However, what we want is incontrovertible data. It would be a huge kick in the teeth if s/c interim data “trended to efficacy” with a p> 0.05 but less than 0.10.