Grey Matter Atrophy and its Relationship with White Matter Lesions in Patients with Myelin Oligodendrocyte Glycoprotein Antibody-associated Disease, Aquaporin-4 Antibody-Positive Neuromyelitis Optica Spectrum Disorder, and Multiple Sclerosis.

OBJECTIVE: To evaluate: (1) the distribution of gray matter (GM) atrophy in myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD), aquaporin-4 antibody-positive neuromyelitis optica spectrum disorder (AQP4+NMOSD), and relapsing-remitting multiple sclerosis (RRMS); and (2) the relationship between GM volumes and white matter lesions in various brain regions within each disease. METHODS: A retrospective, multicenter analysis of magnetic resonance imaging data included patients with MOGAD/AQP4+NMOSD/RRMS in non-acute disease stage. Voxel-wise analyses and general linear models were used to evaluate the relevance of regional GM atrophy. For significant results (p < 0.05), volumes of atrophic areas are reported. RESULTS: We studied 135 MOGAD patients, 135 AQP4+NMOSD, 175 RRMS, and 144 healthy controls (HC). Compared with HC, MOGAD showed lower GM volumes in the temporal lobes, deep GM, insula, and cingulate cortex (75.79 cm3); AQP4+NMOSD in the occipital cortex (32.83 cm3); and RRMS diffusely in the GM (260.61 cm3). MOGAD showed more pronounced temporal cortex atrophy than RRMS (6.71 cm3), whereas AQP4+NMOSD displayed greater occipital cortex atrophy than RRMS (19.82 cm3). RRMS demonstrated more pronounced deep GM atrophy in comparison with MOGAD (27.90 cm3) and AQP4+NMOSD (47.04 cm3). In MOGAD, higher periventricular and cortical/juxtacortical lesions were linked to reduced temporal cortex, deep GM, and insula volumes. In RRMS, the diffuse GM atrophy was associated with lesions in all locations. AQP4+NMOSD showed no lesion/GM volume correlation. INTERPRETATION: GM atrophy is more widespread in RRMS compared with the other two conditions. MOGAD primarily affects the temporal cortex, whereas AQP4+NMOSD mainly involves the occipital cortex. In MOGAD and RRMS, lesion-related tract degeneration is associated with atrophy, but this link is absent in AQP4+NMOSD. ANN NEUROL 2024.

The diagnostic workup of children with the radiologically isolated syndrome differs by age and by sex.

BACKGROUND: Cerebrospinal fluid (CSF) and spinal MRIs are often obtained in children with the radiologically isolated syndrome (RIS) for diagnosis and prognosis. Factors affecting the frequency and timing of these tests are unknown. OBJECTIVE: To determine whether age or sex were associated with (1) having CSF or spinal MRI obtained or (2) the timing of these tests. METHODS: We analyzed children (≤ 18 y) with RIS enrolled in an international longitudinal study. Index scans met 2010/2017 multiple sclerosis (MS) MRI criteria for dissemination in space (DIS). We used Fisher's exact test and multivariable logistic regression (covariates = age, sex, MRI date, MRI indication, 2005 MRI DIS criteria met, and race). RESULTS: We included 103 children with RIS (67% girls, median age = 14.9 y). Children ≥ 12 y were more likely than children < 12 y to have CSF obtained (58% vs. 21%, adjusted odds ratio [AOR] = 4.9, p = 0.03). Pre-2017, girls were more likely than boys to have CSF obtained (n = 70, 79% vs. 52%, AOR = 4.6, p = 0.01), but not more recently (n = 30, 75% vs. 80%, AOR = 0.2, p = 0.1; p = 0.004 for interaction). Spinal MRIs were obtained sooner in children ≥ 12 y (median 11d vs. 159d, p = 0.03). CONCLUSIONS: Younger children with RIS may be at continued risk for misdiagnosis and misclassification of MS risk. Consensus guidelines are needed.

Cancer and multiple sclerosis: 2023 recommendations from the French Multiple Sclerosis Society.

BACKGROUND: Epidemiological data reveal that 45% of persons with multiple sclerosis (PwMS) in France are more than 50 years. This population more than 50 is more susceptible to cancer, and this risk may be increased by frequent use of immunosuppressive drugs. Consequently, concerns have arisen about the potential increased risk of cancer in PwMS and how patients should be screened and managed in terms of cancer risk. OBJECTIVE: To develop evidence-based recommendations to manage the coexistence of cancer and multiple sclerosis (MS). METHODS: The French Group for Recommendations in MS collected articles from PubMed and university databases covering the period January 1975 through June 2022. The RAND/UCLA method was employed to achieve formal consensus. MS experts comprehensively reviewed the full-text articles and developed the initial recommendations. A group of multidisciplinary health care specialists then validated the final proposal. RESULTS: Five key questions were addressed, encompassing various topics such as cancer screening before or after initiating a disease-modifying therapy (DMT), appropriate management of MS in the context of cancer, recommended follow-up for cancer in patients receiving a DMT, and the potential reintroduction of a DMT after initial cancer treatment. A strong consensus was reached for all 31 recommendations. CONCLUSION: These recommendations propose a strategic approach to managing cancer risk in PwMS.

