Benjamin P. Levy, MD: Let’s move on to the last session. We’ve talked a lot about different genotypes wedded to targeted therapies and this whole concept of precision medicine. Not all non—small cell lung cancer is 1 disease. We’ve learned that it can be parsed out into more and more genotypes. We’ve learned, even from the LCMC data, that patients with a genotype who get a targeted therapy live longer than patients with a genotype who don’t.

Zofia Piotrowska, MD: Absolutely.

Benjamin P. Levy, MD: There’s a wealth of information and a groundswell of drugs. We’ve really moved into this era where we are really understanding that the genomic characterization of a tumor in lung cancer informs us on targeted therapies. Obviously, not every patient is going to get a targeted therapy, but there’s a growing list of these mutations. I think it’s nice that we end up trying to decide how we should collect this information. What are the best practices for molecular testing? I don’t think we should go center by center, but Zofia, maybe you can start us off? How should this be done? The data now are too good for it not to be done.

Zofia Piotrowska, MD: It’s true. What’s really important from all these different mutations and rearrangements that we’re seeing is that we really need to be sure that we’re casting a broad enough net with our testing to capture these different rearrangements and alterations. Ultimately, I think that NGS with a broad panel that covers all these different genes of interest is essential and important.

On a practical level, NGS testing can take some time to do. Our NGS panel can take a couple of weeks to return. That can sometimes be difficult. PD-L1 testing comes back early. Patients are anxious to start treatment. Physicians are anxious to start treatment. I do think it’s important to wait for those data, if at all possible. We have patients at our center who are never-smokers, who have some of these features that make these mutations more likely or probable. We do have some more rapid testing that can be done to look for the kind of lowest-hanging fruit. For example, we have a rapid EGFR PCR—based assay that comes back in just a few days. We have ALK and ROS1 IHC so that you capture those larger pieces of the pie early. But ultimately, I think NGS testing is really going to be the gold standard. It is going to help us capture these mutations. It can easily be modified to add future genes of interest as well.

Benjamin P. Levy, MD: It sounds great in practice, but what are some of the hurdles? What are the hurdles that are out there in the community, or even in academia, in terms of getting tissue testing done? Jonathan?

Jonathan W. Riess, MD, MS: Zofia alluded to some of them. We do PD-L1 testing in-house, but we typically send out for the next-generation sequencing platform, for broad genomic profiling. That can take some time to come back. It can take some time for the pathologist to send out the tissue. Ideally, the guidelines talk about having results within 2 weeks. The turnaround for the NGS panel, as well as time for it to receive the tissue, insurance authorization, and so forth, can sometimes be limiting. For patients who are asymptomatic, or are minimally symptomatic, who can wait, my general approach has been a broad next-generation sequencing profile on tissue. For patients who cannot wait, who are never-smokers and have a high probability for having EGFR, ALK, ROS1, I will often send for EGFR PCR, ALK IHC or FISH, and ROS1 FISH to try to get those results. I have found, particularly with the FISH and IHC tests, that you can get results much faster.

For example, I had an ALK patient who was in the ICU with pericardial tamponade. ALK patients seem to be associated more with having effusions—pericardial and pleural. We had an inkling that she could have an ALK mutation, so we sent that off. We harassed pathology to send the tissue out, and we had a result in less than 48 hours. We were able to get her on alectinib, and she did great. That’s an example of a situation in which I’ll deviate from broad genomic profiling.

Sometimes you wind up in a situation where there’s a bone biopsy and you can’t get a good NGS panel because of the decalcification process. In a situation like that, I’ll try to get plasma initially. Sometimes I can make decisions based on that. Of course, a negative plasma test doesn’t necessarily rule out that they won’t have anything. But if it’s positive for something, that can really help you as well. So those are some special situations where I may not always step to the broad genomic profiling first, to try to accelerate things.

Alexander Drilon, MD: We’ve talked a lot about whom we think about screening with NGS, and we mentioned the never-smoking, younger patients. One striking thing about one of the newer drivers that we’ve learned about is that MET exon 14 alterations are almost comparable in frequency to ALK rearrangements. However, these tend to occur in older patients—a median age of greater than 70. And there’s a higher proportion of smokers.

Many of us are moving toward a paradigm where we’re not being biased in terms of clinical features. We are broadly testing lung cancers. Just to sum up the importance of that, crizotinib can work well for these MET exon 14 alterations. It’s in the NCCN guidelines. There are now newer drugs, like tepotinib, in which we’re seeing response rates greater than 60%.

Transcript Edited for Clarity