To evaluate the long-term safety and potential neuroprotective effects of human fetal neural precursor cell transplantation in individuals with progressive multiple sclerosis. This retrospective observational study assessed long-term follow-up data (median 5.7 years) from participants in a first-in-human phase I trial (NCT03269071) investigating neural cell transplantation in progressive multiple sclerosis. Clinical and radiological available data were collected and compared to baseline values and reference populations. Long-term follow-up data were available for 11 of 12 patients. No serious adverse events related to cell transplantation occurred, confirming long-term safety. Although physical disability progressed over time, cognition remained stable compared to baseline. Five patients developed new T2-hyperintense brain lesions without associated contrast enhancement. Long-term gray matter atrophy rates were lower than those observed during the initial two years post-transplantation. Rates of gray and white matter atrophy approximated those of age-matched healthy individuals and patients with early-stage multiple sclerosis, with a trend for a lower atrophy rate compared to a control cohort with progressive disease. These structural findings were consistent with stable levels of circulating biomarkers of neurodegeneration. The findings support the long-term safety of fetal neural cell transplantation in progressive multiple sclerosis. Although exploratory in nature, the stability of cognitive function, reduced brain atrophy, and consistent biomarker profiles support the hypothesis of a possible neuroprotective effect that warrants further investigation in a controlled clinical trial setting.

BACKGROUND AND PURPOSE:The T1-weighted/T2-weighted ratio, a proposed surrogate measure of myelin integrity/damage, has debated specificity in multiple sclerosis. We investigated compartment-specific T1-weighted/T2-weighted alterations across multiple sclerosis phenotypes and age at onset, and their associations with disability and cognition. MATERIALS AND METHODS:In this multicenter study, 918 multiple sclerosis and 272 healthy controls from an Italian neuroimaging initiative underwent standardized 3D T1- and T2-weighted MRI (3D or 2D dual-echo sequences depending on site), together with clinical and cognitive assessments. T1-weighted/T2-weighted ratios were computed using a standardized pipeline and multiple sclerosis z-scores were derived across brain regions using linear mixed-effects models fitted on healthy controls. Associations with disease duration, disability, and Symbol Digit Modalities Test were analyzed. RESULTS:Compared with healthy controls, patients showed lower T1-weighted/T2-weighted ratios in lesions, normal-appearing white matter, and cortical grey matter (all FDR-adjusted p-values<.001), and higher ratios in the thalamus and deep grey matter (all FDR-adjusted p-values≤.01). Differences were more pronounced in secondary progressive and relapsing-remitting multiple sclerosis (all FDR-adjusted p-values<.001) than clinically isolated syndrome and primary progressive. Pediatric-onset patients showed lower normal-appearing white matter and cortical grey matter ratios (all FDR-adjusted p-values≤.02) than adult- and late-onset patients, which conversely showed higher deep grey matter ratios compared to the other groups (all FDR-adjusted p-values≤.01). Higher disability was associated to lower lesion, normal-appearing white matter, cortical grey matter ratios (all p≤.05) and higher thalamic ratios (p=.006). Longer disease duration correlated with lower lesion (p=.02), normal-appearing white matter ratios (p=.007) and higher deep grey matter ratios (p=.002). Worse cognitive performance was associated with lower cortical (p=.01), and higher thalamic and deep grey matter ratios (all p≤.004). CONCLUSIONS:T1-weighted/T2-weighted mapping can provide complementary, compartment-specific information on tissue integrity and clinical outcomes across the MS spectrum.

Introduction:Multiple sclerosis (MS) is a complex disease characterized by diverse clinical presentations and progression patterns. Accurate classification and prediction of disease severity are crucial for personalized treatment. We applied machine learning (ML) to demographic, clinical and MRI data to distinguish MS patients from healthy controls (HC), classify MS phenotypes and predict disability using the Expanded Disability Status Scale (EDSS) score. Methods:We included 1,554 MS patients and 520 HC from the Italian Neuroimaging Network Initiative repository, all with neurological assessment and brain T2-/3D T1-weighted MRI. Derived MRI features included total and regional T2 lesion volumes (LV), and normalized tissue volumes from cortical and subcortical grey matter (GM), white matter, cerebellum and brainstem. ML models, including support vector machines, multi-layer perceptron networks, Random Forest and Gradient Boosting were trained for classification and prediction tasks. SHAP analysis ranked the most influential variables. Results:ML models achieved 89-96% accuracy in distinguishing MS patients from HC, driven mainly by T2 LV and brainstem/cerebellar GM volumes. Relapsing vs progressive MS was classified with 92% accuracy, with EDSS, age, thalamic and cortical GM volumes as key predictors. EDSS prediction achieved an intra-class correlation of 0.56-0.76; most relevant contributors were T2 LV, sex, cortical/cerebellar GM and thalamic volumes. Discussion:ML models demonstrated high accuracy in detecting MS, differentiating phenotypes, and predicting disability. Integrating demographic, clinical and MRI measures emerges as an effective strategy for patients' classification and disease severity assessment.

