Risk-management of natalizumab: a drug utilization study on duration of therapy and cause of discontinuation in Emilia-Romagna region

European Medicines Agency (EMA) recently recommended additional measures to better manage the risk of progressive multifocal leukoencephalopathy (PML) in patients treated with natalizumab (Tysabri). Although benefits of treatment with natalizumab continue to outweigh risks for patients with highly active relapsing-remitting multiple sclerosis (RRMS), EMA warned that the risk of PML (EMA/37607/2010) increases after two years of treatment and clinicians should to discuss the risk of natalizumab with patient both treatment starts and again after two years. However, to date, any evidence on Tysabri therapy duration is available. The aim of the study is to analyse actual duration of therapy with natalizumab and any cause of discontinuation in patients with RRMS in the Emilia-Romagna region from Tysabri Registry (Mancardi et al.,2008). From the Tysabri Registry, an accurate program of surveillance of Italian Agency of Medicine, patients treated with Tysabri in Emilia-Romagna region were selected from January 2007 (start-date of registry) until 30 June 2010. Drug prescriptions were analysed in order to identify start and end of natalizumab treatment for each patient. Discontinuation therapy was defined when a drug prescription was not been renewed after 30 days from latest one. The probability of discontinuation of treatment was analyzed by the Kaplan-Meier survival curve, defining as "event" the permanent discontinuation of treatment. Moreover, a distribution of drug interruption causes was performed. From 198 patients with at least 1 natalizumab prescription, 73% (145 subjects) were still on treatment at the end of the follow-up period, while the remaining 27% (53) discontinued Tysabri. Among patients who discontinued therapy, the median therapy length was 12 months, with a maximum of 35 and a minimum of 1 month. In the first two years of therapy, the frequency of interruption was low and uniform throughout the period, while in the third year increased the rate of therapy withdraw and this difference is statistically significant (p<0.0001). For 42/53 patients that interrupted therapy it was possible to know the cause of discontinuation: 41% for therapy completed course, 26% for adverse events, 14% for lack of treatment effectiveness, 5% for pregnancy and side effects, respectively, and remaining 9% for other causes. To our knowledge, this is the first drug utilization analysis of natalizumab and it could provide information about its real clinical practice use and its risk management approach. Most of therapy discontinuations are probably caused by the uncertain risk benefit profile of the long-term course of treatment.