Regulatory perspectives on data safety monitoring boards: protecting the integrity of data.

The use of interim analyses and data safety monitoring boards (DSMBs) can assist greatly in the timely determination of whether or not a medicine has an acceptable benefit-risk profile. Regulatory authorities regard the appropriate use of interim analyses favourably, but will consider the extent to which the conduct of interim analyses and the involvement of DSMBs may have compromised the evidence of efficacy and safety from a clinical trial. Issues of particular concern, which may potentially introduce bias, include the dissemination of interim data and the rules by which a trial might be terminated early. If data from trials which employ a DSMB are to be considered reliable and scientifically valid, it is the responsibility of the trial sponsor to demonstrate that the DSMB is set up and run appropriately and to verify that any bias introduced has had no important effect on the conclusions.