Sage Journals: Discover world-class research

Abstract

Background

Dynamic contrast-enhanced magnetic resonance imaging is a promising biomarker allowing for in vivo quantification of blood–brain barrier permeability.

Objectives

To explore the relationship between blood–brain barrier permeability, optic neuritis disease severity, and multiple sclerosis conversion in optic neuritis.

Methods

Gjedde-Patlak models from dynamic contrast-enhanced magnetic resonance imaging were used to estimate blood–brain barrier permeability (K_i) in 78 optic neuritis patients. The 2017 McDonald criteria were used to diagnose multiple sclerosis with a minimum follow-up time of 2 years.

Results

Normal-appearing white matter K_i correlated with the number of magnetic resonance imaging criteria for dissemination in space (Spearman's ρ = 0.3, p = 0.0074), but not with visual acuity, color vision, and inter-eye difference in retinal nerve fiber layer thickness. Normal-appearing white matter K_i did not differ between patients with and without oligoclonal bands (p = 0.067), but patients with brain contrast-enhancing lesions had higher normal-appearing white matter K_i than those without (p = 0.04). Early multiple sclerosis-converters diagnosed at optic neuritis onset (n = 36) had higher normal-appearing white matter K_i than non-converters (n = 29) (p = 0.01), but this was not the case for late multiple sclerosis-converters (n = 13) (p = 0.57). Normal-appearing white matter K_i did not significantly predict overall multiple sclerosis conversion (p = 0.068, AUC = 0.652).

Conclusions

Normal-appearing white matter K_i was associated with magnetic resonance imaging biomarkers of multiple sclerosis, but not with biomarkers of optic neuritis disease severity. Normal-appearing white matter K_i was increased at, but not before, the multiple sclerosis diagnosis.

Keywords

Dynamic contrast-enhanced MRI blood–brain barrier multiple sclerosis optic neuritis neuroinflammation biomarkers

Introduction

Increased blood–brain barrier (BBB) permeability is a hallmark of active lesions in multiple sclerosis (MS). Using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) we have previously shown that subtle increases in BBB permeability are also present in the normal-appearing white matter (NAWM) of MS patients when compared to healthy controls (HCs).¹ Although applications of DCE-MRI are evolving,² data on NAWM BBB permeability quantification in MS are still scarce^1,3–11 and our understanding of the factors influencing it is still limited.

Optic neuritis (ON) is a common form of clinically isolated syndrome^12,13 and the 15-year risk of MS following ON has been estimated to be 50%.¹⁴ In a previous study, NAWM BBB permeability predicted the conversion of ON to MS according to the 2010 McDonald criteria.^4,15 Furthermore, measurements of BBB permeability 6 months after treatment initiation have been found to predict two-year treatment responses to natalizumab or fingolimod⁹ as well as alemtuzumab,⁶ highlighting DCE-MRI-based measurements of BBB permeability as a promising biomarker of neuroinflammation.

In this study, to gain a better understanding of the factors associated with increased BBB permeability during ON, we investigated the relationship between BBB permeability and biomarkers of ON disease severity. Furthermore, we assessed the diagnostic performance of BBB permeability measurements, when MS was diagnosed according to the 2017 McDonald criteria.¹⁶

Materials and methods

Study participants

This study presents baseline data from an open-label, non-randomized single-center clinical trial (ClinicalTrials.gov identifier: NCT03451955) examining the effects of a gluten-free diet (GFD) in patients with ON. Patients in the intervention group adhered to a strict GFD for 6 months, whereas patients in the control group maintained their habitual diet. Patients were recruited from the Optic Neuritis Clinic at Copenhagen University Hospital Rigshospitalet-Glostrup between January 2018 and September 2021. Inclusion and exclusion criteria ensured the recruitment of magnetic resonance imaging (MRI) compatible patients with symptomatic ON and no prior MS diagnosis within the age range of 18–59 years and in the absence of conflicting disorders, pregnancy, or lactation. Meningioma was the only conflicting disorder leading to patient exclusion. Causes of ON-resembling symptoms in excluded patients where the diagnosis of ON could not be verified were headache with presbyopia, dry eyes, anterior ischemic optic neuropathy, and nonarteritic anterior ischemic optic neuropathy. Patients with identified causes of ON that were not MS (i.e. sarcoidosis, myelin oligodendrocyte glycoprotein, or aquaporin-4 seropositivity) were excluded. Data from 78 patients with verified ON that had DCE-MRI performed as close as possible to ON onset were included.

