Loss of Neocortical Oligodendrocytes in Parkinson's Disease Patients—A Stereological Study

ABSTRACT

Parkinson's disease (PD) is a neurodegenerative disorder, clinically characterized by motor‐ and non‐motor symptoms. Misfolding of alpha‐synuclein into pathological substrates, followed by neuronal spread, seems to be a key event in the spatiotemporal course of the disease. Previously, few studies have yielded insight into altered cell numbers in post‐mortem brains from PD patients. Beside nigral neuronal loss, no quantitative studies have reported significant loss of neurons in PD brains, suggesting that the neuropathological progress in PD have minor impact on neuronal survival outside of the substantia nigra. Reports further show impairment of intra‐ and inter‐cortical white matter tracts, observations that signify a role of oligodendrocytes in PD pathogenesis. Hence, our objective of the present study was to investigate oligodendroglia morphology (total numbers and volumes) in the neocortex of post‐mortem brains from PD patients and control subjects using stereological methods. Our findings show an explicit 34% reduction in the total number of oligodendrocytes in the frontal cortex of PD patients, with no such changes in the other cortical lobes. Furthermore, the results displayed no group differences in the mean volume of oligodendrocytes in the entire neocortex or its subregions. The reduction in frontal oligodendrocytes may be related to the demonstrated impairment of white matter tracts, however, further studies are needed to clarify the mechanistic and functional relevance of oligodendropathy in PD.

Significance Statement

This study is the first to use validated quantitative methods to estimate the total number and volume of oligodendrocytes in the neocortex and its subregions in brains from patients diagnosed with Parkinson's Disease (PD). We report a region‐ and cell‐specific 34% reduction in the number of oligodendrocytes in the frontal cortex of PD patients. We suggest that this pathological change may be associated with the previously described impairment of white matter tracts in PD brains and points to a selective vulnerability of the frontal cortico‐cortical connectivity.

In this study we used stereology on human brains to show a region‐ and cell‐specific 34% reduction in the number of oligodendrocytes in the frontal cortex of PD patients. This reduction was driven by a loss of oligodendrocytes with a nuclear size of 60–80 μm³ and 80–100 μm³ which implies that myelinating oligodendrocytes may be disproportionally affected in PD. We suggest that the observed pathological change may be associated with the previously described impairment of white matter tracts in PD brains, and points to a selective vulnerability of the frontal cortico‐cortical connectivity. The graphical abstract was created in https://BioRender.com.

1. Introduction

Parkinson's disease (PD) is a neurodegenerative disorder clinically characterized by bradykinesia, resting tremor, rigidity, postural instability, and impairment of gait. In addition, patients often experience a wide range of non‐motor symptoms including autonomic dysregulation, REM sleep behavior disorder (RBD), cognitive impairment, and depression (Bloem et al. 2021; Postuma et al. 2015). Pathologically, the hallmark lesions of PD are dopaminergic neuronal loss in the substantia nigra and accumulation of alpha‐synuclein (α‐syn) deposits in neurons (Lewy bodies) and cell processes (Lewy neurites) throughout the central‐, peripheral‐, and enteric nervous system (Braak and Braak 2000; Koga et al. 2021; Poewe et al. 2017; Wakabayashi et al. 1989). The distribution of Lewy body pathology was first described by Braak and colleagues showing an initial spread from the medulla and olfactory bulbs throughout the CNS as PD progresses (Braak et al. 2003, 2004). More recently, Horsager and colleagues (Horsager et al. 2020), proposed the “brain‐first body‐first”‐hypothesis, suggesting two phenotypes of PD, depending on the initial site of α‐synucleinopathy. Thus, misfolding of α‐syn into pathological species and subsequent neuronal spread seems to be a key event in the disease.

Besides the loss of nigral neurons in PD (Pakkenberg et al. 1991), quantitative studies find no significant changes in neuronal numbers in neocortical (Pedersen et al. 2005), hippocampal (Joelving et al. 2006), and cerebellar (Rusholt et al. 2020) brain regions, as well as no changes in total glial cell numbers in the hippocampus (Joelving et al. 2006). Contrary to this apparent focal neurodegeneration, previous studies have demonstrated impairment of intra‐ and inter‐cortical white matter tracts (Dean et al. 2016; Atkinson‐Clement et al. 2017; Cousineau et al. 2017), suggestive of a role for oligodendrocytes in PD pathogenesis and raising the possibility that their degeneration may contribute to motor and cognitive symptoms in the patients. The aim of this study was therefore to use design‐based stereology to quantify the total number and volume of oligodendrocytes in post‐mortem neocortex and its four major subregions (frontal‐, temporal‐, parietal‐, and occipital lobes) in brains from PD patients and control subjects.

