Select Topic:
Browse by Series:

Patient Selection for Immunotherapy in Gastric Cancer

Panelists: Johanna C. Bendell, MD, Sarah Cannon Research Institute; Ian Chau, MD, Royal Marsden Hospital; Yelena Y. Janjigian, MD, Memorial Sloan Kettering Cancer Center; Manish A. Shah, MD, Weill Cornell Medicine, New York-Presbyterian Hospital; Kohei Shitara, MD, National Cancer Center Hospital East, Japan
Published Online: Monday, Aug 07, 2017



Transcript:

Johanna C. Bendell, MD: This is the biggest question. So with the single-agent checkpoint inhibitors, we see a response rate—let’s just round it to 10% overall. But it’s that tail end of the curve that’s making the difference and showing us the difference in the hazard ratio. Now the important part is finding out who’s the tail end of the curve. Manish, we saw some data that were presented that suggest some patient populations—and we have a couple that we might be thinking of off the top of our heads—extrapolating a little bit from lung data that PD-L1 positivity status where we see some data from both nivolumab and pembrolizumab, so let’s talk about that. And then let’s also talk about MSI and maybe some of the others.

Manish A. Shah, MD: Right. That’s interesting that you brought that up. So the MSI-high tumors clearly have benefit to single-agent therapy, and actually it’s amazing that the FDA approved it across the board for all MSI-high tumors. That really is practice changing. In the third-line setting for gastric, both for pembrolizumab and nivolumab, they actually didn’t require a certain PD-L1 staining level, unlike the lung cancer data. From Dr Fuchs’ data of the third-line pembrolizumab study, you actually see a perhaps slightly higher response rate. But actually, in PD-L1–negative patients, you had some patients who had durable responses that were equally durable.

I think—again, speaking to the heterogeneity of the different diseases—it’s not clear that the PD-L1 status itself will be important in the third-line setting. The way to think of it is that we don’t know how to identify these patients who are going to respond. So if you are doing a clinical trial comparing a PD-1 inhibitor versus another active regimen, then you may need to enrich your population by some mechanism, like PD-L1 staining. The Merck group are actually now using a ratio of the percent-positive tumor cells and stromal cells over the total number of tumor cells that are in the slide, which is a little bit complex, but this is their ratio that they developed. I think it’s to be determined what’s the best way to select patients, and—actually borrowing from Yelena and others here in the group—I think that it may be actually non-IHC-based in the future. It may be something like the mutational load.

Johanna C. Bendell, MD: Yes, so tumor mutational burden, MSI status, etc, are important in knowing for your patients with gastric cancer, and hopefully we’ll start to get approvals in Asia as well as the United States. I don’t know about NICE.

Ian Chau, MD: NICE sees different things that you have to look at in the end. You have to talk about cost-effectiveness. But I would say that NICE has approved pembrolizumab and nivolumab in certain cancers. I suppose I’ve got a couple of points I want to perhaps highlight. We are very excited obviously about the recent FDA approval of pembrolizumab in MSI-solid tumors. As we can see in the testing on the KEYNOTE-059 trial of over 200 patients, only 4% were MSI-positive. In metastatic gastric cancer, the proportion is small, and out of that 4% of patients, only half of them responded. So we talk about if you test 100 patients, you treat the 4, and only 2 of them will respond, obviously we still see the response, we still need to see what survival they’re getting out of this. I suppose it’s more of a reality check. We are excited about it. We think that is a group we want to highlight, but you also need to perhaps not; you can be enthusiastic but not raise the expectation so high for the patients that they think, “Oh, you checked it; then I’m MSI now. I’m not good to suddenly have a curative treatment.” I don’t think we can, in any way, say that yet.

One other point that I want to perhaps discuss and highlight is the side effects. Certainly, we’ve seen from the ATTRACTION-2 study that the side effects are very mild with nivolumab, and it’s what we have noticed. But we know immune-related side effects do come a bit later, and your patient is not on study for very long because they’re being treated third-line or fourth-line. You’re not going to see that side effect kick in. So I think as we move on to earlier-line PD-1, PD-L1, that immune-related side effect management will become more and more important for the oncologists. Certainly, as GI oncologists, we’re probably less familiar outside clinical. The GI oncologist who does not or has not participated in these clinical trials is probably less familiar to manage them. And I think that’s an important thing that people need to know, that we do enter a new era of toxicity as well.