Pseudocystic inflammatory demyelinating lesions in multiple sclerosis: A clinical, radiological, and pathological description.

BACKGROUND: Pseudocystic inflammatory demyelinating lesions (PIDLs) are poorly described in MS and might represent a diagnostic challenge. OBJECTIVES: We described the clinical, radiological, pathological, and follow-up characteristics of 13 PIDL in 9 MS patients. METHODS: We constituted a single-center retrospective case series of PIDLs in MS, defined on MRI as expansive cyst-like lesions, with a fluid-signal content, and a diameter of 1 cm or more. RESULTS: PIDL often occurred at first event (56%), were often asymptomatic (69%), and encircled by a hypo-T2 diffusion-restricted rim and a thin ring-like gadolinium enhancement (100%) on magnetic resonance imaging (MRI). Associated typical MS lesions were constant. Biopsies from two PIDLs displayed classical features of active MS, except for unusual edema. CONCLUSION: PIDLs are clinically unremarkable and associated with a good outcome. Their easily recognizable MRI features could help avoid biopsy.

Teriflunomide and Time to Clinical Multiple Sclerosis in Patients With Radiologically Isolated Syndrome: The TERIS Randomized Clinical Trial.

Importance: Radiologically isolated syndrome (RIS) represents the earliest detectable preclinical phase of multiple sclerosis (MS) punctuated by incidental magnetic resonance imaging (MRI) white matter anomalies within the central nervous system. Objective: To determine the time to onset of symptoms consistent with MS. Design, Setting, and Participants: From September 2017 to October 2022, this multicenter, double-blind, phase 3, randomized clinical trial investigated the efficacy of teriflunomide in delaying MS in individuals with RIS, with a 3-year follow-up. The setting included referral centers in France, Switzerland, and Turkey. Participants older than 18 years meeting 2009 RIS criteria were randomly assigned (1:1) to oral teriflunomide, 14 mg daily, or placebo up to week 96 or, optionally, to week 144. Interventions: Clinical, MRI, and patient-reported outcomes (PROs) were collected at baseline and yearly until week 96, with an optional third year in the allocated arm if no symptoms have occurred. Main outcomes: Primary analysis was performed in the intention-to-treat population, and safety was assessed accordingly. Secondary end points included MRI outcomes and PROs. Results: Among 124 individuals assessed for eligibility, 35 were excluded for declining to participate, not meeting inclusion criteria, or loss of follow-up. Eighty-nine participants (mean [SD] age, 37.8 [12.1] years; 63 female [70.8%]) were enrolled (placebo, 45 [50.6%]; teriflunomide, 44 [49.4%]). Eighteen participants (placebo, 9 [50.0%]; teriflunomide, 9 [50.0%]) discontinued the study, resulting in a dropout rate of 20% for adverse events (3 [16.7%]), consent withdrawal (4 [22.2%]), loss to follow-up (5 [27.8%]), voluntary withdrawal (4 [22.2%]), pregnancy (1 [5.6%]), and study termination (1 [5.6%]). The time to the first clinical event was significantly extended in the teriflunomide arm compared with placebo, in both the unadjusted (hazard ratio [HR], 0.37; 95% CI, 0.16-0.84; P = .02) and adjusted (HR, 0.28; 95% CI, 0.11-0.71; P = .007) analysis. Secondary imaging end point outcomes including the comparison of the cumulative number of new or newly enlarging T2 lesions (rate ratio [RR], 0.57; 95% CI, 0.27-1.20; P = .14), new gadolinium-enhancing lesions (RR, 0.33; 95% CI, 0.09-1.17; P = .09), and the proportion of participants with new lesions (odds ratio, 0.72; 95% CI, 0.25-2.06; P = .54) were not significant. Conclusion and Relevance: Treatment with teriflunomide resulted in an unadjusted risk reduction of 63% and an adjusted risk reduction of 72%, relative to placebo, in preventing a first clinical demyelinating event. These data suggest a benefit to early treatment in the MS disease spectrum. Trial Registration: ClinicalTrials.gov Identifier: NCT03122652.

Impact of COVID-19 vaccination or infection on disease activity in a radiologically isolated syndrome cohort: The VaxiRIS study.

