>> what happens to osip survival rate p-value, if like irressa, long term survivors have a mutant egf'r. If you remove that subset of patients from the traveca trial, is the p-value still significant? <<
The Tarceva trial did not screen for the mutant form of EGFR –such a screening tool is not even available yet. Moreover, the Tarceva trial did not test tissue samples for the mutant EGFR, because the research about the mutant EGFR was just published and is still very new. Hence, there is really no way to adjust the p-value calculated from the Tarceva trial to take into account differential responses to the normal and the mutant forms of EGFR.
In future clinical trials of Tarceva and Iressa, differential responses to the two forms of EGFR could be (and probably will be) incorporated into trial protocols.
P.S. The reply to your query by ‘eventhorzg’ in #14223 on the Yahoo OSIP board is well worth reading by anyone with an interest in this topic.
“The efficient-market hypothesis may be
the foremost piece of B.S. ever promulgated
in any area of human knowledge!”