FLOT4-AIO: Results of a Phase III trial evaluating the effect of neoadjuvant chemotherapy followed by surgical resection on survival in patients with limited metastatic gastric or gastroesophageal junction (GEJ) carcinoma

DR ENZINGER: I think that that probably was the most important study that was presented at ASCO this year. It, arguably, and I think for most of us, has changed the standard of care. It’s based on Salah Al-Batran’s work in metastatic disease where he has taken the AIO regimen, which is 24-hour 5-FU and leucovorin plus oxaliplatin, that they’ve developed — so it’s the German version of FOLFOX — and added docetaxel to that regimen.

So FLOT is 5-FU/leucovorin/oxaliplatin and docetaxel. The regimen has shown promise in the metastatic disease setting and, obviously, now has been moved into the adjuvant setting. For years now, since David Cunningham’s publication in New England Journal of Medicine in 2006, the MAGIC study, that had been the standard of care for many of us, giving 3 cycles of ECF before and after a gastrectomy. This study now looks at ECF or ECX — I think it was investigator’s choice — versus the FLOT regimen.

And I must say, the improvement in disease-free survival/overall survival is truly impressive. There are significant improvements in both parameters, hazard ratios that are highly significant. The toxicity is not a freebie. FLOT clearly is toxic. And I think that many of us are now going to learn the intricacies and toxicities of this regimen. And it’s quite clear that there will be substantial toxicity with this regimen and this should be given probably only to the strongest and healthiest patients.

But the difference in toxicity versus ECF/ECX is not that great. When you look at serious adverse events, I think it was 61% in one arm, 62% in the other arm, so very similar toxicity. There’s certainly more nausea and vomiting with an epirubicin-based regimen. Otherwise, there was more neuropathy with docetaxel, more, I think, infections and some fatigue. So I think that ultimately it’s going to be a new standard of care for patients with good performance status.

It had, I think, 430-something patients. This was conducted entirely in Germany. Some people talking about how German patients were more hardy than patients of other countries, but I think we’re going to very quickly learn if this is doable or not in our American and other European patients.

DR LOVE: So yes, the median OS was 35 months with ECF/ECX and 50 months with FLOT, hazard ratio 0.77, p-value, significant p = 0.012). Three-year overall survival was 48% for ECF/ECX and 57% for FLOT. Also significantly improved PFS with the FLOT regimen, with a median of 30 months with FLOT and 18 months with ECF/ECX.

So are you actually using this regimen in this situation?

DR ENZINGER: Honestly, I’ve not. So I don’t have any first-hand experience yet. Back in the day — “back in the day” is probably 5, 6 years ago — there’d be a new regimen that would come out and we would immediately just handwrite the new regimen in and start giving it.

In this day and age, where we have pathways and we have computer bills — in our particular case, it’s called a Beacon-billed. Pivoting is not as quick anymore. I have to get approval from various pharmacy committees. And I have to get the computerized system to bill the new regimen. So we had not been using it for metastatic disease, and so it’s going to take us a few weeks to bill this for the adjuvant setting.

DR LOVE: And it kind of sounds like maybe — is this going to be maybe similar to FOLFIRINOX in pancreatic cancer?

DR ENZINGER: I think so, yes. I think we’re all a little worried about the toxicity, but I suspect in the end it’s going to be tolerated well enough that most of our patients, particularly at academic centers where we have the better performance status, younger patients are going to be able to tolerate it. We’ll see how the older patients do with this.

Perspective on the results of KEYNOTE-059 Cohort 1: Efficacy and safety of pembrolizumab monotherapy in patients with previously treated gastric cancer

DR LOVE: So getting back to ASCO, like a lot of solid cancers and hematologic cancers now, the continuing big story has been checkpoint inhibitors, a lot of room, I think, for optimism, and certainly in terms of gastric cancer. I want to ask you about a couple of big papers. Your colleague Charlie Fuchs presented the KEYNOTE-059 study, the cohort one, of pembrolizumab monotherapy. Can you comment on that and in light of the monotherapy checkpoint data in general?

DR ENZINGER: Absolutely. Our institution was the largest contributor to that study, so we actually have a lot of first-hand experience with the patients on that trial. What I will say is that probably the most noticeable personal anecdote I can tell you is that you almost know within 3 weeks if this is going to work or not. You either have a patient who muddles along and really nothing much changes, or suddenly they feel the best they have felt since being diagnosed with cancer. So it’s almost like a Lazarus effect for some patients. Patients are almost rejuvenated. They regain their energy, their appetite. So when it works, it really works and we can see it very quickly the improvements.

What’s more difficult is the patients who have stable disease and it’s not quite certain if they’re responding or not. There are sometimes questions of pseudoprogression or even patients with stable disease who have late responses. So there is that group of patients as well. But when it works, boy does it work.

And it’s amazing how long it works. On that trial now, I have patients — I had, I think, 6 patients with a partial response on that trial. And I think every single one of them had duration of response of greater than a year, most of them a year and a half and someone now getting to 2 years. So it really changes your conception, even though the chances of responding to these agents is unfortunately less than we had anticipated. When it does work, it works so well that one really has to consider that as part of when you choose your next line of therapy.

DR LOVE: So, of course, we’re seeing that story, obviously, in a lot of other cancers. And a bunch of general questions always come up. You mentioned the prolonged responses that are so exciting. Have you had any patients where you stopped the therapy? I don't know what this particular trial called for.

DR ENZINGER: Two years. So we’re coming up on that for a few patients. I already have patients asking me, “So what are we going to do? What’s going to happen? Can I come off the study and pay for the drug myself?” And I think the issue really is, this is completely different than cell cycle poisons. And we really don’t know if you really need 2 years. I mean, maybe you only need 3 months of therapy or 6 months of therapy. After all, the theory is that you’re just upregulating the immune system and uncovering these cancers that are hiding. And does it really require ongoing therapy? And I think that that’s a very important question and particularly in these economic times will need to be answered.

DR LOVE: Yes. I mean, it sounds weird, but I remember hearing that story coming out with breast cancer with tamoxifen and people not knowing any of this. They get out there, and they’re doing great. And they don’t mind getting treated. And, of course, obviously a little different with this kind of a treatment, but similarly. I have heard a number of cases in different cancers of people who had autoimmune complications that required them to stop therapy, so prematurely, and a number of cases I’ve heard of, people did well.

DR ENZINGER: Yes. So I always tell the patients that when they have an autoimmune reaction that they actually have a very high chance of responding. And often they respond after they come off treatment and even if they’ve been given high-dose steroids, which in theory is terrible for the immune system. But some people actually feel that there may be an advantage to giving steroids to these patients. But the bottom line is that patients who come off for autoimmune phenomenon can do very well and can have long durations of response.

Activity of anti-PD-1 antibodies alone or in combination with ipilimumab for advanced gastroesophageal cancers

DR ENZINGER: So I’ve seen this, particularly with the nivolumab and ipilimumab combination, which obviously has a much higher rate of hospitalization than single-agent PD-L1 inhibition. The CTLA-4 agents, or at least this one, is certainly more toxic. When you combine, you probably are getting some improvements in response rate and efficacy, but you pay for it. It’s definitely you pay for what you get.

DR LOVE: Yes. And actually, I was going to ask you next about that, the other person, actually, on this program, Dr Janjigian, actually also presented at ASCO, data on what you were just talking about. And this is a trial looking at nivolumab plus or minus ipilimumab. First, can you talk about what we know about nivolumab and, for that matter, other PD-1/PD-L1 agents?

DR ENZINGER: Sure.

DR LOVE: You talked about pembrolizumab.

DR ENZINGER: Absolutely.

DR LOVE: What you see in other cancers is, you kind of see similar results. What do we know about single-agent therapy? And maybe you can also talk about this trial looking at the combination.

DR ENZINGER: Sure. The early data with pembrolizumab was the KEYNOTE-012 study. And the early data with nivolumab was the CheckMate 032 study. Basically, these were both exploratory trials that were conducted in multiple different cancers, and then single disease cohorts were presented. The first data was, again, with KEYNOTE-012 in pembrolizumab, showing activity in gastric adenocarcinomas. Initially, they were talking about a 33% investigator-related response rate. And then when they went to central review, it was down to about 22% to 25%.

They subsequently published, I think in Lancet Oncology — or was it Lancet? I forget — and showed again that there were clearly a — half of the patients had stable or better disease. A quarter of the patients had major responses, long duration of responses and good progression-free and overall survival curves.

CheckMate 032, in many ways, was more interesting, because it not only had a PD-L1 inhibitor — a PD-1 inhibitor, I should say, and also had a CTLA-4 inhibitor, looking at the combination. But I think that in many ways, it was very interesting to look at the degree of PD-L1 positivity. The patients that were negative, the patients that were 1+/5+ and looking at how this degree of positivity modulated the chances for response, but even the patients that were negative still had a chance of responding to therapy.

So that was the early data with these 2 agents. Since then, the companies that make these agents have gone a little bit different routes. One was able to put forward a Phase III study, that has now been presented and published showing a small, I think, 1.15-month overall survival advantage, but also a subgroup of patients that actually did quite well with improvements in 12-month disease-free survival — 12-month progression-free survival and overall survival. That study is the first study that shows, in a randomized fashion, a survival benefit for a checkpoint inhibitor in gastric cancer.

The other route has been looking at a single-arm — actually 3 arms, but 1 arm that was the registration study, looking at efficacy in third line, demonstrating a good response rate, improvement — a good progression-free survival and overall survival. And that’s the KEYNOTE-059 study that you had referred to earlier.

Again, both studies I think clearly show that these agents are active. I think that they are not as active as we had hoped for and perhaps not as active as in other diseases. I think that we’re learning that we need to check all of our patients for MSI and PD-L1, so that we can strategize as to when we bring in immunotherapy. Is it an MSI-high patient where we should be using immunotherapy early on? Or is it a patient who’s MSI stable and PD-L1-negative where we probably should be using ramucirumab and paclitaxel first and leaving immunotherapy for third line?

So I think that we’re beginning to understand the fine details of how to use these agents, when to use them, what line of therapy to use them in. Both companies have ongoing studies. Probably most interesting to me is the adjuvant study. They’ve been willing to go in the adjuvant setting, where they’re giving patients the CROSS regimen that we talked about earlier, the neoadjuvant chemoradiation regimen for esophageal cancer, and then giving a year of nivolumab therapy for those patients who have residual disease on resection. That’s a fascinating study to me, and I’m very hopeful that it’ll be positive.

On the other hand, there are a robust number of Phase III studies in esophageal and gastric cancer that will clearly delineate its efficacy in the front-line setting, second-line setting and third-line setting. They’ve really covered all the bases. And gastric cancer, and particularly esophageal cancer, have been a hinterland for us oncologists for such a long time, very frustrating all the studies being conducted in the other major cancers. And now, finally, we’re seeing interest, very good studies, very large definitive studies being performed finally in esophageal and gastric cancer.

DR LOVE: You were talking about the issue of when to integrate it, how to use biomarkers before or after ramucirumab/taxane, for example. And a question that comes up again, so many different cancers, not lung cancer anymore because now they’re using it first line. But, for example, bladder cancer this discussion goes on, Hodgkin lymphoma the discussion goes on, which is the older, very elderly patient who really isn’t a candidate for chemotherapy. I don't know how often you see that with metastatic gastric. I assume pretty commonly. Would you like to give a checkpoint inhibitor to somebody, particularly where you really don’t want to give them chemo?

DR ENZINGER: Yes. And you certainly have patients who on their own are coming specifically for immunotherapy up front, because they’ve had some relative or a parent who’s had a horrible experience with chemotherapy. And many of them are eager to avoid chemotherapy at all costs.

