.00 (2.00–9.00) / / 4.00 (2.00–7.00) 3.00 (1.00–5.00) .283
Bradykinesia / 8.00 (5.00–16.00) / / 10.00 (5.00–13.25) 7.00 (4.00–12.00) .201
ON state
UPDRS part 3 / / / / 15.00 (9.00–20.00) 10.00 (5.00–18.00) .034*
Tremor / / / / 1.00 (0.00–2.00) 0.00 (0.00–1.00) .061
Rigidity / / / / 3.00 (1.00–6.00) 1.00 (0.00–3.25) .090
Bradykinesia / / / / 6.00 (3.00–10.00) 4.00 (1.75–7.00) .038*
Open in a new tab

Abbreviations: H‐Y, Hoehn and Yahr stage; LED, levodopa equivalent dosage; PD, Parkinson's disease; NC, normal control; UPDRS, Unified Parkinson's Disease Rating Scale.

*

p‐Value <.05.

3.2. Disease‐related connectomes

The NBS identified the increased and decreased connectomes consisting of 111 and 84 functional connections, respectively, and the number of connections belonging to each intranetwork and internetwork pair was quantified according to 10 canonical networks (Figure 3). The increased connectome primarily consisted of the subcortical network (SC, 111 connections), motor network (Mot, 40 connections), visual I network (V I, 24 connections), and cerebellum (CBL, 21 connections) (Figure 3a). The top five network pairs with the most connections in the increased connectome were between SC and Mot, SC and V I, SC and CBL, SC and salience network (SAL), SC and visual association network (VAs), and within SC which were selected for further analyses (Figure 3a,b). The brain region with the most connections in the increased connectome was the left thalamus (61 connections) whose connected regions primarily belonged to Mot and CBL (Figure 3e).

FIGURE 3.

FIGURE 3

Open in a new tab

Increased and decreased disease‐related connectomes. (a, c) Matrix and radar plots for increased (a, in red) and decreased (c, in blue) connectomes. The number of internetwork and intranetwork connections is summarized in the matrix. Boxes with thick borders in the matrix are the top five network pairs with the most connections. The radar plots show the number of connections in 10 networks (both internetwork and intranetwork connections were included). (b, d) Chordal graph plots for the increased (b) and decreased (d) connectome. The circle represents networks and chords depicted intranetwork and internetwork. (e, f) The brain regions with the largest number of connections in the increased (in red) and decreased (in blue) connectomes. CBL, cerebellum; DMN, default mode network; FP, frontoparietal network; L, left; MF, medial frontal network; Mot, motor network; R, right; SAL, salience network; SC, subcortical network; VAs, visual association network; V I, visual I network; V II, visual II network.

The decreased connectome was mainly composed of the Mot (63 connections), SAL (27 connections), and medial frontal network (MF, 24 connections) (Figure 3c). The top five network pairs with the most connections in the decreased connectome were between Mot and MF, Mot and SAL, Mot and DMN, Mot and V I, and within Mot, which were selected for further analyses (Figure 3c,d). The brain region with the most connections in the decreased connectome was the left SMA (10 connections) whose connected regions mainly belong to Mot and SAL (Figure 3f).

3.3. The global and network‐level FC differences among NC, drug‐naïve, and treated patients

In the increased connectome, there were significant differences among three groups in global FC (q‐value = 0.007), FC between Mot and SC, between V I and SC, between SC and SAL, and between SC and CBL (q‐value <0.050), summarized in Table 2. Post hoc analyses revealed that drug‐naïve patients had significantly higher global FC and FC in all top five network pairs than NC (p‐value <.050, Figure 4a–g). Compared to drug‐naïve patients, treated patients showed significantly lower global FC (p‐value = .022, Figure 4a) and lower FC between V I and SC (p‐value = .016, Figure 4d). No significant difference was observed between treated patients and NC in global or any network‐level FC (p‐value >.05, Figure 4a–g). Based on the aforementioned definition, normalization effects were observed in the increased connectomes (global and all top‐five network pairs).

TABLE 2.

