Select Topic:
Browse by Series:

Genomic Testing: GI Tumors

Panelists: Benjamin Levy, MD, Johns Hopkins Sidney Kimmel Cancer Center, Sibley Memorial Hospitale; Mark Agulnik, MD, Northwestern University Feinberg School of Medicine; Marcia Brose, MD, PhD, University of Pennsylvania; Edward Kim, MD, Levine Cancer Institute, Atrium Health ; John Marshall, MD, Georgetown University; Philip Agop Philip, MD, PhD, FRCP, Barbara Ann Karmanos Cancer Institute
Published: Monday, May 20, 2019



Transcript: 

Benjamin P. Levy, MD: I promised you would we talk about next-generation sequencing [NGS]. Finally, we’re there. We’ve touched upon a lot of these points already, but I think it’s important to reinforce them and contextualize them in terms of what we’re doing within each of our disease states. Next-generation sequencing is clearly moving forward. The depth and breadth of what we can capture is fascinating, in terms of what we can get. Obviously, when we find something, we can give the patient the right drug. We’ll start with the colorectal people. Is next-generation sequencing done on every single patient? Philip, you mentioned stage IV versus stage III. I learned something. We’re doing it a lot in lung, obviously, but is this something you’re doing on all of your patients?

John L. Marshall, MD: Well, maybe I’d broaden it just a bit. I think we should quit calling it next-generation sequencing. It’s really molecular profiling because it’s not just DNA. That’s really what we think of, truly, with NGS. There’s some sort of molecular characterization from the sort of oligo or a la carte ordering that’s required, as we talked about, based on the disease, to this now broader testing, which is this big wide net, most of which we catch a bunch of krill that we’re never going to eat, and now to multiomics and how are we going to measure the complexity. So we’re in the middle age of this in some way.

But certainly, for colorectal cancer, all patients, all stages need to know microsatellite instability [MSI]. So how you do that? If you want to just do a la carte and do an IHC [immunohistochemistry] for every patient, that’s fine. More and more, we need to know tumor mutational burdens to kind of confirm this and the like. So honestly, for me, for colorectal in all stages, I’m sending broad profiling. Certainly, for metastatic disease, we need BRAF and all of the RAS genes. We need MSI, NTRK, and the other rare things, HER2, that might target us to some other therapies in our world. And so, I think for colorectal, specifically, broad profiling is appropriate for all stage IVs, and MSI for all patients.

Benjamin P. Levy, MD: Philip, anything to add?

Philip Agop Philip, MD, PhD, FRCP: Not more to add, but again, while John was referring to colorectal, I also do upper GI [gastrointestinal]. I also do neuroendocrine tumors. Believe it or not, it’s helping me in some patients with neuroendocrine tumors, regarding how to manage them. So I really firmly believe that every patient with stage IV disease should go through it. And the real question is the timing of it. When do you do it? The timing is important. Sometimes people are waiting until the patient has a progression after frontline treatment. In the era of targeted therapies, in general, but also in trying to put patients on clinical trials and things like that, I believe it has to be done at the time of the diagnosis or whenever you find that the patient has recurrent or stage IV disease. Do you agree with that?

John L. Marshall, MD: I do.

Benjamin P. Levy, MD: And what about pancreas?

John L. Marshall, MD: So I’m there. About 24% of pancreas cancer will have some sort of actionable mutation, depending on how you define that. But now we’re expecting an approval for a PARP [poly ADP ribose polymerase] inhibitor in the BRCA-positive population, and that’s not an inconsequential number of pancreatic patients for maintenance therapy. And so, how are you going to get that? There’s some push that now all pancreatic cancer patients need to have germline testing. If you look at gastric, you’ve got HER2, MSI. Bile duct: You think of a sort of obscure disease, from your guys’ perspective, and a lot of them out there have highly actionable targets in the I/O [immune oncology] space. A lot of the NTRK fusions have been seen there, and other fusions—FGF fusions, for example.

