Benjamin Levy, MD: Let’s wrap up here with rare genotypes. Again, EGFR and ALK are the beginning of the story. I think we’ve said quite nicely in the first section how important it is to do comprehensive genomic profiling to test for these rare mutations. We have drugs now, and I think when you find one and you give a drug, you understand how important it is to identify them. Some of the data we had today was on MET exon 14 skipping mutations. We had a couple of drugs that were new, at least to a lot of us, in terms of initial activity and safety. Josh, do you want to walk us through the capmatinib and tepotinib data from today?

Joshua Bauml, MD: Absolutely. MET exon 14 skipping mutations are a mutation in the intron that modifies the splicing, and so you end up losing exon 14 when it is expressed. And that can make it difficult to identify these on some DNA-based NGS [next-generation sequencing] assays. But historically, at least in America, we’ve been treating many of these patients with crizotinib. It’s important to remember that it’s in America that we’ve been doing that because our colleagues outside of America, in Europe and other areas, have not had access to crizotinib for this disease. And so capmatinib and tepotinib are being presented and there are no data on patients who were previously treated with crizotinib. So how to apply these data to what we’re doing in our practice is a little bit difficult.

But what we saw with these agents are that both capmatinib and tepotinib were associated with good response rates and toxicity profiles for patients with MET exon 14 skipping non—small cell lung cancer. They both had responses in the CNS [central nervous system], so that is another distinction. And I think one of the things that does need to be remembered about MET exon 14 skipping mutations is that these tend to happen in different patient populations than some of the other ones. So similar to BRAF, these mutations can be seen in patients with a rather heavy smoking history. And they also tend to occur in older persons.

One of the interesting things about both the data from capmatinib and tepotinib is that if you look at patients who are treatment-naïve, or patients who are treated in later lines, there was a substantial drop off in the objective response rate for those patients, which is atypical when you consider the fact that they had not received a therapy of this type before.

One of the thoughts that was put forth by Dr Karen Reckamp, MD, today during her presentation is that perhaps because these patients are older, they’re just not strong enough to really get to that later response. But it is supporting the idea that we need to do this testing early, not limiting it to only patients who were never smokers, and make sure we get patients access to these sorts of drugs.

Robert Doebele, MD, PhD: To weigh in on the smoking issue, although it’s higher numbers than we see for ALK and EGFR with the smoking, remember that we see a 40-pack-year smoking histories in our EGFR and ALK-positive patients too.

Joshua Bauml, MD: Absolutely.

Robert Doebele, MD, PhD: So, yes, we’re seeing a higher rate in the MET just like we do in BRAF, but really all patients potentially have these mutations.

Joshua Bauml, MD: We’re in agreement that everyone should be tested. But it’s just a matter, I think it’s interesting actually. If you think about someone who has a greater than 25-pack-year history of smoking, their likelihood of having an EGFR mutation is less than 5%, which is about the incidence of MET. It sounds like a pretty much equal chance of having MET or EGFR at that point.

D. Ross Camidge, MD: There was a really cool study from Memorial Sloan Kettering Cancer Center. Do you remember that one? If you look at the absolute number of EGFR mutants in smokers and never smokers, it’s the same because it’s a lower frequency and a bigger number.

Benjamin Levy, MD: Well speaking of testing, I want to clarify a few things. You know, the difference between MET exon 14 skipping mutations versus MET amplification. Ross, you explained this to me before the panel. I want you to explain it to me again. Sorting out what is the predictive biomarker that we need to look for, and what’s the difference here between amplification and MET exon 14 skipping mutations?

D. Ross Camidge, MD: There’s evidence that both can be independent drivers of a MET oncogene addicted cell. You can have a MET exon 14 in the absence of gene amplification, and they respond, and you can have high level MET amplification, in the absence of a MET exon 14 skipping mutation, and they respond. But there is some overlap. And importantly, among the group that’s overlapped, you don’t have to have both, it doesn’t change your response. There’s a certain redundancy there. And about 20% of MET exon 14 have some level of amplifications.

Benjamin Levy, MD: Do people test for amplification in their NGS? Has that come back in their in-house NGSs routinely?

D. Ross Camidge, MD: You have to do FISH [fluorescence in situ hybridization] for the same challenge of, it’s even more challenging in the blood, but NGS assay’s ability to quote amplification accurately I think is challenging. It obviously depends on the assay you’ve got in your bioinformatics.

D. Ross Camidge, MD: Do you still do FISH, or do you NGS for the amplification?

Zofia Piotrowska, MD: We do NGS. I would say it’s probably variable. I often do get it, largely because I also get it on my EGFR-resistant cases, so the checkbox comes in naturally. But I do think it picks up some additional patients. Not a lot, but it has the potential to pick up additional patients who may benefit.

Benjamin Levy, MD: We do FISH too, I just don’t know what to do when I get that result back that says: MET amplified.

D. Ross Camidge, MD: Well you get a look at the ratio to CYP7.

Benjamin Levy, MD: What’s the cutoff that would...?

D. Ross Camidge, MD: Probably about, well, as a primary driver, probably about 4. In the setting of acquired resistance, because it’s a subclone, that number will appear lower because it’s diluted by some other cells. It’s probably 3 in the EGFR-mutated setting.

Transcript Edited for Clarity