Global and network‐level FC differences among NC, drug‐naïve, and treated patients in Group 2

Imaging metrics F value Partial η 2 p‐Value Q value Post hoc p value (LSD)
Drug‐naïve PD vs. treated PD Drug‐naïve PD vs. NC Treated PD vs. NC
Increased connectome
Global 6.730 0.096 .002* 0.007* 0.022* <0.001* 0.202
Within SC 2.947 0.044 .056 0.066 0.276 0.017* 0.189
Between Mot and SC 5.449 0.079 .005* 0.014* 0.051 0.001* 0.196
Between V I and SC 5.210 0.076 .007* 0.014* 0.016* 0.003* 0.554
Between VAs and SC 2.611 0.040 .077 0.084 0.211 0.024* 0.311
Between SAL and SC 3.998 0.059 .021* 0.030* 0.137 0.006* 0.189
Between SC and CBL 4.550 0.067 .012* 0.020* 0.067 0.003* 0.262
Decreased connectome
Global 6.167 0.089 .003* 0.009* 0.597 0.006* 0.001*
Within Mot 3.548 0.053 .032* 0.041* 0.868 0.017* 0.029*
Between MF and Mot 7.443 0.105 .001* 0.006* 0.583 0.003* <0.001*
Between DMN and Mot 9.518 0.130 <.001* 0.002* 0.709 <0.001* <0.001*
Between Mot and V I 0.630 0.010 .534 0.534 0.491 0.660 0.268
Between Mot and SAL 4.531 0.067 .013* 0.020* 0.615 0.020* 0.006*
Open in a new tab

Abbreviations: CBL, cerebellum; DMN, default mode network; MF, medial frontal network; Mot, motor network; NC, normal control; PD, Parkinson's disease; SAL, salience network; SC, subcortical network; V I, visual I network; VAs, visual association network.

*

p‐Value <.05.

FIGURE 4.

FIGURE 4

Open in a new tab

Violin plots show intergroup differences in global FC and FC of the top five network pairs in the increased (a–g) and decreased (h–m) connectomes. (a–g) In the increased connectome, compared to NC (in blue), drug‐naïve patients (in red) had significantly higher global FC (p‐value <.001, a) and higher FC in all top five network pairs (p‐value <.050, b–g), while treated patients (in orange) showed no significant difference in all imaging metrics (p‐value >.050, a–g). (a, d) Compared to drug‐naïve patients, treated patients showed significantly lower global FC (p‐value = .022) and FC between V I and SC (p‐value = .016). (h–k, m) In the decreased connectome, compared to NC, both drug‐naïve and treated patients demonstrated significantly lower global FC and FC within Mot, between MF and Mot, between DMN and Mot, and between SAL and Mot (p‐value <.050), but no difference was found between drug‐naïve and treated patients in any of these imaging metrics. (L). No significant difference was found in FC between Mot and VI among three groups. CBL, cerebellum; DMN, default mode network; FC, functional connectivity; MF, medial frontal network; Mot, motor network; NC, normal control; PD, Parkinson's disease; SAL, salience network; SC, subcortical network; VAs, visual association network; V I, visual I network. *: p‐value <.05, **: p‐value <.01, ***: p‐value <.001. LSD‐adjusted p‐values in the post hoc analyses are presented. All values presented were controlled for age and gender.

In the decreased connectome, significant intergroup differences were observed in global FC (q‐value = 0.009, Figure 4h) and FC of all top five network pairs (q‐value <0.050, Figure 4i–k,m) except for FC between Mot and V I (q‐value = 0.534, Figure 4l), summarized in Table 2. Both drug‐naïve and treated patients demonstrated significantly lower FC than NC (p‐value <.050) in these significant imaging metrics; however, no difference was found between drug‐naïve and treated patients. FC normalization was absent in the decreased connectomes (global or any of top‐five network pairs).

3.4. The global and network‐level FC differences between OFF‐ and ON‐state

In the increased connectome, compared to OFF‐state, all patients in ON‐state showed significantly increased global FC (q‐value = .046, Figure 5a) and increased FC between SC and Mot (q‐value = .034, Figure 5c), and between SC and VAs (q‐value = .046, Figure 5e). No significant FC changes were found in any other imaging metrics after the levodopa challenge test.

FIGURE 5.

FIGURE 5

Open in a new tab

Boxplots for differences between OFF‐ and ON‐states in global and network‐level FC of the increased (a–g) and decreased (h–m) connectomes. (a, c, e) In the increased connectome, all patients in ON‐state had significantly higher global FC, FC between SC and Mot, and FC between SC and VAs than that in OFF‐state (q‐value <0.050). (b, d, f, g) In the increased connectome, no significant differences were found between OFF‐ and ON‐state in FC within SC, between SC and V I, between SC and SAL, and between SC and CBL. (h, j, k) In the decreased connectome, all patients in ON‐state had significantly higher global FC, FC within Mot, and FC between DMN and Mot than in OFF‐state (q‐value <0.050). (i, l, m) In the decreased connectome, no significant differences were found between OFF‐ and ON‐state in FC between MF and Mot, between Mot and V I, and between SAL and Mot. CBL, cerebellum; DMN, default mode network; FC, functional connectivity; MF, medial frontal network; Mot, motor network; NC, normal control; PD, Parkinson's disease; SAL, salience network; SC, subcortical network; VAs, visual association network; V I, visual I network. *: q‐value <0.05. FDR‐corrected q‐values are presented. All values presented were controlled for age and gender.

