On a Vertex call (my memory is not that good but I believe it was at the SIG conference) they pointed out some reasons why they believed Boceprevir results were cherry picked (my words) because of the way they prescreened.