Pozzilli, Valeria; Prados Carrasco, Ferran; Kim, Neena; Abdel-Mannan, Omar; Nistri, Riccardo; Goebl, Philipp; Hemingway, Cheryl; ... Hacohen, Yael; + view all Pozzilli, Valeria; Prados Carrasco, Ferran; Kim, Neena; Abdel-Mannan, Omar; Nistri, Riccardo; Goebl, Philipp; Hemingway, Cheryl; Biswas, Asthik; Mankad, Kshitij; Sudhakar, Sniya; Parida, Amitav; Wright, Sukhvir; Wassmer, Evangeline; Eyre, Michael; Lim, Ming; Rossor, Thomas; Eshaghi, Arman; Barkhof, Frederik; De Meo, Ermelinda; Ciccarelli, Olga; Hacohen, Yael; - view fewer (2025) Slowly Expanding Lesions in Pediatric-Onset Multiple Sclerosis. JAMA NEUROLOGY 10.1001/jamaneurol.2025.2619. (In press).

Text Ciccarelli_Slowly Expanding Lesions.pdf Access restricted to UCL open access staff until 12 August 2026. Download (536kB)

Abstract

Introduction: Slowly expanding lesions (SELs) are chronic active lesions showing constant and concentric expansion, detectable through longitudinal MRI analysis. Over time, they exhibit progressive reduction in T1 signal intensity, reflecting axonal damage. Initially described in progressive multiple sclerosis (MS), SELs also occur in relapsing-remitting MS (RRMS), and have been associated with disability progression in adults. We aimed to investigate whether SELs in pediatric-onset MS are associated with greater disability, reduced brain volume growth, and elevated serum NfL levels. / / Methods: This prospective analysis included pediatric-onset MS patients (<18 years) from the POINTMS-Children (Predicting Optimal Individualized Treatment Response in MS-children) observational UK multicenter study (2019-2024), which recruited patients within 3 months of starting a new disease-modifying treatment. Eligible patients had (1) a study entry MRI and (2) at least three follow-up brain scans with 3DT1, 3DFLAIR, and T2 sequences (Supplementary tables). Participants underwent brain MRIs, EDSS, cognitive assessments, and blood tests every 6 months. SELs were identified according to established guidelines as lesions showing a constant and concentric expansion over three timepoints. MRI data from 317 children in the NIH-funded normative brain development study were used to model growth trajectories. Z-scores were calculated by comparing patients’ volumes to these trajectories. Impaired brain volume growth was defined as ≤ 5th percentile. Longitudinal changes in clinical, NfL, and MRI parameters were analyzed using mixed-effects models, while predictors identified via regression analyses. SELs were dichotomized (≥4) per literature-defined cut-off predicting disability progression. Further methodological details are provided in Supplementary methods. / / Results: Forty patients were included (median age 15.6 years; 85% female; median EDSS: 1.0) and 34/40 were treated with anti-CD20 therapy (Table). Median follow-up was 18.3 months (IQR 17–20), during which 92.5% achieved NEDA-3. EDSS (β = -0.02, CI [-0.03, -0.01], p < 0.001) and processing speed (β = 0.32, CI [0.02, 0.62], p = 0.03) improved over time. NfL levels decreased (β = -1.30, CI [-1.99, -0.62], p < 0.001). Total brain and regional (cortical grey matter and thalamus) volume z-scores remained stable. At least one SEL was identified in 34/40 (85%) (Figure). Median number of SELs per patient was 3 (IQR: 2–5), representing 11% of total T2 lesions, and 7% of total lesion volume. Up to 15% of SELs were in the periventricular area. Higher baseline T2 lesion count (β = 0.03, CI [0.02, 0.05], p < 0.001) and lower total brain volume z-scores (β = -0.15, CI [-0.30, 0.00], p = 0.05) predicted a higher number of SELs. Each 1-unit decrease in total brain volume z-scores doubled the odds of ≥4 SELs (OR = 0.50, CI [0.25, 0.88]). Impaired brain volume growth (β = 0.57, CI [0.15, 0.99], p = 0.01) was associated with a higher number of SELs. No associations were found between SEL number or volume and longitudinal changes in EDSS, cognitive scores, or serum NfL levels. / / Discussion: In this study of treated 40 pediatric MS patients, 85% had ≥1 SEL, a frequency comparable to adult RRMS (86%) and lower than secondary progressive MS (99%). Our findings indicate that chronic lesion