Atypical femur fractures associated with use of bisphosphonates and denosumab.

progression [2]. In addition, the time gain is very large in cancers for which the response rate is used as a surrogate end point such as hematological malignancy [3]. Second, we are not discussing a surrogate end point for less serious disease like cardiovascular disease [4], but are discussing a surrogate end point supporting clinical trials of innovative new drugs of high unmet medical need used for treating patients with serious life-threatening diseases. The emphasis is on the importance when a validated surrogate end point of clinical benefits is proven to be useful in such a situation. In the FDA’s accelerated approval program, clinical benefit was confirmed in post-approval trials for 26 of the 47 indications. It is ‘not only’ 26; patients could benefit from earlier availability of the drug with more than half of new indications (26 of 47 indications) [5]. As Dr Braillon pointed out in the last, validation of surrogate end points is of course important, and decision on the acceptance of surrogate end points must be made strictly. Surrogate end points should carry the risk of error. Apart from this, we do not find exactly equivalent system to FDA’s accelerated approval or breakthrough therapy designation in our domestic regulatory environment, there is only a limited place to discuss surrogate end points in Japan [6]. We understand his approach ‘Festina Lente’ he quoted, but we would prefer ‘never hesitate to do good and make haste to do what is right’.