I would caution those people, though, that monotherapy in MSI-stable, PD-L1-negative groups, the response rate and the chance of a response are so low that I probably would favor a mild chemotherapy instead, a weekly taxane by itself or with ramucirumab. Capecitabine, single conventional chemotherapy is more likely to be efficacious and probably not more toxic.

MSI testing for patients with gastric cancer

DR LOVE: So you also brought up, which, to my assessment, is probably the most asked question and issue right now in GI cancers in general — and cancer in general, I think — really is a major, major issue, which is this MSI story, and now with an FDA approval based on the marker and not the tumor.

DR ENZINGER: Yes.

DR LOVE: So first of all, I’m curious how you’re approaching MSI testing at Dana-Farber. I’ve heard of centers where they’re now testing everybody with GI cancers of any type for MSI. I mean, it’s only a biomarker, but I guess you have to take into consideration the cost when it’s such a low — I think in breast cancer it’s, like, 0.01% or whatever — but so efficacious. How are you approaching the issue of MSI testing? And what’s the overlap between MSI in gastric cancer? Initially I was hearing, “A quarter of gastric cancer is MSI high.” Then I was hearing, “That’s a quarter up front. It’s actually only 5% in metastatic disease.” So how does this all come together with really one of the biggest stories in oncology right now?

DR ENZINGER: Sure. You’re absolutely right. It is a huge story in cancers that typically have a poor chance of responding. And we suddenly have something that works remarkably well and has little toxicity.

But to get back to your initial question, how do we do this at the Farber, based on data that is unique to colorectal cancer — and again, that’s looking for Lynch syndrome and the unique properties of colorectal cancer — we test all of our patients for MSI with colorectal cancer. We have talked extensively about doing it for other GI malignancies. And at this point as a group, we haven’t been willing to do this yet, automatically, for our patients, partly because we have something called OncoPanel, which is a version of the FoundationOne testing.

So, actually, the investigators or scientists at Dana-Farber were the originators of FoundationOne. And our OncoPanel is generally one generation ahead of what FoundationOne has. But that includes MSI, so since we have a commitment to our patients that everybody gets OncoPanel testing, we eventually do get that data. It usually takes us about 2 months to get that back. And since we really haven’t made a commitment yet to using immunotherapy in the front-line setting yet for our patients, that 2-month turnover is acceptable to us at the current time.

So at our institution, all of the colorectal cancer patients get MSI testing up front. The other patients get it through OncoPanel, and then we generally like to confirm it through immunohistochemistry. In some patients where there’s some urgency, we will send the tissue blocks for MSI testing by IHC. But as I said, since the role of checkpoint inhibitors in MSI is currently second/third line, we feel comfortable right now not testing up front as yet.

DR LOVE: Although again, with that high a response rate — you think about a patient who presents without visceral disease, who’s not in crisis and it’s MSI high —

DR ENZINGER: In most of the GI malignancies, it’s 2% to 3% of the population, so you got to kiss a lot of frogs before you get your prince.

DR LOVE: What is the story, though, with gastric cancer? Again, originally I was thinking, “Maybe it’s the MSI patients who are responding.” And then I hear the story that, “Actually, in metastatic disease you don’t see that much MSI.” It sounds like colon, in a way.

DR ENZINGER: Yes. So we looked at that and it was, in fact, in KEYNOTE-059, the MSI patients that were driving this. And really, it’s not. So there are clearly patients who are MSI stable who respond to pembrolizumab. And yes, there is a higher chance of responding, but the numbers are so small that I think it was that there were only — was it 8 patients in KEYNOTE-059 that were MSI high? It’s a very small number. It may even be half of that. I forget.

But the number that I now remember is about 3% of metastatic gastric cancer patients are MSI high. So I think that we have to be very careful about telling our patients about this up front. The chances that this is not going to work out for them is extremely high. And I think it’s better to leave it as a wonderful discovery down the line rather than saying, “Hey, you’ve been diagnosed with gastric cancer. Let’s check for MSI. And you have a 3 out of 100 chance of benefiting from this.”

Incorporation of anti-PD-1 immune checkpoint inhibitors into the therapeutic algorithm for advanced gastric cancer

DR LOVE: How are you sequencing checkpoint inhibition into your metastatic disease algorithm?

DR ENZINGER: So I would prefer sticking with the current data, which is in patients who have had platinum and 5-FU and who have also received a taxane with or without ramucirumab. But I do, as I said, test for PD-L1 and MSI in my patients. And if somebody comes back strongly positive for PD-L1 or is MSI high, I will move up the checkpoint inhibitor into second line. And again, the patients I see, it’s not unusual for them to ask for this to be done up front. So the world we live in is, patients are consumers and they’ve heard of this and they know to ask.

DR LOVE: So right now, pembrolizumab is approved. What about nivolumab?

DR ENZINGER: And the question is if nivolumab is going to get approval based on East Asian data, which would be a first, since there were no American or European patients included in the trial.

DR LOVE: At this point, what’s your assessment in terms of the relative efficacy, at least of those 2 agents?

DR ENZINGER: I don't know for sure, but I think that probably the 2 agents are of fairly similar efficacy. I think nivolumab and pembrolizumab probably have similar efficacy.

Comparison of neoadjuvant chemotherapy regimens for locally advanced esophageal cancer

DR LOVE: So I want to actually just talk about a few papers. And one that we found that you wrote that I don’t — a journal I don’t usually read, Diseases of the Esophagus, and I guess it was just published, “A Retrospective Comparison of Neoadjuvant Chemotherapy Regimens for Locally Advanced Esophageal Cancer.”

Can you talk a little bit about what you were trying to get across in that paper and what you think are some of the key clinical issues in this particular issue?

DR ENZINGER: I think that again, there have been a variety of neoadjuvant regimens that have been used with or without radiation therapy for esophageal cancer. And I think that there are a number of issues that need to be considered with these regimens, particularly toxicity, efficacy, as well as the ability for patients then to move on to surgery, which is ultimately the goal for most of these patients.

So we have, over the years, been experimenting with various regimens to see which ones can achieve the highest response rates, which have the least toxicity and ultimately which ones we feel we should move forward with. Ultimately, we found that the CROSS regimen is really quite well tolerated, and it has become our standard since the publication in The New England Journal of Medicine now several years ago, probably 4 or 5 years ago now.

And we found that really, compared to older regimens that we had used, that there was improved efficacy/decreased toxicity. And I think that I’m not alone in really coming to the conclusion that although weekly paclitaxel and carboplatin’s perhaps not as sexy as some of the other regimens that others have proposed, it really offers a nice balance between the ability to shrink the cancer to achieve partial responses and even CRs but also to get the patient to surgery. And we’ve really seen a much higher rate of patients being able to get to surgery than with many of the other regimens. So I think that it’s really the total package that this CROSS regimen, this weekly paclitaxel and carboplatin, has provided for all of us.

Results of the Phase II FAST study: Progression-free and overall survival advantage with the addition of the anti-CLDN18.2 antibody claudiximab (IMAB362) to epirubicin/oxaliplatin/capecitabine as first-line therapy for patients with advanced CLDN18.2-positive gastric and GEJ adenocarcinoma

DR LOVE: So another paper/topic I wanted to ask you about is, you’ve written a lot about novel agents that are being investigated in gastric cancer. And it’s kind of a little bit hard to find anything super exciting outside the checkpoint inhibitors, of course, which we’re going to talk about. But I was curious about this so-called anticlaudin antibody, because that kind of seems a little bit cool and promising. There was a paper in the Journal of Hematologic Oncology in May reviewing this, but maybe you can talk a little bit about what we know about this agent.

DR ENZINGER: Sure. The study, known as FAST, was presented by Salah Al-Batran at ASCO now more than a year ago, ASCO 2016. I was the discussant at that session. And basically the study was quite exciting. It showed that this agent, IMAB362, could significantly improve progression-free survival and overall survival in patients that expressed claudin 18.2. And claudin 18.2 is a molecule that is partially responsible for the tight junctions that could occur between cells.

There are a variety of theories on how this particular antibody works. The thought is that it binds to this molecule and ultimately leads to some cell disintegration and that this ultimately then stimulates an ADCC reaction and also a CDC reaction. So they’re postulating a variety of mechanisms in which this particular antibody works. I think ultimately we truly don’t fully understand what is going on here. I know that we would like to, but I think that in many ways this is still a project in process.

But the bottom line is that it had a very impressive randomized Phase II result, almost doubling in progression-free survival and overall survival and, in fact, an enhanced effect in patients that were high claudin expressers. So I think that it clearly provided the data to move on to a large randomized Phase III study.

DR LOVE: I mean, I kind of think of things in kind of basic ways, probably not that accurate scientifically. But when you think about this in your mind, do you think about something like trastuzumab against anti-HER, or is that the model?

DR ENZINGER: Yes. So again, we actually have looked at data with trastuzumab. And I think that although it’s thought to work through the HER2 pathway — and we actually have some very interesting data from one of our studies looking at pretreatment biopsies and post-treatment biopsies. And there is actually a lot of evidence to suggest that it may actually work through immune activity.

DR LOVE: Right.

DR ENZINGER: So it may be a poor man’s checkpoint inhibitor or immunomodulating agent. I think that a lot of these antibodies may actually work in the same way. They latch onto some aspect of the cancer cell and somehow allow the immune system to realize that there’s a foreign agent or a foreign body there and then activate some sort of an immune process. So I think that a lot of these agents that supposedly work through one pathway or another may ultimately be found to be immunomodulating agents.

DR LOVE: Again, keeping it at a med student level, which I feel like it’s even hard for me to be nowadays, again, when you again conceptualize it and think about it in terms of antibody therapy, I mean, do you think rituximab is in the same ballpark?

DR ENZINGER: I do. I think that cetuximab certainly could have similar efficacy or similar pathway. I mean, there is much better data with cetuximab to suggest that it works through RAS and that particular pathway. But the company that makes cetuximab has, from the beginning, talked about an ADCC activity. And I suspect that that is at least part of its activity.

Status of the Phase III BRIGHTER trial evaluating weekly paclitaxel with or without napabucasin (BBI608) for previously treated advanced gastric or GEJ adenocarcinoma

DR LOVE: So speaking of new agents in Phase III, the last couple of years we’ve been talking with hopeful optimism about the strategy of what’s been called antistem cell therapy. And specifically there’s an agent out there, napabucasin, which I was trying to get excited about. And now I see maybe I shouldn’t get excited about, because the so-called BRIGHTER trial, we’ve seen a press release. What is napabucasin, and what do we know at this point? And where is it heading?

DR ENZINGER: Yes. BBI — what is it? 608 is an agent that is supposed to affect stemness. And I think I must applaud the company. It’s probably the most novel pathway or the most novel way in which we’re theoretically attacking these cells. And I think the idea of targeting some sort of a stem cell that’s quiescent and doesn’t respond well to cell cycle poisons is a very ideal way in targeting cancer.

I think, ultimately, again, it’s the question of if that’s really how this drug works and if it really does work. There’s a variety of stemness pathways. It’s STAT3, WNT and — I mean, there’s a whole host of pathways that seem to be involved in this. And I think that I was, for instance, disappointed in the colorectal data that was generated with this drug as well.

So I think that if I were to handicap some of these ongoing Phase III studies based on the colorectal data and the fact that stemness should probably be very similar between colorectal cancer and gastric cancer, I probably wouldn’t handicap this drug very highly. But it’s clearly a — it’s an interesting mechanism. And if we could somehow target this, I think that potentially this would be a very important pathway.

So I think that we’ll see. We have a robust Phase III study. It’s, I think, of reasonable design of paclitaxel with or without BBI-608. And we’ll see if it’s positive or not. But I think that as you pointed out, some of the preliminary data is a little worrisome.

