Drug agencies and pharmaceutical companies are responsible for continuously monitoring drug safety profiles to implement necessary safety measures. Retrospective analyses of safety-related regulatory actions (SRRAs) are known to be helpful as a tool for evaluating postmarketing safety in Western countries. However, few studies have examined the relationship between SRRAs and postmarketing safety in Japan.Objective
The aim of this study was to evaluate postmarketing safety risk by examining cases of adverse drug reactions appended to the clinically significant adverse reactions (CSARs) section in package inserts as a marker of Japanese SRRAs.Methods
We identified new molecular entities (NMEs) approved in Japan between 1 January 1999 and 31 December 2010, and classified them into two groups: NMEs already approved overseas (Overseas group) and NMEs initially or only approved in Japan (Japan group). Each NME included in this study was investigated for CSARs appended into the package insert based on the ‘Revision of Precautions Section of Package Insert’ released by the end of 2011. The potential of exposure to unknown postmarketing safety risk was expressed as the incidence rate of appended CSARs. In addition, the Kaplan-Meier method was used for analysis of the time until the first CSAR was appended.Results
Among 315 NMEs, we identified 943 appended CSARs that were mandated by the Pharmaceutical and Medical Devices Agency. The incidence rate of CSARs appended for all NMEs was 0.42/drug-year (95% confidence interval [CI] 0.40, 0.45) and individual incidence rates were 0.41/drug-year (95% CI 0.38, 0.44) in the Overseas group, and 0.48/drug-year (95% CI 0.42, 0.55) in the Japan group. In the Overseas group, when subgrouping drugs on the basis of the length of each drug lag, incidence rates of CSARs tended to be greater for shorter drug lags. Using the Overseas group as a reference, the incidence rate ratio (IRR) in the Japan group was evaluated as 1.17 (95% CI 1.01, 1.36). Standardized IRRs for each background factor and the lower limit of the 95% CIs were greater than 1. CSARs were most frequently appended 2 and 3 years after the approval of the NME. Furthermore, CSARs were likely to be appended earlier for NMEs initially or only approved in Japan.Conclusion
The present results suggest that appended CSARs can offer a useful indicator of Japanese SRRAs. Furthermore, NMEs with little or no foreign safety information are more likely to have unknown safety risks recognized more frequently at an early stage of postmarketing. Although this study focused on Japan, unknown risk generally increases with the elimination of drug lag, regardless of location. Reinforcement of a safety monitoring system is thus vital.