BACKGROUND: Vaccination in patients with multiple sclerosis (MS) treated with immunosuppressive drugs is highly recommended. Regarding COVID-19 vaccination, no specific concern has been raised. OBJECTIVES: We aimed to evaluate if COVID-19 vaccination or infection increased the risk of disease activity, either radiological or clinical, with conversion to MS in a cohort of people with a radiologically isolated syndrome (RIS). METHODS: This multicentric observational study analyzed patients in the RIS Consortium cohort during the pandemic between January 2020 and December 2022. We compared the occurrence of disease activity in patients according to their vaccination status. The same analysis was conducted by comparing patients' history of COVID-19 infection. RESULTS: No difference was found concerning clinical conversion to MS in the vaccinated versus unvaccinated group (6.7% vs 8.5%, p > 0.9). The rate of disease activity was not statistically different (13.6% and 7.4%, respectively, p = 0.54). The clinical conversion rate to MS was not significantly different in patients with a documented COVID-19 infection versus non-infected patients. CONCLUSION: Our study suggests that COVID-19 infection or immunization in RIS individuals does not increase the risk of disease activity. Our results support that COVID-19 vaccination can be safely proposed and repeated for these subjects.

A Weakly Supervised Gradient Attribution Constraint for Interpretable Classification and Anomaly Detection.

The lack of interpretability of deep learning reduces understanding of what happens when a network does not work as expected and hinders its use in critical fields like medicine, which require transparency of decisions. For example, a healthy vs pathological classification model should rely on radiological signs and not on some training dataset biases. Several post-hoc models have been proposed to explain the decision of a trained network. However, they are very seldom used to enforce interpretability during training and none in accordance with the classification. In this paper, we propose a new weakly supervised method for both interpretable healthy vs pathological classification and anomaly detection. A new loss function is added to a standard classification model to constrain each voxel of healthy images to drive the network decision towards the healthy class according to gradient-based attributions. This constraint reveals pathological structures for patient images, allowing their unsupervised segmentation. Moreover, we advocate both theoretically and experimentally, that constrained training with the simple Gradient attribution is similar to constraints with the heavier Expected Gradient, consequently reducing the computational cost. We also propose a combination of attributions during the constrained training making the model robust to the attribution choice at inference. Our proposition was evaluated on two brain pathologies: tumors and multiple sclerosis. This new constraint provides a more relevant classification, with a more pathology-driven decision. For anomaly detection, the proposed method outperforms state-of-the-art especially on difficult multiple sclerosis lesions segmentation task with a 15 points Dice improvement.

Rituximab De-escalation in Patients With Neuromyelitis Optica Spectrum Disorder.

BACKGROUND AND OBJECTIVES: Exit strategies such as de-escalations have not been evaluated for rituximab in patients with neuromyelitis optica spectrum disorder (NMOSD). We hypothesized that they are associated with disease reactivations and aimed to estimate this risk. METHODS: We describe a case series of real-world de-escalations from the French NMOSD registry (NOMADMUS). All patients met the 2015 International Panel for NMO Diagnosis (IPND) diagnostic criteria for NMOSD. A computerized screening of the registry extracted patients with rituximab de-escalations and at least 12 months of subsequent follow-up. We searched for 7 de-escalation regimens: scheduled discontinuations or switches to an oral treatment after single infusion cycles, scheduled discontinuations or switches to an oral treatment after periodic infusions, de-escalations before pregnancies, de-escalations after tolerance issues, and increased infusion intervals. Rituximab discontinuations motivated by inefficacy or for unknown purposes were excluded. The primary outcome was the absolute risk of NMOSD reactivation (one or more relapses) at 12 months. AQP4+ and AQP4- serotypes were analyzed separately. RESULTS: We identified 137 rituximab de-escalations between 2006 and 2019 that corresponded to a predefined group: 13 discontinuations after a single infusion cycle, 6 switches to an oral treatment after a single infusion cycle, 9 discontinuations after periodic infusions, 5 switches to an oral treatment after periodic infusions, 4 de-escalations before pregnancies, 9 de-escalations after tolerance issues, and 91 increased infusion intervals. No group remained relapse-free over the whole de-escalation follow-up (mean: 3.2 years; range: 0.79-9.5), except pregnancies in AQP+ patients. In all groups combined and within 12 months, reactivations occurred after 11/119 de-escalations in patients with AQP4+ NMOSD (9.2%, 95% CI [4.7-15.9]), from 0.69 to 10.0 months, and in 5/18 de-escalations in patients with AQP4- NMOSD (27.8%, 95% CI [9.7-53.5]), from 1.1 to 9.9 months. DISCUSSION: There is a risk of NMOSD reactivation whatever the rituximab de-escalation regimen. TRIAL REGISTRATION INFORMATION: Registered on ClinicalTrials.gov: NCT02850705. CLASSIFICATION OF EVIDENCE: This study provides Class IV evidence that de-escalation of rituximab increases the probability of disease reactivation.