Case: A 72-year-old man with previously treated metastatic gastric cancer receives ramucirumab/paclitaxel

DR LOVE: So let’s go through some of your cases, beginning with your 72-year-old man.

DR ENZINGER: Okay. Absolutely. So let me tell you about this gentleman, very nice man, recently retired, an administrator in a local educational system, a gentleman who presented with metastatic gastric cancer up front. He had a mass in the cardia. He had multiple liver lesions, several significantly enlarged lymph nodes, clearly was unresectable up front.

And our standard front-line treatment for metastatic gastric cancer is the FOLFOX regimen. At the time, we had an interesting trial that looked at FOLFOX with or without ziv-aflibercept, which is an angiogenesis inhibitor. And this randomized Phase II study was placebo controlled, double blinded. And the gentleman enrolled in this study and did remarkably well.

In the end, it turned out that he was randomized to the placebo arm and had done remarkably well on FOLFOX, which is frankly not unusual. Patients do fabulously well on this regimen. Generally what I do is, I have them on FOLFOX until they develop raging neuropathy, and then I keep them on the modified de Gramont regimen for months and years, in some cases. Eventually he did progress, though.

And we ultimately switched him to paclitaxel and ramucirumab, because by the time that he actually had progressed, his peripheral neuropathy had resolved. And he was once again able to receive this combination, again had a great response and then after about a year or so had progressive disease.

DR LOVE: Before you go on, I’m kind of curious what you would have done if he still had significant neuropathy at that point.

DR ENZINGER: Yes, that’s interesting, because — and if you look at the textbooks or the trials, you have either Grade 3 neuropathy or you don’t. In most cases, patients tell you, “I still have some neuropathy.” “Does it go away?” “No.” “Does it bother you?” “Not really. It bothers me at certain times.”

So the neuropathy that I deal with typically in my practice is not this debilitating neuropathy. It’s a persistent neuropathy. And, frankly, my experience is that even when they have Grade 1/Grade 2 neuropathy, starting up paclitaxel and ramucirumab doesn’t really worsen it. It certainly doesn’t get better, but it’s certainly not a sudden dramatic worsening of neuropathy. My experience is that it just continues on. And generally the patient progresses or comes off therapy for other reasons before this becomes an issue.

So the fact that some people say, “If you have peripheral neuropathy, you should just use ramucirumab by itself” really hasn’t — I haven’t seen that in my practice, because I, frankly, haven’t seen a worsening in neuropathy when I use the combination after FOLFOX.

DR LOVE: Would you combine ramucirumab with another chemotherapeutic agent?

DR ENZINGER: I’m hesitant to do so, in fact, because I’ve not had a problem with the neuropathy with paclitaxel. I know that others have combined it with irinotecan or switch out docetaxel for paclitaxel or have given it with a variety of other agents. I’ve been hesitant to do so, and again, partly because I’m hampered by our system, where we have pathways, we have computer bills. So I’m more tied to where the data is than, say, somebody out in private practice.

Angiogenesis inhibitors in gastric cancer

DR LOVE: So we’ll talk about what happened to this man at that point. But just to pick out a little bit there, though, on the ramucirumab — because you mentioned he was on this trial looking at ziv-aflibercept, another anti-angiogenic. And I remember there were actually trials, I think, of bevacizumab in gastric. And obviously it wasn’t very promising, because it never came into practice, and yet ramucirumab, like this patient, has now really become a standard.

Can you talk about in general what the clinical research data has shown with using these various anti-angiogenic strategies, bev, aflibercept, regorafenib — apatinib is another one — and regorafenib and why it is that we landed in the ramucirumab?

DR ENZINGER: Absolutely. I think we all know that the age of angiogenesis inhibition started with bevacizumab. And after the initial successes in colorectal cancer and lung cancer and, to a certain degree in breast cancer, they decided to test this agent with standard therapy in gastric cancer. And the study known as AVAGAST combined standard platinum/5-FU therapy with or without their agent, bevacizumab.

So they showed a significant improvement in response rate, progression-free survival and even showed a 2-month improvement in overall survival, but it was just not statistically significant. And in the end, I think that they were undone by the fact that the Japanese and other East Asians were using multiple lines of therapy after this front-line treatment and that really, the response rate/progression-free survival could not be pulled through due to the mediating effects of second- and even third-line therapy for these patients that were primarily from Asia.

And this has ultimately led to the effect that a lot of pharmaceutical companies limit the number of patients on their trials that come from East Asia, at least in this particular arena, so that no more than 25% or 30% of patients are allowed to come from China and Japan. And I think that really, we have learned from this experience that in gastric cancer, we’re no longer talking about 1 line of therapy. We’re talking about a disease that can receive 2 lines and now 3 lines of therapy with the checkpoint inhibitors.

So unfortunately, the company was not willing to redo the study with the appropriate modifications. And I think that’s where ramucirumab took advantage of this particular hole and in many ways does have some advantages over bevacizumab and tested first as a single agent versus best supportive care in cisplatin/5-FU-refractory disease and then in combination with paclitaxel.

And I think that clearly there was an advantage of using this agent in second and even third line, because I do believe that there is some way in which cancer cells become angiogenesis dependent. And when you stop an angiogenesis inhibitor, I think that it’s not necessarily the further inhibition, it’s the fact that you’re withdrawing this agent. And I think, for instance, in colorectal cancer, I don’t think that necessarily giving further bevacizumab or ramucirumab, it’s the efficacy of the drug. I think that actually when you withdraw it, the patients actually do worse.

And I think that that was really, I think, the advantage that ramucirumab had in the second- and third-line setting. That being said, now, the early data with the RAINFALL study with ramucirumab, where they’re reporting a progression-free survival advantage for that study, I think it’s promising that ramucirumab will probably get a front-line indication with cisplatin and 5-FU.

RAINFALL: A Phase III trial of cisplatin/fluoropyrimidine with or without ramucirumab as first-line therapy for metastatic gastric or GEJ adenocarcinoma

DR LOVE: Can you review the design of that study and what we’ve seen from it?

DR ENZINGER: Certainly. That was a study that was completed in January or February of this year. And now they’ve reported progression-free survival data that met this primary endpoint. It was a study that looked at cisplatin/5-FU with or without ramucirumab. And the patients who have received ramucirumab have a statistically significant improvement in progression-free survival. And the progression-free survival advantage is such that it is highly unlikely that the overall survival advantage will not be statistically significant.

So the thought is that it is very likely that there will be an OS advantage with ramucirumab and that we will see FDA approval for this agent now in all lines of therapy. The big question, of course, is, do you continue ramucirumab in multiple lines of therapy and with other agents showing efficacy in this disease? Where do you use ramucirumab? Do you use it up front, second line? And I think that things are finally getting interesting in this disease, where we have multiple competing drugs that are actually efficacious.

DR LOVE: And when it was used in that trial, was it a bev-like approach, where you give the chemo and the anti-angiogenic and then the anti-angiogenic as maintenance?

DR ENZINGER: I think so. I had a few patients on that study. And I don’t remember us giving any maintenance therapy with ramucirumab. To be quite honest with you, I don’t recall.

DR LOVE: I’ll look it up and we’ll see.

DR ENZINGER: It’s embarrassing, because I had a few patients on the study. And I think that my patients unfortunately didn’t quite make it to the maintenance phase.

DR LOVE: But I guess — I mean, again, I mean, in the second-line setting, quote, there’s no maintenance, but I imagine there are people who get neuropathy, where you drop the paclitaxel.

DR ENZINGER: Oh, absolutely. Yes.

DR LOVE: It kind of works out like it really is maintenance anyhow.

DR ENZINGER: That’s right.

DR LOVE: But you give it until progression?

DR ENZINGER: Correct.

DR LOVE: Right. So that is interesting. And we’ll see whether that plays out. It was interesting. I referred to your, quote, colleague, Charlie Fuchs. And, of course, he did a lot of the work on ramucirumab, but he’s not your colleague anymore. So he’s gone now.

DR ENZINGER: No. He’s now the Director of the Yale Cancer Center.

DR LOVE: So he got stolen by the Mets or whatever.

DR ENZINGER: He’s actually a Yankees fan.

DR LOVE: Oh, he’s a Yankee?

DR ENZINGER: So be careful about that.

Incidence of pseudoprogression with the use of immune checkpoint inhibition in gastric cancer

DR LOVE: Anyhow, so let’s continue on with this patient, because I think the next thing that happened was really fascinating. So he gets the paclitaxel/ramucirumab, as you say. He does well for about a year. And he actually had a partial response. And then what happened?

DR ENZINGER: Unfortunately, as these things go, I mean, he eventually progressed. And I think it was not unexpected. When you look at the RAINBOW study, it’s really an amazing overall survival, I mean, but ultimately all these patients progress.

And this gentleman did progress as well, and we ultimately put him on a study of pembrolizumab.

DR LOVE: And what happened?

DR ENZINGER: He felt better, in fact quite a bit better. But his first CAT scan showed a 180% progression. So we were very worried, as you can imagine.

DR LOVE: Where did he have disease?

DR ENZINGER: He had disease mostly in the lymph nodes but also in the liver. And what worried us was it wasn’t just the lymph nodes that had increased in size but also the liver.

DR LOVE: But yet clinically he was feeling better?

DR ENZINGER: He was feeling better, not a lot better. That was also the concern. Slightly better.

DR LOVE: But definitely not worse.

DR ENZINGER: Definitely not worse. So we took a chance, and we actually continued him on therapy. And his most recent scan about 2 weeks ago shows instead of 180% progression shows a 97% progression. So he’s gotten half of that back again. So he’s not yet at a major response, but he’s heading in the right direction.

DR LOVE: When you say “97%,” are you saying —

DR ENZINGER: From baseline.

DR LOVE: So does that mean he’s about the same as baseline?

DR ENZINGER: No. No, no. Double.

DR LOVE: He’s double.

DR ENZINGER: So we went from baseline to 180% progression.

DR LOVE: But 180% means more than double.

DR ENZINGER: Yes.

DR LOVE: Okay.

DR ENZINGER: Almost quadrupling.

DR LOVE: Right.

DR ENZINGER: And now he’s 97% worse than baseline, so he’s heading back in the right direction.

DR LOVE: So it sounds like he’s maybe having something that supposedly never happens, or almost never happens, which is pseudoprogression. Is that your thinking?

DR ENZINGER: Yes. So I think that as these things go, a lot of people bandy about pseudoprogression and is it pseudoprogression? And the reality is that it’s much less common than we would hope or we would think. Our best estimate is that it’s probably around 5% and perhaps even less in esophageal and gastric cancer. I think that, particularly in KEYNOTE-059, there were virtually no patients who had pseudoprogression.

So I think that we’ll have to see how this pans out, but the bottom line, it’s palliative therapy. And if the man is feeling better and his disease is heading in the right direction, who am I to discontinue him off something that offers an improvement in quality of life and is actually keeping his cancer from growing?

DR LOVE: So note to myself, I’m going to keep in contact with you and find out what happens.

DR ENZINGER: That’s right. Hopefully it’ll continue to be a positive story.

DR LOVE: That’ll be cool, if he actually turns into having a response. And it’s funny. Of course, we first started hearing about checkpoint inhibitors from the melanoma people. And they were obviously the first people to bring up this issue of pseudoprogression. And what they started to say was, if the patient feels fine in general, keep going. And then we started to hear other specialties saying the same thing. I guess it is intuitively logical to do that.

DR ENZINGER: I think you basically have to throw out everything you’ve learned about oncology. I think that we’re having difficulty understanding exactly how to count a response or a win with these new agents. I heard recently at a meeting of a patient who had no response to one of the checkpoint inhibitors and, after 12 cycles, suddenly had a PR.