The diagnosis of ON was verified by a professor of neurology with a specialty in ON and MS. The 2017 McDonald criteria¹⁶ were used to diagnose MS with a follow-up time ranging from 24 to 69 months following ON onset. All patients were offered baseline DCE-MRI and lumbar puncture (LBP) as close as possible to ON onset as well as follow-up MRI and LBP (same protocols as baseline) 6 months after baseline DCE-MRI. Results from follow-up MRI and LBP are not presented but were incorporated in the evaluation of future fulfillment of the 2017 McDonald criteria for MS. Furthermore, ON patients who did not receive the diagnosis of MS at baseline were offered yearly MRIs for the first 2–3 years after ON. Patients were classified as “non-converters” if they did not fulfill the 2017 McDonald criteria for MS during the study's follow-up period, as “early MS-converters” if they fulfilled the criteria at ON onset, and as “late MS-converters” if they did not meet the criteria at ON onset but did so later during the follow-up period.

DCE-MRI

Participants were scanned on a 3 T Philips Achieva dStream MRI scanner (Philips Medical Systems) using a 32-element phased-array head coil. The contrast agent (gadobutrol 1 mmol/mL) was injected automatically (Spectris) as two boluses of each 0.045 mmol/kg bodyweight. The methodologies applied to acquire FLAIR, T1, and T2 sequences are provided in Supplemental Material MRI methods. Radiological descriptions of MRI scans with regards to quantification of brain white matter lesions (0–1, 2–8 and ≥ 9), MRI criteria for dissemination in space (DIS in four regions: spinal cord as well as periventricular, cortical or juxtacortical, and infratentorial regions of the brain) and evaluation of contrast-enhancement (CE) in the brain were performed in accordance with the current guidelines¹⁶ by an experienced specialist of MRI in MS. Brain CE-lesions refer to CE-lesions in tentorial and infratentorial brain; excluding enhancement of the optic nerve. BBB permeability was assessed using DCE-MRI.⁷ Methods applied for the acquisition of dynamic sequences in this article have previously been described by Cramer et al.¹⁷

Input function

An arterial input function was extracted as the time-concentration curve from intravascular voxels in the internal carotid artery. Additionally, the signal in the sagittal sinus was extracted and used to correct for partial volume errors and in-flow effects in the input function as described in Cramer et al.¹⁷

Regions of interest and permeability estimation

Five axial T₂-weighted MRI images (acquired as described by Cramer et al.)¹⁷ with similar placement of imaging planes as the DCE-MRI sequence but better in-plane resolution were obtained to allow for drawing of regions of interest (ROIs) in the NAWM and thalamus. Two ROIs were manually drawn in the periventricular white matter (one with frontal and one with posterior placement) of each hemisphere (four in total). Care was taken to avoid lesions and areas in their immediate proximity. Please see Cramer et al.¹⁷ and Supplemental Figure 1 for placements as well as corresponding K_i maps. Additionally, two ROIs were drawn covering both thalami to get an estimate of BBB permeability in a region consisting predominantly of gray matter. ROI placement was performed by three experienced operators blinded to the clinical data. The following equation which is valid for a sequence with centric phase encoding was used to convert the DCE time series from signal to contrast agent concentration:

S (t) = S_{0} \cdot \sin (α) (1 - e^{- TD (\frac{1}{T_{1}} + r_{1} \cdot C_{C A} (t))})

where α is the flip angle, TD is the saturation time delay, and r₁ is the vendor-specific contrast agent relaxivity (4 s⁻¹ mM⁻¹ for gadobutrol). T₁ and S₀ were determined from the multiple saturation delay data prior to the introduction of the contrast agent. Modeling of pharmacokinetic values was based on mean signal-time tissue curves obtained from all voxels in each ROI. Each tissue type was represented by the mean value of all ROIs from that tissue type.^1,4

The influx constant K_i—a measure of BBB permeability—was calculated using MATLAB-based (Mathworks) software developed in our unit. K_i values presented in this article were calculated using the Patlak model as described by Cramer et al.¹⁷

Lesion load estimation

Total lesion volume (mm³) was estimated by an in-house developed segmentation model as described by Hindsholm et al.¹⁸ The segmentation model used the FLAIR, T2, and T1 (pre-contrast) scans (see Supplemental Material MRI methods for sequence parameters) for inferring a lesion mask, where the total lesion volume was then calculated.