In the decreased connectome, global, within Mot, and between DMN and Mot FC were significantly enhanced (q‐value = 0.046, Figure 5h,j–k) after the levodopa challenge test in all patients.

No significant interactive effect was found between medication groups (drug‐naïve and treated patients) and levodopa challenge states (OFF‐ and ON‐states) on all imaging metrics (Supplemental Table 2).

3.5. The association between motor symptoms and FC

In the decreased connectome, higher global FC, FC within Mot, and between MF and Mot was associated with lower UPDRS part 3 and rigidity score in all patients (q‐value <0.050, Figure 6a). Besides, FC between DMN and Mot was negatively correlated with UPDRS part 3 score (r = −.303, q‐value = 0.022). In the increased connectome, no imaging metrics were found associated with motor symptoms (q‐value >0.050, Figure 6a).

FIGURE 6.

FIGURE 6

Open in a new tab

Correlation heatmaps for partial correlation between imaging metrics and motor symptoms (a), and between imaging metrics and medication usage (b). (a) In the decreased connectome, higher global FC, FC within Mot, and between MF and Mot was associated with lower UPDRS part 3 and rigidity score in all patients (q‐value <0.050), and FC between DMN and Mot was negatively correlated with UPDRS part 3 score (q‐value = 0.022). In the increased connectome, no imaging metrics were found associated with motor symptoms (q‐value >0.050). (b) In the increased connectome, higher LED was significantly correlated with reduced global FC, FC between Mot and SC, between V I and SC, and between SAL and SC in ON‐state (q‐value <0.050), but no significant correlations were found between medication and imaging metrics in OFF‐state. CBL, cerebellum; DMN, default mode network; FC, functional connectivity; LED, levodopa equivalent dosage; MF, medial frontal network; Mot, motor network; SAL, salience network; SC, subcortical network; UPDRS, Unified Parkinson's Disease Rating Scale; VAs, visual association network; V I, visual I network. *: q‐value <0.05. **: q‐value <0.01. FDR‐corrected q‐values are presented.

3.6. The association between medication usage and FC

In the increased connectome, higher LED was significantly related to reduced global FC, FC between Mot and SC, between V I and SC, and between SAL and SC in ON‐state (q‐value<0.050, Figure 6b), but not in OFF‐state. In the decreased connectome, no significant medication effect (duration or LED) was found associated with global or network‐level FC (Supplemental Figure 1).

3.7. Additional analyses

Results using PCA‐generated timecourses (Supplemental Table 3, Supplemental Figures 2–5) and Brainnetome atlas (Supplemental Table 4, Supplemental Figures 6–9) were similar but not identical to the original results.

4. DISCUSSION

This study explored the influence of DRT, including long‐term and instant treatment, on brain functional networks and motor symptoms. Two disease‐related connectomes were identified and normalization effects of DRT on the connectomes were observed. The global FC of the increased connectome was abnormally augmented in drug‐naïve patients and normalized after long‐term DRT; however, FC of the decreased connectome in drug‐naïve patients was only improved after the levodopa challenge test but not after long‐term treatment. Besides, FC of the increased connectome was negatively associated with LED, and FC of the decreased connectome was positively correlated with scores of motor symptoms.

The increased connectome primarily consisted of Mot, CBL, V I, and SC, and the left thalamus demonstrated the highest contribution (most connections) among all brain regions. Aberrantly increased FC in this connectome was normalized after long‐term DRT, and lower FC was associated with higher LED in ON‐state. This increased connectome was consistent with the rs‐fMRI‐based PD‐related pattern characterized by increased activity in basal ganglia, thalamus, cerebellum, and SMA (Vo et al., 2017), reflecting excessive neural activity within the cerebello‐thalamo‐cortical circuit (CTC) (McGregor & Nelson, 2019; Wu & Hallett, 2013). Abnormal increased FC between the cerebellum, basal ganglia, thalamus, and motor cortex was well‐documented, especially during the early‐middle stage and in untreated patients (Agosta et al., 2014; Simioni et al., 2016; Vancea et al., 2019; Wu et al., 2009). Increased FC within CTC possibly played a compensating role for long‐term dopaminergic denervation, which was characterized by enhanced bursting and oscillatory neural activity (Blesa et al., 2017; Sen et al., 2010). Such aberrant neural activity was triggered by synaptic changes to offset chronic dopaminergic depletion, including upregulation of dopamine synthesis and release, downregulation of dopamine transporter, and increased receptor expression (Blesa et al., 2017; Rebelo et al., 2021). This increased connectome was normalized under long‐term DRT but not after the levodopa challenge test, which demonstrated CTC restoration was based on long‐term and repeated dopaminergic medication. Long‐term DRT modulates subcortical synapse plasticity and normalizes neuronal firing rate and receptor expression. Therefore, we assumed the pathologically increased neural activity of SC was alleviated by chronic and repeated dopamine replenishment. In summary, aberrantly increased FC of the increased connectome might be pathological and play a compensating role for chronic dopamine depletion, which was normalized in patients under long‐term DRT.