Neuromyelitis optica spectrum disorder (NMOSD) is an autoimmune astrocytopathy where disability is largely relapse-driven. Cerebrovascular risk factors (cVRFs) may exacerbate astrocytic and small-vessel injury. Therefore, we investigated their impact on MRI damage and disability in NMOSD. Twenty-eight aquaporin 4 immunoglobulin G-positive NMOSD patients and 56 age- and sex-matched healthy controls (HC) underwent 3T brain and cervical MRI, cVRF and neurological assessment, including the Expanded Disability Status Scale (EDSS). Brain T2-hyperintense white matter lesion volume (WMLV), normalized global and regional volumes, normal-appearing WM (NAWM) fractional anisotropy (FA) and mean diffusivity (MD), diffusion tensor imaging analysis along the perivascular spaces (DTI-ALPS) index, spinal cord lesion volume (SCLV) and normalized cross-sectional area (nCSA) were quantified. cVRF were present in 57

In multiple sclerosis (MS), functional network abnormalities arise as structural damage accumulates. However their biological basis and spatial distribution remain unclear. This study investigated the associations between MS-related functional network abnormalities and physiological gene expression using the Allen Human Brain Atlas (AHBA). Five-hundred fifty-eight MS patients and 214 healthy controls (HC) underwent neurological assessment and 3 T MRI; 491 patients also completed a neuropsychological evaluation. Resting-state functional MRI was used to generate degree centrality maps to identify network topography alterations. Spatial correlations between centrality abnormalities (p < 0.01 uncorrected) and the expression of 3634 MS-related genes was evaluated using AHBA and the Multimodal Environment for Neuroimaging and Genomic Analysis. Genes showing significant associations (p < 0.001, R2 ≥ 0.15) underwent pathway enrichment analysis (p < 0.05, Bonferroni-corrected). Compared to HC, MS patients showed higher centrality mainly in the default-mode network (DMN), linked to genes regulating inflammation resolution and immune functions, and lower centrality in regions mostly located in the salience network and cerebellum, associated with genes implicated in cytokine response. Compared to HC and relapsing-remitting MS, progressive MS patients showed higher centrality in DMN and cerebellar regions, correlating with genes related to epigenetic and mitochondrial functions. Of the MS cohort, 144 (29.3%) patients were cognitively impaired. Compared to cognitively preserved MS and HC, they showed higher centrality in DMN and mesial temporal lobe regions, negatively correlated with expression of DNASE1, regulating DNA degradation, and CP, encoding ceruloplasmin, involved in iron homeostasis and potentially iron-driven neurodegeneration. Physiological regional gene expression spatially correlates with MS-related functional network alterations. Biological factors may shape regional vulnerability or resilience to MS pathology, influencing functional reorganization.

Sleep modulates brain-fluid transport and glymphatic clearance. In multiple sclerosis (MS), lower diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) index has been reported, but whether sleep quality contributes to DTI-ALPS variability is unknown. We assessed whether sleep disturbances are associated with a lower DTI-ALPS index in MS. 77 MS patients and 84 age- and sex-matched healthy controls (HC) underwent 3 T brain MRI. In MS patients, disability was assessed with the expanded disability status scale (EDSS), sleep quality with the Pittsburgh sleep quality index (PSQI; >5 defining sleep disturbances), and fatigue with the modified fatigue impact scale (MFIS). DTI-ALPS index was derived from fractional anisotropy (FA) maps after lesion exclusion. Sleep disturbances were reported in 32 (41.6

