EGFR & ALK/ROS1 Testing in Squamous NSCLC

Transcript:Mark Socinski, MD: Hello, and thank you for joining this OncLive Peer Exchange® titled “2017 Update on Metastatic Squamous Non—Small Cell Lung Cancer.” Advanced squamous cell carcinoma of the lung is a particularly challenging disease to treat and is further complicated by factors such as patients’ smoking history, comorbid disease, and age. Recently, our research efforts have yielded important advances, adding a significant number of therapeutic options for patients now being treated through multiple lines of therapy. In this OncLive Peer Exchange®, I am joined by a panel of experts in lung cancer medical oncology. Together, we will highlight the most important studies, including those from the 2016 World Conference on Lung Cancer, and provide insight on how to interpret the new data.

I am Dr. Mark Socinski, and I am executive medical director at the Florida Hospital Cancer Institute in Orlando, Florida. Participating today on our distinguished panel are Dr. Tracey Evans, associate professor of clinical medicine at the University of Pennsylvania in Philadelphia; Dr. David Jackman, assistant professor at the Harvard Medical School, medical director of Clinical Pathways at Dana-Farber, and senior physician at the Lowe Center for Thoracic Oncology in Boston, Massachusetts; Dr. Edward Kim, chair of Solid Tumor Oncology and Investigational Therapeutics and the Donald S. Kim Distinguished Chair for cancer research at the Levine Cancer Institute in Carolinas HealthCare System in Charlotte, North Carolina; and Jared Weiss, section chief of Thoracic and Head and Neck Oncology at the UNC Lineberger Comprehensive Cancer Center in Chapel Hill, North Carolina. Colleagues, thank you for joining us today. Let’s now begin.

Ed, I’m going to start with you. Certainly in the past decade or so, in the entity we’re now talking about today, nonsquamous to be adenocarcinoma has brought with it a number of different testing issues and strategies; but it’s a bit different in squamous. There are a number of issues with establishing the initial diagnosis. How confident are you that it’s squamous? There still is room in selected patients for testing. I’d like to get your thoughts in terms of how you think about that.

Edward S. Kim, MD: We go through a process in our institute where we try to discuss any changes to our clinical pathways. And we do that in monthly meetings so that we can get alignment across our system. Even with the group I worked with before, at MD Anderson, there were 20 medical oncologists who did thoracic oncology, and we never audited charts. But I would find it hard to believe that all 20 did the exact same thing in clinic. And so, this is the struggle we try to achieve in a department, in a system, and now across the country. I’m not even going to go overseas on this; we like easy things. You get a breast cancer specimen, reflexively the pathologist knows to do HER2, ER, PR, and that’s easy.

In head and neck, it’s becoming HPV. But in lung cancer, we’ve got the non—small cell versus small cell. And non–small cell then had to get divided up into nonsquamous versus squamous. Even within nonsquamous, there are certain histologies that you may not want to do testing for. What we’ve seen is really great. I love the fact that we have EGFR mutations that we test, and we test those in our adenocarcinoma subsets. We will also look at less smoking histories in squamous. Now, that’s very hard sometimes for a pathologist to discern because it’s not necessarily in the requisition. So, we really have to rely on our physicians to specifically ask for that. And we won’t reflexively do that because it’s tough when you have tissues at the path coming from pulmonary or other areas.

We like to test for ALK/ROS1 and EGFR mutations as part of the standard biomarker testing diagnostic workup in patients with nonsquamous and specifically adenocarcinomas, regardless of smoking status. In the exception with squamous, in those patients who have either light- or never-smoking histories, we will run those tests as well. PD-L1 has brought that together. And I know we’ll discuss some of this later, but now it allows you to talk to every non—small cell lung cancer patient and say, “We want to test biomarkers for you, and this may be an important one to also test.” So, this is how we’ve put it together.

There are also local labs that do these tests. There are larger commercial labs that run these tests in a wide variety of numbers of genes—anywhere from 50 genes to 600 and more genes. We do like to run large panels, but, again, it’s not as easy because these are not necessarily all FDA-approved tests, and so reimbursements can be different. That’s one of the challenges that we have with wanting to run larger up-front genomic testing. So, our standard of practice is just to run the ones we need for diagnosis—meaning the EGFR mutation, ALK translocation, ROS1—in selected squamous patients, but definitely the PD-L1 in all of our patients who have squamous histology.

Mark Socinski, MD: I’m commonly asked by community oncologists about this issue of mixed histologies. There’s rather limited biopsy, and the pathologists may wax and wane about some adenocarcinoma and some squamous features. What’s your perspective on the mixed histology, where that fits in?

Edward S. Kim, MD: Yes, I hate seeing those, too. I hate seeing mixed histology, and I hate seeing very small amounts of tissue or cells. Those are the 2 most frustrating parts. I think the only thing more frustrating is when you do a procedure and nothing is there; there are no specimens, and now you don’t even have a diagnosis.

We tend to err on the side of overtesting; that’s where I would like to do it. And so, if it’s one of those mixed cases, we usually will try to get the whole panel. But if it is more predominant in the squamous histology, it looks like a squamous on the scans, there’s a large smoking history, and there’s limited tissue, then we will prioritize those markers in a manner that is more PD-L1ish. But a lot of that needs the clinical input that a pathologist can give.

Mark Socinski, MD: With regard to this, how many things do you test for? I’ve heard some people provide the rationale that EGFR mutations do occur in squamous at a much lower frequency, but maybe not any lower than, say, ROS1 in adenocarcinoma. So, why not look regardless of histology? And then, the issue of how much do you look for it. How far do you cast that net? Are you going to find some of the other very uncommon alterations in which there may have been only a case report on something? How do you frame that in your practice?

Edward S. Kim, MD: It’s challenging, and that’s a really great point because you always have that case report and someone on faculty will always bring up the one esoteric case report.

Mark Socinski, MD: Yes, they always do.

Edward S. Kim, MD: What we try to do is make sure that you can be as less variable as possible. And so, overtesting is not a good thing. Then again, you might find that one here or there. That’s why we try to build in these practices where we have certain things that we think are in line with what practicing medicine in 2017 is. Because the same argument goes down a slippery slope—I want to test for KRAS in everybody, and I want to start testing for BRAF because we might be able to get an off-label drug for that. And so, that’s where we try to draw a line on what is it, FDA-approved or not, and whether there’s an actual therapy that could help that patient. That’s the general gist of the way we go. But I agree, it’s very challenging. You want to do as much as you can for a patient, but that’s also why we don’t do a 650-gene panel up front on everybody.

Transcript Edited for Clarity