Inside the Clinic: Interdisciplinary Care for Stage IV NSCLC - Episode 3

How Much Tissue Do You Need for NSCLC Diagnosis?

Transcript:

Benjamin P. Levy, MD: In my former institution, when I sent patients to get biopsies, the pulmonologists or interventional radiologists, in addition to the interventional pulmonologists, would ask, “How much tissue do you need?” and I would just say, “A boat load. I just need a lot.” Unfortunately, or fortunately, we need a lot of tissue for the different testing that we’re now doing. And, Andrea, I just wanted to get your thoughts. For better or for worse, tissue is precious and that’s why we get as much as we can because we need to do certain testing on that tissue. So, can you talk a little bit about the testing we do for lung cancer tissue and how we prioritize this as we get it and you receive it, and then you execute kind of the next step, which is getting all of the different tests done?

Andrea Richardson, MD, PhD: Well, from my perspective, a surgical sample is a boat load of tissue. A cytopathology sample is a tiny, tiny, tiny bit of tissue. And one of the advances over the last few years is the molecular pathology laboratories have gotten much more efficient at extracting the DNA from those small amounts of cells and are able to do full panel sequencing. We can take a section and stain it for PD-L1, which is the biomarker for a lot of the checkpoint immunotherapy drugs. And together with the point mutation sequencing for EGFR and other oncogenes, we are now able to get a more broad tumor mutation burden measure that is looking like it’s a good predictor for response to the immunotherapies.

Benjamin P. Levy, MD: Yes, so the list just keeps growing of all the genes that we need to be testing for: EGFR, ALK, BRAF, ROS, RET, HER2. We now have PD-L1 as a bona fide biomarker, which may or may not be here to stay. And then, as you mentioned, tumor mutation burden is coming down the pike now as a potential biomarker. How do we do this? You mentioned this. I guess with comprehensive genomic profiling, we can do it all in one big fell swoop. We don’t need to do it sequentially. But how much tissue do you really need? When you’re sending those samples up to the molecular pathologist, how much tissue do we really need?

Andrea Richardson, MD, PhD: They can do it from as little as 50 cells.

Benjamin P. Levy, MD: Is that 2 slides, 3 slides, 4 slides, or even 1 slide?

Andrea Richardson, MD, PhD: That’s one-tenth of 1 slide.

Benjamin P. Levy, MD: Wow, OK.

Andrea Richardson, MD, PhD: So, usually we will get in the order of 500 cells. It’s more than enough to do the studies. Most of the laboratories now have multiplexed all of the sequencing, so that in one sequencing run, you have the primers for all of the different oncogenes of interest and you sequence all of them all at once.

Benjamin P. Levy, MD: Right. And then the other thing that we’ve wrestled with at our center is, do we test everyone? Do we test early-stage as well? Will insurance cover it? I remember meeting about a year ago and sitting down and saying, “We should really just test everyone, given the clinical trials that we have.” Do you think this is the best way? Is there a best way institutionally to make sure that molecular testing gets done? What’s the message to the community out there?

Andrea Richardson, MD, PhD: Well, I think that it’s an important team decision. You and I have talked about it. The issues that come into play are the lag time between diagnosis and a patient starting treatment. If we have it standardized so that the pathologist knows to send every lung cancer case for molecular testing, those results get turned around much faster than if we are waiting to hear from you after the patient has seen you and then you have to wait another 2 weeks to get the results.

Benjamin P. Levy, MD: That is, I think, the main reason I found, in my former institution, that if the testing is driven by the oncologist making that request, there’s a significant amount of turnaround time that it takes. I have to call you and you then have to set it up, rather than once Andrew procures the tissue, it goes to you and we reflexively send it out. I think that that has made drastic changes in how quickly we receive the result. Because, as you know, most of our decisions now with lung cancer are based on these results. I’m sure you could provide more perspective than I could. But it used to be just non—small cell lung cancer, and now it is adenocarcinoma and squamous cell. And I think we’ve moved into that every patient at our center, at least those patients with an adenocarcinoma, gets tested for genomic profiling and PD-L1. And any never-smoker that’s a squamous cell, we try to test for genetic testing as well as a PD-L1. I don’t know if there’s a right way. I think we have done a nice job of having an institutional reflexive testing comprehensive for all patients. And if they’re early-stage, you could make the argument, “Well, they don’t necessarily need that testing done.” But we have a lot of clinical trials, and there are a lot of clinical trials nationally that these patients could be referred to. So, I think this is another example of a team approach, as we try to get the information we need to make a diagnosis.

Transcript Edited for Clarity