Visual tests and optical coherence tomography

Visual acuity was evaluated using retroilluminated ETDRS (Early Treatment of Diabetic Retinopathy Study) charts (Precision Vision). Test of color vision using Velhagen pseudoisochromatic plates was performed in a daylight cabinet (Macbeth, The Judge, v.2). Optical coherence tomography (OCT) was performed by trained personnel using Cirrus 4000 HD-OCT (Carl Zeiss). Data on visual acuity, color vision, and retinal nerve fiber layer (RNFL) thickness of the affected eye were excluded from two patients with previous ON affecting the same eye. Data on inter-eye differences in RNFL were excluded from four additional patients with previous ON affecting the other eye and one patient with bilateral ON.

Fluid biomarker analyses

All patients were offered an LBP with routine analyses of cerebrospinal fluid (CSF) leukocyte and erythrocyte concentrations, IgG-index, Qalb (CSF to plasma ratio of albumin), and oligoclonal bands (OCB) evaluated by isoelectric focusing and immunoblotting. One patient refused to have an LBP, as OCB had previously been demonstrated. This patient was considered OCB positive, but no other data were carried forward. Furthermore, LBP was not performed in one patient (late MS-converter) due to an Arnold Chiari malformation. Data on CSF leukocytes, IgG-index, and Qalb were excluded from two patients due to erythrocyte contamination (> 400 µL⁻¹). In the case of CSF leukocytes and CSF IgG below the limits of quantification (CSF leukocytes < 3 million/L: n = 22, CSF IgG < 10 mg/L: n = 3), CSF leukocytes and CSF IgG were sat to 2 million/L and 8 mg/L, respectively.

Statistics

Normality was assessed using the Shapiro-Wilk normality test. None of the investigated parameters were normally distributed. The Mann–Whitney U-test was used to test for differences between numeric, non-normally distributed variables. Pearson's chi-squared test was used to test for differences between categorical variables. In the case of categorical variables with categories with less than five observations, Fisher's exact test was used. Associations between BBB permeability and numerical biomarkers were explored using Spearman's correlation analyses. For linear regression models, a pseudocount of 0.0175 was added to all NAWM K_i values before transformation with the natural logarithm. This was done to meet the assumptions of homoscedasticity and normality of the residuals in the linear regression model. The pseudocount of 0.0175 was selected as it was the smallest pseudocount that would ensure no zero values while still placing all NAWM K_i values on the positive scale. The respective pseudocount for thalamic K_i was 0.0709. The same pseudocounts were also applied in logistic regression analyses, before transformation using the logarithm with base two. The validity of the logistic regression model was verified by a professional statistician from the Biostatistical Department at the University of Copenhagen. All statistical analyses were performed using R¹⁹ Versions 4.3.1 and 4.3.2 and packages readxl,²⁰ dplyr,²¹ ggpubr,²² table1,²³ ggplot2,²⁴ viridis,²⁵ and pROC.²⁶

Results

Patient characteristics

The baseline characteristics of the 78 patients with verified ON are presented in Table 1. None of the patients received steroid therapy or disease-modifying treatments prior to baseline DCE-MRI or LBP. Ten patients (12.8%) had history of past relapses, but none of them had previously fulfilled the 2017 McDonald criteria for MS. As there was no significant difference in NAWM K_i between ON patients with and without history of past relapses (0.0342 [0.0278, 0.0399] vs. 0.0239 [0.0154, 0.0405], p = 0.20, Mann–Whitney U-test), all patients were included in the further analyses. Out of 49 MS-converters (62.8%), 73.5% fulfilled the 2017 McDonald criteria for relapsing-remitting MS already at ON onset (early MS-converters), whereas the remaining 26.5% fulfilled them within a median of 427 days (min = 46, first quartile = 267, third quartile = 670, max = 1477 days) (late MS-converters). At baseline, 92.3% of 13 late MS-converters did not demonstrate DIS, whereas OCB were present in 76.9% of late MS-converters. Only one ON patient (late MS-converter) presented with DIS without DIT or OCB. Following the ON that led to study inclusion, four patients were treated for CIS; three with dimethyl fumarate and one with peginterferon beta-1a (three converted to MS within the follow-up time).

Table 1.

Demographic and clinical characteristics of ON cohort.