BACKGROUND:Cognitive impairment is common in multiple sclerosis (MS), yet the application of diagnostic frameworks of Neurocognitive Disorders (NCDs) is limited. Additionally, the integration of multimodal data for predicting cognitive outcomes using artificial intelligence (AI) remains underexplored. This study aimed to characterize NCDs in MS and predict cognitive worsening using an explainable deep learning model trained on MRI and clinical data. METHODS:Two-hundred twenty-four MS patients and 115 healthy controls (HC) underwent 3.0 T MRI and clinical assessment at baseline. MS patients also completed neuropsychological testing, including estimation of z-cognitive reserve, at baseline and after a median follow-up of 3.4 (interquartile range = [2.0; 6.1]) years. MS patients were classified as Mild or Major NCD according to the Diagnostic and Statistical Manual of Mental Disorders criteria at baseline, and as "stable" or "worsened" based on cognitive changes at follow-up. A deep learning model was trained on baseline T1-weighted MRI, demographic, clinical, and brain volumetric data to predict cognitive decline, with explainability methods used to interpret the model's decisions. RESULTS:At baseline, 4% of patients had Mild and 11% Major NCD. At follow-up, 12% showed cognitive decline. The deep learning model predicted follow-up cognitive status with 90% accuracy. Explainability models identified the most relevant predictors, in order of importance: cortical gray matter volume, age, thalamic and hippocampal volumes, T2 lesion volume, and z-cognitive reserve. CONCLUSIONS:The proposed multimodal AI approach demonstrated robust performance and highlighted relevant brain regions associated with cognitive worsening, underscoring its potential for personalized cognitive assessment and monitoring in MS.

BACKGROUND:Vascular risk factors (VRFs) are associated with diffuse brain damage in multiple sclerosis (MS). We investigated VRFs' impact on motor performance and sensorimotor network (SMN) structural integrity in MS patients. METHODS:In this retrospective cross-sectional study, 268 MS patients and 180 healthy controls were classified according to VRF presence (VRF[+]) or absence (VRF[-]). Clinical measures included Expanded Disability Status Scale (EDSS), Timed 25-Foot Walk test, and 9-Hole Peg Test. SMN integrity was assessed using white matter lesion volumes, cortical thickness of primary motor and somatosensory cortices, normalized deep gray matter volume (NDGMV), anterior (ACMA) and posterior cerebellar motor area volumes, mean upper cervical cord area, and diffusion metrics of the middle and superior cerebellar peduncles, medial lemniscus (ML), and corticospinal tracts. Group differences and MS-VRF interactions were examined. In MS, associations between clinical and magnetic resonance imaging (MRI) measures were explored, focusing on VRF influence. RESULTS:MS-VRF[+] patients showed higher EDSS scores and worse motor performance than MS-VRF[-] (false discovery rate p value [pFDR] ⩽ 0.004). Significant MS-VRF interactions were observed for motor performance, ML mean diffusivity, NDGMV, and ACMA volumes (pFDR ⩽ 0.039). In MS, EDSS was associated with the VRF × ML fractional anisotropy interaction term (β = 1.931, p = 0.042). CONCLUSIONS:VRFs may contribute to increased disability, motor impairment, and region-specific SMN structural damage in MS patients.

BACKGROUND AND OBJECTIVES:The choroid plexus (ChP) regulates CSF production and CNS homeostasis. In multiple sclerosis (MS), ChP enlargement occurs early in the disease course, yet its normative lifespan trajectory and relationship with MS-specific features remain poorly defined. We aimed to define the normative ChP volume model, to characterize its trajectory across the healthy lifespan, and to assess ChP enlargement in MS using z-scores, evaluating associations with demographic, clinical, MRI, and genetic variables, including human leukocyte antigen (HLA) and non-HLA polygenic risk scores (PRSs). METHODS:This monocentric retrospective cross-sectional study included 461 healthy controls (HCs) and 727 patients with MS (age 18-70 years) who underwent 3T brain MRI and neurologic assessment. ChP volumes were quantified using ASCHOPLEX, normalized for head size, and modeled in HCs. We derived age-related normalized ChP volume (NChPV) trajectory across adulthood. Z-scores for patients with MS were computed. PRSs were calculated from established MS susceptibility loci, encompassing both HLA and non-HLA regions. RESULTS:In HCs, normalized brain volume, lateral ventricle volume, and its squared term were independently associated with NChPV (R2 = 0.54). Model-derived NChPV trajectory remained stable until age 35 and then increased nonlinearly (p-FDR<0.002), with annualized growth rates rising from 0.24% at age 35 to over 0.7% in later decades. Patients with MS had significantly higher NChPV z-scores than controls (estimated mean [EM] z-score = 0.452, p-FDR<0.001), consistently across phenotypes (p = 0.744) and ages, with no evidence of age-dependent variation (β = 0.02 × 10-2, p = 0.957). In cross-sectional analyses, NChPV z-scores, already elevated 1 year after disease onset (EM z-scores = 0.252, p-FDR<0.001), increased further up to 5 years after disease onset (EM z-scores = 0.463, p-FDR<0.001), before plateauing. Higher z-scores were associated with greater T2-hyperintense white matter lesion volume (β = 0.012 × 10-2, p < 0.001) and HLA genetic burden (standardized β = 0.097, p = 0.038), but not non-HLA PRSs (p ≥ 0.177). DISCUSSION:We provided a normative ChP volume model, characterized its trajectory across the healthy lifespan, and showed early, consistently higher ChP volume in patients with MS across ages and disease durations in cross-sectional analyses, linked to inflammatory lesion burden and HLA-mediated genetic risk. The ChP may represent a noninvasive biomarker of neuroinflammation and HLA-related immunogenetic background in MS.