So I think that it ultimately really comes down to the fact that if you’re forced to stop, you’re forced to stop. I mean, if the patient’s going down the tubes and you still have another agent that you can use, then you’ve got to salvage them. But if they seem to be doing okay and there’s not significant progression of disease, I think that waiting it out is certainly reasonable, particularly since just coming off of chemotherapy makes a lot of the patients feel better.

So again, I think we forget that we’re not treating a CAT scan. We’re treating a patient. And quality of life, although we traditionally had difficulty measuring this in clinical trials, is ultimately why we treat patients. It’s to improve quality of life as well as to prolong life. And if that is what’s working, then why not continue it?

DR LOVE: So we’ll see what happens with this patient, but a really interesting situation.

Checkpoint inhibitor-associated autoimmune toxicities

DR LOVE: I want to hear about your 53-year-old man. And actually it kind of sounds like he went down a similar course in getting, actually, on the same trial, so FOLFOX plus or minus aflibercept until disease progression. Was he on placebo or treatment?

DR ENZINGER: He was also on placebo. Yes.

DR LOVE: Okay. So then he also got paclitaxel/ramucirumab. Did he respond to that, incidentally?

DR ENZINGER: Also responded well. Not in the sense that he had a PR, but he literally had a minor response and then maintained that for a year and a half.

DR LOVE: So the thing I want to ask you about him, because so far it kind of — then he gets pembrolizumab and was on it for a year and a half. So that sounds pretty good.

The thing I want to ask you about, and I’ve kind of heard a few cases of this, but it wasn’t really in the initial thing when we started to hear about autoimmune toxicity. I don’t remember hearing much about Type 1 diabetes, although you certainly wouldn’t be surprised. And now I’ve kind of been hearing a few cases, and I see that’s what happened to this patient.

DR ENZINGER: Absolutely. And so I think that we’re using these agents enough that we’re beginning to see all of the rare reactions. I mean, recently we had to reissue all of our consent forms for Stevens-Johnson syndrome. Traditionally, when I was first using these agents, particularly nivolumab and ipilimumab, we were seeing pneumonitis. It was pneumonitis and colitis that was primarily what was getting these patients into trouble. But since then, we’ve been seeing all sorts of different things, partly also because we’re getting more and more liberal with using these agents.

For instance, there’s now data showing that in certain cases, you can use these with patients who have rheumatoid arthritis. And when they start getting a flare of the rheumatoid arthritis, you have to back off these agents. But we’re seeing all sorts of problems like this patient, who literally, in the last few weeks of being on pembrolizumab, his blood glucose started going up into the hundreds, then in the 200s, and then really where he presented one day to me with a blood glucose of 500, a gentleman who was thin, otherwise no family history of diabetes. And suddenly I have to put him on an insulin drip to get his disease under control, an emergency endocrinology consult. How often do you get those, right?

But absolutely. And I think one of the things that, when consenting patients, I’d always been saying, “Oh, these drugs have virtually no side effects. You don’t have to worry about this. There’s some rare autoimmune phenomenon.” But I think we’re seeing particularly thyroiditis. There’s a lot of that occurring. Fortunately it’s not that dangerous, but you certainly see hypothyroidism, hyperthyroidism. And I think that, for instance, in KEYNOTE-059, I think 17% of patients had some form of autoimmune syndrome, even though it was typically mild.

So I think that we’re going to have to learn how to treat these. And, for instance, at our institution we already have a service that focuses on immunomodulatory-related autoimmune phenomenon. So this is a service that specializes in autoimmune disease that is secondary to our treatments.

DR LOVE: That’s wild. What kind of docs do you have on the service?

DR ENZINGER: Rheumatologists and infectious disease specialists. Primarily the rheumatologists and the endocrinologists.

DR LOVE: Hmm. Interesting. And, like, for example, when you see a patient with diarrhea, you’re thinking about colitis, do you send them to them?

DR ENZINGER: Yes. They get the patient if it’s severe enough to be hospitalized.

DR LOVE: Hmm. That’s really interesting.

DR ENZINGER: But we definitely have some of the members from rheumatology and endocrinology already specializing in this and see it as an opportunity to build their own research practice.

Case: A 58-year-old man presents with HER2-positive GEJ cancer and receives CAPOX with bevacizumab and trastuzumab on a clinical trial

DR LOVE: So let’s hear about your 58-year-old man.

DR ENZINGER: Sure. This gentleman, a really nice guy, the guy — like, Mike Ditka kind of a guy, right? Loves his pretzels and beer and really is just a really down-to-earth guy, no nonsense, never complains.

This guy had presented with HER2-positive GE junction cancer. And at the time, we were running a study in which we were giving oxaliplatin/capecitabine/bevacizumab and trastuzumab. And the reason for that was that there was this great preclinical data suggesting synergy between these 2 pathways.

There’s several preclinical studies looking at LD50s and looking at xenografts. And there’s a lot of data suggesting that these pathways can be synergistic. And we moved forward with this idea and ultimately treated 36 patients with this combination.

And as the studies that we’ve talked about so far — like, AVAGAST has a 46% response rate. The ToGA study, which is the trial that got trastuzumab approved for gastric cancer, has, I think, a 46% response rate. This study had an 81% response rate. Virtually every patient responded to this treatment. And the only patients, in retrospect, that we found that didn’t respond had MEK mutations — I’m sorry, MET mutations — and ultimately responded to MET inhibitors. But really, almost every patient responded to this treatment.

And the duration of response was incredible. We had a progression-free survival of about a year, overall survival of more than 2 years. The study has taken so long, because up until recently, I still had a third of the patients on the study. So we’re now finally — or unfortunately — getting to the point where we have enough progressers that we’re going to publish this. But there clearly is something to this combination that I think needs to be explored further.

DR LOVE: That's interesting, because I think I remember hearing about this in breast cancer. Dennis Slamon was involved with the trial. And my recollection was it crashed.

DR ENZINGER: That just goes to show you that we know that something that works in breast cancer doesn’t necessarily work in gastric cancer. So why not the reverse?

But as I said, this has been remarkable. And I’m hopeful that we can confirm these results in a Phase III study.

DR LOVE: That’s interesting. I mean, bev has some toxicity, but not necessarily the worst thing in the world.

DR ENZINGER: No.

DR LOVE: Is this something you do or would do outside a trial setting?

DR ENZINGER: I would if I could get it reimbursed by the insurance companies. If I could, I would definitely do it. I would give this as my standard front-line treatment for HER2 patients. I think that what we found was very little toxicity. And ultimately, the responses and duration of responses are so dramatic that I can’t see this as being a fluke of just us treating the patients. This was a multicenter study from 4 academic centers. It was blinded. We had a tumor metrics core that was blinded to the agents. There was 1 patient I remember, 29% response rate. They wouldn’t give it to me. So it was a rigorous, centrally reviewed study of response.

DR LOVE: That’s really fascinating — that’s amazing, how many people I interview about gastric cancer. I’ve never heard that. I guess it’s never been published, so probably people don’t even know about it.

DR ENZINGER: We presented it at ASCO, I think, about 4 years ago.

DR LOVE: Oh. That's interesting.

DR ENZINGER: It was a poster discussion, I think.

DR LOVE: Interesting.

DR ENZINGER: Yes. And Ian Chau was actually the one who discussed it along with — I think it was — no, I’m sorry. That was another study of mine. So this was a poster.

DR LOVE: Okay. So we’ll check that out.

DR ENZINGER: Yes.

DR LOVE: So he was on this for 5 years or —

DR ENZINGER: Yes.

DR LOVE: Wow! And did well, no toxicity issues?

DR ENZINGER: No toxicity. And I will say, what we did, in part to be able to continue patients for this length of time is, we would drop agents as needed, so you’d come off the oxaliplatin usually within the first several months. Some of the patients ended up having low LVFs, so we had to drop the trastuzumab. Others had thromboses, so we would drop the bevacizumab. So this study allowed the various agents to be dropped as needed. But as long as the disease was kept from growing, they were allowed to continue on the study.

DR LOVE: And so with him, he was on it for 5 years.

DR ENZINGER: In the last year and a half, he was just on bevacizumab and trastuzumab.

DR LOVE: Interesting. So you stopped the 5-FU also?

DR ENZINGER: Yes, we had to because he had low platelets. I think we burned out his marrow over a 4-year course.

DR LOVE: Amazing. So then you faced the challenging issue of second-line therapy in this patient who did great, a very controversial situation, because all the general oncologists who treat breast cancer, they just want to keep trastuzumab going and give chemotherapy. But you also have the issue of ramucirumab. There are some people who try to combine all 3, although most people don’t do that. But it really is a vexing problem.

Perspective on anti-angiogenic therapy

DR LOVE: And I know he went on a trial, but I am curious, before you talk about the trial that he went on, what you would do in general or how you handled this or what you would have done with him outside a trial. Do you ever keep trastuzumab — in this case he’s on bev, too — going and give different chemo or T-DM1?

DR ENZINGER: So you’re absolutely right. This was the problem, was that this was a gentleman who had problems with counts. And so what conventional agent do you give? You’ve already exposed him to an angiogenesis inhibitor. You’ve already exposed him to trastuzumab. And that’s why I was really so excited that we could get him on our clinical trial.

But the answer to what would I use standardly for this person, clearly I would give him a checkpoint inhibitor next, because that’s the one that he would most likely be able to tolerate.

DR LOVE: So why not paclitaxel/ramucirumab?

DR ENZINGER: Because his platelets were too low.

DR LOVE: Interesting.

DR ENZINGER: So he had trouble with this platelets. And this is why he came off the capecitabine. And I should point out that this is a gentleman we’re sustaining his platelets with romiplostim. So to keep the platelets up.

DR LOVE: But in general, how do you approach second-line therapy? Do you ever keep the trastuzumab going? Do you go with paclitaxel/ramucirumab? Would you use trastuzumab/paclitaxel/ramucirumab?

DR ENZINGER: Yes. So the answer is, I have never continued trastuzumab in second and subsequent lines of therapy, in part because at our institution, I have to show data and prove the benefit or efficacy. And there really is no good data. I think there was a recent publication that offers some evidence. But the bottom line is that unlike breast cancer where they actually have the data now, we really don’t have it in gastric cancer. And I think that up until recently, we’ve been trying to keep ourselves going by participating in trials like the GATSBY study and now, for instance, this study looking at a variety of companies are producing second-line anti-HER2 agents. And I think that that’s probably where we should be going rather than trying to continue trastuzumab into subsequent lines of therapy.

Phase I/II trial of the anti-HER2 monoclonal antibody margetuximab in combination with pembrolizumab for relapsed/refractory advanced HER2-positive gastric or GEJ cancer

DR LOVE: So this patient goes on a trial of pembrolizumab with margetuximab. What is that, incidentally?

DR ENZINGER: So it’s an FC-optimized antibody to HER2. And what the company says is that the drug seems to produce a greater ADCC activity than does trastuzumab. They have a Phase I study where they gave this agent, margetuximab, to patients with trastuzumab-refractory disease. I believe they had like 4 out of 19 partial responses. And I think the majority of patients actually had some sort of a response. So based on the preclinical data and on this Phase I study, they’re moving forward and looking at trastuzumab-refractory disease in both breast and gastric cancer.

DR LOVE: So how did he do on this therapy?

DR ENZINGER: He just started it. So we’ve been participating in this trial. But he has just started it. So I wish I could tell you. I wish I had some exciting news, but stay tuned.