	All ON	MS-converters	Non-converters	p-value^a
	(n = 78)	(n = 49)	(n = 29)	p-value^a
Sex, female	53 (67.9%)	33 (67.3%)	20 (69%)	0.88
Age (years)	32 [26, 40]	31 [26, 42]	33 [27, 36]	0.98
History of past neurological symptoms	10 (12.8%)	7 (14.3%)	3 (10.3%)	0.74
Visual acuity of affected eye	0.50 [0.031, 0.80] ⁿ ⁼ ⁷⁶	0.36 [0, 0.80] ⁿ ⁼ ⁴⁸	0.63 [0.19, 1.00] ⁿ ⁼ ²⁸	0.043
Velhagen pseudoisochromatic test of affected eye (error scores)	10 [4, 21] ⁿ ⁼ ⁷⁶	15 [4.75, 21] ⁿ ⁼ ⁴⁸	6.5 [3.75, 12] ⁿ ⁼ ²⁸	0.012
RNFL thickness of affected eye (µm)	107 [94.5, 129] ⁿ ⁼ ⁷⁶	107 [96, 123] ⁿ ⁼ ⁴⁸	105 [83.8, 132] ⁿ ⁼ ²⁸	0.49
Inter-eye difference in RNFL thickness^b	10 [1, 29] ⁿ ⁼ ⁷¹	11 [3, 31] ⁿ ⁼ ⁴⁵	5.5 [–5, 26.8] ⁿ ⁼ ²⁶	0.34
Days from ON onset to DCE-MRI	20 [15, 33]	20 [16, 33]	21 [14, 34]	0.89
Brain white matter lesions (n)				4.3 × 10⁻⁶
0–1	33 (42.3%)	11 (22.4%)	22 (75.9%)
2–8	25 (32.1%)	19 (38.8%)	6 (20.7%)
≥ 9	20 (25.6%)	19 (38.8%)	1 (3.4%)
Total lesion load (mm³)	704 [359, 1290]	1120 [710, 2220]	304 [211,506]	1.8 × 10⁻¹¹
Brain contrast-enhancing lesions, present	13 (16.7%)	13 (26.5%)	0 (0%)	0.0014
Days between DCE-MRI and LBP	7 [1, 11]	6 [1, 8]	7 [3, 11]	0.89
Oligoclonal bands, present	56 (73.7%) ⁿ ⁼ ⁷⁶	45 (95.8%) ⁿ ⁼ ⁴⁷	11 (37.9%)	9 × 10⁻⁸
CSF leukocytes (× 10⁶/L)	5 [2, 8.25] ⁿ ⁼ ⁷²	6 [3, 10.5] ⁿ ⁼ ⁴³	3 [2, 6]	0.020
IgG-index	0.59 [0.47, 0.79] ⁿ ⁼ ⁷²	0.68 [0.55, 0.91] ⁿ ⁼ ⁴³	0.48 [0.44, 0.54]	9 × 10⁻⁵
Qalb (× 10⁻³)	4.4 [3.3, 5.4] ⁿ ⁼ ⁷²	4.5 [3.2, 5.3] ⁿ ⁼ ⁴³	4.2 [3.3, 5.6]	0.90

Abbreviations: BBB: blood–brain barrier; CSF: cerebrospinal fluid; DCE-MRI: dynamic contrast-enhanced magnetic resonance imaging; LBP: lumbar puncture; MS: multiple sclerosis; NAWM: normal-appearing white matter; ON: optic neuritis; Qalb: CSF to plasma ratio of albumin; RNFL: retinal nerve fiber layer. The laboratory's upper cut-off values were: CSF leukocytes ≥ 5 (× 10⁶/L), IgG-index > 0.67 and Qalb ≥ 14 (× 10⁻³). Numerical parameters are not normally distributed and are presented as median [first quartile–third quartile]. Categorical variables are described as numbers (percentages). Data from eyes previously affected by ON have been excluded from all visual and OCT measures.

For the difference between MS-converters and non-converters.

Inter-eye difference was calculated as RNFL thickness of the affected eye minus RNFL thickness of the non-affected eye.

