Refining EGFR Classifications and Impact on Treatment for NSCLC

Transcript:

Estelamari Rodriguez, MD, MPH: Dr Levy, I went to the recent Texas Lung [Cancer] Conference, where Dr [John] Heymach presented this very interesting topic of exon-based grouping classification for EGFR mutation. It opened my eyes to thinking outside the box about a very common mutation like EGFR, which we’ve tested for a long time. We’ve made decisions based on the report that we get on how the trials were done. But can we think about it differently? Can you comment on this concept of exon-based grouping classification and how we can use that to pick treatments for patients?

Benjamin Levy, MD: We’ve always looked at EGFR through the prism of exons. You’ve got exon 19 and exon 21, which are sensitizing mutations. You have the uncommon mutations, which are point mutations in exon 18 and exon 21. Then you’ve got exon 20. We’ve always looked at this as, “This is the exon. How do you treat it?” What Dr Heymach et al from [The University of Texas] MD Anderson [Cancer Center] has shown is that there’s extraordinary heterogeneity within the exons. Exons may not be able to predict best who’s going to respond to which tyrosine kinase inhibitor [TKI]. They’ve come up with ways that may better classify the EGFR mutation based on this structure and front function of the tyrosine kinase protein that drives cell signaling.

What they’ve found is that doing a structure-function-based classification may be able to better predict sensitivity to TKIs. I can set the stage, and maybe you can go into a little more of this. I’m happy to discuss this with you as well. But it’s an outside-the-box way of thinking about things. Can this be recapitulated in everyday clinics and our molecular reports? I don’t know. But it’s a way to say, “This structure classification is better for osimertinib.” We’re moving the field forward and thinking, “Let’s not think about just exons. Let’s think about classifying these based on sensitivity TKI.” This is perhaps more meaningful to patients in terms of how we treat them. These data were presented on refining the EGFR classification. I’ve set the stage. What are your thoughts on the data, and are there implications to the study that we can move forward with?

Estelamari Rodriguez, MD, MPH: What I took away from it is that we’ve been stuck in this EGFR world of doing clinical trials for the most common mutations. We’ve done that for a long time in the metastatic setting. We’re doing that in the way we test patients for neoadjuvant and adjuvant treatments. We’re still in these 2 common mutations. They do the best, and those are the ones we want to treat because our trials will show positive results. But we’re leaving a lot of patients who have uncommon mutations, like EGFR [exon] 20 insertions. In this series we saw that the responses and the time-to-treatment failure were very different. Patients who had these common mutations—the classic mutations del19 atL858R—had a median survival in this series from MD Anderson and Moffitt [Cancer Center] of about 16 months. When they [have] the atypical mutations, they do worse. The next step is to find out how can we do better.

This is what the data will push us toward. If you have an atypical mutation, like exon 20 orexon 18, can you select a treatment based on the structure of this mutation? Can you select a better drug? What opened my eyes is that we can start thinking about trials for these patients. We’re excluding patients because they haven’t done well, [but that’s] because of the way we’ve gone about it. We concentrate on the mutations that we understand [because] the drug will bind and respond. When this structure of the tyrosine kinase changes, the drugs don’t attach as well. That’s the way I think about it: can we select better drugs for these patients?

More important, can we include patients in different trials who will use that information to the advantage of selecting drugs? They found that exon 1and exon 21 mutations—hundreds of mutations—can be exploited if we know exactly how they change the structure. Then you can select drugs that will do better. Afatinib is 1 of those drugs that has been tested in trials. They’ve seen a signal that this drug may do better with some of these mutations. But it has to do with how the drug does with the pocket, how it alters the alteration of the P-loop and the C-helix. It tells you about how the drug will attach and how effective it will be.

Benjamin Levy, MD: It’s a fascinating paper. It was published in Nature a year or 2 ago, looking at things through a different prism. There are 4 subtypes that they talk about: classical-like, T790-like, P-loop-compressing-like, and an exon 20 loop insertion. We’re not familiar with these terms yet. I don’t know if we can recapitulate this in everyday practice. But when you start to think about clinical trials, this may be the way to go. I’m excited to see where this heads.

Estelamari Rodriguez, MD, MPH: Most of our trials, especially looking at the evolution of treatment for EGFR, have concentrated on common mutations, and we’ve found better drugs like osimertinib for those common mutations. But what’s the difference between uncommon EGFR and common EGFR mutations? How do you use that information to select treatment options for patients?

Benjamin Levy, MD: You framed it nicely. We know that [with] exon 19, not all EGFR mutations are the same, which makes it complicated. Look at the molecular report; it’s a challenge. Not all EGFR mutations are the same. Exon 19 and exon 21 are the most common—80% to 90% of all EGFR mutations are those sensitizing mutations. Maybe a little less, 75%. Then you get into this sliver of the pie, these uncommon mutations, depending on where the mutation occurs along the gene. This is relevant clinically because some of these uncommon mutations may not respond to drugs that we routinely use on exon 19 and exon 21, most commonly osimertinib. The ones we think about are G719X—that’s a very common 1—L861Q, and S768I. These are the ones you’ll see. You have to look these up online or consult a friend, somebody in the field who can guide you on this because there have been a lot of data on approaching these a little differently. It’s important to think of uncommon EGFR mutations as still druggable. You’ve got to look at this through a different lens.

Transcript edited for clarity.