ONCAlert | 2017 San Antonio Breast Cancer Symposium

Immunotherapy on the Horizon in Gastric Cancer

Greg Kennelty
Published Online:10:41 AM, Fri August 26, 2016

Zev Wainberg, MD

While the utilization of immunotherapy in gastric cancer may not be as clear cut as it is in melanoma or lung cancer, Zev Wainberg, MD, illustrates some trials where the therapy type is breaking through.

In an interview with Targeted Oncology, Zev Wainberg, assistant professor of medicine, UCLA Health, discusses pembrolizumab (Keytruda), durvalumab, avelumab, and apatinib in gastric cancer, as well as the FAST trial and the JAVELIN trial.

The FAST trial examined IMAB362, a monoclonal antibody, in gastric cancer. The drug improved median progression-free survival (PFS) by 3.1 months and median overall survival by 4.8 months in combination with a 3-drug chemotherapy regimen. This regimen was compared with chemotherapy alone in results of the FAST trial, where it was shown that the former's benefit was more pronounced among patients with high expression levels of CLDN18.2.

The JAVELIN trial is currently investigating avelumab in patients across multiple solid tumor types who progressed on more than 2 lines of prior therapy, or had received 1 line of chemotherapy without progression. Data from February 2015 show that, of the 75 patients with gastric cancer/gastroesophageal junction that were treated with avelumab, responses were observed in 7 patients. PD-L1 expression was evaluable in 55 patients, including in 3 patients with a response. Median PFS in patients who had received 2 or more prior lines of therapy was 36 weeks (95% CI, 6.0-36.0) for PD-L1–positive patients and 11.6 weeks for PD-L1−negative patients. Among patients who had received 1 prior line of chemotherapy, the median PFS was 17.6 weeks for PD-L1–positive patients and 11.6 weeks for PD-L1–negative patients.

TARGETED ONCOLOGY: Can you tell us about the clinical trial you're involved in with pembrolizumab?

Wainberg: That's been studied in several different settings, including the third-line setting, for which preliminary evidence of activity has already been presented in small patient numbers at previous meetings. That data has been reported now a few times, with response rates and some of the activities of this agent in a group of patients that have not historically responded very well. These are the third-line typical patients, or advanced patients with metastatic gastric cancer.

The bigger study that's being done is to confirm those results in a very large number of patients with that profile. Additionally, there are randomized trials going on with that agent in both the first-line setting—so newly diagnosed patients with metastatic gastric cancer, either by itself or in combination with chemotherapy—and in second-line metastatic gastric cancer, randomized, and with chemotherapy as the control arm.

We have, to my knowledge, at least 3 very large studies looking at pembrolizumab in gastric cancer. These are 3 very large, international trials with about 100 patients each trials.

TARGETED ONCOLOGY: So far, from what you've seen in each of these trials that already happened, does it seem like a promising agent in this space?

Wainberg: I think the hints of activity in these small groups suggest that it is promising, or this group of therapies is promising. It remains to be seen who these patients are; certainly not a large number of them, but rather the minority of them, do respond. There is this group of patients that respond, and these things really need to be reproduced in a much larger set of data to be confident that it's real.

TARGETED ONCOLOGY: Has PD-1 positivity been investigated in this space?

Wainberg: That's one of the factors, and PD-L1 positive staining is one of the criteria for at least 2 of the trials that I mentioned. It hasn't been worked out to the extent that it has in lung cancer, and we don't even know, to be fair, what percentage of patients are PD-L1 positive in gastric cancer. It is one of the criteria that are used, and on some level, there does seem to be this suggestion that if one doesn’t have staining of the antigen, the likelihood of the response is very low.

TARGETED ONCOLOGY: What about PD-L1 agents? Do you see potential for that class of drugs?

Wainberg: On some level, there have already been reported results from the JAVELIN trial with avelumab, and there have been reported results of durvalumab with gastric cancers. We're seeing a little more than anecdotal responses in this group of patients that have 10% or 20% responses. I don't think it's going to be pembrolizumab that's going to produce the responses here. I think if it's a true phenomenon in gastric cancer, and we've already seen it with nivolumab as well, then it won't be these one or two drugs. It's going to be across the board.

The question is if the study designs will be adequate to answer the questions they're asking. Different companies use different strategies, so for example, with pembrolizmuab, their strategy is PD-L1 positive, randomized to chemotherapy, or randomized with chemotherapy, whereas some of the other trials are using nuances of that strategy with either combinations with CTLA4 inhibitors.