BBB permeability and disease biomarkers in ON

When examining the relationship between BBB permeability and ON severity, we found no correlations between NAWM K_i and visual acuity of the affected eye (Spearman's ρ = –0.0055, p = 0.96, Supplemental Figure 2A), color vision of the affected eye (Spearman's ρ = 0.085, p = 0.47, Supplemental Figure 2B), or inter-eye difference in RNFL thickness (Spearman's ρ = 0.04, p = 0.74, Supplemental Figure 2C). Furthermore, NAWM K_i was not significantly related to the presence of OCB (p = 0.067, Figure 1A), nor did it correlate with IgG-index (Spearman's ρ = 0.15, p = 0.21, Supplemental Figure 2D) or CSF leukocytes (Spearman's ρ = –0.00049, p = 1, Supplemental Figure 2E). Interestingly, although Qalb is commonly used as a biomarker of BBB permeability, it did not correlate with NAWM Ki (Spearman's ρ = –0.028, p = 0.82, Supplemental Figure 2F). When examining the relationship between BBB permeability and other MRI biomarkers, we observed higher NAWM K_i in ON patients with brain CE lesions compared to those without (p = 0.04, Figure 1B). Furthermore, we found a positive correlation between NAWM K_i and the number of fulfilled MRI criteria for DIS (Spearman's ρ = 0.3, p = 0.0074, Figure 1C). To expand on this observed relationship with DIS, we estimated total lesion volumes, which revealed a similar positive correlation with NAWM K_i (Spearman's ρ = 0.26, p = 0.022). Respective analyses for thalamic K_i are presented in Supplemental Figure 3.

Figure 1.

Relationship between blood–brain barrier permeability in NAWM expressed as the influx constant K_i from Gjedde-Patlak models obtained using dynamic contrast-enhanced MRI and the presence of oligoclonal bands (A), the presence of CE lesions in the brain (B) and the number of MRI criteria for DIS (C) in patients with optic neuritis. Early MS-converters fulfilled the 2017 McDonald criteria of MS already at ON onset. Late MS-converters did not fulfill the 2017 McDonald criteria of MS at ON onset but did so later during the follow-up time of the study (minimum 2 years). Non-converters are patients with idiopathic ON. Abbreviations: NAWM: normal-appearing white matter; MRI: magnetic resonance imaging; CE: contrast-enhancing; DIS: dissemination in space; MS: multiple sclerosis; ON: optic neuritis.

BBB permeability and diagnosis of MS

MS-converters had higher NAWM K_i compared to non-converters (0.0294 [0.0204, 0.0428] vs. 0.0185 [0.0140, 0.0330], p = 0.026), mainly driven by a difference between early MS-converters and non-converters (0.0300 [0.0220, 0.0502] vs. 0.0185 [0.0140, 0.0330], p = 0.01, Figure 2A). When investigating thalamic K_i, there was no difference between MS-converters and non-converters (0.0278 [0.0181, 0.0480] vs. 0.0320 [0.0157, 0.0397], p = 0.62), but early MS-converters had higher thalamic K_i than late MS-converters (0.0316 [0.0220, 0.0531] vs. 0.0223 [0.0049, 0.0321], p = 0.043, Figure 2B). DCE-MRI scans were performed at a median of 20 days from ON onset (min = 8, Q1 = 15, Q3 = 33, max = 94). Neither NAWM K_i nor thalamic K_i correlated with time from ON onset to DCE-MRI (Supplemental Figure 4). Likewise, we observed no significant correlation between NAWM K_i and age (Spearman's ρ = –0.18, p = 0.11). In a univariate linear regression model with log-transformed NAWM K_i as the dependent variable and timing of MS diagnosis as an independent variable, early MS-converters had 60.8% (6.2%–143.5%, p = 0.026) higher NAWM K_i compared to non-converters (adjusted R² = 0.04). The coefficient for late MS-converters did not reach significance (p = 0.35). In univariate logistic regression analyses, neither NAWM K_i (AUC = 0.652, p = 0.068) nor thalamic K_i (AUC = 0.466, p = 0.284) were significant predictors of the risk of MS (Figure 3).

Figure 2.

Blood–brain barrier permeability expressed as the influx constant K_i from Gjedde-Patlak models obtained using dynamic contrast-enhanced MRI in NAWM (A) and thalamus (B) according to the timing of MS diagnosis. “Early MS” denotes early MS-converters who fulfilled the 2017 McDonald criteria of MS already at ON onset. “Late MS” denotes late MS-converters who did not fulfill the 2017 McDonald criteria of MS at ON onset but did so later during the follow-up time of the study (minimum 2 years). “Not MS” denotes non-converters with idiopathic ON. Mann–Whitney U-test was used to test for differences between groups. Abbreviations: MRI: magnetic resonance imaging; NAWM: normal-appearing white matter; MS: multiple sclerosis; ON: optic neuritis.

