DR LOVE: So if you think about, I guess, the extremes of the clinical presentation, you have esophageal squamous cancer, you have GE junction cancer, I guess usually adenocarcinoma, and then you have gastric cancer. So at least in my mind, I think that’s kind of how clinicians begin. But when you’re sort of sorting it out, how do you think through location, histology and then what you’re going to do in terms of neoadjuvant and post-neoadjuvant treatment?

DR WAINBERG: Yeah. So esophageal squamous, esophageal adeno often present in similar clinical ways. They present with obstructions, they present with dysphagia, odynophagia, weight loss, lower chest pain, upper abdominal pain. Whereas gastric cancer, true gastric body cancer really presents a little differently. It’s more associated with weight loss, gastric outlet obstruction, early satiety, those kind of clinical symptoms. So there are some subtle distinctions between how these 2 cancers manifest and get to the GI doc. The GI doc obviously has got to find the diagnosis and make the diagnosis correctly. When we’re approaching it from a localized fashion and they’ve completed their staging and they’re, okay, they’re nonmetastatic, let’s call it, whether you want to call it someone who is resectable right now or you want to call it someone who could be resectable, we know at least they’re nonmetastatic. And those patients usually are either launched on a path of neoadjuvant chemoradiation. And I believe personally that’s still the best approach for tumors that are at or above the GE junction. And anything below the GE junction, predominantly stomach tumors, I still think the standard of care there is neoadjuvant chemotherapy with FLOT. So the dividing line for me is still somewhere midway through the GE junction recognizing that this is usually a multidisciplinary discussion, recognizing that there’s a lot of opinions in either side of the direction. I usually use that as the dividing line. And patients above it, I generally prefer neoadjuvant chemoradiation with carbo/paclitaxel.

DR LOVE: To what extent does histology affect your approach to this situation?

DR WAINBERG: Histology doesn’t affect it in so far as it doesn’t trump staging. For example, in patients who have a T1 or T2N0 adenocarcinoma of the esophagus or squamous cell carcinoma of the esophagus, those patients are resected. There’s no real data and no real value, if you believe in your clinical staging at least, to any — to give them neoadjuvant therapy. Where it gets a little sticky is in the circumstances where it’s a T2N0 or T3 and early T3N0 which is a lot of times what the GI endoscopists will give us. And in those terms truthfully, I do tend, if it’s a T2N0 that’s subjectively symptomatic, I tend to give a lot of those patients neoadjuvant as well nowadays.

DR WAINBERG: With either neoadjuvant chemoradiation or chemo. Yeah.

DR LOVE: In terms of esophageal cancer in terms of how you manage those patients, again, does histology matter?

DR WAINBERG: Not in so far as neoadjuvant or perioperative approaches, Neil. It really — histology comes down to a little more in the metastatic setting.

DR LOVE: Right.

DR WAINBERG: And so in the circumstance of locally advanced, it’s still CROSS, it’s still adjuvant nivolumab independent of histology.

DR LOVE: Okay. So I think somewhere along the way, probably even at diagnosis, or you tell me what your typical biomarker workup would be in a patient with localized disease.

DR WAINBERG: So with localized disease, to be fair, a lot of the biomarker stuff is, you know, it’s interesting, it’s academic. We don’t have drug approvals. So to be honest with you, it doesn’t really change the management that much. And to be real practical here, I like getting the biomarker stuff because I think it’s really important, particularly MSI-high. That is the distinction. I would say if a patient has localized gastric cancer or even localized GE junction adeno, because in squamous cell, it almost never happens. But those 2, I do think it’s important to know that and I do think those patients ought to be treated with immunotherapy from the get-go. With the exception of that, and that includes HER2, that includes CPS score, that includes whatever other stuff we’re going to talk about, it doesn’t have at this juncture any practical implication for treatment.

We collect it because in practical terms, sometimes it’s easier to collect it and have the data later for when they recur.

DR LOVE: I would think a PD-1 assay would be something you’d want to know. Maybe it’s not going to change what you do, but it seems like it’d be relevant. You tell me.

DR WAINBERG: So at the moment, we don’t have any use, we don’t have any data to support the use of neoadjuvant pembro/nivo, any of the PD-L1 inhibitors in the neoadjuvant context. So I don’t use it even if they do have a high PD-L1 score, personally. I don’t see how that would necessarily change their clinical outcome. That may change. All these trials are ongoing. I may say something totally different to you 6 months from now. But I think standing here right now, I like getting the data, as I mentioned, but even if they have a high PD-L1 score, I’m not sure — I am not going to use neoadjuvant nivo or pembro yet.

