Drug-eluting stents (DES) were first approved in the US in 2003 based on randomized controlled trials comparing DES to bare metal stents. Since approval of the first generation DES, the strategy for showing safety and efficacy of a new DES has changed and concerns about potential late safety events have surfaced. Therefore, the requirements for approval of a new DES have changed. One company’s experience with designing trials for a first, second and third generation DES, including the choice of endpoints, sample size calculations and interaction with FDA statisticians, will be discussed.