The radiologically isolated syndrome: revised diagnostic criteria.

The radiologically isolated syndrome (RIS) was defined in 2009 as the presence of asymptomatic, incidentally identified demyelinating-appearing white matter lesions in the CNS within individuals lacking symptoms typical of multiple sclerosis (MS). The RIS criteria have been validated and predict the transition to symptomatic MS reliably. The performance of RIS criteria that require fewer MRI lesions is unknown. 2009-RIS subjects, by definition, fulfil three to four of four criteria for 2005 dissemination in space (DIS) and subjects fulfilling only one or two lesions in at least one 2017 DIS location were identified within 37 prospective databases. Univariate and multivariate Cox regression models were used to identify predictors of a first clinical event. Performances of different groups were calculated. Seven hundred and forty-seven subjects (72.2% female, mean age 37.7 ± 12.3 years at the index MRI) were included. The mean clinical follow-up time was 46.8 ± 45.4 months. All subjects had focal T2 hyperintensities suggestive of inflammatory demyelination on MRI; 251 (33.6%) fulfilled one or two 2017 DIS criteria (designated as Groups 1 and 2, respectively), and 496 (66.4%) fulfilled three or four 2005 DIS criteria representing 2009-RIS subjects. Group 1 and 2 subjects were younger than the 2009-RIS group and were more likely to develop new T2 lesions over time (P < 0.001). Groups 1 and 2 were similar regarding survival distribution and risk factors for transition to MS. At 5 years, the cumulative probability for a clinical event was 29.0% for Groups 1 and 2 compared to 38.7% for 2009-RIS (P = 0.0241). The presence of spinal cord lesions on the index scan and CSF-restricted oligoclonal bands in Groups 1-2 increased the risk of symptomatic MS evolution at 5 years to 38%, comparable to the risk of development in the 2009-RIS group. The presence of new T2 or gadolinium-enhancing lesions on follow-up scans independently increased the risk of presenting with a clinical event (P < 0.001). The 2009-RIS subjects or Groups 1 and 2 with at least two of the risk factors for a clinical event demonstrated better sensitivity (86.0%), negative predictive value (73.1%), accuracy (59.8%) and area under the curve (60.7%) compared to other criteria studied. This large prospective cohort brings Class I evidence that subjects with fewer lesions than required in the 2009 RIS criteria evolve directly to a first clinical event at a similar rate when additional risk factors are present. Our results provide a rationale for revisions to existing RIS diagnostic criteria.

Pregnancy and multiple sclerosis: 2022 recommendations from the French multiple sclerosis society.

OBJECTIVE: The objective of this study was to develop evidence-based recommendations on pregnancy management for persons with multiple sclerosis (MS). BACKGROUND: MS typically affects young women in their childbearing years. Increasing evidence is available to inform questions raised by MS patients and health professionals about pregnancy issues. METHODS: The French Group for Recommendations in Multiple Sclerosis (France4MS) reviewed PubMed and university databases (January 1975 through June 2021). The RAND/UCLA appropriateness method was developed to synthesise the scientific literature and expert opinions on healthcare topics; it was used to reach a formal agreement. Fifty-six MS experts worked on the full-text review and initial wording of recommendations. A group of 62 multidisciplinary healthcare specialists validated the final proposal of summarised evidence. RESULTS: A strong agreement was reached for all 104 proposed recommendations. They cover diverse topics, such as pregnancy planning, follow-up during pregnancy and postpartum, delivery routes, locoregional analgesia or anaesthesia, prevention of postpartum relapses, breastfeeding, vaccinations, reproductive assistance, management of relapses and disease-modifying treatments. CONCLUSION: The 2022 recommendations of the French MS society should be helpful to harmonise counselling and treatment practice for pregnancy in persons with MS, allowing for better and individualised choices.

Clinical and MRI measures to identify non-acute MOG-antibody disease in adults.