2. Material and Methods

2.1. Patients

Brains from 22 individuals were obtained from the brain bank at the Centre for Neuroscience and Stereology, Copenhagen University Hospital—Bispebjerg and Frederiksberg, Denmark. The brain bank is approved by the Danish Data Protection Agency (j. no.: p‐2023‐15,234) and the study has been approved by the Research Ethics Committee of the Capital Region (j. no.: H‐19079007). A brain from a patient diagnosed with PD was excluded because of technical reasons resulting in the inclusion of nine brains from PD patients (4 male/5 female; mean age 76 (71–83) years) and 12 brains from age‐ and gender‐matched control subjects (6 male/6 female; mean age 74 (66–83) years), with no history of neurological disorder, in the study. The brains were collected between 1974 and 1994 (mean post‐mortem interval (PMI): PD patients = 34 (12–60) hours; control subjects = 29 (10–60) hours). For each PD patient, the clinical data were collected retrospectively from the medical records by a movement disorder specialist (Department of Neurology, Copenhagen University Hospital—Bispebjerg and Frederiksberg, Copenhagen, Denmark). The study was performed in accordance with recognized standards (Declaration of Helsinki) and the Danish laws regarding the use of post‐mortem human tissue in research at the time. Finally, all PD patients fulfilled the UK Parkinson's Disease Society Brain Bank diagnostic criteria for clinically probable idiopathic PD (Calne et al. 1992). Demographic‐ and autopsy‐related data are shown in Table 1, and clinical data on PD patients are shown in Table 2.

TABLE 1.

Demographic and autopsy‐related data on PD patients and control subjects.

	n	Male/Female	Age (years)	Brain weight (g)	Hemisphere (left/right)	PMI (hours)
Control	12	6/6	74 [66–83]	1285 [970–1420]	6/6	29 [10–60]
PD	9	4/5	76 [71–83]	1210 [1060–1420]	4/5	34 [12–60]

Note: Mean values with range in brackets.

Abbreviations: n, number of individuals; PD, Parkinson's disease; PMI, post‐mortem interval.

TABLE 2.

Clinical characteristics of PD patients.

Patients	1	2	3	4	5	6	7	8	9	Mean
Age at disease onset	66	72	59	67	74	68	45	77	NA	66
Disease duration (years)	9	2	13	12	3	6	26	3	> 20	10
Motor symptoms
Rigidity	S	R	R > L	L > R	S	L > R	S	S	NA
Bradykinesia	NA	S	S	S	U	U	R	U	NA
Tremor	NA	R > L	R > L	L > R	S	−	−	S	NA
Postural instability	NA	+	NA	+	+	NA	NA	+	NA
Parkinsonian gait	NA	+	+	+	+	NA	+	+	NA
Cognitive impairment	NA	+	+	NA	NA	NA	+	NA	+
Treatment
Amantadine	NA	NA	+	+	+	−	+	−	NA
Bromocriptine	NA	NA	+	−	+	−	−	−	NA
Levodopa	+	NA	+	+	+	−	+	+	+
Trihexyphedyl	NA	NA	−	+	−	−	−	−	NA
Benzohexol	NA	NA	−	−	−	−	−	+	NA
Surgery (year)	NA	NA	−	−	−	−	1965	−	NA
Response to treatment	+	NA	+	NA	NA	NA	+	+	NA
Side effects a	+	NA	+	+	NA	NA	+	−	NA
Comorbidity	St IHD Depr	SF	Spi	St HA IHD Art	HA	NA	NA	−	NA
Other neurological symptoms	NA	NA	NA	CN3	−	GAp	NA	−	NA
Neuropathology	LB	LB/LN	LB	LB	LB	LB	LB	LB	LB
LPC	NA	Bs	Limb	Bs	Bs	Limb	Neo	Bs/Limb	Bs
ABC‐score	A0	A0	A1	A0	A0	A0	A0	A0	A0
	B1	B1	B2	B0	B1	B0	B1	B1	B2
	C0	C0	C2	C0	C0	C0	C0	C0	C0

Note: ABC‐score = co‐AD pathologies in accordance with the National Institute on Aging‐Alzheimer's Association guidelines (Hyman et al. 2012); −, not applicable.