DR LOVE: But it’s interesting, though, thinking back to your comments earlier about the mechanism of action of monoclonal antibodies, combining this monoclonal antibody with a checkpoint inhibitor, kind of lines up with what you’re talking about.

DR ENZINGER: Right. I mean, there’s obviously some very promising preclinical data. There’s some wonderful theoretical data, or theoretical concepts that need to be proven here. But I think that this is something that definitely needs to be explored.

Second-line therapy options for patients with metastatic gastroesophageal cancer

DR LOVE: So let’s talk a little bit about second-line therapy of metastatic disease.

DR ENZINGER: Sure. Sure. So I think that for the longest period of time, it was unclear if there was a second-line therapy benefit in gastric and, for that matter, in esophageal cancer. People would argue this. And, frankly, randomized studies in Europe and elsewhere, patients would get 1 line of therapy and then would move on to hospice.

So I think the East Asians felt a little bit differently about this. But all of this more or less changed with the presentation and then publication of the COUGAR-02 study, which was ultimately published in Lancet Oncology in 2014. Basically, this study took patients with metastatic gastric cancer that had progressed on cisplatin and 5-FU and randomized them to paclitaxel versus best supportive care.

The study showed a small but statistically significant survival advantage, 3.6 months for best supportive care versus 5.2 months for docetaxel. And really, this confirmed for many of us who believed in second-line therapy for this disease that there clearly was an advantage to giving some therapy beyond front-line treatment.

Very soon thereafter, Charlie Fuchs presented and then published data from the REGARD study. They at the time had been looking at using ramucirumab in this particular setting. And it also looked at patients with cisplatin and 5-FU-refractory disease. At the time, there was no proof of second-line therapy, so they felt very comfortable using placebo as a control. So patients were randomized to placebo versus ramucirumab.

And ultimately, this study proved to be positive, with a statistically significant benefit for patients on the ramucirumab arm. These patients notably had very little toxicity. In fact, I always like telling people when you look at the toxicity table on this particular study, there was less toxicity on the ramucirumab arm than on the placebo arm. But when you looked at progression-free survival and overall survival, even though the difference in median overall survival was not that great, there were nice improvements in 6-month overall survival, 12-month overall survival, about 10% and 7%, respectively.

And again, people said, “This agent isn’t that active,” but on the other hand, it’s not very toxic. So I think that it does certainly have its use, particularly when you go to the next study, the RAINBOW study, which now combines these 2 positive studies. You’ve now combined COUGAR and you combine REGARD and you get RAINBOW, where now in this study patients, again with platinum- and 5-FU-refractory disease, are randomized either to a taxane or to a taxane plus ramucirumab.

And again, the paclitaxel arm has a survival that’s actually similar, if not better, than what we’ve seen in COUGAR-02, but the ramucirumab arm, where the combination is even better, a 2.3-month improvement in median survival, a nice 10% improvement in overall survival. And building on COUGAR and building on REGARD, I think that RAINBOW clearly has demonstrated the new second-line standard of care. And really, this is what I give to all of my patients as long as they’ve got a reasonable performance status.

Activity of anti-PD-1 checkpoint inhibitors for recurrent or metastatic gastric or GEJ adenocarcinoma

DR ENZINGER: Fortunately, we’re able now to move beyond second-line therapy. I think the checkpoint inhibitors are now going to move us into having a viable third-line option. The first data that we saw with checkpoint inhibition in gastric cancer was the KEYNOTE-012 study. This trial used an outdated dose and schedule of pembrolizumab. But ultimately in melanoma, we’ve seen that the standard straight 200-mg dose is just as efficacious. And I think that we can translate this data to the currently FDA-approved doses.

But what we saw in KEYNOTE is that there’s clearly a benefit for a subgroup of patients. Half of the patients had stable or better disease. But, more importantly, about a quarter of the patients had a major response. And everybody who had a major response — almost everybody had a response duration that was greater than 6 months, extending into a year, year and a half. And progression-free and overall survival were quite nice in this study.

When you look at a median overall survival, 11.4 months, you can argue, this is a very select group of patients. But, on the other hand, this really beats what we would see, certainly, with best supportive care, so I think clearly a viable third-line option that’s further supported now with the Ono 4538 study. This is a study from Japan, Korea and Taiwan looking at patients with multiply refractory gastric cancer and treating them with nivolumab versus best supportive care in the standard nivolumab dose.

They showed a small but statistically significant improvement in overall survival, 4.14 months with placebo versus 5.32 months progression — overall survival, really a very small improvement in progression-free survival but still statistically significant. What you see here on the curves, you see that most of the patients don’t respond to this agent, and then you see a subgroup where there’s a nice separation of the curves. There’s clearly a subgroup of patients with refractory disease that respond to checkpoint inhibitors and for a durable amount of time.

So I think now in 2017, we now have 3 good, clear lines of therapy for this disease. And I think there’s a chance, with some ongoing studies, that we could have even more lines of therapy for gastric cancer. Really moving from only 1 line of therapy now to 3 effective lines of therapy is really a huge advance for this very refractory, aggressive cancer.

Case: A 47-year-old man with heavily pretreated advanced gastric cancer experiences a prolonged response with nivolumab and ipilimumab on the Phase I/II CheckMate 032 trial

DR JANJIGIAN: So he started on first-line therapy, and at that time he was nutritionally relatively compromised. And the data for a combination of fluoropyrimidine and platinum therapy is the strongest in terms of response rate and overall survival. We consider addition of a third drug, cytotoxic chemotherapy, in select patients. But at that time, due to his disease status and his nutritional status, I did not believe that a third drug would be warranted, and so he started therapy with 5-FU and platinum, or a FOLFOX combination.

I think it’s important to note his tumor was HER2-negative and, therefore, addition of biologic agents in the first-line setting was not warranted. Trastuzumab was the only option in terms of biologic therapy and, because his tumor was HER2-negative, we opted to treat him with 5-FU and platinum combination.

The patient actually did quite well for some time and has experienced response. He developed significant neuropathy related to oxaliplatin and, approximately after 8 cycles, we discontinued oxaliplatin and we continued the patient on 5-FU maintenance.

And so he actually did quite well on 5-FU maintenance and has remained on therapy for approximately a year and a half. So in first-line therapy, that’s a pretty good run. And again, excellent functional status. And then suffered disease progression with, again, symptomatic obstruction and pain and discomfort.

At that point, he went on to receive second- and even third-line therapy. So in second-line therapy, we’d begun to use paclitaxel/ramucirumab combination. And then he received irinotecan-based therapy. And what I tell my patients is, once the disease becomes chemotherapy refractory after first-line therapy, we’re selecting for treatment-refractory clones. So the duration of response and benefit to second- and third-line therapies are, by definition, shorter.

So although the patient transiently responded to second- and third-line therapy, the response was short lived and again was punctuated by symptomatic disease progression and deterioration of his quality of life. At that point we opted to do palliative radiation therapy, so his primary tumor was radiated, again with little benefit. And that’s when this whole emotional factor of frustration and loss of faith in your oncologist and in medicine in general. And it’s an emotional and totally understandable reaction, but that’s what we were struggling with.

Right around that time, the Cancer Genome Atlas started to become more of an entity that we were aware of. And through research funding and a research program, I was able to sequence his tumor. And although the next-generation sequencing in metastatic gastric and esophageal cancers have not been, at that point, fully explored, interestingly what we found is that the patient’s tumor had a somatic PIK3CA mutation.

And the PIK3CA mutations, although potentially could be a target, in a heavily pretreated tumor I would not have recommended a PIK3CA inhibitor, particularly in tumors that are not HER2 driven that are HER2-negative. But what was interesting is that in the gastric TCGA, 80% of EBV, Epstein-Barr virus-positive tumors, are PIK3CA mutant, which had me wondering if his tumor was EBV-positive.

EBV is not a common pathogen in the West. And it’s generally considered to be a rare entity and, therefore, it’s not routinely tested. There are some characteristics, histologic abnormalities that may trigger a pathologist to test for EBV. It’s a relatively straightforward test, but EBV involves testing for EBV DNA by florescent hybridization, so it’s a FISH type of a test.

But on small biopsies, sometimes it’s hard to clearly see these lymphoid infiltrates and abnormalities that are characteristics of EBV infections. And so some patients don’t have sufficient tumor testing for EBV, et cetera. Luckily in this patient, we were able to get an EBV test, which was positive. And in parallel, also I strongly encouraged him to get enrolled in a clinical trial that we had open at the time, which was before all the immunotherapy data emerged in esophageal and gastric adenocarcinomas.

And he was randomized to receive a combination of nivolumab and ipilimumab on CheckMate 032 study, which is a trial that I recently presented at ASCO. And likely due to his EBV biology and this unique viral entity that can predispose patients to gastric cancer, he experienced complete response radiographically, and on repeated biopsies there’s no longer tumor visible on endoscopy.

So now he’s been 31-plus months on therapy. And ipilimumab was discontinued after the first 4 doses, per protocol. And we’re in this quandary as to do we need to consider therapy? And he’s doing well. He has minimal, if any, side effects other than having to come in and see his oncologist every 2 weeks, which, in itself, can be an anxiety-provoking endeavor. But he is hesitant to discontinue therapy, again, because he’s so worried and he vividly remembers what it was like to be sick from his cancer.

Incidence of microsatellite instability (MSI)-high tumors in advanced gastroesophageal cancer

DR LOVE: I’m kind of curious. And this maybe was a couple of years ago, but did he have MSI testing?

DR JANJIGIAN: Yes. So, actually, NSG — next-generation sequencing — is the 450-gene panel, which is compared to matched tumor. And this is the so-called MSKCC IMPACT, but there are other commercial entities that can do this. And what you do is, it’s a surrogate for mismatch repair, or MMR testing. And what you do is, you get a mutational burden, which can calculate the number of mutations per the base pairs that are tested but also can give you this MSI score.

So this was not an MSI tumor. In MSI tumors in metastatic gastric — and this was another case that we can discuss — generally you see specific types of frame shift and mutations, but when you’re getting their report, often it’s a 2-page report, all with mutations. So this patient also only had 2 alterations in his tumor, a PIK3CA and a KRAS mutation, actually a classic KRAS mutation, which is otherwise very rare in gastric. And it was not an MSI tumor. Clearly there were very few mutations.

DR LOVE: So I’m kind of curious, first of all, we’ll get back to this case in a second. But at this point, do you consider MSI testing as some sort of standard with metastatic gastric?

DR JANJIGIAN: Absolutely. And so actually this is a paper that is soon to be published based on our initial experience with close to 350 patients in the metastatic setting that we sequenced prospectively at MSKCC. And MSI gastric cancer can be detected on next-generation sequencing and should certainly be done in a standard prospective way, particularly these patients. And we have this data that suggests these patients tend to be platinum resistant. They progress rapidly on platinum-based therapies. And now with the FDA approval of pembrolizumab in chemotherapy refractory in MSI solid tumors, the MSI testing in gastric cancer should be routine.

DR LOVE: And in your series, what fraction of these patients are MSI high?

DR JANJIGIAN: MSI tumors in metastatic series is rare. It’s approximately 5%. And this is actually similar to the data that was presented from the KEYNOTE-059 study, which was recently presented at ASCO in the metastatic setting. Again, only approximately 5% of tumors are MSI high in metastatic disease. This proportion is higher in early-stage tumors. But again, these patients are important to find, because you are changing their lives, potentially, with these therapies, the immunotherapies.

DR LOVE: Yes. Actually, I just heard at ASCO — I didn’t realize that, because I always thought MSI was fairly high in gastric. I think it was Charlie Fuchs, was at a symposium we were doing. And he was saying what you just said, that it’s kind of like colon. You see it more in early disease.