MRI and clinical features of myelin oligodendrocyte glycoprotein (MOG)-antibody disease may overlap with those of other inflammatory demyelinating conditions posing diagnostic challenges, especially in non-acute phases and when serologic testing for MOG antibodies is unavailable or shows uncertain results. We aimed to identify MRI and clinical markers that differentiate non-acute MOG-antibody disease from aquaporin 4 (AQP4)-antibody neuromyelitis optica spectrum disorder and relapsing remitting multiple sclerosis, guiding in the identification of patients with MOG-antibody disease in clinical practice. In this cross-sectional retrospective study, data from 16 MAGNIMS centres were included. Data collection and analyses were conducted from 2019 to 2021. Inclusion criteria were: diagnosis of MOG-antibody disease; AQP4-neuromyelitis optica spectrum disorder and multiple sclerosis; brain and cord MRI at least 6 months from relapse; and Expanded Disability Status Scale (EDSS) score on the day of MRI. Brain white matter T2 lesions, T1-hypointense lesions, cortical and cord lesions were identified. Random forest models were constructed to classify patients as MOG-antibody disease/AQP4-neuromyelitis optica spectrum disorder/multiple sclerosis; a leave one out cross-validation procedure assessed the performance of the models. Based on the best discriminators between diseases, we proposed a guide to target investigations for MOG-antibody disease. One hundred and sixty-two patients with MOG-antibody disease [99 females, mean age: 41 (±14) years, median EDSS: 2 (0-7.5)], 162 with AQP4-neuromyelitis optica spectrum disorder [132 females, mean age: 51 (±14) years, median EDSS: 3.5 (0-8)], 189 with multiple sclerosis (132 females, mean age: 40 (±10) years, median EDSS: 2 (0-8)] and 152 healthy controls (91 females) were studied. In young patients (<34 years), with low disability (EDSS < 3), the absence of Dawson's fingers, temporal lobe lesions and longitudinally extensive lesions in the cervical cord pointed towards a diagnosis of MOG-antibody disease instead of the other two diseases (accuracy: 76%, sensitivity: 81%, specificity: 84%, P < 0.001). In these non-acute patients, the number of brain lesions < 6 predicted MOG-antibody disease versus multiple sclerosis (accuracy: 83%, sensitivity: 82%, specificity: 83%, P < 0.001). An EDSS < 3 and the absence of longitudinally extensive lesions in the cervical cord predicted MOG-antibody disease versus AQP4-neuromyelitis optica spectrum disorder (accuracy: 76%, sensitivity: 89%, specificity: 62%, P < 0.001). A workflow with sequential tests and supporting features is proposed to guide better identification of patients with MOG-antibody disease. Adult patients with non-acute MOG-antibody disease showed distinctive clinical and MRI features when compared to AQP4-neuromyelitis optica spectrum disorder and multiple sclerosis. A careful inspection of the morphology of brain and cord lesions together with clinical information can guide further analyses towards the diagnosis of MOG-antibody disease in clinical practice.

Pregnancy and neuromyelitis optica spectrum disorders: 2022 recommendations from the French Multiple Sclerosis Society.

BACKGROUND: In 2020, the French Multiple Sclerosis (MS) Society (SFSEP) decided to develop a national evidence-based consensus on pregnancy in MS. As neuromyelitis optica spectrum disorders (NMOSD) shares a series of commonalities with MS, but also some significant differences, specific recommendations had to be developed. OBJECTIVES: To establish recommendations on pregnancy in women with NMOSD. METHODS: The French Group for Recommendations in Multiple Sclerosis (France4MS) reviewed PubMed and universities databases (January 1975 through June 2021). The RAND/UCLA appropriateness method, which was developed to synthesise the scientific literature and expert opinions on health care topics, was used to reach a formal agreement. Fifty-six MS experts worked on the full-text review and initial wording of recommendations. A sub-group of nine NMOSD experts was dedicated to analysing available data on NMOSD. A group of 62 multidisciplinary healthcare specialists validated the final proposal of summarised evidence. RESULTS: A strong agreement was reached for all 66 proposed recommendations. They cover diverse topics, such as pregnancy planning, follow-up during pregnancy and postpartum, delivery routes, loco-regional analgesia or anaesthesia, prevention of postpartum relapses, breastfeeding, vaccinations, reproductive assistance, management of relapses, and disease-modifying treatments. CONCLUSION: Physicians and patients should be aware of the new and specific evidence-based recommendations of the French MS Society for pregnancy in women with NMOSD. They should help harmonise counselling and treatment practise, allowing for better individualised choices.

Multiple sclerosis clinical forms classification with graph convolutional networks based on brain morphological connectivity.

Multiple Sclerosis (MS) is an autoimmune disease that combines chronic inflammatory and neurodegenerative processes underlying different clinical forms of evolution, such as relapsing-remitting, secondary progressive, or primary progressive MS. This identification is usually performed by clinical evaluation at the diagnosis or during the course of the disease for the secondary progressive phase. In parallel, magnetic resonance imaging (MRI) analysis is a mandatory diagnostic complement. Identifying the clinical form from MR images is therefore a helpful and challenging task. Here, we propose a new approach for the automatic classification of MS forms based on conventional MRI (i.e., T1-weighted images) that are commonly used in clinical context. For this purpose, we investigated the morphological connectome features using graph based convolutional neural network. Our results obtained from the longitudinal study of 91 MS patients highlight the performance (F1-score) of this approach that is better than state-of-the-art as 3D convolutional neural networks. These results open the way for clinical applications such as disability correlation only using T1-weighted images.