Figure 3.

Receiver operating characteristic curves of univariate logistic regression models with diagnosis of multiple sclerosis within the follow-up time of the study as dependent variable and K_i in NAWM or thalamus as independent variables. K_i values were log2-transformed with the use of a pseudocount (0.0175 for NAWM and 0.0709 for thalamus) prior to logistic regression. Abbreviations: AUC: area under the curve; NAWM: normal-appearing white matter.

Discussion

NAWM BBB permeability was increased in early MS-converters compared to non-converters (p = 0.01). On the contrary, late MS-converters did not have significantly different NAWM BBB permeability from non-converters (p = 0.57). We observed that only patients fulfilling three (p = 0.018) or four (p = 0.046) MRI criteria for DIS had increased NAWM BBB permeability compared to patients with zero MRI criteria for DIS and found a correlation between NAWM BBB permeability and the number of fulfilled MRI criteria for DIS (Spearman's ρ = 0.3, p = 0.0074). Similarly, NAWM BBB permeability correlated with total lesion load (Spearman's ρ = 0.26, p = 0.022) and was also higher in ON patients with brain CE lesions compared to ON patients without brain CE lesions (p = 0.04). The above confirms that NAWM BBB permeability is increased already early in the disease course of MS and suggests that NAWM BBB alterations are associated with MS-related brain pathology.

As the BBB may even respond to systemic inflammation,²⁷ it could be hypothesized that ON-related neuroinflammation would be adequate to alter NAWM BBB permeability. We did not observe any relationships between BBB permeability and OCB, IgG-index, or CSF leukocytes, although CSF leukocytes were previously found to correlate with NAWM BBB permeability.⁴ Similarly, we found no correlations between BBB permeability and visual acuity, color vision, or inter-eye differences in RNFL thickness suggesting that BBB permeability is not related to the degree of optic nerve pathology.

Although Qalb is considered a biomarker of BBB permeability, we and others have found Qalb and DCE-MRI-based measures of BBB permeability to be dissociated.^4,5,28 K_i is believed to reflect BBB changes in capillaries or postcapillary venules, however, during subtle, non-disruptive BBB perturbances as in ON, the choroid plexus is expected to be the main route of albumin entry to the CSF (albumin is ∼100 times larger than gadolinium-DOTA).²⁹

Contrary to our previous publication⁴ where NAWM BBB permeability predicted MS conversion according to the 2010 McDonald criteria,¹⁵ NAWM BBB permeability did not reach statistical significance as an independent predictor of MS conversion according to the 2017 McDonald criteria (AUC = 0.652, p = 0.068). The 2017 McDonald criteria have higher sensitivity and lower specificity for clinically definite MS resulting in faster diagnosis in a higher number of patients.^30,31 Therefore, it could be hypothesized that the lower predictive power of K_i in this study⁴ could be due to the classification of more subjects with a less active disease course, consequently exhibiting lower degrees of neuroinflammation, brain pathology, and presumably lower K_i, as MS-converters when using the updated diagnostic criteria. The present study did indeed show a higher two-year MS-conversion rate (62.8% vs. 44.7%).⁴ Implementation of routine testing for MOG antibodies has also allowed for improved differential diagnosis since the recruitment of our past ON cohort.⁴ Lastly, we cannot rule out potential selection bias related to patients' willingness to participate in a time-consuming trial or potential influences of the GFD on the risk of MS.

In summary, this is one of the largest datasets with K_i to be analyzed to date. Our data support the presence of NAWM BBB alterations at the time of MS diagnosis and enhance our understanding of the relationship between BBB permeability and neuropathological processes during ON.