Across the board in GI, I think early. The question often is, do you do it from the beginning and repeat? In our world, we do a lot of metastasectomies. And so, a lot of our original pathology was done on the primary. We’re either sampling or removing a lot of metastases and it’s very tempting to send it again. We have a case report right now of a patient who’s actually got small bowel cancer and has lived long enough for 4 different biopsies and resections. We’ve profiled each one of those, mostly through research. Each is different.

Benjamin P. Levy, MD: That’s incredible.

John L. Marshall, MD: In some genes or in others, different sites, different times. But, there you go. There’s a clear change over time that might, one day, influence our therapies.

Philip Agop Philip, MD, PhD, FRCP: The sampling is also something to keep in mind. Like in pancreatic cancer, the fine needle aspirate is frequently done. The yield is different. Some places use core biopsies even if they do an endoscopic ultrasound, for example, and that follows other sites of the disease where other, like biliary, although difficult to get, may be a good sample. So these are the things that are sometimes challenging. Do you go back and do another one? And then, what is the value of the liquid biopsies we talked about?

John L. Marshall, MD: I’ll throw one more at you because this was a case we had recently in our tumor board. They tried to get a biopsy of a pancreas mass 3 times and just couldn’t get a sample. This is not uncommon for us in pancreas cancer. And the thought that just came to mind was, “Well, let’s do a Guardant [Health liquid biopsy] on him.” Of course, this patient had a RAS mutation in their blood. We thought a little bit about that, but is that enough to make a diagnosis? And so, then I was thinking, “OK, are we going to stop doing biopsies and just do circulating tumor DNA markers?”

Benjamin P. Levy, MD: It’s a fascinating field, regarding what to make of alterations that you pick up on plasma in the absence of tissue. We can go on and on about that, in terms of how to interpret that. I would say that in lung, at least, the world is changing. We are using it as a supplement. I’ve used it once, full disclosure, in a similar situation. This was an [elderly] lady. We couldn’t get tissue. I informed her that we would probably find something, or not find something; but if we did, it would be tough to know. It was an EGFR mutation in the plasma and we did act on it. I’m not advocating for that, but I think there’s an increasing role for these liquid biopsies.

Transcript Edited for Clarity

SELECTED
LANGUAGE
Slider Left
Slider Right


Transcript: 

Benjamin P. Levy, MD: I promised you would we talk about next-generation sequencing [NGS]. Finally, we’re there. We’ve touched upon a lot of these points already, but I think it’s important to reinforce them and contextualize them in terms of what we’re doing within each of our disease states. Next-generation sequencing is clearly moving forward. The depth and breadth of what we can capture is fascinating, in terms of what we can get. Obviously, when we find something, we can give the patient the right drug. We’ll start with the colorectal people. Is next-generation sequencing done on every single patient? Philip, you mentioned stage IV versus stage III. I learned something. We’re doing it a lot in lung, obviously, but is this something you’re doing on all of your patients?

John L. Marshall, MD: Well, maybe I’d broaden it just a bit. I think we should quit calling it next-generation sequencing. It’s really molecular profiling because it’s not just DNA. That’s really what we think of, truly, with NGS. There’s some sort of molecular characterization from the sort of oligo or a la carte ordering that’s required, as we talked about, based on the disease, to this now broader testing, which is this big wide net, most of which we catch a bunch of krill that we’re never going to eat, and now to multiomics and how are we going to measure the complexity. So we’re in the middle age of this in some way.

But certainly, for colorectal cancer, all patients, all stages need to know microsatellite instability [MSI]. So how you do that? If you want to just do a la carte and do an IHC [immunohistochemistry] for every patient, that’s fine. More and more, we need to know tumor mutational burdens to kind of confirm this and the like. So honestly, for me, for colorectal in all stages, I’m sending broad profiling. Certainly, for metastatic disease, we need BRAF and all of the RAS genes. We need MSI, NTRK, and the other rare things, HER2, that might target us to some other therapies in our world. And so, I think for colorectal, specifically, broad profiling is appropriate for all stage IVs, and MSI for all patients.

Benjamin P. Levy, MD: Philip, anything to add?

