Expanding Targets in NSCLC - Episode 5

Molecular Testing Approaches in NSCLC

Transcript: Bruce E. Johnson, MD: It’s critically important for patients, at their time of diagnosis with advanced non—small cell lung cancer, particularly nonsquamous non–small cell lung cancer, to undergo a broad next-generation sequencing panel. And that’s for 2 reasons. One is, it gives some information about things you can potentially target. Of our patients who get conventional chemotherapy versus those who get targeted agents, the vast majority of them prefer to get the targeted agents. You can take a pill once a day and have relatively modest adverse effects. Your expectation is that things are going to go on for years rather than for a few months to a year or 2 years. So that becomes critically important.

The second thing about next-generation sequencing panels is that it also gives some part of predictability about how likely they are to respond to checkpoint inhibition. The more mutations, the more likely they are to respond. So it also informs that.

Another part of that, though: To do the next-generation sequencing panel, you have to make sure that there’s a big enough biopsy that you can characterize the tumor. For those of us who work in referral centers, as we work with our interventional radiologists, pulmonary physicians, and thoracic surgeons to make sure we have adequate amounts of tumor tissue, we have to be sure that we get adequate information about the next-generation sequencing panels.

In addition, there are a number of immunohistochemical tests that are important, that you need to have. You prefer to have the tissue and look at both the tumor and the normal pairs. So you need to have enough tissue from each patient. The amount of cancer tissue that you need to accurately characterize a tumor for modern-day treatment means you have to have adequate-size biopsies.

There is advice for the nonsquamous non—small cell lung cancer patients that recommends a large next-generation sequencing panel. From my personal experience, we work with the people who do the billing, and it’s a challenge to get these covered. And so it is 1 of the barriers that we have characterizing all the patients. And in the US, where we have a variety of payments and sources, there are different polices for different companies. Some think it’s important and cover it. Others think it’s not important and almost never cover it. And so it’s a bit of an unknown about how well you think this is going to work, and if the patient is potentially stuck with a bill after it’s done.

The ability to molecularly profile circulating DNA has expanded our diagnostic capabilities. One of the things that we’ve learned by doing this is that when you pick it up in the plasma, by and large, it’s almost always present in the tumor.

There are a few cases—we assume this is because the tumor DNA can be shed from either all the different tumors or some of them—for which you biopsy and there may be some heterogeneity. But by and large, the circulating DNA reflects what’s going on in the tumor.

The part that’s a challenge is that the patients who have the amount of circulating DNA that’s easy to measure are the ones with pretty advanced tumors. And so if you have a low-volume tumor, in general, there’s less circulating free DNA. Therefore, it’s an increasing challenge to be able to detect the mutation.

The part that is very helpful is if you get a positive result. That is, if you test the circulating DNA and you find 1 of the common oncogenic drivers for which there’s a targeted therapy—which includes the epidermal growth factor receptor mutations, the ALK rearrangements, the ROS rearrangements, the NTRK rearrangements, and the BRAF mutations—you can be pretty confident that it’s present in the tumor. But the problem is, depending on how sensitive the technique, you’ll end up not being able to detect a mutation in up to half the patients.

Therefore, the testing is not complete. One of the things for me, personally, is that on all my patients, I try very hard to be able to get the tumor tissue for testing. In addition, if you don’t have enough for the next-generation sequencing panels, you also don’t have enough for PD-L1 [programmed death-ligand 1] testing and can’t really do PD-L1 testing on circulating free DNA. So 1 of the other important markers that we have is not available.

I believe that the standard method of detection thus far should be a biopsy and doing next-generation sequencing on the tumor. From a practical application that we’ve known from studying our patients, in somewhere between 5% and 20% of patients the tumor is not very accessible. For instance, if you’ve had a surgical resection for which you remove the primary tumor and then you turn up with brain metastasis later, it becomes quite a challenge. We rarely will go in and take out somebody’s brain lesions, unless there’s a single 1 that turns up—1 or 2 brain lesions. Sometimes you’ll surgically resect them. But if you end up with multiple brain metastases, you usually don’t end up biopsying them. And so there are sites, and it can be critically important to be able to measure that in that particular clinical situation.

You’re going to encounter times when it’s very difficult to be able to get an adequate biopsy, and that’s when being able to test the circulating free DNA can really come in handy.

The other part that happens too, that we run into once in a while, is that the lesion can be very close to the heart. A lot of the lung cancer lesions arise centrally, and sometimes it can be very dangerous to go in and biopsy a lesion that’s close to the heart.

Transcript Edited for Clarity