Panelists: John L. Marshall, MD, Georgetown University Hospital; Dirk Arnold, MD, PhD, Instituto CUF de Oncologia; University of Hamburg; Fortunato Ciardiello, MD, PhD, Seconda Università di Napoli; Paul R. Helft, MD, Indiana School of Medicine of the Indiana University Melvin and Bren Simon Cancer Center; Tanios Bekaii-Saab, MD, Mayo Clinic Cancer Center in Arizona
John L. Marshall, MD: I want to end this session by discussing MSI and the data that have come out. Just very recently, the FDA approved pembrolizumab for any MSI-high solid tumor. Obviously, this has foundations in the MSI-high colon space. Dirk, can you tell us a little bit about MSI and colon cancer, ie, frequencies, inherited? Do you want to give us a broad overview, and then we’ll talk about how to figure this out?
Dirk Arnold, MD, PhD: MSI is a molecular marker, one of the pathognomonic ways colorectal cancer exists or develops—and it’s a minority of patients who follow this in locally advanced colorectal cancer. The rate of MSI-positive patients, or MSI-high patients, is between 8% and 12%. In metastatic disease, it’s even less. It’s a small proportion of patients who are characterized by the mechanism that they bear many mutations in their genome. These mutations are really a high number of mutations which make them, let’s say, kind of susceptible to checkpoint inhibitors. And, therefore, this MSI status stands for hypermutation. Maybe there is not a correlation of 1-by-1, but this stands for any efficacy of these checkpoint inhibitors, namely with pembrolizumab, with a PD-1 inhibitor.
John L. Marshall, MD: It’s not very common, but we’ve already said we would know it in all our patients. This is just one of those things. I will postpone any discussion about re-biopsy, but I do think there is some evidence that this might change over time, believe it or not, which will play into the next question.
Tanios Bekaii-Saab, MD: But, I think that’s relevant because, frankly, it’s not just MSI. It’s the tumor mutational burden and, in some patients who may be MSI-indeterminate and the tumor mutational burden is high, we’ve actually used some of these agents.
John L. Marshall, MD: What are other ways to measure? We talked about immunohistochemistry in looking for the 4 proteins. You’re kind of hinting at tumor mutational loads.
John L. Marshall, MD: For most of the major testers, the companies present to you.
John L. Marshall, MD: PD-L1, MSI, and TML—each of these is different, and I will share with you that they do not overlap. You can have some tumors that are PD-1–positive and are not TML-high, and the like.
Tanios Bekaii-Saab, MD: They do now. But, PD-1 staining is probably the least useful.
John L. Marshall, MD: For colon cancer.
Tanios Bekaii-Saab, MD: For colon cancer and other cancers.
John L. Marshall, MD: In lung cancer, I think it’s a staple on some level.
Tanios Bekaii-Saab, MD: Borderline, in some ways, because there are still some PD-1–negative patients who do benefit, although not as much.
John L. Marshall, MD: And TML in colon cancer seems to correlate very nicely with MSI, almost overlapping in Venn diagrams. Fortunato, you’ve got a patient with metastatic colon cancer, with frontline, MSI-high disease. They’ll send you the drug before the approval in Europe. What are you going to do?
Fortunato Ciardiello, MD, PhD: Actually, the answer is, we are lucky because we are the only center in Italy participating in the pembrolizumab versus chemotherapy trial in a first-line setting. So, for me, it would be easy. I can enroll the patient in this trial for either first-line or second-line therapy, according to their trial with pembrolizumab. That is the only motivation for the patient to be tested—he hopes that, in this way, he will get pembrolizumab either in the first- or second-line setting.
John L. Marshall, MD: That study had a crossover in it.
Fortunato Ciardiello, MD, PhD: It has a crossover. It will be very important, also, because who knows if it’s better to do chemotherapy before, or not. This trial would be much more relevant, we think, now, because we can place when to use PD-1 inhibition in the continuum of care with these patients.
John L. Marshall, MD: It really becomes a very important study because our patients are going to knock on our doors and say, “I want that drug.” And we’re going to say, “We don’t know the answer.”
Fortunato Ciardiello, MD, PhD: This study is even more relevant, now. After FDA approval, it will give a real answer of when to use it.