DR LOVE: Okay. And then in terms of postop management of these patients, both the patients who get CROSS chemoradiation as well as the patient who gets FLOT, how do you approach that?

DR WAINBERG: So in CROSS patients, we have the CheckMate, you know, we have the CheckMate data 577 study. We have a clear role for adjuvant nivolumab in patients who have had chemoradiation, surgery and have some residual disease. Those patients, it’s an absolute, in my opinion, unless there’s a contraindication, good reason to give them nivo. And that’s independent of squamous or adeno. And that’s independent of PD-L1 score, in my opinion. It might work a little better. If they have a high TPS score, it might work a little better if they have squamous cell, but it doesn’t mean you want — you would withhold it in the context of adenocarcinoma of the esophagus. For patients with stomach cancer who commit to neoadjuvant chemotherapy, and that’s at times FLOT, the current standard is to continue that FLOT after surgery if they need more FLOT. It’s not to change the chemo. It’s not to add immunotherapy at this juncture. We’re waiting for very, very big studies that may change that paradigm. But as of now, that’s where we are.

DR LOVE: So I’m just kind of curious in terms of the use of adjuvant nivo after the CROSS regimen. What kind of data do we have on PD-1 negative patients?

DR WAINBERG: There I would use primarily with TPS score so I don’t want to make it confusing, but let’s consider them PD-L1 negative. In the adenocarcinoma PD-L1 negative patients in particular, there wasn’t as much benefit with adjuvant nivo there compared to placebo, to be fair. In squamous cell, it’s a different story. In squamous cell which made up about, if I recall, 75 or 70% of the patient population in that CheckMate 577 study even in the PD-L1 low patients or TPS low patients, there was benefit to nivolumab. So I think that the numbers end up being a little small in that context with the adenocarcinomas only. So therefore, in that context and certainly when we’re talking about a potentially curative scenario, I do practice and give it to those patients regardless of PD-L1 score. I feel a lot better about it if I know they’re TPS positive or high.

DR LOVE: Where do we stand in terms of trials looking at post-FLOT use of IO?

DR WAINBERG: Yeah. So there’s actually some datasets that have been — one data set, a large dataset out of Asia surprisingly was announced to be negative which I believe was one of the ATTRACTION studies, ATTRACTION-5 which looked at after patients who had gastrectomies, chemo plus or minus nivolumab, and did not meet its endpoint. It’s a public announcement. We don’t have the fine details. I’m sure we’ll see that soon. That does make me a little concerned theoretically that the concept of adjuvant chemoimmunotherapy may not hold as well in gastric cancer as it did in esophageal cancer. That being said, the next wave of studies is very different because they looked at it neoadjuvantly and adjuvantly. And there’s 2 large ones that we’re waiting for the data.

At least 2 to my knowledge. One is the KEYNOTE-585 study which adds pembrolizumab to FLOT — or I should say, any perioperative chemotherapy, mainly 5-FU platinum-based. There was some patients getting FLOT on there. And the other one is MATTERHORN which is a durvalumab study which added durva to FLOT chemotherapy exclusively. So those 2 trials and maybe others that I’m not aware of will answer the question a little more.

DR LOVE: And that’s neoadjuvant IO or post-neoadjuvant IO?

DR WAINBERG: They give it both. So they give it with chemo neoadjuvantly and then they give it adjuvantly. In some studies, it’s compared at 6 months. In some, it’s a year. But it’s essentially continued at the completion. They get the 8 cycles of FLOT which is 4 cycles, but 4 months of FLOT basically and then continuing on the immunotherapy on the backend for another 6 months to a year.

DR LOVE: Yeah. So like what happens in triple-negative breast cancer. Okay. Let’s switch over to metastatic disease, first-line therapy. Again, the same model of the 3 locations and how you think it through.

DR WAINBERG: Okay. So squamous cell of the esophagus metastatic disease.

We have 3 options that I would consider reasonable at this juncture. Number 1, chemotherapy plus immunotherapy which includes wither chemo plus nivo based on CheckMate 648 or based on KEYNOTE-590, FOLFOX or 5-FU based chemo with pembrolizumab. Both good options. Both clearly benefiting patients in squamous cell of the esophagus, particularly those with PD-L1 high and particularly those who have high TPS scores as well. Now so you also have a third option which is ipi/nivo. And ipi/nivo also is a good option in my opinion and based on CheckMate 648 in squamous cell exclusively. So those are the 3 standard options for someone who has metastatic squamous cell carcinoma. How people go about it, which one they choose, it’s clinical context to determine these factors by and large, not so much the datasets we have. But those are the 3 options that most people prefer.

DR LOVE: And how about adeno of the esophagus?