Supplemental Material

sj-pdf-1-mso-10.1177_20552173251346979 - Supplemental material for Blood–brain barrier permeability in relation to disease severity and timing of multiple sclerosis diagnosis in optic neuritis

Supplemental material, sj-pdf-1-mso-10.1177_20552173251346979 for Blood–brain barrier permeability in relation to disease severity and timing of multiple sclerosis diagnosis in optic neuritis by Moschoula Passali, Maria Højberg Knudsen, Knud Josefsen, Julie Christine Antvorskov, Amalie Monberg Hindsholm, Ulrich Lindberg, Jette Lautrup Frederiksen, Henrik Bo Wiberg Larsson and Stig Præstekjær Cramer in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-docx-2-mso-10.1177_20552173251346979 - Supplemental material for Blood–brain barrier permeability in relation to disease severity and timing of multiple sclerosis diagnosis in optic neuritis

Supplemental material, sj-docx-2-mso-10.1177_20552173251346979 for Blood–brain barrier permeability in relation to disease severity and timing of multiple sclerosis diagnosis in optic neuritis by Moschoula Passali, Maria Højberg Knudsen, Knud Josefsen, Julie Christine Antvorskov, Amalie Monberg Hindsholm, Ulrich Lindberg, Jette Lautrup Frederiksen, Henrik Bo Wiberg Larsson and Stig Præstekjær Cramer in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Supplemental Material

sj-docx-3-mso-10.1177_20552173251346979 - Supplemental material for Blood–brain barrier permeability in relation to disease severity and timing of multiple sclerosis diagnosis in optic neuritis

Supplemental material, sj-docx-3-mso-10.1177_20552173251346979 for Blood–brain barrier permeability in relation to disease severity and timing of multiple sclerosis diagnosis in optic neuritis by Moschoula Passali, Maria Højberg Knudsen, Knud Josefsen, Julie Christine Antvorskov, Amalie Monberg Hindsholm, Ulrich Lindberg, Jette Lautrup Frederiksen, Henrik Bo Wiberg Larsson and Stig Præstekjær Cramer in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Footnotes

Acknowledgements

We thank Ms. Helle J. Simonsen for her assistance with acquiring and analyzing MRI data. Furthermore, we acknowledge MD Josefine Britze, MD Mathias F. Schmidt, and Ms. Asma I. Ali for clinically examining the recruited patients and thank all participants as well as the Danish Multiple Sclerosis Society (DMSS) for making this study possible.

Consent to participate

Study participants received written and oral information and signed written informed consent before participation. Study consent forms were approved by Copenhagen's Regional Committee on Health Research Ethics.

Consent for publication

Not applicable.

Data availability statement

The data can be provided by the corresponding author in an anonymized form upon reasonable request from any qualified investigator.

Declaration of conflicting interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: M Passali received funding from the DMSS. MH Knudsen received funding from Sanofi Genzyme and the DMSS and non-financial support from Merck. K Josefsen, JC Antvorskov, AM Hindsholm, and U Lindberg report no disclosures. J Frederiksen served on scientific advisory boards for and received funding for travel related to these activities as well as honoraria from Merck Serono, Sanofi-Aventis, Roche, Novartis, and Chiesi. HBW Larsson received funding from the DMSS. SP Cramer received funding from Sanofi Genzyme and the DMSS. This study has been funded by the DMSS (grant numbers A35733, A38270, and A42545).

Ethical approval

All procedures were conducted in accordance with the 1975 Declaration of Helsinki. Patients included in this study participated in our GFD clinical trial and our ON biomarker study. Both protocols were approved by Copenhagen's Regional Committee on Health Research Ethics (H-17019986 and H-4-2014-095).

ORCID iDs

Moschoula Passali

Maria Højberg Knudsen

Amalie Monberg Hindsholm

Ulrich Lindberg

Jette Lautrup Frederiksen

Henrik Bo Wiberg

Stig Præstekjær Cramer

Supplemental material

Supplemental material for this article is available online.

References

Cramer

Simonsen

Frederiksen

, et al. Abnormal blood-brain barrier permeability in normal appearing white matter in multiple sclerosis investigated by MRI. Neuroimage Clin 2014; 4: 182–189.

Heye

Culling

Valdés Hernández

MDC

, et al. Assessment of blood-brain barrier disruption using dynamic contrast-enhanced MRI. A systematic review. Neuroimage Clin 2014; 6: 262–274.

Ingrisch

Sourbron

Morhard

, et al. Quantification of perfusion and permeability in multiple sclerosis: dynamic contrast-enhanced MRI in 3D at 3 T. Invest Radiol 2012; 47: 252–258.

Cramer

Modvig

Simonsen

, et al. Permeability of the blood-brain barrier predicts conversion from optic neuritis to multiple sclerosis. Brain 2015; 138: 2571–2583.

Taheri

Gasparovic

Shah

, et al. Erratum: quantitative measurement of blood-brain barrier permeability in human using dynamic contrast-enhanced MRI with fast T1 mapping. Magn Reson Med 2011; 65: 1036–1042.