Abbreviations: Art, arthritis; Bs, Brainstem predominant; CN3, occulomotorius paresis; GAp, global aphasia; HA, hypertensio arterialis; IHD, ischemic heart disease; L, left; LB, Lewy Bodies; Limb, Limbic predominant; LN, Lewy neurites; LPC, Lewy pathology consensus criteria²³; NA, not available; NAD, no abnormality detected; Neo, Neocortical predominant; PD, Parkinson's disease; R, right; S, symmetrical; SF, Spanish flu; Spi, lumbal spinal stenosis; St, stroke; U, universal.

^a Hallucinations, dyskinesia.

2.2. Neuropathology

To verify the neuropathological diagnosis of PD, tissue samples were taken from the medulla, substantia nigra, amygdala, cingulate gyrus, neocortex (frontal‐, temporal‐, parietal‐ and occipital lobes) and hippocampus of each brain. Tissue blocks were embedded in paraffin and sliced into 10‐μm‐thick sections for hematoxylin and eosin (HE) staining. Additionally, we performed immunohistochemistry using antibodies against α‐synuclein (Enzo Life Sciences Cat# BML‐SA3400‐0025, RRID:AB_2050693), phospho‐tau (Thermo Fisher Scientific Cat# MN1020, RRID:AB_223647) and beta‐amyloid (Agilent Cat# M0872, RRID:AB_2056966). Following the Lewy pathology consensus criteria (Attems et al. 2021), Parkinson's type neuropathology was confirmed for all nine PD brains. As neuropathology was performed retrospectively and not at brain donation, resulting in long‐term fixation of the tissue, our staging is restricted to immunohistochemistry for α‐synuclein in six brains and visible Lewy bodies on HE‐stained sections in the remaining three brains. According to the National Institute on Aging‐Alzheimer's Association guidelines (ABC‐score) (Hyman et al. 2012) mild co‐pathology in the form of neurofibrillary tangles positive for hyperphosphorylated tau was observed in seven brains and amyloid plaques for a single brain (for details see Table 2).

2.3. Tissue Preparation

The protocol for tissue processing is described in detail elsewhere (Pedersen et al. 2005). In short, brains were stored in 10% neutral buffered formalin (pH = 7.2) for at least 5 months. One cerebral hemisphere from each brain was sampled systematically at uniform random, resulting in an even distribution of right and left hemispheres in each group (see Table 1). To distinguish the four anatomical lobes (frontal‐, temporal‐, parietal‐, and occipital cortices), the surface of each hemisphere was painted with different colors of waterproof ink. The hemispheres were embedded in 4% agar and sliced coronally into slabs of 4.5 or 7.0 mm with a random start. Each slab was photographed for subsequent estimation of the volumes of each cortical lobe. Every other slab was then cut systematically to produce 25–30 transcortical 4‐mm wide rods for each lobe (Regeur and Pakkenberg 1989). Every 3rd transcortical rod was subsampled, resulting in 8–10 rods per lobe. These rods were randomly rotated around the vertical axis, and then embedded in Historesin (2‐hydroxyethyl methacrylate, Kulzer, Germany). From each Historesin‐embedded rod, a 35‐μm‐thick section was cut from its center and stained with a modified Wolbach's Giemsa stain for cell counting using optical disectors (Gundersen et al. 1988). During preparation of the rods, extra rods were collected and measured for shrinkage due to processing. No significant shrinkage was detected between groups. In the present study, the described preparation of the tissue was already performed (Pedersen et al. 2005).