DR WAINBERG: So adeno of the esophagus is a little more, you know, they’re bunched in with the gastric cancers predominantly. You’ve got primarily an option with chemo plus nivo based on CheckMate 649 which included adenocarcinoma of the esophagus. You’ve also got an option of KEYNOTE-590 which is, as I mentioned, chemo plus pembro. So those are the 2 main options. Now that’s where the CPS score, in my opinion, matters a lot more than in patients with squamous cell. In squamous cell carcinoma of the esophagus, while PD-L1 score predicts benefit, the prediction of benefit is much less — is much stronger in squamous cell than it is adenocarcinoma, if that makes sense. So what I mean by that is that the CPS low patients don’t get as much benefit from these checkpoint inhibitors in adeno whereas in squamous, there is a little more benefit.

DR LOVE: What about let’s bring in gastroesophageal and gastric cancers first line.

DR WAINBERG: Okay. So gastric cancer first-line in HER2-negative, the dividing line for me is still CPS 5. If you have a CPS greater than 5, in my opinion, the data is strong and the survival and hazard ratios are clearly beneficial for chemo plus nivo. The CPS less than 5, in my opinion, despite having an FDA approval, I just don’t use it there. So I’m personally — I use CPS 5 as the dividing line. Now the situation is dependent a little bit on your comfort level with the CPS diagnostic that’s been obtained. Your situation is comfortable not offering immunotherapy to that patient. And so it really depends on individual characteristics. Now to add a level of sort of complexity which we try not to do, Neil, but as you mentioned, this is a complex situation, this disease. There was a presentation last week on an ESMO virtual of the KEYNOTE-859 study which by and large, if you just look at the totality of it, was a very large study of upwards of 1,500 patients. They more or less mirrored, in my opinion, what we saw in CheckMate 649 which is that there is a benefit of survival in the all-comer patient population. However, the hazard ratios and survival improvements are significantly better as we go up in the CPS such that one might expect that just like in CheckMate 649, the patients with CPS 1 through 5 did not garner a benefit. But I don’t — we haven’t yet seen all the details in that respect. But what we saw with KEYNOTE-859 is, in my opinion, a very, very similar study to what we saw with CheckMate 649.

DR LOVE: And, again, what about metastatic first-line therapy of carcinoma of the GE junction?

DR WAINBERG: Those are bunched in with gastric cancer. So GE junction adenocarcinoma is always pretty much included on any gastric cancer study. So there were about one-third of patients or so in the CheckMate 649 who had what was characterized as GE junction, a little less so in the KEYNOTE-859 because that was a primarily Asian study where GE junction is a little less common than gastric cancer. But by and large, I think the data is extrapolatable to both of those circumstances.

DR LOVE: So would it be accurate to say then when you look at gastroesophageal cancers metastatic first-line therapy that in general if you have a patient who has a squamous cell which is going to be esophageal generally, they should get an IO, regardless of PD-1? And for people who are non-squamous or adeno, is a question mark. And in your case, you feel that these people in general should not get it unless their CPS is 5? Is that sort of your algorithm?

DR WAINBERG: That summarizes my opinion very well. Yeah.

You do know, Neil, I would say that that’s a subject of great controversy. Others believe everybody should get nivo.

DR LOVE: But then if you sort of look at it from that more global point of view, I guess the other question is going to be where is zolbetuximab going to fit in?

DR WAINBERG: Yeah.

DR LOVE: And I was having this weird thought that PD-1 and claudin 18.2 maybe is going to start to become like ER and HER2 of breast cancer, like you’re going to need it on everybody with metastatic disease. Is that the case?

DR WAINBERG: Yeah, I agree. I think we’ll be in a position probably in a year from now or so where we’ll definitely want that information and we’ll — we should have it. You’re absolutely right. It’s a positive study in overall survival. So by definition, I think we should try to get that information.

DR LOVE: So let’s talk a little bit about zolbetuximab. Could you summarize a little bit your own view of the agent, how it works and what we know about it so far in terms of efficacy and tolerability?

DR WAINBERG: So it’s a first-in-class monoclonal antibody blocking claudin 18.2 which is a protein that is overexpressed, we use that term in this context, in about 30% of gastric cancer and some other cancers too, by the way, a little bit of pancreas, a little bit of squamous carcinoma of the lung. But what’s unique about claudin 18.2 as an adhesion molecule is that its presence in normal tissue, but only in the stomach mucosa such that the stomach mucosa by far has the only normal expression of this antigen. Now a number of years ago, probably even a decade ago, this was identified as an interesting target. And even in the TCGA datasets, there’s a group of patients that are suggested to be claudin 18.2 positive.

