First, you're wrong about a 50 mg arm alone being sufficient. It damn well can be sufficient if the numbers are there for p value and clinical effect.
Second, what is the point of designing separate arms here in this "potentially pivotal" trial if the arm numbers doom it to fail? Why not only go with the clinically inadequate 30 mg dose instead? Why bother with a second arm? Why design a three armed trial "potentially pivotal" trial where, no matter how you report it, it can't be pivotal? How stupid do you think the biostatisticians are?
News 
Market Data 
Discover 