Philip Agop Philip, MD, PhD, FRCP: Not more to add, but again, while John was referring to colorectal, I also do upper GI [gastrointestinal]. I also do neuroendocrine tumors. Believe it or not, it’s helping me in some patients with neuroendocrine tumors, regarding how to manage them. So I really firmly believe that every patient with stage IV disease should go through it. And the real question is the timing of it. When do you do it? The timing is important. Sometimes people are waiting until the patient has a progression after frontline treatment. In the era of targeted therapies, in general, but also in trying to put patients on clinical trials and things like that, I believe it has to be done at the time of the diagnosis or whenever you find that the patient has recurrent or stage IV disease. Do you agree with that?

John L. Marshall, MD: I do.

Benjamin P. Levy, MD: And what about pancreas?

John L. Marshall, MD: So I’m there. About 24% of pancreas cancer will have some sort of actionable mutation, depending on how you define that. But now we’re expecting an approval for a PARP [poly ADP ribose polymerase] inhibitor in the BRCA-positive population, and that’s not an inconsequential number of pancreatic patients for maintenance therapy. And so, how are you going to get that? There’s some push that now all pancreatic cancer patients need to have germline testing. If you look at gastric, you’ve got HER2, MSI. Bile duct: You think of a sort of obscure disease, from your guys’ perspective, and a lot of them out there have highly actionable targets in the I/O [immune oncology] space. A lot of the NTRK fusions have been seen there, and other fusions—FGF fusions, for example.

Across the board in GI, I think early. The question often is, do you do it from the beginning and repeat? In our world, we do a lot of metastasectomies. And so, a lot of our original pathology was done on the primary. We’re either sampling or removing a lot of metastases and it’s very tempting to send it again. We have a case report right now of a patient who’s actually got small bowel cancer and has lived long enough for 4 different biopsies and resections. We’ve profiled each one of those, mostly through research. Each is different.

Benjamin P. Levy, MD: That’s incredible.

John L. Marshall, MD: In some genes or in others, different sites, different times. But, there you go. There’s a clear change over time that might, one day, influence our therapies.

Philip Agop Philip, MD, PhD, FRCP: The sampling is also something to keep in mind. Like in pancreatic cancer, the fine needle aspirate is frequently done. The yield is different. Some places use core biopsies even if they do an endoscopic ultrasound, for example, and that follows other sites of the disease where other, like biliary, although difficult to get, may be a good sample. So these are the things that are sometimes challenging. Do you go back and do another one? And then, what is the value of the liquid biopsies we talked about?

John L. Marshall, MD: I’ll throw one more at you because this was a case we had recently in our tumor board. They tried to get a biopsy of a pancreas mass 3 times and just couldn’t get a sample. This is not uncommon for us in pancreas cancer. And the thought that just came to mind was, “Well, let’s do a Guardant [Health liquid biopsy] on him.” Of course, this patient had a RAS mutation in their blood. We thought a little bit about that, but is that enough to make a diagnosis? And so, then I was thinking, “OK, are we going to stop doing biopsies and just do circulating tumor DNA markers?”

Benjamin P. Levy, MD: It’s a fascinating field, regarding what to make of alterations that you pick up on plasma in the absence of tissue. We can go on and on about that, in terms of how to interpret that. I would say that in lung, at least, the world is changing. We are using it as a supplement. I’ve used it once, full disclosure, in a similar situation. This was an [elderly] lady. We couldn’t get tissue. I informed her that we would probably find something, or not find something; but if we did, it would be tough to know. It was an EGFR mutation in the plasma and we did act on it. I’m not advocating for that, but I think there’s an increasing role for these liquid biopsies.

Transcript Edited for Clarity
View Conference Coverage
Online CME Activities
TitleExpiration DateCME Credits
Clinical Practice Connections™: From Diagnosis to Emerging Immunotherapeutic Options: Understanding the Burden and Risks in Peanut AllergySep 28, 20191.0
Enduring CME activity from the School of Breast Oncology®: 2018 Mid-Year Video UpdateSep 28, 20192.0
Publication Bottom Border
Border Publication
x