IMMUNOPATHOLOGY OF INTRATHECAL INFLAMMATION IN MULTIPLE SCLEROSIS: CLINICAL, RADIOLOGICAL AND IMMUNOLOGICAL CORRELATES

Intrathecal inflammation is a key factor that drives disability accumulation in Multiple Sclerosis (MS), since early disease phases. Clinical, radiological, and cerebrospinal fluid markers that correlate with chronic intrathecal inflammation are advocated. i) To better clarify its clinical correlates, a post-mortem autopsy cohort study was performed on 269 progressive MS cases. The presence of early active lesions (EALs) and the extent of perivenular inflammation were examined. A subset of patients (n=87) already underwent characterization for the presence/absence of meningeal infiltrates and a detailed count of the CD20+ B cells and CD3+ T cells in perivenular infiltrates (n=22). EALs were detected in 22% of the examined cases, while high levels of perivascular inflammation were detected in 52% of cases. High levels of both active lesions and focal perivenular inflammation, enriched in B cells and associated with meningeal inflammation, within the white matter at post-mortem were associated with rapid disease evolution from onset and to the terminal stages. ii) Several observations suggested that cortical lesions are a frequent and early phenomenon in MS and one of the major drivers of disability accumulation, being associated with chronic intrathecal inflammation. We evaluated 199 relapsing- remitting MS (RRMS) patients who underwent brain and spinal cord MRI examinations at the time of diagnosis, including assessment of cortical lesions (CLs) by Double Inversion Recovery sequences. All patients had regular clinical median follow-up for 18 years (range 15-22). Confirmed disability worsening (CDW) was assessed, based on EDSS increase by ≥1.5, 1 or 0.5 points according to baseline EDSS (0, <5.5 or ≥5.5, respectively). Progression independent of relapse activity (PIRA) was referred as a 6-month CDW in which the Expanded Disability Status Scale (EDSS) worsening was sustained in all following assessments. The time to PIRA events, conversion to SPMS and to the EDSS score milestones of 4.0 and 6.0 was recorded. Eighty-three patients (41.7%) had PIRA after a mean of 7.6±3.6 yrs. Patients with PIRA had increased CLn and volume compared to those without PIRA events. After random survival forest, higher CLn, CLv, and EDSS were factors best associated with the risk of experiencing PIRA. Multivariate regression (with MRI, demographics, clinical variables including therapies) confirmed a higher number of CLs, a larger CLv and higher EDSS as predictive of PIRA. Higher CLv, CLn, baseline EDSS and presence of SC lesions were best predictors of EDSS ≥4 and ≥6 with the above- mentioned models. ROC analysis estimated at least 3 CLs as the optimal cut-off to identify patients more likely to develop SPMS (≥3, HR 12.2 [CI 3.55- 41.8],p< 0.001). Finally, 2-year accumulation of new CLs predicted SPMS (HR 1.3 [CI 1.1-1.5], p<0.001), and achievement of EDSS≥4 and ≥6 (HR 1.28[CI 1.09- 1.51],p=0.002, and HR 1.39 [CI 1.12-1.73],p=0.003, respectively). Concluding, early accumulation of focal cortical damage is an independent predictor of sustained disabillity progression and long-term disability. iii) Candidate cerebrospinal fluid (CSF) markers of intrathecal inflammation are advocated. 67 CSF markers were evaluated at the time of diagnosis with Multiplex assay in patients with both primary progressive MS and RRMS. CXCL12 and Osteopontin (OPN) revealed significantly increased in those patients experiencing a progressive course of the disease. OPN resulted best associated with accumulating cortical atrophy and confirmed disability worsening after two years of follow-up in a group of 107 patients with RRMS. Finally, the predictive value of CSF markers on developing PIRA events was evaluated in a retrospective study. Many markers, particularly related to TNF superfamily (sTNFR1, LIGHT) as well as OPN were increased in those patients developing PIRA events (23 out of 80), and were confirmed in those with PIRA events and no radiological activity during the follow-up. This susuggest that a specific intrathecal inflammatory profile characterize patients at higher risk of PIRA since early disease phases. Further studies are needed to evaluate the efficacy of anti-inflammatory treatments, particularly those acting intrathecally, in preventing PIRA events and subsequent disability.