A private company in Germany did what’s called the FAST trial which did a randomized Phase II trial that plus chemotherapy plus or minus zolbetuximab in claudin 18.2 positive, it was a positive study from the Phase II perspective, met its progression free survival and overall survival endpoints. Stylus acquired the drug, launched these randomized Phase III trials, 2 randomized Phase III trials, SPOTLIGHT and GLOW. SPOTLIGHT was just presented now at GI ASCO and reported a 2-month or so improvement in overall survival in the group of patients who got zolbetuximab compared to chemotherapy plus placebo.

DR LOVE: But also, in terms of hazard rate, my recollection is for progression free survival, it was like about 0.75. But also, survival I think was about, which is kind of unusual, about 0.75.

DR WAINBERG: Yeah.

DR LOVE: And both of those are pretty decent numbers, particularly the survival. Although it’s kind of, to me, it seemed a little unusual to have them show up at the first analysis like that. You don’t see too many trials like that. But any other thoughts in terms of the granularity in terms of what we know about the responses and the efficacy data?

DR WAINBERG: I think there’s a lot of granularity here actually. Number 1, it was a huge effort. And we have to keep in mind that they screened something like 2,000 patients or so to get 500 patients or so and randomized them into 2 groups because you have to remember that it was a central screening effort that looked specifically for this biomarker which is fairly proprietary. It’s not an easy thing to do in any of our institutions. It’s centrally checked. And then they reported back the information to the investigator. So we’re talking about a fairly selected group of patients already, #1. #2, there was no increase in response rate in the group of patients who got chemotherapy plus zolbetuximab over chemotherapy plus placebo. And that’s interesting. I thought it was interesting. I think there’s scientific reasons why claudin 18.2 antibodies don’t confer response rate having to do with perhaps the mechanism of action of the protein. And to some extent, it’s not a kinase — it’s not a single transduction type of drug. It’s an adhesion molecule. So one might not necessarily expect an improvement in response rate. That being said, the primary endpoint of the study was progression free survival. And so they did meet that. And coinciding with that, you’re absolutely right, was the secondary endpoint of overall survival which is a bit unusual that they would both happen at the same time. But I think speaks to the fact that the drug does have significant activity in that select group of patients.

And we’re not used to seeing in gastric cancer any circumstance where a drug that doesn’t improve response rate has a corresponding improvement in PFS and OS. That’s not been the case with the HER2 inhibitors. That’s not been the case with the PD-L1 inhibitors. So there’s something quite different in this regard.

DR LOVE: What about tolerability issues? You mentioned the fact that you see claudin 18.2 in the stomach. I’ve heard that these people maybe have GI toxicity. What do you see?

DR WAINBERG: Yeah.

So this is a class effect. And it’s important for everybody to keep that in mind that these drugs cause nausea and they cause vomiting. And it’s particularly increased on the first time they get the drug. There’s an element of tachyphylaxis here such that on subsequent dosing of the drug, that toxicity is diminished. However, there’s a very high rate of nausea which is sometimes under reported in the presentations. We haven’t seen the manuscript yet. But there is a significant concern for that toxicity. And to be fair, in my experience at least with these drugs and I’ve had a few of them, the standard antiemetics aren’t as useful. Because you’re talking about something that’s not a centrally mediated nausea and vomiting problem. It’s a GI nausea problem. It’s in the GI tract. So there are tricks of the trade to diminish the toxicity including slowing down the infusion rate, premedicating sometimes with steroids and/or lorazepam. But it’s not as responsive, let’s say, to our classic ondansetron, prochlorperazine type of pre-medications, Neil.

DR LOVE: What about like PPIs or, you know, local things for the stomach?

DR WAINBERG: So a lot of these folks are already on those drugs in the context of their cancer. They’ve been diagnosed with reflux at some point.

DR LOVE: Right.

DR WAINBERG: And a lot of them are already on these PPIs. And so I think one would hope that that might help too.

DR LOVE: Any evidence that you see greater benefit with higher levels of claudin 18.2?

DR WAINBERG: Yeah. So we haven’t seen all the data. The cutoff here was 75%, so that’s already a pretty high cutoff. And that was a little bit higher than the Phase II study. They had a group of patients there who were 40%. Here, you had exclusively 75%. It’s a pretty high cutoff. So I don’t know if we’re going to see gradations above 75%.

DR LOVE: That’s interesting.

DR WAINBERG: But maybe we’ll see that at some point.

DR LOVE: So any crystal ball predictions of where we’re going to be in a couple years? I guess could you imagine people getting an IO and zolbetuximab?