One intriguing finding is that the global FC in the increased connectome was enhanced after the instant levodopa challenge, which was negatively related to long‐term medication dosage. Although no significant interactive effects were found between long‐term and instant medication on all imaging metrics in this research, multiple clinical evidence supported levodopa responsiveness changed after long‐term treatment continued (Albin & Leventhal, 2017; Cilia et al., 2020; Hershey, 2003). Therefore, such interactive influence on neural activity and motor symptoms still needs further exploration.

The decreased connectome was mainly composed of Mot, MF, and SAL, and was reinstated after acute dopaminergic administration instead of long‐term DRT. Meanwhile, higher FC of the decreased connectome was associated with lower UPDRS part 3 and rigidity score. This lower FC in the decreased connectome reflected inadequate neuron activity in higher‐order networks due to dopaminergic denervation. Consistent with our finding, a significantly decreased FC in the prefrontal cortex, motor area, and DMN was reported in both treated and drug‐naïve patients and was enhanced after the levodopa challenge test (Michely et al., 2015; Wu et al., 2009; Zeng et al., 2022). Besides, lower activation in the motor area was significantly correlated with worse akinesia and rigidity symptoms in treated patients (Cao et al., 2020), and the neuronal response to muscle stretch was also reduced in the motor cortex of parkinsonian monkeys with severe rigidity (Pasquereau & Turner, 2013). The proper cooperation of these higher‐order networks is fundamental for motor planning, initiation, coordination, and adjustment, therefore, lower connectivity between and within these networks may impair voluntary movements (Krajcovicova et al., 2012; Xu et al., 2019). The levodopa challenge test generated dopamine signaling in a relatively short time scale and was responsible for the fine motor coordination (Albin & Leventhal, 2017); therefore, FC of the aforementioned networks was increased with better motor performance after levodopa challenge test. FC of the decreased connectome was not offset under long‐term DRT, and this possibly interpreted why DRT failed to slow down the motor progression. To sum up, the reduced FC of the decreased connectome in drug‐naïve and treated patients showed malfunctions of multiple higher‐order networks and was associated with motor impairments, which was only offset after instant levodopa challenge.

In additional analyses, connectomes generated using PCA‐generated and average timecourses were similar and the normalization effect of long‐term DRT was also found in FC between SAL and SC in the increased connectome. However, these supplementary results were not identical to original findings. The results using the PCA‐based method are sparser and more restricted than those using average timecourses, as PCA only captures the first temporal principal component and abandons the rest principal components. These discarded components might contain some subtle but possibly meaningful information. In this exploratory study, average timecourses were used for identifying more generalized disease‐related connectomes.

There were several limitations in this study. First, a major limitation of our study is the fixed OFF–ON order of MRI data acquisition. All patients were scanned in the OFF state, followed by scanning in the ON state after the levodopa challenge test. Scanning in a randomized order was not possible due to practical and ethical reasons. Therefore, we cannot exclude that results are at least partially due to order effects. A further major limitation is that levodopa was administered to patients but not to NC; however, this would be necessary to disentangle the general medication effect from the specific effects of levodopa administration in PD. Thus, we are not able to relate our findings specifically to levodopa in PD. Further studies might also apply levodopa to NC (Grimm et al., 2020; Kelly et al., 2009) to check for potential interactions between levodopa (OFF/ON) and group factor (PD/NC). Second, it was a cross‐sectional study and multi‐central longitudinal data is needed for verifying our preliminary findings. Third, only functional MRI was analyzed in this research, it is crucial to combine other neuroimaging techniques and biological markers for a comprehensive understanding. Last, this is a real‐world study, and parts of the treated patients included were prescribed different kinds of dopaminergic medication, including dopamine agonists and monoamine oxidase type‐B inhibitors, although LED and medication duration were recorded to measure the overall dopaminergic medication effects and propensity score matching was used to minimize the discrepancies of confounding factors.

5. CONCLUSION

Two distinct disease‐related connectomes were observed in this study. The abnormally increased connectome was normalized in long‐term DRT treated patients, while the decreased connectome was offset after levodopa challenge test. Our preliminary findings provided novel insights into the influence of DRT on neuron activity and its relationship with motor symptoms.

CONFLICT OF INTEREST STATEMENT

The authors declare no competing conflict of interest.

