So I’ve been invited to be on the faculty of a lung cancer conference in Kauai next month (yes, a good gig, but this is the first year that the flights are so expensive that I can’t bring my family to this normally very family-friendly event), and my topic is to argue in a debate about whether molecular testing for EGFR should be routinely used in clinical practice.

In the initial draft of the agenda, I had been assigned the “Pro” side of the argument, to say that we should be ordering EGFR mutation and/or FISH tests on patients routinely to determine whether and when we should be using EGFR-based therapies. I think this was an attempt to be provocative, since they would probably know that I am not a proponent of this approach. I called the organizers and requested that I trade positions with Dr. Roman Perez-Soler from Albert Einstein College of Medicine in NYC, who is not militant but I think far more in favor of molecular testing than I am. We all agreed it made sense to switch positions, which is fortunate because I was otherwise going to present the data from the standpoint of trying to convince myself that there were strong arguments to be made in favor of routine molecular testing, then show that I couldn’t convince myself this was a good idea even when I needed to, and therefore still argue the “Con” side. Then there would be actually be two arguments against molecular testing, which is fine with me.

I’m putting together my thought process and my slide set now (we need to submit slides weeks before the meeting so that they can be printed in a syllabus book to be distributed at the meeting). First, I thought I’d outline some areas where I think there is pretty clear consensus. First, though immunohistochemistry for EGFR is the most widely available test, this approach is also done in many different ways and is the least reliably associated with better or worse outcomes with EGFR inhibitors. Second, EGFR activating mutations (in the DNA of the EGFR gene) are very highly associated with tumor shrinkage, although not in everyone with EGFR mutations and there is not a clear improvement in survival – that is, patients with EGFR mutations may live just as long with or without an EGFR inhibitor. Third, the technique of EGFR fluorescence in situ hybridization, or FISH, which measures the number of copies of the gene in the cancer cells, is less clearly associated with response than mutations have been, but this test may be useful in predicting survival for the EGFR oral agents (called tyrosine kinase inhibitors) like tarceva (erlotinib) and iressa (gefinitib), and the EGFR FISH test may be especially promising in predicting who will benefit from the EGFR monoclonal antibodies, IV drugs like erbitux (cetuximab) (see prior post).

At this point, the major oncologist groups that put out guidelines don’t recommend routine molecular testing, but the real question is whether there is enough evidence to support it that they should. My answer is no, for now, because I’d like to see at least one of the following criteria be met before I’d say there is good reason to do this testing as a standard approach.

1) A negative test should preclude there being a clinical benefit from getting the drug – otherwise, it makes more sense to give it to everyone
2) A positive test would need to produce a greater benefit if given early vs. later, if later is the standard
3) A positive test would lead you to give an EGFR inhibitor to people in a different clinical setting, such as before or after surgery, or along with chemo and radiation for locally advanced lung cancer

I’ll discuss all of these points in some detail, but my sense of the literature right now is that none of these is true. At the same time, there are a couple of additional factors, including the added time delay and cost of the tests, and the fact that we often don’t have adequate biopsied tissue to do molecular testing.

We’ll cover these topics more thoroughly, and I’ll be very interested in hearing people’s assessments. Your questions and objections will help me to sharpen my thought process in anticipation of the actual debate. More later.