DR WAINBERG: Well so they’re doing that study now. It’s called ILUSTRO. I’ll point out they also have a second study which we’re going to see very soon in one of these virtual plenary sessions called GLOW which is an Asian study looking at primarily capecitabine/oxaliplatin with or without zolbetuximab. And we haven’t seen the data yet, but we know that it’s a positive study. Presumably, positive for the same endpoints. Now simultaneously with that, they’re doing a study called the LUSTRO which looks at the combination of chemotherapy and nivolumab and zolbetuximab because there is a little bit of overlap. It’s not dramatic, but it depends on what you’re reading. It’s somewhere around 5 to 10% of patients. And it may be higher in the real-world when we get out there. So time will tell. I predict, if I had a crystal ball, that once the diagnostic gets approved, we’ll be testing for claudin 18.2 and PD-L1 and then there will be this circular debate for some time about which one to use or both. But a lot will depend a little more on the granularity of these combination studies. To add a level of complexity, of course, we’re hoping to add to that a la carte testing of protein-based immunohistochemistry with something called FGFR2b which we’re hoping, we’re doing the Phase III studies now, Neil, that it would also add another protein-based biomarker to the list. A lot more to come in that regard.

DR LOVE: Right. I forgot about that. Yeah. So the Venn diagram of those 3 would be pretty interesting to see. But that’s really interesting.

DR WAINBERG: That’s exactly right, Neil. It’s a Venn diagram. It’s not a, you know, I’ve been talking about this now the last month. We’re not going to be looking at slices of a pie like they do perhaps in lung cancer or where the crossover to the biomarkers is minimal. I think we’re going to be looking more like a Venn diagram when all is said and done here.

DR LOVE: Wow. Really interesting. So let me come back to one other thing that you commented on, but maybe you could elaborate a little bit on. So the issue in first-line metastatic of IO in PD-1 less than 5 because you said that you generally don’t offer it to patients in that. And that’s very controversial because, as you said, the FDA approved it in that space. What’s your thinking about it? Obviously, you must believe that the downside of an IO in this situation is greater than the minimal, you know, small chance of an upside. Does it matter to you whether it’s 1 to 4 or 0, for example?

DR WAINBERG: Yeah, it’s a really good question. Number 1, I have to feel good about the CPS score. So I’ll be honest with you, if it’s a cruddy CPS report, I’m feeling a little bad withholding an IO in that patient, to be totally honest. So I’ve got to feel good about it. Now if it’s good and I feel good about it and it’s under 5, I have the benefit and I’m sure some of your other speakers have benefit of having studies to put these patients on because we recognize that the value of an IO there is minimal. Maybe not 0, but minimal. So those are perfect study candidates. That’s the group of patients we should be studying in different contexts. Now I you didn’t have a study and you’re just talking about a CPS 1 through 5, I think it really depends on where you fall. I personally, and I haven’t seen yet the comparable pembro data with the KEYNOTE-859 study, but in the CheckMate 649 study, if you just look exclusively at those 1 through 5’s, there’s really no benefit of those drugs — that drug, in my opinion. So you’re going to ask yourself, is there harm? Probably not, but you never do know. And we’ve all been in the circumstance before where we’ve caused IO toxicities. So I am personally not using it in that context. I also practice in the real-world. And if a patient comes in having their idea in mind that they need immuno — chemoimmunotherapy and they don’t leave it up to me and they’re like, doctor, I want immunotherapy. Then I do include it, to be candid, because I feel that the patients know a lot of times what they’re looking for. And it is FDA approved, and so I don’t feel too bad about it, to be honest.

And I know that’s where the community fins itself, in a very difficult circumstance. An so you shouldn’t worry about it too much. It’s not wrong to use that, in my opinion.

In MSS patients, if we’re just thinking about it logically, the addition of nivo is a benefit for all patients and a significant improvement in the way we’ve treated this cancer. But it is not bringing us curing these cancer patients.

Maybe there is a little tail of the curve there. But if you look at other studies, you’ll also see tails of the curve once in a blue moon. So I don’t look at it quite as black and white as that.

DR WAINBERG: I don’t withhold it from patients. And I mean that sincerely. I have a patient right now who was saying to me this is what I want. And I’m like fine. I’ve got no quarrels with that. I don’t want to look at it as I know all the answers. That’s not the way we should look at it as oncologists.

DR LOVE: Do you entertain the possibility of using IO later on in those patients?

DR WAINBERG: Well if you’re going to do it, I think, in a low CPS patient, probably the best time to do it is up front.

So if you —

DR LOVE: Right.