Johanna C. Bendell, MD: And especially the delayed onset. Some people can be off that treatment and months later come with autoimmune side effects from study.

Transcript Edited for Clarity

Slider Left
Slider Right


Transcript:

Johanna C. Bendell, MD: This is the biggest question. So with the single-agent checkpoint inhibitors, we see a response rate—let’s just round it to 10% overall. But it’s that tail end of the curve that’s making the difference and showing us the difference in the hazard ratio. Now the important part is finding out who’s the tail end of the curve. Manish, we saw some data that were presented that suggest some patient populations—and we have a couple that we might be thinking of off the top of our heads—extrapolating a little bit from lung data that PD-L1 positivity status where we see some data from both nivolumab and pembrolizumab, so let’s talk about that. And then let’s also talk about MSI and maybe some of the others.

Manish A. Shah, MD: Right. That’s interesting that you brought that up. So the MSI-high tumors clearly have benefit to single-agent therapy, and actually it’s amazing that the FDA approved it across the board for all MSI-high tumors. That really is practice changing. In the third-line setting for gastric, both for pembrolizumab and nivolumab, they actually didn’t require a certain PD-L1 staining level, unlike the lung cancer data. From Dr Fuchs’ data of the third-line pembrolizumab study, you actually see a perhaps slightly higher response rate. But actually, in PD-L1–negative patients, you had some patients who had durable responses that were equally durable.

I think—again, speaking to the heterogeneity of the different diseases—it’s not clear that the PD-L1 status itself will be important in the third-line setting. The way to think of it is that we don’t know how to identify these patients who are going to respond. So if you are doing a clinical trial comparing a PD-1 inhibitor versus another active regimen, then you may need to enrich your population by some mechanism, like PD-L1 staining. The Merck group are actually now using a ratio of the percent-positive tumor cells and stromal cells over the total number of tumor cells that are in the slide, which is a little bit complex, but this is their ratio that they developed. I think it’s to be determined what’s the best way to select patients, and—actually borrowing from Yelena and others here in the group—I think that it may be actually non-IHC-based in the future. It may be something like the mutational load.

Johanna C. Bendell, MD: Yes, so tumor mutational burden, MSI status, etc, are important in knowing for your patients with gastric cancer, and hopefully we’ll start to get approvals in Asia as well as the United States. I don’t know about NICE.

Ian Chau, MD: NICE sees different things that you have to look at in the end. You have to talk about cost-effectiveness. But I would say that NICE has approved pembrolizumab and nivolumab in certain cancers. I suppose I’ve got a couple of points I want to perhaps highlight. We are very excited obviously about the recent FDA approval of pembrolizumab in MSI-solid tumors. As we can see in the testing on the KEYNOTE-059 trial of over 200 patients, only 4% were MSI-positive. In metastatic gastric cancer, the proportion is small, and out of that 4% of patients, only half of them responded. So we talk about if you test 100 patients, you treat the 4, and only 2 of them will respond, obviously we still see the response, we still need to see what survival they’re getting out of this. I suppose it’s more of a reality check. We are excited about it. We think that is a group we want to highlight, but you also need to perhaps not; you can be enthusiastic but not raise the expectation so high for the patients that they think, “Oh, you checked it; then I’m MSI now. I’m not good to suddenly have a curative treatment.” I don’t think we can, in any way, say that yet.

One other point that I want to perhaps discuss and highlight is the side effects. Certainly, we’ve seen from the ATTRACTION-2 study that the side effects are very mild with nivolumab, and it’s what we have noticed. But we know immune-related side effects do come a bit later, and your patient is not on study for very long because they’re being treated third-line or fourth-line. You’re not going to see that side effect kick in. So I think as we move on to earlier-line PD-1, PD-L1, that immune-related side effect management will become more and more important for the oncologists. Certainly, as GI oncologists, we’re probably less familiar outside clinical. The GI oncologist who does not or has not participated in these clinical trials is probably less familiar to manage them. And I think that’s an important thing that people need to know, that we do enter a new era of toxicity as well.

Johanna C. Bendell, MD: And especially the delayed onset. Some people can be off that treatment and months later come with autoimmune side effects from study.

Transcript Edited for Clarity
View Conference Coverage
Online CME Activities
TitleExpiration DateCME Credits
Publication Bottom Border
Border Publication