BACKGROUND:Only a few longitudinal studies, with follow-up duration up to 3.6 years, have assessed the substrates of relapses and disability in aquaporin-4 (AQP4) IgG-positive neuromyelitis optica spectrum disorders (NMOSD), primarily focusing on clinical and conventional MRI variables. OBJECTIVE:We evaluated the association of clinical and MRI measures, including a comprehensive multimodal advanced MRI protocol, with key long-term clinical outcomes in AQP4 IgG-positive NMOSD patients over a median follow-up of 8.5 years (interquartile range [IQR] = 3.5; 13.3). METHODS:Forty-six NMOSD patients and 77 healthy controls underwent 3T brain MRI. In patients, neurological evaluations were collected at baseline and every six months. Time to first relapse and 6-month confirmed disability worsening (6m-CDW) were recorded. Lesion volume and topography, global and regional brain volumes, normal-appearing white matter and gray matter fractional anisotropy and mean diffusivity, choroid plexus volume, enlarged perivascular spaces number and glymphatic system function were assessed. RESULTS:During the follow-up, 30% patients experienced at least one clinical relapse, while 22% had 6m-CDW. Higher number of previous attacks (hazard ratio [HR] = 1.17, 95%-confidence interval [CI] = 1.04; 1.32) and presence of anterior optic pathway lesions (HR = 4.92, 95%-CI = 1.12; 21.852) were risk factors independently associated with a shorter time to a first clinical relapse; lower thalamic volume was marginally associated (HR = 0.93, 95%-CI = 0.87; 1.00). Presence of cervical cord lesions (HR = 3.93, 95%-CI = 1.16; 13.31) and lower normalized cortical volume (HR = 0.94, 95%-CI = 0.89;0.99) were risk factors independently associated with a shorter time to 6m-CDW; higher number of previous attacks contributed marginally (HR = 1.19, 95% = 0.99; 1.43). High efficacy treatments (HETs) delayed time to first relapse and 6m-CDW. CONCLUSIONS:Previous attacks, optic nerve/spinal cord involvement, cortical/thalamic atrophy and the use of HETs associate with long-term outcomes in NMOSD.

BACKGROUND AND OBJECTIVES:The choroid plexus (CP) regulates immune functions and produces most CSF that circulates in the brain parenchyma through perivascular spaces, part of the glymphatic system. In multiple sclerosis (MS), CP enlargement and glymphatic dysfunction are associated with inflammatory activity, clinical disability, and brain damage, but their interrelation is unclear. We investigated whether glymphatic system dysfunction mediates the association between CP enlargement and brain damage in patients with MS. METHODS:Brain fluid-attenuated inversion recovery, 3-dimensional T1-weighted, diffusion-weighted, and susceptibility-weighted sequences were obtained from 146 patients with MS and 72 healthy controls (HC). Glymphatic function was assessed using the diffusion along the perivascular space (DTI-ALPS) index, and CP volume was measured automatically. RESULTS:Patients with MS showed significantly higher white matter (WM) lesion and CP volumes (p < 0.001), and lower DTI-ALPS index, brain, WM, thalamic, and cortical volumes than HC (p ≤ 0.048). In patients with MS, higher CP volume correlated with a lower DTI-ALPS index (r = -0.305, false discovery rate p value = 0.001). Both measures were associated with higher total, periventricular, and juxtacortical (JC) WM lesion volumes (CP volume: r from 0.285 to 0.340, p-FDR ≤ 0.001; DTI-ALPS index: r from -0.301 to -0.444, p ≤ 0.001), and lower brain, thalamic, cortical, and WM volumes (CP volume: r from -0.246 to -0.405, p-FDR ≤ 0.006; DTI-ALPS index: from 0.269 to 0.497, p-FDR ≤ 0.003). The DTI-ALPS index partially mediated the associations of normalized choroid plexus volume with total, periventricular, and JC T2-hyperintense WM lesion volumes (standardized-β ranging from 0.073 to 0.115, relative effect ranging from 25.2% to 33.6%) and normalized brain, thalamic, cortical, and WM volumes (standardized-β ranging from -0.086 to -0.125, relative effect ranging from 25.3% to 52.7%). DISCUSSION:In MS, enlarged normalized CP volume may contribute to brain damage accumulation possibly through the promotion of a chronic proinflammatory state and the mediation of glymphatic system dysfunction.