DR WAINBERG: Because then, there’s really no data to support it. And as you know, we don’t really have approvals in that context. It’ll be harder to pay for and this, that and the other thing. So I think if you’re — if you want to use it up front, it’s perfectly reasonable in that situation after that discussion, but I would probably do it up front.

DR LOVE: Well anyhow. I’m really looking forward to see that GLOW study.

DR WAINBERG: GLOW has a press release. We know they have a survival advantage. What’ll be, you know, we’re going to see the fine print, right? How big it is and what the difference is of the — hazard ratios were in PFS and OS compared to the SPOTLIGHT trial. It’s exclusively Asian.

I think it’s very validating of what we already know which is that the drug adds survival advantage in 2 independent studies.

DR LOVE: So I want to talk about HER2-positive GI cancers as well, but just one final question because this is going to relate to the issue of HER2-positive disease which is in HER2-negative disease, how are you approaching second-line therapy outside of a trial?

DR WAINBERG: We don’t have that many trials. And we’re struggling in second-line, Neil. We’ve — everybody is focused on front-line. And so second-line in HER2-negative disease is paclitaxel/ramucirumab. And it’s been that way for a decade. And I don’t — I’m not aware of too many trials looking at combinations in second-line, unfortunately. I think it’s a missed opportunity personally to explore novel drugs. They don’t have to be IO drugs. But right now, pac/ram.

DR LOVE: That’s interesting. We’re starting to see a lot of antibody drug conjugates coming in in other cancers. Any exciting antibody drug conjugates coming in gastroesophageal?

DR WAINBERG: Yeah. There’s certainly, we’ll talk about it in a minute, but the HER2 ADC space is very popular.

DR LOVE: Sure, yeah.

DR WAINBERG: Claudin 18.2 has got a few ADCs that are being developed by a number of companies.

DR LOVE: That’s right. Right.

DR WAINBERG: And studies are ongoing. I also personally think TROP2 is probably a relevant antigen just like it is in breast cancer or triple-negative breast cancer here such that the ADCs are being investigated in gastric cancer as well. Those 3 are all that come to mind.

DR LOVE: Yeah, bladder cancer also, lung cancer. Yeah. That was one — that’s what I was actually wondering about like Dato-DXd. All right. Let’s talk about HER2-positive gastroesophageal cancer.

Can you kind of provide a little bit of an overview of, first of all, when you do HER2 testing, what kind of HER2 testing you do, do you use liquid biopsies and how do you define HER2 positivity?

DR WAINBERG: Tissue biopsy is still preferred. And there’s no reason not to get HER2 from the diagnosis. It makes it easier on your life, like we talked about earlier. But the implications of treatment are more in the metastatic. But certainly, it’s an easy test. The IHC diagnostic and Reflex to FISH has been around for a while yet. So that’s our standard for gastric cancer and it has been. A liquid biopsy, of course, if you don’t have tissue is, in my opinion, you know, pretty good. But ideally, tissue-based diagnostic would be critical here.

If you have someone with HER2-positive disease that is localized, as of this time, there’s no role for anti-HER2 therapy. The study that we did a few years ago that was done by RTOG did not show any benefit. You can quarrel about some of the challenges in that kind of study, but there was no clear benefit in adjuvant trastuzumab. Some of the other studies have hinted at some signals. But certainly, nothing that is standardized for localized gastroesophageal HER2-positive disease.

DR LOVE: Yeah. Before we get into metastatic disease, the reason I brought up liquid biopsy is kind of a thing that’s interesting to observe that I see the GI people often repeating HER2 by liquid biopsy when they get to second-line, for example. And I just think it’s — and they — I hear a lot of people taking that information and if they go from HER2-positive to HER2-negative, acting on it. And I wonder, we’re going to get into second-line therapy in a second, but just in terms of testing, I wonder if that’s your practice. Because it really is not the practice in breast cancer. In breast cancer, once they’re HER2-positive, they’re HER2-positive even if they test HER2-negative again. But it’s a different approach in GI cancers. Is that your approach?

DR WAINBERG: Yeah, that is. And that’s based primarily on a bunch of studies that have shown that for whatever reason, there is a group of patients that do not retain HER2 beyond progression. And it could be a quarter of them. It could be 30%. Studies have looked at this now in a bunch of smaller studies, but even some randomized Phase II studies that have shown that’s the case. Now primarily, that’s been done based on tissue. Those studies are tissue diagnostics. However, because we recognize in the real-world the challenges of repeat biopsying everybody, people have gravitated to ctDNA in that context. I agree. My personal preference is still if they progress on front-line trastuzumab-based therapy, to consider repeat biopsy when making that second-line decision. I try to do it. Let’s put it that way. And if I can’t, I do ctDNA as a surrogate.