Supporting information

SUPPLEMENTAL MATERIAL 1. Xxxx

Click here for additional data file. (9.1MB, docx)

SUPPLEMENTAL MATERIAL 2. Xxxx

Click here for additional data file. (2.2KB, sps)

Wu, C. , Wu, H. , Zhou, C. , Guan, X. , Guo, T. , Cao, Z. , Wu, J. , Liu, X. , Chen, J. , Wen, J. , Qin, J. , Tan, S. , Duanmu, X. , Zhang, B. , Huang, P. , Xu, X. , & Zhang, M. (2023). Normalization effect of dopamine replacement therapy on brain functional connectome in Parkinson's disease. Human Brain Mapping, 44(9), 3845–3858. 10.1002/hbm.26316

Chenqing Wu and Haoting Wu contributed equally to this work.

DATA AVAILABILITY STATEMENT

The materials used and/or analyzed during the current study are available from the corresponding author (i.e., Prof Minming Zhang) on reasonable request.

REFERENCES

  1. Agosta, F. , Caso, F. , Stankovic, I. , Inuggi, A. , Petrovic, I. , Svetel, M. , Kostic, V. S. , & Filippi, M. (2014). Cortico‐striatal‐thalamic network functional connectivity in hemiparkinsonism. Neurobiology of Aging, 35, 2592–2602. [DOI] [PubMed] [Google Scholar]
  2. Albin, R. L. , & Leventhal, D. K. (2017). The missing, the short, and the long: Levodopa responses and dopamine actions: DRT and DA actions. Annals of Neurology, 82, 4–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Austin, P. C. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46, 399–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ballarini, T. , Růžička, F. , Bezdicek, O. , Růžička, E. , Roth, J. , Villringer, A. , Vymazal, J. , Mueller, K. , Schroeter, M. L. , & Jech, R. (2018). Unraveling connectivity changes due to dopaminergic therapy in chronically treated Parkinson's disease patients. Scientific Reports, 8, 14328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Benjamini, Y. , & Hochberg, Y. (1995). Controlling the false discovery rate—A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society, Series B, 57, 289–300. [Google Scholar]
  6. Benjamini, Y. , & Yekutieli, D. (2001). The control of the false discovery rate in multiple testing under dependency. The Annals of Statistics, 29(4), 1165–1188. [Google Scholar]
  7. Blesa, J. , Trigo‐Damas, I. , Dileone, M. , del Rey, N. L.‐G. , Hernandez, L. F. , & Obeso, J. A. (2017). Compensatory mechanisms in Parkinson's disease: Circuits adaptations and role in disease modification. Experimental Neurology, 298, 148–161. [DOI] [PubMed] [Google Scholar]
  8. Cao, C. , Li, D. , Zhan, S. , Zhang, C. , Sun, B. , & Litvak, V. (2020). L‐Dopa treatment increases oscillatory power in the motor cortex of Parkinson's disease patients. NeuroImage: Clinical, 26, 102255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Caspers, J. , Rubbert, C. , Eickhoff, S. B. , Hoffstaedter, F. , Südmeyer, M. , Hartmann, C. J. , Sigl, B. , Teichert, N. , Aissa, J. , Turowski, B. , Schnitzler, A. , & Mathys, C. (2021). Within‐ and across‐network alterations of the sensorimotor network in Parkinson's disease. Neuroradiology, 63, 2073–2085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cilia, R. , Cereda, E. , Akpalu, A. , Sarfo, F. S. , Cham, M. , Laryea, R. , Obese, V. , Oppon, K. , Del Sorbo, F. , Bonvegna, S. , Zecchinelli, A. L. , & Pezzoli, G. (2020). Natural history of motor symptoms in Parkinson's disease and the long‐duration response to levodopa. Brain, 143, 2490–2501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cote, S. R. , & Kuzhikandathil, E. V. (2015). Chronic levodopa treatment alters expression and function of dopamine D3 receptor in the MPTP/p mouse model of Parkinson's disease. Neuroscience Letters, 0, 33–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. De Micco, R. , Agosta, F. , Basaia, S. , Siciliano, M. , Cividini, C. , Tedeschi, G. , Filippi, M. , & Tessitore, A. (2021). Functional connectomics and disease progression in drug‐naïve Parkinson's disease patients. Movement Disorders, 36, 1603–1616. [DOI] [PubMed] [Google Scholar]