2.4. Stereology

Applying the optical disectors (Gundersen et al. 1988), we estimated the numerical densities of oligodendrocytes in the cortical gray matter using the newCAST software (www.visiopharm.com, RRID:SCR_014570). Stained sections were examined using an Olympus BX50 microscope, equipped with a 60× oil‐immersion lens (numerical aperture = 1.40), providing a final magnification of 1470×. The microscope was equipped with a high‐resolution color camera (Basler, Germany), a motorized X‐Y‐stage (Märzhäuser, Germany), and a digital microcator (Heidenhain VRZ401, Germany) for measuring the Z‐position. Disectors measuring 37.500 μm³ (area = 2500 μm², height = 15 μm) for the frontal‐, temporal‐, and parietal lobes, and 18.750 μm³ (area = 1250 μm², height = 15 μm) for the occipital lobe were superimposed on the magnified image of the tissue on a computer screen. The average section thickness was 35 μm, and the upper guard zone 5 μm, leaving a 15 μm guard zone at the bottom. Disectors were uniformly randomly distributed in two dimensions with an X‐Y step length of 1000 μm. In each brain, a mean of 252 (151–348) disectors were applied, with means of 456 oligodendrocytes (291–708) counted. The total number of oligodendrocytes in the four different cortical lobes was calculated by multiplying their numerical density by the reference volume, with doubling to obtain bilateral numbers. Reference volumes had been estimated using point counting and Cavalieri's principle, as part of an earlier study (Pedersen et al. 2005). The volume of oligodendrocytes was measured using the vertical rotator principle (Jensen and Gundersen 1993) and a 100× oil‐immersion lens (numerical aperture = 1.40), providing a final magnification of 2700×. Disectors measuring 60,000 μm³ (area = 4000 μm², height = 15 μm) were uniformly randomly distributed with an X‐Y step length of 1750 μm. A mean number of 77 cells (51–123) was used for oligodendrocytic volume estimations in each cortical lobe. Manual cell counting and volume estimations were conducted by a single experimenter who was blinded to the groups (control/disease).

2.5. Cell Identification

Oligodendrocytes were identified visually by identifying cytological characteristics. To ensure uniformity and systematic identification, a validated algorithm was used (García‐Cabezas et al. 2016) defining oligodendrocytes by their round, darkly stained nucleus.

2.6. Statistics

Stereological data were analyzed using two‐tailed Students t‐test corrected for multiple comparisons (n = 5); the four brain regions and the entire neocortex (Bonferroni‐Dunn method, alpha value = 0.05). All data were normal distributed according to the Shapiro Wilk normality test. For all analyses, outliers were removed when data points were outside the range of mean ±2 standard deviations. As one outlier was removed from each dataset in the following groups: PD patients (estimation of cell numbers in the frontal lobe), PD patients (estimation of oligodendrocytic volume in the temporal lobe) and control subjects (estimation of oligodendrocytic volume in the occipital lobe), we repeated the statistical analyses keeping the outlier in and reaching similar significant results. For analyses of the average size distribution of oligodendrocytes, we performed 2‐way ANOVA followed by Bonferroni's multiple comparisons test (alpha value = 0.05). The coefficients of variation (CV) and the coefficients of error (Gundersen CE (Gundersen et al. 1999)), expressing the precision of the cell number estimates ranged from: Control: CV = 0.11–0.24; CE = 0.10–0.22; PD: CV = 0.26–0.41; CE = 0.10–0.20. Data analysis and graphical presentations were performed using GraphPad Prism 10 (www.graphpad.com, RRID:SCR_002798).

3. Results

The stereological estimates and the p‐values of the total number of oligodendrocytes, their mean volume and average size distribution are shown in Figures 1, 2, 3 and Table 3.

FIGURE 1.

Total number of oligodendrocytes. Estimated total number of neocortical oligodendrocytes in the frontal‐ (A), temporal‐ (B), parietal‐ (C), and occipital (D) cortices as well as the entire neocortex (E) of control subjects (CS) and Parkinson's disease (PD) patients. Outlier is shown as a gray circle in A. Error bars represent mean values ± standard deviation (SD). **p < 0.01.

FIGURE 2.

Mean volume of oligodendrocytes. Estimated mean volume of neocortical oligodendrocytes in the frontal‐ (A), temporal‐ (B), parietal‐ (C), and occipital (D) cortices as well as the entire neocortex (E) of control subjects (CS) and Parkinson's disease (PD) patients. Outliers are shown as gray circles in B and D. Error bars represent mean values ± standard deviation (SD).

FIGURE 3.

Size distribution of oligodendrocytes. Average size distribution of oligodendrocytes in the frontal‐ (A), temporal‐ (B), parietal‐ (C), and occipital (D) cortices as well as the entire neocortex (E) of control subjects (open circles) and Parkinson's disease (black circles) patients. Error bars represent mean values ± standard deviation (SD). *p = 0.041, **p = 0.010, ***p < 0.001.

TABLE 3.

Stereological estimation of oligodendrocytes in control subjects and PD patients.