development begins early in the disease course, even with effective treatment. A higher baseline lesion burden and reduced brain growth were associated with increased SEL formation. Although SELs reflect a neurodegenerative process independent of relapses, their presence was not accompanied by concurrent disability accrual, as measured by EDSS or cognitive scores. This may reflect a greater recovery capacity in younger patients or the beneficial effects of early initiation of anti-CD20 therapy in this cohort. A limitation is the relatively small sample size and short follow-up period, which may have limited the ability to detect longer-term clinical correlations. Methods: This prospective analysis included pediatric-onset MS patients (<18 years) from the POINTMS-Children (Predicting Optimal Individualized Treatment Response in MS-children) observational UK multicenter study (2019-2024), which recruited patients within 3 months of starting a new disease-modifying treatment. Eligible patients had (1) a study entry MRI and (2) at least three follow-up brain scans with 3DT1, 3DFLAIR, and T2 sequences (Supplementary tables). Participants underwent brain MRIs, EDSS, cognitive assessments, and blood tests every 6 months. SELs were identified according to established guidelines as lesions showing a constant and concentric expansion over three timepoints. MRI data from 317 children in the NIH-funded normative brain development study were used to model growth trajectories. Z-scores were calculated by comparing patients’ volumes to these trajectories. Impaired brain volume growth was defined as ≤ 5th percentile. Longitudinal changes in clinical, NfL, and MRI parameters were analyzed using mixed-effects models, while predictors identified via regression analyses. SELs were dichotomized (≥4) per literature-defined cut-off predicting disability progression. Further methodological details are provided in Supplementary methods. / / Results: Forty patients were included (median age 15.6 years; 85% female; median EDSS: 1.0) and 34/40 were treated with anti-CD20 therapy (Table). Median follow-up was 18.3 months (IQR 17–20), during which 92.5% achieved NEDA-3. EDSS (β = -0.02, CI [-0.03, -0.01], p < 0.001) and processing speed (β = 0.32, CI [0.02, 0.62], p = 0.03) improved over time. NfL levels decreased (β = -1.30, CI [-1.99, -0.62], p < 0.001). Total brain and regional (cortical grey matter and thalamus) volume z-scores remained stable. At least one SEL was identified in 34/40 (85%) (Figure). Median number of SELs per patient was 3 (IQR: 2–5), representing 11% of total T2 lesions, and 7% of total lesion volume. Up to 15% of SELs were in the periventricular area. Higher baseline T2 lesion count (β = 0.03, CI [0.02, 0.05], p < 0.001) and lower total brain volume z-scores (β = -0.15, CI [-0.30, 0.00], p = 0.05) predicted a higher number of SELs. Each 1-unit decrease in total brain volume z-scores doubled the odds of ≥4 SELs (OR = 0.50, CI [0.25, 0.88]). Impaired brain volume growth (β = 0.57, CI [0.15, 0.99], p = 0.01) was associated with a higher number of SELs. No associations were found between SEL number or volume and longitudinal changes in EDSS, cognitive scores, or serum NfL levels. / / Discussion: In this study of treated 40 pediatric MS patients, 85% had ≥1 SEL, a frequency comparable to adult RRMS (86%) and lower than secondary progressive MS (99%). Our findings indicate that chronic lesion development begins early in the disease course, even with effective treatment. A higher baseline lesion burden and reduced brain growth were associated with increased SEL formation. Although SELs reflect a neurodegenerative process independent of relapses, their presence was not accompanied by concurrent disability accrual, as measured by EDSS or cognitive scores. This may reflect a greater recovery capacity in younger patients or the beneficial effects of early initiation of anti-CD20 therapy in this cohort. A limitation is the relatively small sample size and short follow-up period, which may have limited the ability to detect longer-term clinical correlations.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item