With avelumab, they're doing an interesting strategy called a switch-maintenance strategy, to see if the drug has an effect in gastric cancer, maybe as a maintenance drug.

TARGETED ONCOLOGY: Do you think these treatments will be useful in combination?

Wainberg: In gastric cancer, we know it's not going to be like it is in melanoma, and probably not like lung cancer either. It's going to be a small group, and the big focus is going to be on what the group is. Whether that's PD-L1 positive, there are subtypes of gastric cancer that might be more responsive to immunotherapy drugs like Epstein-Barr Virus (EBV). For example, there's a subgroup of patients with gastric cancer that are part of the EBV group. There's a subgroup of them that are microsatellite unstable (MSI), which is the area in colon cancer where immunotherapy really works. We really need to tease out whether these patients with gastric cancer are among those 2 groups.

TARGETED ONCOLOGY: What about outside of immunotherapy? Are there other agents you're excited about?

Wainberg: There was a very promising drug presentation at ASCO from Germany from a small company called Ganymed with a trial called the FAST trial. The target is something called a Cloudin antigen, and that looked really promising in the phase II study results. So they've already, I think, launched a phase III trial in that setting. That looks really promising based on the results.

One drug I'm involved with that we're running a phase III trial in gastric cancer for is sponsored by Gilead Pharmaceuticals, and the drug is MMP9, which is an antibody against that target. That one is already launched, we just launched a phase III trial, and we're combining that drug with chemotherapy versus chemotherapy as the control arm.

TARGETED ONCOLOGY: What is the reasoning behind looking at that particular drug?

Wainberg: MMP9 has a lot of roles to play in the tumor stromal interaction, as well as an antigen in improving, and decreasing on some level, the immune response. The monoclonal antibody has been shown in phase I studies and in other phase II studies to have promising activity, better than one would expect with the control, affecting the tumor stromal environment in a favorable way. There is promise, and there aren't that many of these out there, and that's what makes it a unique and interesting target.

The trial itself combines the antibody with standard chemotherapy because the hypothesis is that this drug will not work as a standalone agent, but when combined with chemo, it will allow, on some level, chemotherapy penetration to be greater into the tumor cells. It will deplete the tumor stromal factors that will allow chemotherapy to be at a higher concentration in the tumor cell.

That's one of the big hypotheses behind how this drug classes. This isn't even really a class of drugs; it's just one drug to my knowledge, which makes it a bit of a higher-risk endeavor in some respects, but in many ways, it'll be very interesting.

TARGETED ONCOLOGY: What point are you at in that trial?

Wainberg: We launched phase III. We did phase I and phase I expansion, and they showed promising results, so the decision was made to essentially go to a phase III.

TARGETED ONCOLOGY: There's been some data with apatinib, another VEGF inhibitor. Do you have any thoughts on it?

Wainberg: On the surface, and if you look at the manuscripts, it looks like an active VEGF agent that has activity in gastric cancer. That's what they're reporting, and that's what seems to be the case in the presentations and applications. I think, in terms of the agent being potentially controversial, if the trials have exclusively been done in China, the question is, should the data be replicated before proceeding? I think we do that with every drug, and to get the FDA approval, those require some degree of certainty among other centers and across international studies.

It seems like an active VEGF inhibitor in that respect. One could imagine that it might work.

TARGETED ONCOLOGY: What would you say are the biggest challenges in gastric cancer that you would like to see tackled?

Wainberg: Immunotherapy has taken a lot of effort and a lot of interest, and I think one of the challenges is going to be in knowing which patients should be getting these drugs. There's been such competition in that field that the rush to get these trials done has made it more challenging to develop them in a clever way. We want to learn from the other malignancies that have had these drugs, and this is not going to be like melanoma or lung cancer. This is most likely going to be a small group that benefits—smaller than the other groups that have FDA indications for PD-1 inhibitors.

So who are these patients? It’s important to really be thorough about things. That's going to be the key, and I hope we can do that.

TARGETED ONCOLOGY: Is that going to take more molecular classification?

Wainberg: The patients are there, and the patients have been enrolled in a lot of these studies, but this has to be similar to what we’ve seen in colon cancer, where it was very clear that it didn't work in all colon cancers, but it worked in MSI colon cancer. So we stopped testing it in all colon cancers, and I think that's the wise thing to do. We would like to see gastric cancer use some of that to see if we can refine the patients, and maybe it's not MSI or EBV. We've got some classification schemes that might be really critical to interrogate thoroughly as to who the patients are that benefited.

Copyright © TargetedOnc 2017 Intellisphere, LLC. All Rights Reserved.