The study I was just discussing, the French trial of Iressa at 250 mg daily for advanced BAC (abstract here), provided interesting clinical information, especially when viewed in the context of previous work on EGFR inhibitors in BAC. But in 2007 we're also interested in the next generation of questions, including trying to identify which patients are more or less likely to benefit from Iressa or other EGFR tyrosine kinase inhibitor therapy. In addition to the clinical portion of the trial, the French investigators evaluated several clinical, pathologic, and molecular variables that were associated with disease control (DCR: response or stable disease) vs. progressive disease (PD) in a separate reported part of the study (abstract here).

From the 88 eligible patients enrolled on the BAC trial, they had tissue submitted from 65, of whom the expert pathologists felt that 50 had BAC or adenoBAC, evenly split between mucinous (M) and non-mucinous(NM) BAC, while the others didn't have tumor tissue in their submitted specimen or had an adenocarcinoma that the experts didn't think could be called BAC (this is typical -- expert reviews of pathology submitted as "BAC" from various hospitals often show high rates of disagreement, with a less strict definition in the "real world"). Not suprisingly, the patients with tissue submitted, like those on the trial in general, had a higher rate of non-smokers (>40%, and from France, no less!), and more than half of the BAC patients were women (pretty much the only lung cancer setting where we see this). The tumor tissue was tested for EGFR by protein expression (immunohistochemistry, or IHC), gene amplification (by a process called FISH, and another called CISH), and also for EGFR mutations; they also checked for ras mutations, which I described in a prior post as being likely associated with a lower likelihood of benefit on EGFR inhibitors. Finally, they checked for thyroid transcription factor-1 (TTF-1), which is a marker of thyroid and lung tissue that helps us determine whether a cancer is actually from the lung or thyroid vs. another part of the body (they can tell the difference between lung and thyroid from other protein stains in the unusual cases where there's a question between those sites as the primary tumor site). About 70% of lung adenocarcinomas express TTF-1.

First, the investigators compared the M-BAC to NM-BAC tumors and found differences in several regards. Although there were no gender differences and never-smokers were found in similar proportions between the two types of BAC, NM-BAC was much more likely to be associated with TTF-1 expression, EGFR protein overexpression (by IHC; about 35%) and gene amplification (by FISH and CISH; 10% range)) and EGFR mutations (12% of the population) than M-BAC tumors. They didn't differ in their frequency of ras mutations (about 1/4 of both groups). So they have some differences that might explain differences in how the different types of BAC tumors respond to EGFR inhibitor therapy.

And when they looked at the characteristics of the patients who achieved disease control vs. those who showed PD, they saw that the patients who had stable disease or better were significantly more likely to be women, never-smokers, have NM-BAC, have a tumor that expresses TTF-1, and also a higher likelihood of having an EGFR mutation. In contrast, those with ras mutations were more likely to be the ones who showed progression. Here's the summary:

(Click to enlarge)

This isn't a large enough study to say anything definitive, but it's a step forward in giving us hints about biological differences between M-BAC and NM-BAC, and it also helps provide some insight about why people with NM-BAC may be more likely to respond better and have longer survival on Iressa and Tarceva trials. In addition, this biological information may be useful outside of BAC. Perhaps the patients with TTF-1 positive tumors are the ones more likely to respond to EGFR inhibitors. We've never really looked at that, but that marker is a routine part of testing lung tumors. It's readily available everywhere, and it doesn't take days or weeks to obtain, unlike the mutation work. There are several other interesting leads here, so we need to follow up and see what holds up in other studies of EGFR inhibitors.