  13. Fan, L., Li, H., Zhuo, J., Zhang, Y., Wang, J., Chen, L., Yang, Z., Chu, C., Xie, S., Laird, A. R, Fox, P. T., Eickhoff, S. B., Yu, C., & Jiang, T. (2016). The Human Brainnetome Atlas: A New Brain Atlas Based on Connectional Architecture. Cereb Cortex N Y NY, 26, 3508–3526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ferrazzoli, D. , Carter, A. , Ustun, F. S. , Palamara, G. , Ortelli, P. , Maestri, R. , Yücel, M. , & Frazzitta, G. (2016). Dopamine replacement therapy, learning and reward prediction in Parkinson's disease: Implications for rehabilitation. Frontiers in Behavioral Neuroscience, 10, 121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fieblinger, T. , Zanetti, L. , Sebastianutto, I. , Breger, L. S. , Quintino, L. , Lockowandt, M. , Lundberg, C. , & Cenci, M. A. (2018). Striatonigral neurons divide into two distinct morphological‐physiological phenotypes after chronic L‐DOPA treatment in parkinsonian rats. Scientific Reports, 8, 10068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Filippi, M. , Basaia, S. , Sarasso, E. , Stojkovic, T. , Stankovic, I. , Fontana, A. , Tomic, A. , Piramide, N. , Stefanova, E. , Markovic, V. , Kostic, V. S. , & Agosta, F. (2021). Longitudinal brain connectivity changes and clinical evolution in Parkinson's disease. Molecular Psychiatry, 26, 5429–5440. [DOI] [PubMed] [Google Scholar]
  17. Ganzetti, M. , & Mantini, D. (2013). Functional connectivity and oscillatory neuronal activity in the resting human brain. Neuroscience, 240, 297–309. [DOI] [PubMed] [Google Scholar]
  18. Grimm, O. , Kopfer, V. , Küpper‐Tetzel, L. , Deppert, V. , Kuhn, M. , Greck, M. , & Reif, A. (2020). Amisulpride and l‐DOPA modulate subcortical brain nuclei connectivity in resting‐state pharmacologic magnetic resonance imaging. Human Brain Mapping, 41, 1806–1818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hayes, M. T. (2019). Parkinson's disease and parkinsonism. The American Journal of Medicine, 132, 802–807. [DOI] [PubMed] [Google Scholar]
  20. Heeger, D. J. , & Ress, D. (2002). What does fMRI tell us about neuronal activity? Nature Reviews. Neuroscience, 3, 142–151. [DOI] [PubMed] [Google Scholar]
  21. Hershey, T. (2003). Long term treatment and disease severity change brain responses to levodopa in Parkinson's disease. Journal of Neurology, Neurosurgery, and Psychiatry, 74, 844–851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hughes, A. J. , Daniel, S. E. , Kilford, L. , & Lees, A. J. (1992). Accuracy of clinical diagnosis of idiopathic Parkinson's disease: A clinico‐pathological study of 100 cases. Journal of Neurology, Neurosurgery, and Psychiatry, 55, 181–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jenkinson, M. , Bannister, P. , Brady, M. , & Smith, S. (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. NeuroImage, 17(2), 825–841. [DOI] [PubMed] [Google Scholar]
  24. Kalia, L. V. , & Lang, A. E. (2015). Parkinson's disease. The Lancet, 386, 896–912. [DOI] [PubMed] [Google Scholar]
  25. Kang, G. A. , Bronstein, J. M. , Masterman, D. L. , Redelings, M. , Crum, J. A. , & Ritz, B. (2005). Clinical characteristics in early Parkinson's disease in a central California population‐based study. Movement Disorders, 20, 1133–1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kelly, C. , de Zubicaray, G. , Di Martino, A. , Copland, D. A. , Reiss, P. T. , Klein, D. F. , Castellanos, F. X. , Milham, M. P. , & McMahon, K. (2009). L‐Dopa modulates functional connectivity in striatal cognitive and motor networks: A double‐blind placebo‐controlled study. The Journal of Neuroscience, 29, 7364–7378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Krajcovicova, L. , Mikl, M. , Marecek, R. , & Rektorova, I. (2012). The default mode network integrity in patients with Parkinson's disease is levodopa equivalent dose‐dependent. Journal of Neural Transmission, 119, 443–454. [DOI] [PubMed] [Google Scholar]
  28. Kriegeskorte, N. , Simmons, W. K. , Bellgowan, P. S. F. , & Baker, C. I. (2009). Circular analysis in systems neuroscience: The dangers of double dipping. Nature Neuroscience, 12, 535–540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lang, A. E. , & Espay, A. J. (2018). Disease modification in Parkinson's disease: Current approaches, challenges, and future considerations: Disease modification in PD. Movement Disorders, 33, 660–677. [DOI] [PubMed] [Google Scholar]
  30. McGregor, M. M. , & Nelson, A. B. (2019). Circuit mechanisms of Parkinson's disease. Neuron, 101, 1042–1056. [DOI] [PubMed] [Google Scholar]