DR JANJIGIAN: The gastric TCGA, Cancer Genome Atlas, in early-stage nonmetastatic tumors, shows that 22% of gastric tumors are MSI. What we see, though — so this was stomach TCGA, stomach cancer TCGA. When we recently published in Nature this year, 2017, the esophagus and the stomach TCGA together — so these are adenocarcinoma of the esophagus and GE junction — even in early-stage tumor, that MSI number went down to approximately 12%.

So in the US, the esophageal GE junction tumors are more common than the distal stomach tumors. And yes, MSI tumors tend to have a more indolent biology. And it’s probably due to the host immune response and ability to suppress metastasis. Once resected, they tend to not metastasize. However, if you’re an MSI patient and you develop metastatic disease, the chemotherapy response tends to be shorter.

DR LOVE: Any thoughts about the EBV connection? When you were talking about him, I was kind of flashing on the data your colleague Craig Moskowitz presented in Hodgkin lymphoma, checkpoint inhibitors, so effective. And EBV, people don’t even know exactly what it means. Do you have any hypotheses about why EBV tumors might be more sensitive to checkpoint inhibitors?

DR JANJIGIAN: Majority of these tumors are PD-L1-positive, and they’re clearly immunogenic. I don’t necessarily think PD-L1 is the best biomarker, immunohistochemistry by PD-L1, but it is a sign of the tumor more visible to the immune system. And these viral tumors, in general — with anal squamous cell carcinoma, although the responses are not as high, but there are 25% response rates. So again, the host factors and ability of the immune system to recognize what it has already seen before.

DR LOVE: I mean, I know it’s different, but I always thought it was kind of weird. In head and neck cancer, from what I understand, the HPV-positive patients seem to respond as well as the smoker/drinker patients.

DR JANJIGIAN: Right.

DR LOVE: A completely different mechanism, yet —

DR JANJIGIAN: Completely. So that kind of puts a hole in that hypothesis. And a lot of it we don’t know. I mean, this particular patient got a combination of therapy, of nivolumab/ipilimumab. And one could hypothesize that perhaps monotherapy would have been enough, but we really don’t know, because anti-PD-1 and anti-CTLA-4 tends to have higher responses.

DR LOVE: Yes. I was thinking that. As you were presenting this case, I was thinking to myself, “I wonder if he would have” — kind of like the high PD-1 in melanoma.

DR JANJIGIAN: Right.

DR LOVE: Maybe you don’t really need both.

Results of CheckMate 032: Efficacy and tolerability of nivolumab alone or in combination with ipilimumab for advanced chemotherapy-refractory gastric, esophageal or GEJ cancer

DR LOVE: I was curious. You said he did really well, yet he got the combination, which kind of has a bad reputation. And you just reported data on that combination. What did you see? It was just 1 arm, but what did you see in terms of efficacy and, particularly, in terms of tolerability/toxicity?

DR JANJIGIAN: Right. So the data you’re referring to is the CheckMate 032 study, which is a 160-patient esophagogastric cohort, which was completed as part of a larger basket study, where patients with small cell cancer and other tumors were enrolled, ovarian cancer. And so what we saw is, nivolumab monotherapy was very well tolerated. And the combination of nivolumab with ipilimumab had higher Grade 3/4 toxicities even at the lower doses of ipilimumab.

And so these toxicities were manageable, and most patients recovered. There were no colitis- or pneumonitis-associated deaths, which is what you’re worried about when you’re giving the high doses of ipilimumab. We did find that the toxicities were mostly asymptomatic LFT abnormalities, blood work abnormalities, such as adrenal insufficiency, and abnormalities that could be intervened on if the patient is watched carefully.

And particularly in the younger patient population, like this patient, is, combination therapies are probably relatively safe. And this is, again, part of what we’ve seen reported with these dual agents. In our disease, in esophagogastric cancer, higher dose of ipilimumab, similar to what was approved in melanoma, seems to be more effective. And if you are going to do combination, you’re better off doing the higher dose, because even the lower dose of ipilimumab, there’s still toxicities associated with it.

DR LOVE: What did you see in terms of efficacy? And how does that compare to what we’ve seen in trials of monotherapy with anti-PD-1?

DR JANJIGIAN: I’ve treated a lot of patients on the study. And the response to double therapy, to dual therapy, they were clearly higher. And the data speaks for itself. It was 24% response rate with the combination therapy, which is heavily pretreated, third-/fourth-line patients. I mean, 45% of the patients on the trial were third line and beyond. So basically your only drug left at that point is your irinotecan, which is not going to be very effective.

And so to see responses in these patients is quite striking. And again, it’s from the data that we see now — and this was not the design of this trial, so it’s hard to interpret it, but it does not clearly translate into improved survival for the combination versus the monotherapy. So this question is further being answered and asked in more prospective Phase III studies.

DR LOVE: Can you go through the numbers specifically in your trial? This was what, a randomized Phase II?

DR JANJIGIAN: It was a Phase IB/II. And basically, the intent of it was to explore these and to study nivolumab monotherapy — nivolumab 1, ipilimumab 3 or nivolumab 3, ipilimumab 1 — and to see the efficacy. The arms were not open all at the same time. Nivolumab was open the longest, so you have the longest follow-up in nivolumab monotherapy. But there were long periods of time where these were randomized arms.

So that being said, there were no formal statistics to compare the arms. So there were no p-values or biostatistical plan. These were supposed to be stand-alone arms to assess efficacy. And what we saw, again — like I mentioned, heavily pretreated population — majority of the patients progressed on 2, and 45% progressed on 3 lines or more. The treatment of nivolumab monotherapy was well tolerated, and in PD-L1-positive patients the responses were quite high.

The response rates are higher and, in PD-L1-positive patients — again, it’s a small number, but 40% response rate in PD-L1-positive patients with nivo/ipi, which is, again, quote, hypothesis generating and provocative.

DR LOVE: Exciting enough to consider outside a clinical trial setting?

DR JANJIGIAN: I would say a third of the time when I submit the approval to the insurance companies, they now approve it. I submit the CheckMate 032 abstract, they see me as the first author. And so I think it’s worth a try. I mean, it’s not proven, certainly, but outside of other trials, if there’s no trials available and there’s no other standard options, I would consider it.

Perspective on duration of immune checkpoint inhibitor therapy

DR LOVE: What about the issue you mentioned — and this comes up in a lot of different cancers — about whether to stop therapy? As you mentioned in this man, he wants to continue. How do you feel about it? If he said, “What do you think I should do,” do you think it’s reasonable to stop?

DR JANJIGIAN: Yes. So, I mean, in him the answer is easy, because he’s in a trial. And so we’re guided by the trial. But I can tell you the culture has changed in these immunotherapy trials. And most patients stop after a year and then are given an option to resume if there’s evidence of clinical or radiographic progression.

DR LOVE: Yes. It’s funny, because I just started hearing that recently, more and more people talking about stopping preemptively, which I hadn’t heard that mu — it was the old, like, tamoxifen. If everything’s okay, let’s keep it going.

DR JANJIGIAN: Right.

DR LOVE: But as you point out, it’s not necessarily a walk in the park. And I guess also sometimes I’m starting to hear about delayed autoimmune problems that you see. Have you observed that yourself?

DR JANJIGIAN: Yes. We do see a higher incidence of hypothyroidism and adrenal insufficiency that patients get, and these are subtle and sometimes present with fatigue. And so if you’re not aware and carefully monitoring them, you can miss it.

Ongoing investigations of immune checkpoint inhibitors in the (neo)adjuvant setting

DR LOVE: Can you talk about the other research that’s been done with using checkpoint inhibitors in metastatic gastric?

DR JANJIGIAN: So there’s several exciting trials that are ongoing. Probably one of the most promising avenues is to try to move these agents in the neoadjuvant or adjuvant setting to try to treat micrometastatic disease. And there are different schools of thought on that, whether or not you need more antigen to be in the body before you can sensitize or respond to immunotherapy. That being said, these are some of the trials, both in neoadjuvant and adjuvant setting, that are being explored both in gastric and resected esophageal adenocarcinomas.

In Asia, there’s a number of trials with squamous histologies. In the US, squamous esophageal cancer is relatively an orphan, even more so orphaned than an adenocarcinoma. But in China and parts of Japan it’s still very common, so that’s where it’s being explored. For the majority of our patients in US, metastatic disease remains the main battle. And so they’re both, pembrolizumab and nivolumab and nivolumab/ipilimumab, are being explored in the first-line setting in combination with chemotherapy in gastric.

Sequencing of chemotherapeutic regimens and immune checkpoint inhibitors for metastatic gastric cancer

DR LOVE: Can you talk about where you see checkpoint inhibitors fitting in right now in the sequence of therapy?

DR JANJIGIAN: Yes. So, I mean, you’ve seen those data coming from Japan from ATTRACTION 2. You’ve seen the data that I presented, that Charlie Fuchs presented at ASCO. There’s clearly a tail on the curve, and there’s a flattening of the curve. And some people are truly — their lives have been changed with these drugs. But that percent is very, very small — only approximately 10% to 15% of patients are truly long-term responders and benefit from these therapies.

And what we’re working on, me personally but also others in the field, are trying to ident — the best way to identify those patients, because as you mentioned, these therapies are not without toxicity and the cost of them are quite high. And so we need to identify the biomarker to best enrich these populations. MSI is certainly a possibility, EBV. And although PD-L1 helps enrich for a population that benefits, there’s certainly PD-L1-negative patients that respond. So that may not be the best and the only biomarker.

DR LOVE: Can you talk about right now where you see checkpoint inhibitors fitting in, the sequence of therapy and what line therapy should they be used?

DR JANJIGIAN: Right. So for MSI patients, I would probably use them in second-line therapy, although again, this will depend a lot on disease burden and the availability of the testing and the drugs. But these patients should be tested in the first-line setting for these biomarkers, so that you can plan their treatment. And the way I envision it is it’s a chess game, and you need to plan your strategy out ahead of time. For unselected patients in the metastatic setting, I see these agents being in the near future FDA approved probably in a third-line setting.

DR LOVE: And from your point of view, where do you think right now, based on the data that we have, it would be optimal to utilize these agents? And I’ll ask you, for example, you have an older patient, frail. You’re a little bit concerned about giving chemotherapy. They don’t have a lot of disease burden. Why not give them it first line? It’s less toxic than chemo.

DR JANJIGIAN: Yes. But it’s not available first line. In the first-line patient, I would probably not do compassionate release treatments, because you have a lot to lose. I mean, these patients do well with monotherapy with 5-FU in the frail patients. And you could even potentially do a dose-reduced combination of 5-FU/platinum. I mean, these drugs do cause side effects. You can develop acute interstitial nephritis after 1 dose of an anti-PD-1 agent, and then you won’t ever get to chemotherapy. And so to do this outside of a clinical trial in the first-line setting, I would probably not advocate for that.

DR LOVE: What about second line, as opposed to paclitaxel/ramucirumab?

DR JANJIGIAN: Medicine is an art. And I always tell my fellows, “You cannot necessarily” — a lot of what we do is not that straightforward. And so I’m seeing patients now who progressed on neoadjuvant chemotherapy, progressed on neoadjuvant chemoradiation, within 2 months of completing this developed liver metastases. So they’re clearly aggressive chemotherapy-refractory disease. And so for me to tell them, “You got to get more chemotherapy,” it’s hard to swallow for the patients. And so we try to get them on a clinical trial with immunotherapy and, if there’s no clinical trial, these agents are available on compassionate release basis.

Case: A 52-year-old woman with MSI-high, Stage IV GEJ adenocarcinoma

DR LOVE: You have your 52-year-old patient, also looks like MSI high. Is that the case?

DR JANJIGIAN: Yes.