Classification of multiple sclerosis clinical profiles using machine learning and grey matter connectome.

Purpose: The main goal of this study is to investigate the discrimination power of Grey Matter (GM) thickness connectome data between Multiple Sclerosis (MS) clinical profiles using statistical and Machine Learning (ML) methods. Materials and Methods: A dataset composed of 90 MS patients acquired at the MS clinic of Lyon Neurological Hospital was used for the analysis. Four MS profiles were considered, corresponding to Clinical Isolated Syndrome (CIS), Relapsing-Remitting MS (RRMS), Secondary Progressive MS (SPMS), and Primary Progressive MS (PPMS). Each patient was classified in one of these profiles by our neurologist and underwent longitudinal MRI examinations including T1-weighted image acquisition at each examination, from which the GM tissue was segmented and the cortical GM thickness measured. Following the GM parcellation using two different atlases (FSAverage and Glasser 2016), the morphological connectome was built and six global metrics (Betweenness Centrality (BC), Assortativity (r), Transitivity (T), Efficiency (E g ), Modularity (Q) and Density (D)) were extracted. Based on their connectivity metrics, MS profiles were first statistically compared and second, classified using four different learning machines (Logistic Regression, Random Forest, Support Vector Machine and AdaBoost), combined in a higher level ensemble model by majority voting. Finally, the impact of the GM spatial resolution on the MS clinical profiles classification was analyzed. Results: Using binary comparisons between the four MS clinical profiles, statistical differences and classification performances higher than 0.7 were observed. Good performances were obtained when comparing the two early clinical forms, RRMS and PPMS (F1 score of 0.86), and the two neurodegenerative profiles, PPMS and SPMS (F1 score of 0.72). When comparing the two atlases, slightly better performances were obtained with the Glasser 2016 atlas, especially between RRMS with PPMS (F1 score of 0.83), compared to the FSAverage atlas (F1 score of 0.69). Also, the thresholding value for graph binarization was investigated suggesting more informative graph properties in the percentile range between 0.6 and 0.8. Conclusion: An automated pipeline was proposed for the classification of MS clinical profiles using six global graph metrics extracted from the GM morphological connectome of MS patients. This work demonstrated that GM morphological connectivity data could provide good classification performances by combining four simple ML models, without the cost of long and complex MR techniques, such as MR diffusion, and/or deep learning architectures.

Signal Intensity Evaluation in the Dentate Nucleus and Subcortical Gray Matter : Effect of Several Administrations of Gadoterate Meglumine in Multiple Sclerosis.

PURPOSE: Several studies reported gadolinium deposition in the dentate nuclei (DN) and the globus pallidus (GP) that was associated to linear GBCA administrations rather than macrocyclic. It is therefore imperative to evaluate and assess the safety of cumulative administration of gadoterate meglumine (macrocyclic). Thus, T1-weighted images (T1WI) of multiple sclerosis (MS) patients longitudinally followed for 4 years were retrospectively analyzed. METHODS: In this study 44 patients, 10 with clinically isolated syndrome (CIS), 24 relapsing-remitting MS (RRMS) and 10 primary-progressive MS (PPMS) were examined every 6 months (first four scans) and then with a 1-year interval (last two scans). Image processing consisted in reorienting unenhanced T1WI to standard space, followed by B1 inhomogeneity correction. A patient-specific template was then generated to normalize T1WI signal intensity (SI) and segment the DN and subcortical GM structures. All structures were then transformed to each patient space in order to measure the SI in each region. The cerebellar peduncles (CP) and semi-oval (SO) white matter were then manually delineated and used as reference to calculate SI ratios in the DN and subcortical GM structures. A linear mixed-effect model was finally applied to longitudinally analyze SI variations. RESULTS: The SI measurements performed in all structures showed no significant increases with the cumulative GBCA administration. CONCLUSION: This study showed no significant SI increases within the DN and subcortical GM structures of longitudinally followed MS patients even with the cumulative administration of the macrocyclic GBCA gadoterate meglumine.

Leptomeningeal enhancement on post-contrast FLAIR images for early diagnosis of Susac syndrome.