Tanios Bekaii-Saab, MD: Yes, I agree. But, besides the clinical trial, over the last year that I’ve been using this in practice, I’ve actually moved it, now, to first-line therapy when they’re not willing to go on the trial. And the reason is very simple—although these are 20% to 30% of the patients who are more in the refractory setting, the responses you see for those patients who respond are dramatic and longer lasting than I could achieve with any chemotherapy that I wish to use in those patients. I’ve actually had patients who were incredibly symptomatic from those tumors, and that, within 2 weeks, their symptoms start resolving. Very, very dramatic effects.
John L. Marshall, MD: You do really see a Lazarus effect.
Tanios Bekaii-Saab, MD: You do.
John L. Marshall, MD: We all have stories, now, of patients who were refractory...
Tanios Bekaii-Saab, MD: I’d like to give this in the first-line setting. I’d like to give it a quick shot and, then, pending the study, and I hope the study confirms that. Otherwise, I still am not thinking that I’m harming patients. I think I’m benefitting them in many ways because they still have access to chemotherapy, at some point.
Fortunato Ciardiello, MD, PhD: Think about the reason why, in first-line lung cancer patients, it’s the same situation. If there’s very high PD-L1 expression, it’s really much better to try PD-1 inhibition and, then, to use chemotherapy after. So, maybe it will be the same situation.
Tanios Bekaii-Saab, MD: Absolutely.
Fortunato Ciardiello, MD, PhD: That’s why the trial will be important.
John L. Marshall, MD: Is there any difference of opinion on this side of the table?
Paul R. Helft, MD: No, not at all. I’m particularly impressed with some of the chemotherapy plus pembrolizumab responses. I haven’t been doing this nearly as long as you, John, but in my almost 20-years, the most amazing response I’ve ever seen came in response to our trial, which my colleague, Dr. Safi Shahda, led. He presented an abstract at the 2017 ASCO Annual Meeting about an unbelievable response in a 27-year-old female patient with Lynch syndrome.
John L. Marshall, MD: Is there any difference in response between Lynch syndrome and acquired disease?
Dirk Arnold, MD, PhD: We don’t know.
Paul R. Helft, MD: There was a nice paper presented at the 2017 ASCO Annual Meeting.
John L. Marshall, MD: It said that they look the same.
Paul R. Helft, MD: Well, no. They actually suggested that, in the metastatic setting, the patients who have hereditary Lynch syndrome do better, on average. They look better. But, I agree—I think that the answer is that we just don’t know for sure.
Tanios Bekaii-Saab, MD: I’m not sure that it’s the PD-1 as much as their disease tends to be a little bit more smoldering.
Paul R. Helft, MD: Right. This was the point of the abstract, which I think came out of the Dutch group.
Tanios Bekaii-Saab, MD: Their disease is less aggressive. We all deal with these patients. They tend to have less aggressive disease.
John L. Marshall, MD: A really subtle point to this is that MSI patients with BRAF mutations are not “inherited.” We pretty much know that, and we need to pull out that subgroup, here, as well. We start to have, even within this small subgroup, some variation there. One of the lessons for our group listening in is that because this is relatively new, if you do not know the MSI status on your metastatic colon cancer patient, you need to know it, right? You send it off. Would this be something everyone in the room would biopsy a patient for if you didn’t have tumor? You’d still biopsy some?
John L. Marshall, MD: This is worth knowing.
Fortunato Ciardiello, MD, PhD: It’s an important opportunity for the patients.
John L. Marshall, MD: Everybody agrees?
Tanios Bekaii-Saab, MD: Absolutely.
Paul R. Helft, MD: I just have one other comment from a clinical perspective, and I’m certainly not an expert on PD-1 inhibitors at this table—there is much more expertise from others than me. We formed a working group across the cancer center to standardize our practice with respect to managing toxicities, which I feel I still don’t have a full understanding of. Again, you all may have a much more sophisticated understanding, but those toxicities are not what we’re used to. By recognizing when they come up and knowing how to manage them, there are lots of opportunities, I think.
Tanios Bekaii-Saab, MD: They get even worse when you start introducing the CTLA-4 inhibitors with the PD-1 inhibitors.
John L. Marshall, MD: If you combine them with chemotherapy, you can’t give steroids as antiemetics. So, now you have more nausea. That’s an example of a trade-off.
Paul R. Helft, MD: Many problems.
John L. Marshall, MD: It does. It’s not simple, and one has to learn to be a better doctor around that in managing the toxicity. I’ve noticed that even in our inpatient, sort of hospitalists, core. Our intensive care unit team meets regularly to talk about management of these toxicities because, on some level, when they get admitted and they’re really sick, we have to determine how much steroid to use and if we’re going to use it or not. It’s created an awareness among our colleagues, as well.