Could you cite a source for wide-spread EGFr mutation status testing? The link below best summarizes why not to do the testing.
You are probably right today's ODAC rejection is not a big deal for Tarceva sale. Physicians who love it would prescribe it off-label for patients not indicated for or tolerable to pemetrexed. Physicians (eg those knuckleheads on ODAC today) who hate it would not in any case. Roche could probably scrap the Avastin in maintenance NSCLC setting from the list of potential submissions.
What EGFR Mutation Testing Does and Does NOT Tell Us H. Jack West, MD, Oncology, Medical, 08:10PM Nov 2, 2009
Medical Director, Thoracic Onc, Swedish Cancer Inst, Seattle, WA
Several weeks ago, two high profile papers appeared in the New England Journal of Medicine that highlight the predictive value of prospective testing for EGFR mutations in advanced NSCLC. In fact, I do agree that these papers, which highlight a response rate of about 70-75% and progression-free survival eclipsing a year in patients with an EGFR mutation, make a strong case for using molecular testing. In particular, the IPASS trial by Tony Mok and colleagues shows that even in Asian never-smokers with an adenocarcinoma, a group that many of us would have considered to be overwhelmingly likely to benefit from an EGFR TKI, the mutation rate is only 60%. Moreover, the Asian never-smokers who don't have an EGFR mutation didn't do especially well with gefitinib.
The IPASS trial showed that molecular factors definitely trump clinical factors in predicting who will benefit from an EGFR TKI. It also makes a compelling argument for first line EGFR TKI therapy instead of chemo (though the evidence suggests that overall survival is quite comparable if patients with an EGFR mutation receive an EGFR inhibitor as later therapy), but it highlights that patients who don't have an EGFR mutation are better served by receiving chemotherapy-based first line treatment.
To me, the EGFR mutation test is most useful for justifying a decision to give an EGFR TKI as first line therapy. It's not clearly important to do EGFR mutation testing before giving it as a second or third line therapy: the BR.21 trial already showed a survival benefit with erlotinib in a molecularly unselected patient population, and the FDA has approved this as a salvage therapy in a broad population of previously treated patients with advanced NSCLC.
EGFR mutation testing tells us that someone with a mutation is highly likely to have a dramatic and often long-lasting response to an EGFR TKI whenever they get it. Being EGFR wild type, however, does NOT mean that someone will get no benefit from an EGFR TKI. Instead, it strongly suggests that they won't get a major and prolonged benefit, but they are in the large proportion of patients who may experience a minor response or prolonged stable disease that translates to a survival benefit in the range of a few months. These are people for whom erlotinib is a candidate to be a single, if not a home run. Both docetaxel and pemetrexed have an approximately 9% response rate in previously treated NSCLC, so if you were only going for a home run, you wouldn't do much better with standard chemo.
Molecular testing is rapidly evolving, which means that we need to be careful not to over-interpret the data. I know that there are people interpreting EGFR wild type as "don't bother with an EGFR TKI, ever", and that's not what the test is for.
Roche is certainly pursuing this strategy see this phase III trial of Tarceva vs.chemo, in 1st line NSCLC patients with EGFR mutation positive (completion expected end of 2010).
Market Data 
Markets 
AI