BACKGROUND: Leptomeningeal enhancement (LME) is a key feature of Susac syndrome (SuS) but is only occasionally depicted on post-contrast T1-weighted images (T1-WI). OBJECTIVE: As post-contrast fluid-attenuated inversion recovery (FLAIR) may be more sensitive, our aim was to assess LME in SuS on this sequence. METHODS: From 2010 to 2020, 20 patients with definite SuS diagnosis were retrospectively enrolled in this multicentre study. Two radiologists independently assessed the number of LME on post-contrast FLAIR and T1-WI acquisitions performed before any treatment. A chi-square test was used to compare both sequences and the interrater agreement was calculated. RESULTS: Thirty-five magnetic resonance imagings (MRIs) were performed before treatment, including 19 post-contrast FLAIR images in 17 patients and 25 post-contrast T1-WI in 19 patients. In terms of patients, LME was observed on all post-contrast FLAIR, contrary to post-contrast T1-WI (17/17 (100%) vs. 15/19 (79%), p < 0.05). In terms of sequences, LME was observed on all post-contrast FLAIR, contrary to post-contrast T1-WI (19/19 (100%) vs. 16/25 (64%), p < 0.005). LME was disseminated at both supratentorial (19/19) and infratentorial (18/19) levels on post-contrast FLAIR, contrary to post-contrast T1-WI (3/25 and 9/25, respectively). Interrater agreement was excellent for post-contrast FLAIR (κ = 0.95) but only moderate for post-contrast T1-WI (κ = 0.61). CONCLUSION: LME was always observed and easily visible on post-contrast FLAIR images prior to SuS treatment. In association with other MRI features, it is highly indicative of SuS.

Evaluation of CSF1R-related adult onset leukoencephalopathy with axonal spheroids and pigmented glia diagnostic criteria.

BACKGROUND AND PURPOSE: Diagnostic criteria for adult onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP) due to colony-stimulating factor 1 receptor (CSF1R) mutation have recently been proposed. Our objective was to assess their accuracy in an independent multicenter cohort. METHODS: We evaluated the sensitivity and specificity of the diagnostic criteria for ALSP (including the "probable" and "possible" definitions) in a national cohort of 22 patients with CSF1R mutation, and 59 patients with an alternative diagnosis of adult onset inherited leukoencephalopathy. RESULTS: Overall, the sensitivity of the diagnostic criteria for ALSP was 82%, including nine of 22 patients diagnosed as probable and nine of 22 diagnosed as possible. Twenty of the 59 CSF1R mutation-negative leukoencephalopathies fulfilled the diagnostic criteria, leading to a specificity of 66%. CONCLUSIONS: Diagnostic criteria for ALSP have an overall limited sensitivity along with a modest specificity. We suggest that in patients suspected of genetic leukoencephalopathy, a comprehensive magnetic resonance imaging pattern-based approach is warranted, together with white matter gene panel or whole exome sequencing.

Ensemble Learning for Multiple Sclerosis Disability Estimation Using Brain Structural Connectivity.

Background: Multiple sclerosis (MS) is an autoimmune inflammatory disease of the central nervous system characterized by demyelination and neurodegeneration processes. It leads to different clinical courses and degrees of disability that need to be anticipated by the neurologist for personalized therapy. Recently, machine learning (ML) techniques have reached a high level of performance in brain disease diagnosis and/or prognosis, but the decision process of a trained ML system is typically nontransparent. Using brain structural connectivity data, a fully automatic ensemble learning model, augmented with an interpretable model, is proposed for the estimation of MS patients' disability, measured by the Expanded Disability Status Scale (EDSS). Materials and Methods: An ensemble of four boosting-based models (GBM, XGBoost, CatBoost, and LightBoost) organized following a stacking generalization scheme was developed using diffusion tensor imaging (DTI)-based structural connectivity data. In addition, an interpretable model based on conditional logistic regression was developed to explain the best performances in terms of white matter (WM) links for three classes of EDSS (low, medium, and high). Results: The ensemble model reached excellent level of performance (root mean squared error of 0.92 ± 0.28) compared with single-based models and provided a better EDSS estimation using DTI-based structural connectivity data compared with conventional magnetic resonance imaging measures associated with patient data (age, gender, and disease duration). Used for interpretation of the estimation process, the counterfactual method showed the importance of certain brain networks, corresponding mainly to left hemisphere WM links, connecting the left superior temporal with the left posterior cingulate and the right precuneus gray matter regions, and the interhemispheric WM links constituting the corpus callosum. Also, a better accuracy estimation was found for the high disability class. Conclusion: The combination of advanced ML models and sensitive techniques such as DTI-based structural connectivity demonstrated to be useful for the estimation of MS patients' disability and to point out the most important brain WM networks involved in disability. Impact statement An ensemble of "boosting" machine learning (ML) models was more performant than single models to estimate disability in multiple sclerosis. Diffusion tensor imaging (DTI)-based structural connectivity led to better performance than conventional magnetic resonance imaging. An interpretable model, based on counterfactual perturbation, highlighted the most relevant white matter fiber links for disability estimation. These findings demonstrated the clinical interest of combining DTI, graph modeling, and ML techniques.

