Major discrepancies concerning risk-benefit assessments and regulatory actions are frequent among regulatory agencies. We explored the differences by scrutinizing a case of gemtuzumab ozogamicin (GO) in patients with acute myeloid leukaemia (AML). Assessment reports of GO were retrieved form the websites of the US Food and Drug Administration (FDA), the European Medicines Agency (EMA) and Japanese regulatory agency, and we also reviewed published clinical trials. While GO was approved by the US FDA under the accelerated approval program in 2000, it was withdrawn from the market in 2010, based on the required post-marketing commitment failure. The EMA refused granting marketing authorization for GO in 2008 on the grounds that there were no randomised controlled trials (RCTs). GO was approved as an orphan drug in Japan in 2005, and the Japanese regulatory authority decided to continue with the approval in 2010 on the condition that post-marketing surveillance is strengthened. Under these situations, promising new results of RCTs appeared in 2011, and the role of GO in AML treatment was refocused worldwide. The stringent regulation may not be suitable in case of an orphan drug of targeted therapy, and more room should be kept to facilitate effective developments of new anti-neoplastic agents.