OBJECTIVES:

Assess the prevalence and incidence of secondary progressive multiple sclerosis (SPMS) in Relapsing Remitting Multiple Sclerosis patients in an Italian MS registerr according to three different definitions of SPMS: the data driven definition (based on the Lorscheider algorithm LA),the EXPAND study inclusion criteria definition and neurologist’s definition (gold standard in clinical practice)

METHODS:

We extracted data from the registerby applying manually sequential filters to the dataset. First we selected RRMS patients fulfilling the inclusion criteria. Annual incidence of SPMS conversion was reported as number of patients converting to SPMS every 100 patients/year.

We tested diagnostic performances of definitions of SPMS (data driven and EXPAND criteria) against neurologist’s definition in terms of calibration, discrimination and goodness of fit (i.e. sensibility, sensitivity, Positive Predictive Value (PPV), Negative Predictive Value (NPV)).

RESULTS:

10,240 RRMS patients were analyzed. According to the neurologist definition 880 (8.59%) were SPMS. By applying data-driven definition 1,806 (17.64%) patients were SPMS, whereas, based on the EXPAND study definition 1,134 (11.07%) patients were SPMS. Annual rate of conversion to SPMS was 0.74 every 100 patients/year (neurologist), 1.57 every 100 patients/year (data-driven) and 0.94 every 100 patients/year (EXPAND criteria). Number of patients identified as SPMS by data-driven and neurologist’s definition together was 1,569, while 604 patients were identified as SPMS by EXPAND and neurologist’s definition. Data driven algorithm showed greater discrimination power (AUC: 0.83 vs 0.65) and had higher sensibility (77.1% vs 38.0%) in comparison to EXPAND definition. Both definitions showed similar specificity (88.0% vs 91.5%).

CONCLUSIONS:

An accurate disease course classification is needed for an efficacious understanding and treatment of SPMS. Real-world data from the Italian MS Register suggests that a data-driven definition based on the LA is more accurate in the diagnosis of SPMS.

The two authors write on behalf of the Italian MS Register.