DR LOVE: So let’s talk about first-line therapy of metastatic HER2-positive disease. That changed recently with the addition of pembrolizumab. Is that your approach? Most of the patients in, I think it was the 811 study, were PD-1 positive, like 85%. So I don’t know if they really — we really know for sure that it helps in that situation. Is that sort of your approach to use the triplet?

DR WAINBERG: Yeah, yeah. And you’ve heard this before. I think that rate of PD-L1 positivity in the HER2-positive group probably doesn’t matter that much. So if you have HER2-positive metastatic gastric cancer, I do believe that at this time, the data supports chemotherapy plus trastuzumab plus pembrolizumab based on the accelerated approval of the 811 study. I think you can use CAPOX, you can use FOLFOX in that context. I in that context don’t care as much about CPS score because if you look at the totality of the literature on HER2-positive disease, they almost all have a good enough CPS score. There is a codependence there between HER2 and PD-L1. That being said, I think we’re all waiting to see what happens with the survival endpoints of that study and whether it’ll meet its progression free survival, overall survival endpoints. Keep in mind that it was an accelerated approval based on a response rate which increased from about 50 to about 70, so a good response. But people are still waiting for survival information. So it’s complicated right now. I think personally for right now, I’m absolutely using it in front line.

DR LOVE: So a lot of the controversy then is, what’s next? So you have the option potentially of T-DXd, but also ramucirumab/paclitaxel.

DR WAINBERG: So those 2 haven’t been compared head-to-head, but they are being compared head-to-head in Europe at this time, Europe and Asia, the DESTINY-Gastric04 study. If someone is HER2-positive right now and I confirm HER2 positivity, I am using T-DXd. I am using trastuzumab deruxtecan based also on the accelerated approval, they have a randomized Phase II study that showed a benefit of survival in Asia in that context and we have complimentary data in the Western world showing response rate and PFS benefits, similar magnitude with trastuzumab deruxtecan. So for now, I am using that in second-line HER2-positive disease and conferring pac/ram to third-line. At that point, you’re outside of the HER2 milieu usually.

DR LOVE: So how about an older patient with non-visceral disease, not that sick? Still T-DXd?

DR WAINBERG: Yeah. In that patient, you know, sometimes I’ll give T-DXd at a lower dose, to be honest with you. Or, you know, everything is context dependent. But sick, older lady, not giving full-dose T-DXd. You’re right.

DR LOVE: What are you seeing in terms of tolerability issues with T-DXd? As time has gone on, I’m hearing more and more about chemo side effects starting out with the breast people, GI, alopecia, using of pre-medication. Is that kind of what you’re seeing in these patients?

DR WAINBERG: All of the above. Yeah, I totally agree. And I think — and breast cancer is going to lead the way here because they have so many indications now for T-DXd. It’s getting used all over the place. But absolutely. I think the chemo payload here causes some chemo-like toxicities and we have to warn patients of that. Antiemetics, hair loss, nausea even, cytopenias. Antibody drug conjugates by definition have potent payloads, and this is a potent payload. And so we have to be mindful of that. That’s not even touching on the idiosyncratic side effects like ILD which I’ve seen a few of those and we’ve all seen a few if you give the drug long enough. Adjudicating that properly is critical now. But it’s very important for us to be mindful that these are not monoclonal antibodies.

DR LOVE: Yeah. I want to ask you, of course, about ILD. But first, one other issue, again, that comes up in breast cancer which is brain mets. And the first question is, do you see more brain mets in HER2-positive gastric cancer than HER2-negative disease? In breast cancer, it seems like there’s more brain mets, more of an issue. And any data or experiences in terms of T-DXd and brain mets? It looks like in breast cancer, it has some activity in the brain. I don’t know if it’s really been looked at in upper GI cancer.

DR WAINBERG: Yeah. So that’s a good question. HER2-positive gastric cancer, just like breast cancer, is more likely to cause brain metastases than HER2-negative gastric cancer. It’s not nearly as common. And keep in mind that the patients, unfortunately, don’t live as long. So that may be a product simply of that. But if you look at the 2 groups side-by-side, you’re more likely to have brain mets in HER2-positive gastric cancer. Now I don’t think we have enough data, Neil, to comment on T-DXd’s role in treating brain metastases in gastric cancer. It simply is not as common and it also hasn’t been studied in the same manner as in breast cancer. I know that the breast cancer folks have a very potent kinase inhibitor, tucatinib, that crosses the blood-brain barrier and is an effective anti-HER2 therapy for breast cancer. The challenge is that drug has not had a very large experience in gastric cancer of testing. So I don’t know if we’ll ever have the ability in a large trial to have that drug at our disposal.

