Reduced Paramagnetic Rim Visibility Signals Lesion Aging and Tissue Degeneration in Progressive Multiple Sclerosis
TL;DR
Underlying tissue destruction in Definite vs Probable PRLs in people with progressive MS is assessed using quantitative MRI to assess underlying tissue destruction in PRLs in people with progressive MS.
Abstract
Background and Objectives Paramagnetic rim lesions (PRLs) in multiple sclerosis (MS) indicate chronic, innate immune-mediated inflammation contributing to disease progression. Rim disappearance or attenuation is interpreted as a treatment response; however, it remains uncertain whether reduced susceptibility signal—likely reflecting decreased iron within the rim compartment—represents true resolution of chronic inflammation, natural lesion aging, and/or advanced tissue degeneration. Cross-sectionally, PRLs show variable rim conspicuity, enabling classification as Definite or Probable. We used quantitative MRI (qMRI) to assess underlying tissue destruction in Definite vs Probable PRLs in people with progressive MS (PwPMS). Methods One hundred six PwPMS (34 primary progressive [PP], 72 secondary progressive; age 56 ± 10 years; median Expanded Disability Status Scale [EDSS] 5.5; 44% on disease-modifying therapy) underwent 3D-EPI and multishell diffusion MRI. PRLs (N = 155) were classified as Definite (N = 114) or Probable (N = 41). Prior clinical MRI scans estimated lesion age using an accelerated failure time model accounting for censoring. Whole-lesion, rim, and core regions were segmented; lesion size, quantitative susceptibility mapping, diffusion tensor imaging, Neurite Orientation Dispersions Density Imaging metrics, and T1 signal intensity were compared after multiple comparison correction. Results In total, 52% of PwPMS had ≥1 PRL; only 33% had Definite PRLs. Probable PRLs were more frequent in secondary vs PP MS (24% vs 9%, p = 0.05). Higher PRL counts were associated with younger age (p < 0.001), and PRLs were more often Definite in younger PwPMS (p < 0.0004). Participants with only Definite PRLs had shorter disease duration than those with ≥1 Probable PRL or no PRLs, and Probable PRLs were significantly older. Definite PRLs showed higher rim susceptibility (22 vs 8 ppb, p < 0.001); Probable PRLs demonstrated greater tissue disruption at qMRI—higher mean diffusivity (MD), axial diffusivity (AD), radial diffusivity (RD), and FISO and lower neurite density index (NDI), and larger lesion size (all p < 0.001). Discussion In progressive MS, PRLs with lower rim susceptibility are more likely classified as Probable but may represent older lesions with greater tissue damage. Because chronic active lesions do not require iron pathologically and rims can fade over time, rim disappearance at MRI cannot be interpreted unequivocally as resolving inflammation and may reflect natural lesion evolution. Rim persistence may be a more informative marker of ongoing compartmentalized inflammation than rim disappearance is of its resolution.
