Perhaps it is best to wait for the actual data.
If the baseline survival in the Tarceva trial was 4-5 months as you recite, then for the trial to take this long to accrue the necessary events (estimate late jan/early feb from CCs), you'd be looking at a hazard ratio of 0.5 or less. In any case, I believe it is relatively clear that the trial did not conform to the 4.5 month vs. 6 month structure.
As for the value of such a drug.... it all depends on the quality of life. At this terminal stage of cancer, my opinion is that the best thing that can happen to patients is that they stay out of bed longer. Once they hit the bed, it's game over. So if Tarceva gives them an extra month or so before they hit the bed... then, well, you decide what that is worth to you. No one is forced to take and pay for Tarceva. If this was a straight chemo regimen, then i'd say that the clinical relevance would be of very limited value. But it's not.
However, this was a well designed and, apparently, a well executed trial. Perhaps it is more and indictment on where we at with the "war on cancer", rather than OSIP/DNA/Roche trying to milk patients for $15-25K for another month of life.