  31. Michely, J. , Volz, L. J. , Barbe, M. T. , Hoffstaedter, F. , Viswanathan, S. , Timmermann, L. , Eickhoff, S. B. , Fink, G. R. , & Grefkes, C. (2015). Dopaminergic modulation of motor network dynamics in Parkinson's disease. Brain, 138, 664–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Noble, S. , Spann, M. N. , Tokoglu, F. , Shen, X. , Constable, R. T. , & Scheinost, D. (2017). Influences on the test–retest reliability of functional connectivity MRI and its relationship with behavioral utility. Cerebral Cortex, 27, 5415–5429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Parkes, L. , Fulcher, B. , Yücel, M. , & Fornito, A. (2018). An evaluation of the efficacy, reliability, and sensitivity of motion correction strategies for resting‐state functional MRI. NeuroImage, 171, 415–436. [DOI] [PubMed] [Google Scholar]
  34. Pasquereau, B. , & Turner, R. S. (2013). Primary motor cortex of the parkinsonian monkey: Altered neuronal responses to muscle stretch. Frontiers in Systems Neuroscience, 7, 98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Postuma, R. B. , Berg, D. , Stern, M. , Poewe, W. , Olanow, C. W. , Oertel, W. , Obeso, J. , Marek, K. , Litvan, I. , Lang, A. E. , Halliday, G. , Goetz, C. G. , Gasser, T. , Dubois, B. , Chan, P. , Bloem, B. R. , Adler, C. H. , & Deuschl, G. (2015). MDS clinical diagnostic criteria for Parkinson's disease. Journal of Movement Disorders, 30, 1591–1601. [DOI] [PubMed] [Google Scholar]
  36. Poulopoulos, M. , & Waters, C. (2010). Carbidopa/levodopa/entacapone: The evidence for its place in the treatment of Parkinson's disease. Core Evidence, 5, 1–10. [PMC free article] [PubMed] [Google Scholar]
  37. Rebelo, D. , Oliveira, F. , Abrunhosa, A. , Januário, C. , Lemos, J. , & Castelo‐Branco, M. (2021). A link between synaptic plasticity and reorganization of brain activity in Parkinson's disease. Proceedings of the National Academy of Sciences, 118, e2013962118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ruottinen, H. M. , & Rinne, U. K. (1996). Entacapone prolongs levodopa response in a one month double blind study in parkinsonian patients with levodopa related fluctuations. Journal of Neurology, Neurosurgery, and Psychiatry, 60, 36–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Saranza, G. , & Lang, A. E. (2021). Levodopa challenge test: Indications, protocol, and guide. Journal of Neurology, 268, 3135–3143. [DOI] [PubMed] [Google Scholar]
  40. Sen, S. , Kawaguchi, A. , Truong, Y. , Lewis, M. M. , & Huang, X. (2010). Dynamic changes in cerebello‐thalamo‐cortical motor circuitry during progression of Parkinson's disease. Neuroscience, 166, 712–719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shen, X. , Tokoglu, F. , Papademetris, X. , & Constable, R. T. (2013). Groupwise whole‐brain parcellation from resting‐state fMRI data for network node identification. NeuroImage, 82, 403–415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Simioni, A. C. , Dagher, A. , & Fellows, L. K. (2016). Compensatory striatal–cerebellar connectivity in mild–moderate Parkinson's disease. NeuroImage: Clinical, 10, 54–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Smith, M. L. , King, J. , Dent, L. , Mackey, V. , Muthian, G. , & Grif, B. (2014). Effects of acute and sub‐chronic L‐dopa therapy on striatal L‐dopa methylation and dopamine oxidation in an MPTP mouse model of Parkinson's disease. Life Sciences, 110(1), 1–7. [DOI] [PubMed] [Google Scholar]
  44. Stebbins, G. T. , Goetz, C. G. , Burn, D. J. , Jankovic, J. , Khoo, T. K. , & Tilley, B. C. (2013). How to identify tremor dominant and postural instability/gait difficulty groups with the movement disorder society unified Parkinson's disease rating scale: Comparison with the unified Parkinson's disease rating scale: PIGD and the MDS‐UPDRS. Movement Disorders, 28, 668–670. [DOI] [PubMed] [Google Scholar]
  45. Suárez, L. M. , Solís, O. , Caramés, J. M. , Taravini, I. R. , Solís, J. M. , Murer, M. G. , & Moratalla, R. (2014). L‐DOPA treatment selectively restores spine density in dopamine receptor D2–expressing projection neurons in dyskinetic mice. Biological Psychiatry, 75, 711–722. [DOI] [PubMed] [Google Scholar]
