New Algorithms Improving PML Risk Stratification in MS Patients Treated With Natalizumab

Background: The use of Natalizumab is limited in MS by the risk of progressive multifocal leucoencephalopathy (PML). The factors currently used to stratify PML risk present some limitations. Objective: To identify new risk factors that improve the PML risk stratification. Methods: We studied 1307 MS patients treated with natalizumab for a median time of 3.28 year. Factors associated with PML onset were explored by uni- and multivariate logistic regression. Results: The multivariate analysis of total cohort identified anti-JC virus antibody indices (OR=18.07) and relapse rate (OR=4.66) as the best predictors for the onset of PML. The risk ranged from < 1/3300 to 1/50 depending on these variables. In the 277 patients with lipid-specific oligoclonal IgM bands restricted to CSF (LS-OCMB) assessment, age at natalizumab onset, anti-JCV antibody indices, and LS-OCMB status predicted PML risk (AUC=0.92). The absence of LS-OCMB was the best individual predictor (OR=26.83). Age was also an important factor (OR=6.74), which contributed to PML risk similarly to anti JC status (OR=6.52). Depending of these three factors the individual risk ranged from <1/10000 to 1/33. Conclusion: Disease activity, LS-OCMB antibody levels and age can help tailor natalizumab therapy in MS patients, as predictors of PML