DR LOVE: Can you just briefly talk about what happened there?

DR JANJIGIAN: Okay. So this is a woman who was getting treated in the community. And she presented with locally advanced, curable GE junction adenocarcinoma, presented with dysphasia and had extensive disease imaging that was negative for metastatic disease. So she was started on chemotherapy with radiation, and her primary tumor was irradiated. And while she was on therapy with a cross regimen, which is carboplatin/paclitaxel with radiation, she developed bulky metastatic disease, so this is, what you would say, chemotherapy-refractory setting.

She was then referred to me for an opinion and management and stayed with me for therapy. She had a large bulky lymph node, which was biopsy-confirmed metastatic adenocarcinoma and was starting to quickly get obstructed despite radiation. She was not able to swallow, required a feeding tube placement. We started her on FOLFOX in the second-line setting, and it just made her tumor angrier. And this is what I was alluding to earlier. MSI tumors tend to really progress quickly on platinum therapy. They are not chemotherapy susceptible to the same degree as the general cohort of patients.

So she had a feeding tube in and got admitted, was quite frail and the discussions were about hospice and best supportive care, unfortunately. So at that time, the nivolumab monotherapy arm was open. And her functional status — she’s a young woman, so despite of all of the comorbidities and issues, she qualified for the study.

And just after 8 weeks of therapy — so it’s nivolumab every 2 weeks, so 4 doses — her lymph node started to melt away and she was able to swallow, first her own secretions, which she was unable to before. And then her eating improved and dysphasia resolved. We actually removed her feeding tube. And radiographically, she had a complete response. I mean, it was quite satisfying. You feel like you’re bringing someone from the other side.

But again, these responses are rare. And she had persistent disease left at the time of the endoscopy. Her disease started to change and become more from adenocarcinoma to this just poorly differentiated carcinoma, just ugly badness that doesn’t have any architecture to it. And she progressed on the monotherapy within 14 months of having excellent response. And it’s interesting. She only progressed in the primary tumor.

So any sites of metastatic disease were still gone and were complete ongoing CR, and only the primary tumor progressed. And so at the time of biopsy, her tumor developed a secondary mutation in the beta 2M, which is a melanoma-acquired resistance mechanism, basically. It’s a loss-of-function mutation, where the drug can no longer bind and inhibit the pathway. So she’s still actually doing okay and alive. We were able to give her another short dose of radiation therapy to the — and I gave her nivolumab/ipilimumab combination. I was able to switch her over.

And again, she didn’t have as good of a response to combination, but the PET scan did get better before radiation even started. And so now it’s been 4 years since she was diagnosed with her disease. And in the absence of other metastatic disease, her surgeon is actually considering taking her to the OR, because she’s a young woman who’s functionally well and has this permanent issue and is at risk for fistulas, because she’s been radiated twice.

And so this is again uncharted territory, and certainly we have to consider it case by case. And I’m not generally a big supporter of surgery in this situation, but it may be her only solution.

DR LOVE: Wow! That’s an incredible case. When along the way did you find out she was MSI high?

DR JANJIGIAN: So we biopsied her and sent her tumor for sequencing. And pretty much as she was starting — so I was going to start her on immunotherapy probably anyway, but her functional status was so — for full disclosure, it was borderline, you can imagine. And I really had to push and advocate for her. And then when I saw the MSI status, I really kind of just basically didn’t take no for an answer.

DR LOVE: It’s amazing. Interview people like this in different fields, I’ve had so many cases I’ve heard in the last month of people who were considering hospice, and then — I mean, just the idea that you’re considering hospice and 3 months later you’re walking around. At a human level —

DR JANJIGIAN: She went back to get her Master’s in education.

DR LOVE: Wow! Amazing. Amazing. The other thing about this case that’s obviously really fascinating is why she progressed in the primary. And I’m curious. Any speculations, biologically? I mean, you mentioned the mutation, but I was wondering. I mean, that was the only area she got radiated, right? Because we have this abscopal thing that somehow it’s better. Could you envision that in some way the radiation was part of the resistance or something?

DR JANJIGIAN: Yes. I have had cases, abscopal-like cases, where the radiation does help, actually. And in her situation, certainly, she didn’t have other sites of disease, so it was hard to assess that. But I have had patients who were progressing, and then we radiated the lesion, and other spots elsewhere started to shrink. Again, it’s a little premature to call it abscopal effect, but there are certain entities like that.

I think for her, possibly due to radiation but also because of just having persistent disease that was bulky and didn’t quite go away completely, it just turned it into a zombie. It was no longer an adenocarcinoma. It was this ugly, almost kind of small cell type of tumor that — it’s like reasoning with a rabid dog. You can’t.

DR LOVE: Is she currently on the combination?

DR JANJIGIAN: No. So she had quite a bit of just swallowing still issues and everything and now is on feeding. Her feeding tube is back in, unfortunately, because of this reradiation and everything. But she’s gaining weight and looks well. And so I gave her 4 doses of ipi, and I just stopped. And now she’s completely off everything, kind of getting stronger and getting ready for an operation.

DR LOVE: Any thoughts about — particularly in somebody with MSI high, about any other kind of — particularly because she had this great response, any other kind of immune manipulation? I’ve heard — I think they’re called IO — what am I thinking about?

DR JANJIGIAN: Yes. IDOs?

DR LOVE: IDO.

DR JANJIGIAN: There’s also LAG-3, IDO —

DR LOVE: Right, LAG-3 —

DR JANJIGIAN: There are all of these.

DR LOVE: So what are some of the other immune things that are being looked at that might be —

DR JANJIGIAN: Yes. I think one of the most exciting areas is the so-called BiTEs, or bispecifics, antibodies that are designed, again, in combination with immunotherapy with checkpoint inhibitors to augment the immune response. And they’re bispecific antibodies that are directed against CEA. Josep Tabernero presented a podium presentation at ASCO this past year, this past summer. And then also, HER2 bispecific antibodies, some of the drugs that are being developed right now. I think it’s only a matter of time that these patients with acquired resistance and recurrences emerge. And we have to plan ahead.

Case: A 60-year-old man with HER2-positive GEJ adenocarcinoma and liver and lung metastases

DR LOVE: So I wanted to ask you about your 60-year-old man with HER2-positive disease, where I see you had the interesting dilemma that we always talk about, which is second-line therapy. What happened with this man?

DR JANJIGIAN: Yep. So in first-line setting, the patient received trastuzumab-based therapy. And then after progression on trastuzumab, the question is what to do next. This patient had no residual neuropathy, excellent functional status. And the question is, do we continue trastuzumab beyond progression?

Certainly in breast cancer, there’s literature to suggest that this is the way to go, but in gastric cancer, time after time what we’ve run into trouble with is taking lessons from breast cancer and pretending that they’re directly translatable to gastric cancer. And I applaud these companies for doing clinical trials in HER2-positive gastric cancer, but the design of the trials needs to be carefully considered.

So as you’re familiar with, trastuzumab is FDA approved in the first-line setting and showed overall survival, response rate and progression-free survival benefit. In the second-line setting, though, after trastuzumab failure, we already had several negative strategies. T-DM1 failed. Lapatinib failed. And again, T-DM1 is trastuzumab cross linked with also cytotoxic agent called DM1.

They compared it in a large trial head to head against paclitaxel, and there was no difference in paclitaxel versus T-DM1, highlighting the aggressiveness of the biology of HER2-positive gastric cancer. And so it’s just giving trastuzumab with another cross-linking agent was not enough. And paclitaxel works quite well in our disease, comparatively, and so it was not any worse than this fancy new HER2-directed agent.

DR LOVE: And was it paclitaxel or docetaxel?

DR JANJIGIAN: Paclitaxel.

DR LOVE: So before you go on, I always pick a bone with the GI investigators about that trial, because you all call it negative. And I know it wasn’t an equivalency trial.

DR JANJIGIAN: Right.

DR LOVE: But to me, it kind of looked equivalent but with less toxicity.

DR JANJIGIAN: Right.

DR LOVE: But nobody uses it.

DR JANJIGIAN: You could say that, but it was not an inferiority trial. It was a small trial, and to do an inferiority trial, you need 3 times as many patients. But you can’t go back and say, “Oh, never mind.” You just can’t play with statistics like that.

DR LOVE: Right. So I would gather you probably gave this man paclitaxel/ramucirumab?

DR JANJIGIAN: Yep. But what I didn’t put in the slides is, it’s interesting. So this is, again, data emerging. And I presented it at ESMO, but also since then, my collaborators in Korea and Japan have replicated this. In patients with HER2-positive disease at the time of trastuzumab progression, up to 20% of tumors lose HER2 expression.

And so actually, in this particular patient, I biopsied him and the tumor was no longer HER2-positive. So in that sense, I was also very reassured that paclitaxel/ramucirumab would be the right approach.

DR LOVE: Interesting. So how did he do on the paclitaxel/ramucirumab?

DR JANJIGIAN: He’s still on it. I mean, again, it’s relatively well tolerated, but certainly a majority of the patients suffer disease progression. And that’s why, in the third-line setting, our options are real limited right now.

Case: A 62-year-old woman with metastatic gastric cancer previously treated with FOLFOX who is now receiving ramucirumab/paclitaxel

DR LOVE: So I want to tell you about a case actually that we had presented by a general oncologist at one of our ASCO meetings recently. But first, before we get to that, I want to hear about your 62-year-old lady, because I think it kind of relates to that. This lady also had gotten FOLFOX for metastatic HER2-negative disease and then went on paclitaxel/ramucirumab. What happened when you gave her that treatment?

DR JANJIGIAN: So in patients who receive oxaliplatin, neuropathy is a factor. And although with paclitaxel you don’t get as much cold sensitivity, the neuropathy can be additive. And I’ve had patients — not this patient, but several patients — who develop significant neuropathy, where paclitaxel either had to be dose reduced or discontinued.

Generally, in a frailer patient with neuropathy as a preexisting factor, I tend to dose paclitaxel every other week even, because of the ramucirumab dosing. And so in this patient, after you achieve maximum disease benefit to paclitaxel, meaning maximum response on scans and then disease stabilization, I discontinued paclitaxel and continued ramucirumab monotherapy as a maintenance approach.

DR LOVE: Yes, but I see that she had a problem with neuropathy.

DR JANJIGIAN: Yes.

DR LOVE: So that was occurring off paclitaxel?

DR JANJIGIAN: No. It happened while she was on paclitaxel/ramucirumab.

DR LOVE: And so once you stopped it, did you continue the ramucirumab?

DR JANJIGIAN: Yes. You continue the ramucirumab. The neuropathy improves, but again, because of her prior oxaliplatin, it doesn’t go away. Generally, we have them see a neurologist. If it’s painful neuropathy — again, mostly from oxaliplatin — pregabalin or agents like that can be considered, gabapentin. And physical therapy helps manage some of these symptoms.

DR LOVE: What do you do with patients who’ve already had a taxane?

DR JANJIGIAN: So those patients are rare. Generally if they’ve had a taxane, it’s in the neoadjuvant setting. And in that case, there’s actually data coming out from Japan saying that you could potentially, in the patients who had carbo/paclitaxel in the neoadjuvant setting, they tend to perhaps still respond to docetaxel, although that’s kind of not necessarily clear, that data.

And so in patients who have not had taxane, paclitaxel/ramucirumab is really the approach. Monotherapy of ramucirumab has been shown to be disease stabilizing, but it has not, in my practice or in published literature, shown to have significant responses. So it’s more of a disease stabilizer.

DR LOVE: That’s why I want to present my case.

DR JANJIGIAN: Go for it.