OBJECTIVE:The aim of this study was to explore the microstructural dynamics of the subventricular zone (SVZ) with aging and their associations with clinical disability and brain structural damage in pediatric-onset multiple sclerosis (MS) patients. METHODS:One-hundred and forty-one pediatric-onset MS patients (67 pediatric and 74 adults with pediatric-onset) and 233 healthy controls (HC) underwent neurological and 3.0 T MRI assessment. Fractional anisotropy (FA) and mean diffusivity (MD) were extracted from the SVZ and the thalamus (as control region). RESULTS:In HC, SVZ FA was higher until age 40 then declined, whereas MD was lower until age 35 before rising (false discovery rate p value [pFDR] ≤ 0.008). Thalamic FA was higher until age 30 and then declined, whereas MD was higher until age 50 (pFDR ≤ 0.007). Pediatric MS patients showed significantly higher SVZ FA than pediatric HC (pFDR < 0.001), while adult patients showed no differences compared to adult HC (pFDR ≤ 0.724). Adult patients had lower thalamic FA and higher MD (pFDR < 0.001). Adults had lower SVZ FA and MD, but higher thalamic MD compared to pediatric patients (pFDR < 0.001). In pediatric MS, higher SVZ FA and MD were associated with higher white matter (WM) lesion volume (LV) and choroid plexus volume and lower brain and thalamic volumes (pFDR ≤ 0.047). In adult patients, higher SVZ MD associated with higher WM LV, lower brain volumes, and lower z-SDMT (pFDR≤0.019). Thalamic microstructural abnormalities were associated with more severe disability and brain damage in both groups (pFDR ≤ 0.018). INTERPRETATION:Our findings suggest that microstructural changes in the SVZ occur early in pediatric MS and are associated with brain structural damage but not with clinical impairment. ANN NEUROL 2025;97:979-992.

BackgroundImmune-mediated processes are implicated in the pathogenesis of fatigue, a common symptom in multiple sclerosis (MS). The choroid plexus (CP) regulates central nervous system (CNS) immune homeostasis and undergoes volumetric modifications possibly contributing to MS-related fatigue. We explored the association between MS-related CP volume changes and fatigue dynamics.MethodEighty-five patients with MS and 68 healthy controls (HC) underwent brain 3T MRI, neurological evaluation and Modified Fatigue Impact Scale (MFIS) at two timepoints (median follow-up=1.4 years). Normalised brain and regional grey matter (GM) volumes were obtained using FSL-SIENAx, FIRST, SIENA and tensor-based morphometry. CP volumes were quantified with in-house methods, and longitudinal changes were analysed using linear mixed models.ResultsAt baseline, 25 (29%) patients with MS had fatigue (f-MS) (MFIS ≥38). Compared with HC, patients with MS had significantly higher brain T2-lesion volume, lower brain, deep GM, cortical volumes and higher CP volume (false discovery rate (FDR)-p ≤0.024). Compared with non-fatigued (nf-MS) patients, f-MS were older, more disabled (FDR-p ≤0.002) and showed numerically higher CP volume (FDR-p=0.076). At follow-up, 41 (68%) nf-MS remained non-fatigued (nf-FU-MS) and 19 (32%) developed fatigue (f-FU-MS). Patients with MS showed higher brain and deep GM atrophy rates versus HC (FDR-p ≤0.048), whereas clinical, lesional and brain volumetric changes were not significantly different among MS groups (FDR-p ≥0.287). CP volume significantly increased in all MS groups compared with HC (FDR-p ≤0.043), with greater enlargement in f-FU-MS versus nf-FU-MS (FDR-p=0.048).ConclusionsLarger CP and greater enlargement are associated with the presence and development of fatigue in MS, likely reflecting dynamic inflammatory states within the CNS, supporting the immunological contribution to MS-related fatigue.