Cortical region	Control	PD	Statistics
			Outlier removal	All data
Total numbers (109)
Frontal	10.3 (7.37–12.8)	6.78 (4.42–10.1) a	t18 = 4.504; p = 0.001	t19 = 2.966; p = 0.040
Temporal	6.04 (5.06–7.27)	5.57 (3.05–9.46)	t19 = 0.738; p > 0.999	—
Parietal	6.09 (3.79–8.67)	5.17 (2.12–8.91)	t19 = 1.181; p > 0.999	—
Occipital	4.16 (3.35–5.65)	3.21 (1.59–5.15)	t19 = 2.146; p = 0.225	—
Entire neocortex	26.6 (20.3–31.0)	21.4 (12.0–34.0)	t19 = 2.205; p = 0.200	—
Volumes (μm3)
Frontal	82.7 (68.5–96.0)	85.9 (69.7–107)	t19 = 0.662; p > 0.999	—
Temporal	74.5 (60.7–87.5)	75.0 (63.0–108) a	t18 = 0.128; p > 0.999	t19 = 0.846; p > 0.999
Parietal	73.2 (57.2–88.3)	76.9 (59.5–94.1)	t19 = 0.825; p > 0.999	—
Occipital	69.1 (54.3–81.3) a	72.6 (52.8–91.5)	t18 = 0.721; p > 0.999	t19 = 0.344; p > 0.999
Entire neocortex	75.5 (63.8–84.0)	78.1 (62.7–99.9)	t19 = 0.757; p > 0.999	—

Note: Mean values with range in parenthesis. The displayed p‐values are the adjusted levels following correction for multiple comparisons.

^a One outlier was removed from the dataset.

3.1. Number of Oligodendrocytes

No significant difference was found in the total number of oligodendrocytes in the entire neocortex between brains from PD patients (21.4 (12.0–34.0) × 10⁹) and control subjects (26.6 (20.3–31.0) × 10⁹) (Figure 1E). However, the results showed a significant 34% decrease in the total number of oligodendrocytes in the frontal cortex of PD patients compared to control subjects (Figure 1A), with no other significant differences in the temporal (Figure 1B), parietal (Figure 1C), or occipital (Figure 1D) cortices.

3.2. Volume of Oligodendrocytes

Estimation of the mean volume of oligodendrocytes showed no significant difference in the entire neocortex between brains from PD patients (78.1 (62.7–99.9) μm³) and control subjects (75.5 (63.8–84.0) μm³) (Figure 2E). Neither did the results display any significant differences in the frontal (Figure 2A), temporal (Figure 2B), parietal (Figure 2C), or occipital (Figure 2D) cortices.

3.3. Size Distribution of Oligodendrocytes

Scatter plots of the average size distribution in the cortical lobes and the entire neocortex are shown in Figure 3A–E. In the frontal lobe, PD patients had a significantly reduced number of oligodendrocytes with nuclear volumes of 60–80 μm³ (p = 0.041) and 80–100 μm³ (p < 0.001) (Figure 3A) compared to controls. A reduced number of oligodendrocytes was also observed in nuclear volumes of 80–100 μm³ (p = 0.010) (Figure 3E) in the entire neocortex of PD patients. No significant changes were seen in the temporal‐ (Figure 3B), parietal‐ (Figure 3C), or occipital lobes (Figure 3D).

4. Discussion

To our knowledge, this is the first quantitative study reporting the total number of oligodendrocytes and their volume in the neocortex of PD brains. The main finding of this study is a 34% reduction in the number of oligodendrocytes in the frontal cortex of PD patients. The reduction was specific for the frontal lobe, as we did not find significant changes in the total number of oligodendrocytes in the entire neocortex or the temporal‐, parietal‐, and occipital cortices. Neither did we find any significant differences in the mean volume of their nucleus.