DR LOVE: So believe it or not — and I saw the scans — this was a patient, second-line gastric cancer. And the doc wanted to use paclitaxel/ramucirumab. And the patient was going on vacation, and she said, “Can we delay the paclitaxel?” And she kept talking him into delaying it until she got her first scan at 3 months and had a significant PR, almost a CR that now continues, I think, 2 and a half years later. So my first question is, it sounds like you’ve never seen that?

DR JANJIGIAN: Excellent. That’s good. I mean, we’ve had some reductions. And I have a clinical trial of actually in HER2-positive disease with ramucirumab/trastuzumab with a run-in of these 2 biologics before we add chemotherapy. And I have seen responses, but there, you don’t know if it’s the combination or if it’s the trastuzumab they’re responding to. But in second- and third-line patients with monotherapy, I have not seen any responses.

DR LOVE: So this doc hadn’t either. So here’s what really makes it interesting: Now she’s, like, 2 and a half years —

DR JANJIGIAN: Wow!

DR LOVE: — into therapy, and she’s got 2 grams of proteinuria. And, I mean, hypertension, but that’s under control. And so his question is, what do you do? I mean, this comes up more, like, maybe in colon cancer, if it ever comes up. But any thoughts about — he was saying her creatinine clearance is fine. Her renal function, just like —

DR JANJIGIAN: I would leave her on the therapy. Because dose reduced perhaps or — but I wouldn’t change it, because honestly, asymptomatic proteinuria in a patient with Stage IV disease, it’s the lesser of the 2 evils.

DR LOVE: He actually dose reduced her, and she still had the proteinuria. And he decided to do exactly —

DR JANJIGIAN: Full disclosure, I don’t monitor proteinuria.

DR LOVE: Yes. That’s probably even a better idea.

DR JANJIGIAN: Just don’t look, or…

Case: A 70-year-old woman with metastatic gastric adenocarcinoma and a history of congestive heart failure

DR LOVE: Anyhow, I wanted to ask you also — let me see here. You had your first case here — yes, right — your 70-year-old lady with a history of heart failure. Can you talk about how she presented and what the issues were with her?

DR JANJIGIAN: So this is an interesting case, because the question comes up. And I still, when I speak to the fellows or for folks in the community, these patients, the combination therapies in elderly patients. So this is a 70-year-old woman with heart failure and — although well compensated — and right upper quadrant pain and was found to have anemia, an albumin of 3.0, which is generally considered to be a negative prognostic factor in patients with metastatic disease.

And so the question here is, what’s the appropriate agents to use? And certainly some of the literature historically has advocated the use of 3-drug therapies in metastatic gastric cancer. It’s important to note that this patient has multiple comorbidities. And drugs like anthracyclines and taxanes would probably exacerbate her functional status. And you may not be able to get additional lines of therapy in her.

And a patient like this is a perfect example where combination therapies of 5-FU and platinum — and again, cisplatin should be avoided. Oxaliplatin has just as good of track record in terms of efficacy. And it may not be very sexy, but FOLFOX is really the best option in some of these patients.

DR LOVE: And so what happened with this lady?

DR JANJIGIAN: So she did well. Again, we had to do multiple dose reductions. And generally in elderly patients, I start at lower doses. But she responded, and clinically her albumin improved and she started to eat. And she tolerated the combination therapy well.

New agents on the horizon in gastric cancer

DR LOVE: So you had a really great paper in the Annals of Oncology, New Agents on the Horizon in Gastric Cancer. And, of course, you went through checkpoint inhibitors that we’ve already talked about. But there were a couple of agents and strategies that you mention in the paper that I wonder if you could elaborate on.

And one is something I’ve heard about for a long time and kind of wondering whether it’s ever going to end up in practice, which is the concept of stem cell inhibitors, particularly the drug napabucasin, which I think used to be called BBI-608. Can you talk a little bit about what these kinds of agents are and what we know about this strategy?

DR JANJIGIAN: Yep. So this whole cancer stemness, the idea is that the progenitor cells to the cancer are hard to eradicate with chemotherapy. And if you target the stem cells and the stemness of it, you will help eliminate and treat the cancer.

So these agents are being explored in combination with chemotherapy. And although this is an interesting avenue of investigation, certainly right now, it’s not prime time. And we’re still awaiting clinical trial data.

One of the other classes of drugs that are up and coming are these claudin inhibitors, claudin 18-2 rearrangements — were actually noted in the genomically stable or this diffuse gastric cancer cohort in the TCGA. And what’s interesting about this protein is that it’s a cell adhesion protein. And again, historically the signet ring cell type gastric cancer, or the E-cadherin-driven gastric cancers, have been very challenging to target. They generally don’t respond as well to chemotherapy. The patients develop peritoneal carcinomatosis and become quite sick. And so these claudin rearrangements could be the way to target these drugs.

The Phase II data that was presented at ESMO was very promising, but again, we’ve been burned by Phase II data in the past multiple times, including MET inhibitors and so on. And so actually, the Phase III trial is being planned even as we speak.

DR LOVE: And there is, like, a predictive marker that’s used?

DR JANJIGIAN: Yep. It’s immunohistochemistry for the protein. And it seems to be a relatively useful biomarker.

DR LOVE: What about tolerability issues both with this agent — I mean, would you consider this — I guess it’s a monoclonal antibody-targeted type therapy? Would you look at it that way?

DR JANJIGIAN: Yes. It’s a monoclonal antibody. And the original study used anthracyclines, again, as a backbone. And I think in combination, actually, between our institution, MD Anderson, and others in the field, we put out a statement in JCO trying to put a stop to the use of anthracyclines in the metastatic setting. And so it’s a difficult backbone to build on. And biologically, there’s really no rationale to use it for the claudin inhibitors. And I believe the drug should be relatively well tolerated in combination with 2-drug cytotoxics.

DR LOVE: We were talking before about napabucasin. What about tolerability issues with that? I’ve heard about GI issues.

DR JANJIGIAN: Yes. I personally haven’t treated anyone with the drug, but I think we only care about tolerability if there’s any sign of the drug being effective. And I think that jury’s still out.

DR LOVE: Do you buy into the mechanism involved here?

DR JANJIGIAN: I think it’s a very provocative and interesting and scientifically exciting mechanism. Whether or not it will be the way to treat our disease, we still have to wait and see.

Case: A 68-year-old woman with locally advanced gastric adenocarcinoma

DR LOVE: So you also have a case here of a 68-year-old lady. Maybe you can go through that one. This is the lady who presented with — it says Case 1, former smoker, epigastric discomfort. Do you have that one there?

DR JANJIGIAN: Yep.

DR LOVE: Okay.

DR JANJIGIAN: Yep. So this Case 1 is locally advanced disease and, again, goes into the difference between the tumor location and driving the type of operation and the type of neoadjuvant therapy that we would recommend. So this was a 68-year-old woman with locally advanced, under endoscopic ultrasound was a T3N1 tumor. And the question is, how would you stage this cancer and how would you approach the treatment?

So for the majority of these patients, it’s important to rule out metastatic disease. And even on the CT scan, often if the CAT scan is negative, we can pick up occult metastatic disease in 10% to 15% of the patients by FDG fluorinated glucose PET scan. And so these patients tend to develop metastatic disease to supraclavicular and retroperitoneal lymph nodes, which you can miss on the scan.

And also, in a patient who presents with weight loss and symptomatic cancer, even in patients with a PET scan negative of metastatic disease, a laparoscopy is necessary to rule out occult peritoneal carcinomatosis, particularly in patients where the tumor, the majority of the tumor, is in the stomach.

So this patient had a perioperative laparoscopy, which was negative, no evidence of peritoneal disease on the PET scan. And then the primary tumor, the standard uptake value was quite bright, SUV of 14. So what do you do for this patient? And generally, there’s a lot of consensus to show that something more than surgery needs to be done. If the tumor was mostly in the esophagus and the higher portion of the GE junction, the consensus is that chemoradiation should be the standard.

In this patient, however, the majority of the tumor was at the bottom portion of the GE junction and extending into the stomach, so your classical gastric cardia proximal tumor, where the radiation generally is not indicated. And there’s some sway on this data. And certain academic centers still advocate radiation for it, but I would not radiate this tumor. And we elected to treat her with preoperative therapy.

Then the question comes up — and actually, since this patient was treated, the data has changed. You probably saw Al-Batran present the FLOT-4 data at the ASCO 2017 meeting. And so in this patient who’s otherwise fit and is able to tolerate 3-drug combination, with the current standard it would be a combination of 5-FU/oxaliplatin and docetaxel or the FLOT-4 regimen.

Before the data came out, the OEO5 data, which was presented by Cunningham, showed that epirubicin, or EOX-based therapy, did not do any better than FOLFOX or 5-FU/platinum. And so for this patient in particular, she received FOLFOX as a preoperative treatment.

DR LOVE: And do you want to elaborate on — you mentioned the editorial that you wrote about the issue of anthracyclines. Can you comment on that?

DR JANJIGIAN: Right. So we need to move this field forward. And the only way to move a field forward, if there’s a consensus among experts. Until last year, there were still, at these meetings and scientific consortiums, a lot of discussion back and forth, what to use as a backbone for clinical trials and so on. And so there’s this perceived disagreement.

So Jaffer Ajani and I thought it was very important to come together and write a statement piece about what we as experts think and agree on. And so it’s basically a summary of pertinent highlights of the recent negative studies, including the CALGB study of FOLFOX versus carboplatin/irinotecan versus ECF. So the consensus statement was basically to summarize and to show that there’s no benefit for addition of epirubicin in the metastatic setting and it should be avoided. And it will help us consolidate all the efforts and advance the field forward, if we can agree on what the backbone should be in metastatic patients.

DR LOVE: So getting back to your patient, what’s her current situation?

DR JANJIGIAN: So she went on to have surgery. So we actually tend to use the PET scan as an assessment of biologic response. This is based on Florian Lordick’s data in the MUNICON trial, showing that the patients who don’t respond to the first cycle or 2 cycles of FOLFOX — and the response is defined by change in SUV on FDG by 35% or more. So the tumors that don’t respond biologically represented more aggressive tumors and should probably either go on to have surgery, or there’s also an Alliance trial that looked at switching therapy.

So this patient had FOLFOX, responded beautifully and then went on to have surgery and had 99% treatment response, again, unusual scenario in this disease. But she had an excellent response. And then there’s a conundrum. So she’s elderly. She had a complete gastrectomy, total gastrectomy, for a proximal tumor. Most of these patients lose up to 10% to 15% of their body weight in the postop setting and so nutritionally quite compromised.

And so you’re seeing this patient who now has a ypT1BN0 tumor — so downstaged from T3N1 tumor — and she’s asking you, “What should we do?” And obviously she is thankful and wants to capitalize on this response and does not want to have her cancer recur, because once it recurs, obviously it’s no longer curable. And in the adjuvant setting, even the FLOT-4 data and other clinical trials data shows that half of the patients only are able to receive therapy, if we’re lucky.

And so in this particular patient, she rallied. She did well. She didn’t gain weight, but her weight stabilized. And then, because she was motivated and she has a grandchild on the way and everything, a lot to lose and at stake here, we elected to do 3 more cycles of FOLFOX.

DR LOVE: And I’m curious what you would predict her prognosis to be, her chance of relapse at this point.

DR JANJIGIAN: It’s a very hard number to predict. And what I tell my patients is, they’re cured and they should feel they are. And everything else is inconsequential. They shouldn’t be Googling numbers and et cetera.

But, obviously, with node-negative disease, her chances of recurrence are substantially lower, but they’re not zero. And there are new and emerging technologies to try to predict when to stop chemotherapy, which patients are less likely to recur. I’m particularly interested in that field and using cell-free DNA as a biomarker to try to hone in better on this risk of recurrence.