Case 1: Molecular Testing in Locally Advanced NSCLC

EXPERT PERSPECTIVE VIRTUAL TUMOR BOARD

Benjamin P. Levy, MD:Sam, as far as routine testing for a patient goes, this is a patient with squamous cell carcinoma, stage III. Ignoring the histology, maybe you can walk us through if this patient should be tested at all since they are stage III. In squamous cell carcinoma, are those patients being tested molecularly? What’s your role as a pathologist in phase III lung cancer?

Samuel Caughron, MD:Sure. When it first comes to the pathologist, to Lonny’s point, we don’t know it’s a cancer yet. Our first step is still to make that diagnosis. But pathologists today have to be thinking past the diagnosis in terms of how is this patient going to be managed? Biomarkers are the future. That’s so important to many of the decisions that are going to be made about how the patient is managed.

This is a squamous histology. However, mixed tumors are out there. I think the trend is toward more reflex testing of all histologies. Then we have a number of biomarkers that are histology agnostic. ForNTRKgene fusions—although it’s low incidence as are all morphologies—as well as MSI [microsatellite instability] and a pan-tumor indication. Our approach is to be more aggressive rather than less, to get that information onto the chart so that the decisions can be made in managing the patient.

Certainly, PD-L1 [programmed death-ligand 1] testing is important. You’ve got a stage III that’s squamous histology, non—small cell lung cancer. We’re going to make sure we’ve got PD-L1 and probably prioritize that above the other biomarkers because of the squamous histology. Looking at the amount of tissue you have, I think there were lymph node samples in this. Presumably, we’ve got plenty of tissue. Then we’d have a discussion because of the squamous histology about how much additional testing may be required or may be desired. We have some hospitals we serve where we have an understanding that even in a patient like this, we’re going to reflexively do the standard hot-spot biomarkers. If those all come back negative, we kick it back to the treating physician to decide if they want to order additional biomarkers back after a conversation with the patient.

You can certainly argue that next-generation sequencing could be done right away. Lung cancer is the area where it’s most valuable and the National Cancer Comprehensive Network Guidelines support, considering and potentially encourage it. Again, with the squamous histology, you might be a little bit more conservative.

In this patient with the squamous morphology, you’re going to want to get PD-L1 testing. Next-generation sequencing may be considered. Microsatellite instability and some of the other pan-tumor markers might also be considered.

Benjamin P. Levy, MD:You mentioned reflective testing and how important it is. Has that sped things up for you guys? We’ve had challenges both at my former center and in my current center, Johns Hopkins Kimmel Cancer Center at Sibley Memorial Hospital in Washington, DC, whether all tissue should be sent off or not. And should it be reflexive, or should the medical oncologist be calling and saying, “Hey, I want this sent off?”

Has it been helpful to do reflective testing because it essentially ensures that every patient is getting testing done? And have you found that to be helpful in terms of expediting the results?

Samuel Caughron, MD:Absolutely. I’m both a fan and an advocate for reflexive testing because I think if you don’t just do it across all tumor types, or across all patients, the pathologist will forget to do it. They can get lost, so you have to have a fairly simple protocol. Also, you need to look at the turnaround time. I’m not sure pathologists are always as sensitive to the clinical timeline and the need to have that information in the decision making.

Within our institution, we were 1 of the first in our community to go live with reflexive testing in non—small cell lung cancer several years ago. And we quickly got feedback from the oncology community and from other institutions that it was great, and they wanted it done the way it’s done in our institution. Sometimes I think pathologists need to get more onboard with this. It’s an opportunity for the multidisciplinary team to come together and discuss what’s going to be done and let the pathologist know it’s important to have that information.

Benjamin P. Levy, MD:In a perfect world, we’ll have binary histological assessment. We’ve got squamous cell carcinoma, or we have adenosquamous carcinoma. But that’s not always the case, right?

Samuel Caughron, MD:No.

Benjamin P. Levy, MD:There are times it’s really tough to tell. I push my pathologist, “Can you tell me what this is? How often does this happen, and why does it happen? Are these poorly differentiated? Are the immunostains not aligning? Is this just tumor biology, that you’re going to get adenosquamous carcinoma or poorly differentiated 10%, 15%, or 20% of the time?” Walk us through your thoughts.

Samuel Caughron, MD:I think part of this goes back into history. It didn’t matter that we called it non—small cell lung cancer. The pathologists knew it for treatment purposes, and that was sufficient. There wasn’t a lot of focus on adenosquamous carcinoma versus squamous cell carcinoma. These are presumably coming from the same pneumocytes as the cell of origin. There’s really a continuum. It shows some features of squamous cell and some features of adenosquamous. Some clearly are well differentiated in 1 category or the other.

But it’s biology; nothing is 100%. These are the same cells just showing different features. I think it’s going to depend on the pathologist. Sometimes, the immunohistochemical stains that we use will suggest 1 or the other. Sometimes it’s clear cut, but often it’s not. I think the guidelines recommend if it can be called clear cut, do so. Otherwise, say favor 1 or the other.

Benjamin P. Levy, MD:One more question. As a medical oncologist, my best friends have become the pulmonologists and the pathologist. I try to call them at all times, “Hey, do you have it—did you get enough tissue?” Then I talk to the pathologist. “Do you have enough tissue?” What’s the communication like? Have the changes in molecular testing altered how frequently you’re speaking to the medical oncologist? Or is it pretty operationalized, where you don’t really need to have that conversation and everything is done in a reasonable way?

Samuel Caughron, MD:We tried to operationalize it. Within our CoC [Commission on Cancer]—accredited cancer program, 1 of the studies we did was looking at the adequacy of specifically lung biopsies for biomarker testing. I think that helped raise awareness across the institution of the need for an adequate tissue sample.

What’s interesting is we found there were not significant increases in the adverse effects where interventional radiologists were doing the biopsies, and they became a little bit more aggressive in the wake of that. Even under optimal circumstances, I think you’re going to end up with a 10% to 20% QNS quantity not sufficient rate. The lung is a difficult organ to get a generous sample. But it’s important to get the tissue when you can.

Transcript edited for clarity.