Question for you, did SNTA qualify "largely driven"? Did new mets constitute 30%, 50% or 70% of PFS events?
No they did not give any numbers - and although I don't any longer remember Safi's exact wording, the word 'largely' is probably mine. The point is that progression was driven by new mets to a much larger extent in SOC arm than it was in the Ganetespib arm. Whether that is 20% vs 5%, 35% vs 10%, or ... I am not sure it matters.
This phenomenon of progression event being largely driven by new mets has not been reported for docetaxel monotherapy. I can see it's confusing if the bolded text is omitted.
Do you have a cite? As I noted before I looked for this kind of data long ago and it appears that the ratio of progression driven by new mets is, at best, rarely reported.