T1/T2 ratio: A quantitative sensitive marker of brain tissue integrity in multiple sclerosis.

BACKGROUND AND PURPOSE: The aim of this study is to determine whether cerebral white matter (WM) microstructural damage, defined by decreased fractional anisotropy (FA) and increased axial (AD) and radial (RD) diffusivities, could be detected as accurately by measuring the T1/T2 ratio, in relapsing-remitting multiple sclerosis (RRMS) patients compared to healthy control (HC) subjects. METHODS: Twenty-eight RRMS patients and 24 HC subjects were included in this study. Region-based analysis based on the ICBM-81 diffusion tensor imaging (DTI) atlas WM labels was performed to compare T1/T2 ratio to DTI values in normal-appearing WM (NAWM) regions of interest. Lesions segmentation was also performed and compared to the HC global WM. RESULTS: A significant 19.65% decrease of T1/T2 ratio values was observed in NAWM regions of RRMS patients compared to HC. A significant 6.30% decrease of FA, as well as significant 4.76% and 10.27% increases of AD and RD, respectively, were observed in RRMS compared to the HC group in various NAWM regions. Compared to the global WM HC mask, lesions have significantly decreased T1/T2 ratio and FA and increased AD and RD (p < . 001). CONCLUSIONS: Results showed significant differences between RRMS and HC in both DTI and T1/T2 ratio measurements. T1/T2 ratio even demonstrated extensive WM abnormalities when compared to DTI, thereby highlighting the ratio's sensitivity to subtle differences in cerebral WM structural integrity using only conventional MRI sequences.

Risk Factors and Time to Clinical Symptoms of Multiple Sclerosis Among Patients With Radiologically Isolated Syndrome.

Importance: Younger age, oligoclonal bands, and infratentorial and spinal cord lesions are factors associated with an increased 10-year risk of clinical conversion from radiologically isolated syndrome (RIS) to multiple sclerosis (MS). Whether disease-modifying therapy is beneficial for individuals with RIS is currently unknown. Objectives: To evaluate the 2-year risk of a clinical event (onset of clinical symptoms of MS) prospectively, identify factors associated with developing an early clinical event, and simulate the sample size needed for a phase III clinical trial of individuals with RIS meeting 2009 RIS criteria. Design, Setting, and Participants: This cohort study used data on prospectively followed-up individuals with RIS identified at 1 of 26 tertiary centers for MS care in France that collect data for the Observatoire Français de la Sclérose en Plaques database. Participants were aged 10 to 80 years with 2 or more magnetic resonance imaging (MRI) scans after study entry and an index scan after 2000. All diagnoses were validated by an expert group, whose review included a double centralized MRI reading. Data were analyzed from July 2020 to January 2021. Exposure: Diagnosis of RIS. Main Outcomes and Measures: Risk of clinical event and associated covariates at index scan were analyzed among all individuals with RIS. Time to the first clinical event was compared by covariates, and sample size estimates were modeled based on identified risk factors. Results: Among 372 individuals with RIS (mean [SD] age at index MRI scan, 38.6 [12.1] years), 354 individuals were included in the analysis (264 [74.6%] women). A clinical event was identified among 49 patients (13.8%) within 2 years, which was associated with an estimated risk of conversion of 19.2% (95% CI, 14.1%-24.0%). In multivariate analysis, age younger than 37 years (hazard ratio [HR], 4.04 [95% CI, 2.00-8.15]; P < .001), spinal cord lesions (HR, 5.11 [95% CI, 1.99-13.13]; P = .001), and gadolinium-enhancing lesions on index scan (HR, 2.09 [95% CI, 1.13-3.87]; P = .02) were independently associated with an increased risk of conversion to MS. Having 2 factors at the time of the index MRI scan was associated with a risk of 27.9% (95% CI, 13.5%-39.9%) of a seminal event within 2 years, increasing to 90.9% (95% CI, 41.1%-98.6%) for individuals with all 3 factors (3 risk factors vs none: HR, 23.34 [95% CI, 9.08-59.96]; P < .001). Overall, with 80% power to detect an effect size of 60% within 24 months, a total of 160 individuals with RIS were needed assuming an event rate of 20%. Conclusions and Relevance: This study found that age younger than age 37 years, spinal cord involvement, and gadolinium-enhancing lesions on index MRI scan were associated with earlier clinical disease and relevant to the number of enrolled patients needed to detect a potential treatment effect.