The finding of a significant decrease in the total numbers of frontal oligodendrocytes corresponds well with previous findings of white matter alterations in brains from PD patients. Imaging studies show significant changes in endpoints describing myelination quality for white matter tracts (e.g., thalamofrontal‐, thalamoparietal‐, and thalamo‐occipital tracts (Dean et al. 2016; Feng et al. 2020)) and pathways (e.g., nigro‐subthalamo‐putaminal‐thamalo‐cortical pathway (Cousineau et al. 2017)). Also, some imaging studies show reductions of the white matter volume in brains from PD patients (Nyatega et al. 2022; Wen et al. 2015), though this is not a consistent finding (Zheng et al. 2022). In addition, molecular changes such as downregulation of myelin genes and impairment of oligodendrocytic developmental pathways have been reported in PD white matter (Bryois et al. 2020; Siokas et al. 2022; Xie et al. 2022) and cingulate cortex (Xie et al. 2022), indicating disrupted myelination. Interestingly, in the study by Xie and colleagues, cell‐type‐specific identification displayed a disproportional damage to myelinating oligodendrocytes. This led us to investigate the oligodendrocytic nuclear volume and their size distribution across the neocortex as it is well‐known that the size of oligodendrocytes varies between sub‐classes and differentiation status. Our results displayed that while there were no significant differences in the mean volume of the total pool (all subtypes) of oligodendrocytes, the significant reduction of oligodendrocytes in the frontal cortex was driven by a loss of oligodendrocytes with a nuclear size of 60–80 μm³ and 80–100 μm³. Overall, these are the two most numerous size groups, constituting 57% of the total estimated oligodendrocytic population in the frontal cortex of PD brains. This result suggest that myelinating oligodendrocytes may be disproportionally affected in PD, supporting previous semiquantitative studies (Azevedo et al. 2022; Fu et al. 2022; Xie et al. 2022). Besides a potential deprived myelin support, the decrease in frontal oligodendrocytes may also lead to a changed metabolic support, which alone or in combination with demyelination could lead to the previously described impairment of the white matter tracts. Still, future studies examining the involvement of oligodendrocytic lineage cells and molecular pathways of oligodendrocytes are warranted.

Cognitive impairment is frequent in PD with up to 83% of PD patients developing dementia after 20 years (Emre et al. 2007; Hely et al. 2008). While cognitive impairment in PD patients has traditionally been linked to neuronal changes, Lewy body pathology and co‐existing alterations related to Alzheimer's, (reviewed by Aarsland et al. 2021, Jellinger 2023), studies increasingly recognize oligodendroglia and demyelination as contributing factors (Dean et al. 2016; Bae et al. 2023; Barba‐Reyes et al. 2025). This is supported by imaging studies in PD patients showing a correlation between white matter microstructural alterations, thought to reflect changes in oligodendrocytic cell numbers or demyelination, and cognitive decline (Auning et al. 2014; Monchi et al. 2024; Zarkali et al. 2024). In our study, cognitive impairment was reported in four of the nine patients, with the cognitive status of the remaining five patients being unknown and preventing us from testing a possible link between oligodendrocytic numbers and cognition. Future studies of a PD cohort stratified for cognitive impairment looking at oligodendroglia in the context of co‐pathology are needed to address this issue.

While the relatively small sample size increases the risk of false‐negative results, possibly masking minor reductions in oligodendrocyte numbers in the other cortical lobes, we observed a significant 34% reduction of oligodendrocytes in the frontal lobe. The use of nuclear morphology and size points toward a preferential reduction of myelinated oligodendrocytes, a finding that will be interesting to address in future studies using immunohistochemical staining techniques for targeting oligodendrocytes explicitly (e.g., anti‐OLIG2) and/or lineage cells (e.g., oligodendrocytic progenitor cells—anti‐NG2; premyelinating oligodendrocytes—anti‐Sox10 or anti‐NKX2.2; myelinating oligodendrocytes—anti‐MOG or anti‐CC1).

In summary, we applied stereological methods to estimate total numbers of oligodendrocytes and their volume in neocortical brain regions from PD patients and matched controls. Our results of an exclusively significant 34% reduction in frontal oligodendrocytes could be related to the previously demonstrated impairment of white matter tracts; however, further studies are warranted to map the causative mechanisms and functional relevance of oligodendrocytic loss in PD.

Supporting information: For declaration of transparency see Data S1.

Author Contributions

All the authors had full access to the data of the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, M.V.O.; Methodology, M.V.O.; Investigation, M.P.H., L.S., S.S.K., M.V.O.; Formal analysis, M.P.H., L.S., S.S.K., M.V.O.; Resources, L.S., S.S.K., M.V.O.; Writing – original draft, M.P.H., L.S., M.V.O.; Writing – review and editing, M.P.H., L.S., S.S.K., M.V.O.; Visualization, M.P.H., M.V.O.; Supervision, M.V.O.; Funding acquisition, M.V.O.

Funding

This work was supported by a grant from the Copenhagen University Hospital—Bispebjerg and Frederiksberg internal funds.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: jnr70146‐sup‐0001‐DataS1.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.