P3 -- It is bad science to analyze subset data from a trial, especially one with small patient enrollment numbers. End of story. This is the downfall of too many cancer drugs, the companies that develop them and the investors that lose money. So, the shame is with people who are knowledgable and supposedly statistically savvy continuing to hype the overinterpretation of data.
I did not say that Howard's thesis was correct. I am pointing out that comparing the survival rates of the control patients is unsound.