Knudsen

Lindberg

Frederiksen

, et al. Blood-brain barrier permeability changes in the first year after alemtuzumab treatment predict 2-year outcomes in relapsing-remitting multiple sclerosis. Mult Scler Relat Disord 2022; 63: 103891.

Cramer

Larsson

HBW

. Accurate determination of blood-brain barrier permeability using dynamic contrast-enhanced T1-weighted MRI: a simulation and in vivo study on healthy subjects and multiple sclerosis patients. J Cereb Blood Flow Metab 2014; 34: 1655–1665.

Varatharaj

Liljeroth

Darekar

, et al. Blood–brain barrier permeability measured using dynamic contrast-enhanced magnetic resonance imaging: a validation study. J Physiol 2019; 597: 699–709.

Cramer

Simonsen

Varatharaj

, et al. Permeability of the blood–brain barrier predicts no evidence of disease activity at 2 years after natalizumab or fingolimod treatment in relapsing–remitting multiple sclerosis. Ann Neurol 2018; 83: 902–914.

10.

Taheri

Rosenberg

Ford

. Quantification of blood-to-brain transfer rate in multiple sclerosis. Mult Scler Relat Disord 2013; 2: 124–132.

11.

Cramer

Hamrouni

Simonsen

, et al. Insights from DCE-MRI: blood–brain barrier permeability in the context of MS relapses and methylprednisolone treatment. Front Neurosci 2025; 19: 1546236.

12.

Sørensen

Frederiksen

Brønnum-Hansen

, et al. Optic neuritis as onset manifestation of multiple sclerosis. Neurology 1999; 53: 473–478.

13.

Petzold

Fraser

Abegg

, et al. Diagnosis and classification of optic neuritis. Lancet Neurol 2022; 21: 1120–1134.

14.

Levin

Lessell

. Risk of multiple sclerosis after optic neuritis. JAMA 2003; 290: 403–404.

15.

Polman

Reingold

Banwell

, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011; 69: 292–302.

16.

Thompson

Banwell

Barkhof

, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 2018; 17: 162–173.

17.

Cramer

Larsson

HBW

Knudsen

, et al. Reproducibility and optimal arterial input function selection in dynamic contrast-enhanced perfusion MRI in the healthy brain. J Magn Reson Imaging 2023; 57: 1229–1240.

18.

Hindsholm

Andersen

Cramer

, et al. Scanner agnostic large-scale evaluation of MS lesion delineation tool for clinical MRI. Front Neurosci 2023; 17: 1177540.

19.

R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing 2021. https://www.R-project.org/

20.

Wickham

Bryan

. readxl: read excel files. R Package Version 1.4.3 2023.

21.

Wickham

François

Henry

, et al. dplyr: a grammar of data manipulation. 2023.

22.

Kassambara

. ggpubr: ‘ggplot2’ based publication ready plots. 2023.

23.

Rich

. table1: tables of descriptive statistics in HTML. 2023.

24.

Wickham

. ggplot2: elegant graphics for data analysis. 2016.

25.

Garnier

Ross

Rudis

, et al.

viridis(Lite) - Colorblind-Friendly Color Maps for R.

viridis Package Version 0.6.5 2024. doi:https://doi.org/10.5281/zenodo.4679423

26.

Robin

Turck

Hainard

, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 2011; 12: 77.

27.

Galea

. The blood–brain barrier in systemic infection and inflammation. Cellul Molecul Immunol 2021; 18: 2489–2501.

28.

Hillmer

Erhardt

Caprihan

, et al. Blood-brain barrier disruption measured by albumin index correlates with inflammatory fluid biomarkers. J Cereb Blood Flow Metab 2023; 43: 712–721.

29.

Cousins

Hodges

Schubert

, et al. The blood–CSF–brain route of neurological disease: the indirect pathway into the brain. Neuropathol Appl Neurobiol 2022; 48: e12789.

30.

Filippi

Preziosa

Meani

, et al. Performance of the 2017 and 2010 revised McDonald criteria in predicting MS diagnosis after a clinically isolated syndrome: a MAGNIMS study. Neurology 2022; 98: e1–e14.

31.

van der Vuurst de Vries

,. et al. Application of the 2017 revised McDonald criteria for multiple sclerosis to patients with a typical clinically isolated syndrome. JAMA Neurol 2018;75:1392–1398.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.65 MB