  46. Suo, X. , Zuo, C. , Lan, H. , Pan, N. , Zhang, X. , Kemp, G. J. , Wang, S. , & Gong, Q. (2022). COVID‐19 vicarious traumatization links functional connectome to general distress. NeuroImage, 255, 119185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Tinaz, S. (2021). Functional connectome in Parkinson's disease and parkinsonism. Current Neurology and Neuroscience Reports, 21, 24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tomlinson, C. L. , Stowe, R. , Patel, S. , Rick, C. , Gray, R. , & Clarke, C. E. (2010). Systematic review of levodopa dose equivalency reporting in Parkinson's disease: Systematic review of LED reporting in PD. Movement Disorders, 25, 2649–2653. [DOI] [PubMed] [Google Scholar]
  49. Vancea, R. , Simonyan, K. , Petracca, M. , Brys, M. , Di Rocco, A. , Ghilardi, M. F. , & Inglese, M. (2019). Cognitive performance in mid‐stage Parkinson's disease: Functional connectivity under chronic antiparkinson treatment. Brain Imaging and Behavior, 13, 200–209. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Vo, A. , Sako, W. , Fujita, K. , Peng, S. , Mattis, P. J. , Skidmore, F. M. , Ma, Y. , Uluğ, A. M. , & Eidelberg, D. (2017). Parkinson's disease‐related network topographies characterized with resting state functional MRI: Disease‐related network analysis with rs‐fMRI. Human Brain Mapping, 38, 617–630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wang, J. , Wang, X. , Xia, M. , Liao, X. , Evans, A. , & He, Y. (2015). GRETNA: A graph theoretical network analysis toolbox for imaging connectomics. Frontiers in Human Neuroscience, 9, 386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wang, Z. , Goerlich, K. S. , Ai, H. , Aleman, A. , Luo, Y. , & Xu, P. (2021). Connectome‐based predictive modeling of individual anxiety. Cerebral Cortex, 31, 3006–3020. [DOI] [PubMed] [Google Scholar]
  53. Wu, J. , Zhou, C. , Guo, T. , Guan, X. , Gao, T. , Bai, X. , Wu, H. , Chen, J. , Wen, J. , Liu, X. , Gu, L. , Song, Z. , Xuan, M. , Gu, Q. , Huang, P. , Pu, J. , Zhang, B. , Xu, X. , & Zhang, M. (2022). Cholinergic relevant functional reactivity is associated with dopamine responsiveness of tremor in Parkinson's disease. Brain Imaging and Behavior, 16, 1234–1245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wu, T. , & Hallett, M. (2013). The cerebellum in Parkinson's disease. Brain, 136, 696–709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wu, T. , Wang, L. , Chen, Y. , Zhao, C. , Li, K. , & Chan, P. (2009). Changes of functional connectivity of the motor network in the resting state in Parkinson's disease. Neuroscience Letters, 460, 6–10. [DOI] [PubMed] [Google Scholar]
  56. Xu, P. , Chen, A. , Li, Y. , Xing, X. , & Lu, H. (2019). The medial prefrontal cortex in neurological diseases. Physiological Genomics, 51(9), 432–442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Yan, C.‐G. , Wang, X.‐D. , Zuo, X.‐N. , & Zang, Y.‐F. (2016). DPABI: Data processing & analysis for (resting‐state) brain imaging. Neuroinformatics, 14, 339–351. [DOI] [PubMed] [Google Scholar]
  58. You, H. , Mariani, L.‐L. , Mangone, G. , Febvre, L. , de Nailly, D. , Charbonnier‐Beaupel, F. , & Corvol, J.‐C. (2018). Molecular basis of dopamine replacement therapy and its side effects in Parkinson's disease. Cell and Tissue Research, 373, 111–135. [DOI] [PubMed] [Google Scholar]
  59. Zalesky, A. , Fornito, A. , & Bullmore, E. T. (2010). Network‐based statistic: Identifying differences in brain networks. NeuroImage, 53, 1197–1207. [DOI] [PubMed] [Google Scholar]
  60. Zeng, W. , Fan, W. , Kong, X. , Liu, X. , Liu, L. , Cao, Z. , Zhang, X. , Yang, X. , Cheng, C. , Wu, Y. , Xu, Y. , Cao, X. , & Xu, Y. (2022). Altered intra‐ and inter‐network connectivity in drug‐naïve patients with early Parkinson's disease. Frontiers in Aging Neuroscience, 14, 783634. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

SUPPLEMENTAL MATERIAL 1. Xxxx

Click here for additional data file. (9.1MB, docx)

SUPPLEMENTAL MATERIAL 2. Xxxx

Click here for additional data file. (2.2KB, sps)

Data Availability Statement

The materials used and/or analyzed during the current study are available from the corresponding author (i.e., Prof Minming Zhang) on reasonable request.

Articles from Human Brain Mapping are provided here courtesy of Wiley

RESOURCES