OBJECTIVES: The European Medicines Agency (EMA) may require a post-authorization safety study (PASS), and in some cases secondary real-world data (RWD) may be appropriate for use. It can be challenging to obtain secondary data in Europe; however, there is precedent on using secondary RWD from the United States (US) to fulfill PASS requirements. Our objective is to provide an example of using such data to fulfill an EMA PASS requirement.

METHODS: A comparative, retrospective cohort study in commercially- and Medicaid-insured patients from the MarketScan^® databases was conducted. Patients newly-prescribed lurasidone for schizophrenia (lurasidone cohort) were compared to patients newly-prescribed other atypical antipsychotics (OAAs; OAA cohort) for the treatment of schizophrenia. Patients from the lurasidone cohort were matched to the OAA cohort by propensity scores on age, gender, Charlson comorbidity index, payer type, number of pre-index hospitalizations, and chlorpromazine dose equivalence. The primary outcomes included important identified and potential risks of lurasidone and OAAs.

RESULTS: A total of 1,440 patients in the lurasidone cohort were matched to an OAA patient: mean (standard deviation [SD]) age (lurasidone: 39.8 [15.37]; OAA: 41.9 [18.65]), males (lurasidone: 56.0%; OAA: 48.6%), and Medicaid-insured (lurasidone: 77.6%; OAA: 81.4%) were similar. The mean (SD) daily dose of lurasidone at index was 52.9 mg (30.64) (chlorpromazine equivalence 330.7 mg [191.50]), which was greater than any other OAA (median: 227.2 mg; range: 126.7-310.8 mg). Incidence rates of important identified and potential risks were not significantly different, except drug interactions (lurasidone: 0.038, OAA: 0.059; P=0.031) and third-trimester pregnancy exposure (lurasidone: 0.030, OAA: 0.005; P=0.005), respectively.

CONCLUSIONS: This study adds to the precedent of using secondary RWD from the US to fulfill an EMA PASS requirement. Among insured patients in the US diagnosed with schizophrenia, incidence rates for important identified and potential risks were similar between patients initiating lurasidone and OAAs.