I have a patient right now, Neil, I have a patient right now, a perfect example, who has got gastric cancer, HER2-positive, has a small brain met, 1 to 2 cm, that small. And I treated it with stereotactic radiosurgery and she’s doing fine. But I’d love to give her some drugs. And she’s progressed on T-DXd, by the way. I’d love to give her something that can cross the blood-brain barrier. And I’ve petitioned for tucatinib in her case, by the way.

DR LOVE: Really?

DR WAINBERG: But it’s not approved. Yeah.

DR LOVE: You’re trying to get tucatinib for her?

DR WAINBERG: I am.

DR LOVE: So you would give it with trastuzumab? In breast, they give tucatinib, trastuzumab and capecitabine. Is that what you would do with her?

DR WAINBERG: That’s exactly what I’m trying to do for her.

The problem is tucatinib is expensive and not affordable and there’s no trials that are so simple. Right now, they halted a lot of their drug development in gastric cancer, to my knowledge. So in colon cancer, they have a great approval. And I think actually there, that’s a big win for patients to have tucatinib/trastuzumab as an FDA approved option now.

DR LOVE: Absolutely. Yeah, they have 2 lines of therapy now. They have tucatinib/trastuzumab and T-DXd for HER2-positive colon. Hard to believe.

Getting back to the ILD, the message that we’ve gotten in breast cancer is Grade 2 or higher. In other words, any symptoms, if they get ILD and they’re not getting T-DXd again. Period. Is that your algorithm in GI cancer?

DR WAINBERG: Yeah, yeah. I think it’s similar. You never want to make the mistake of redosing someone with symptomatic ILD. So I think we’re all mindful of that. I do feel like nowadays, we’ve gotten a lot better at figuring out what ILD is and what it isn’t. And we have pulmonologists who are really helpful helping us adjudicate it properly because it can get a little tricky, especially in patients that have been on IO therapy. So I look at it as, yeah, if I’m fairly convinced that’s symptomatic ILD even Grade 2, I’d be cautious, and not use the drug.

DR LOVE: Well that belief or that algorithm that they established in breast cancer and has great activity in breast cancer, incredible really, but is getting them to be very aggressive about diagnosing it on imaging. So if you ask the investigators, you ask them how often do you want to get imaging of the lungs? Even if they don’t need to get it, they don’t have any lung mets or anything, but just for ILD, they actually want to get it as often as they can. They want to get it like every 2 months for a long time, maybe indefinitely because they want to pick it up before the patient is symptomatic. And, again, I don’t know whether that, even in lung, I don’t see that kind of approach to T-DXd. And I don’t really necessarily hear it with the GI people. But it is — because the general oncologists, they hear it from all 3 of you and they’re trying to kind of justify it. You mentioned that you’ve actually had patients with ILD with T-DXd. Any serious — any deaths?

DR WAINBERG: No, I haven’t personally had a patient die of ILD. I’ve had, I’d say, Grade 3 ILD for certain, and stopped the drug and prolonged recovery. But I think that the management strategies are there. You’ve got a pretty good adjudication now of recognition of what is ILD and what isn’t. You’ve got good management guidelines for management with steroids and all of the other things. You’ve got recognition of the toxicity, first and foremost. So I think the tools are there to help us manage this problem better. I think the differences you’re seeing in breast cancer and maybe other cancers has to do primarily because of tolerance risk that you may have for individual cancers. And I don’t mean that like we want more toxicity or we’ll accept more toxicity. The reality in some of these cancers, and I’ll use gastric cancer as an example, is we don’t have the opportunity to turn to 4 or 5 other options or 3 other options like they may have in metastatic breast cancer. So we have to as clinicians keep that in mind. And I’m not suggesting we should just continue the drug if people are sick and having ILD. But I don’t know if I would stop the drug in the context of radiographic findings that are not associated with symptoms in this disease.

DR LOVE: Oh, yeah. Not without symptoms. But the problem is that if there are symptoms even if they’re, at least that’s the algorithm that they have. But, you know, yes, there are other options in breast cancer, but I don’t know that once you get out to second or third-line therapy, their survival is that much — yeah, certainly gastric cancer has a very dismal prognosis at that point. But I’m sure part of that is a result of that. What do you see in terms of the future of T-DXd in gastroesophageal cancer? I’m sure there are trials looking at it in earlier-stage disease. Neoadjuvant trials we’re seeing in breast cancer. We were talking about neoadjuvant therapy. How is T-DXd being studied right now in gastroesophageal cancer?

DR WAINBERG: So number 1, I think people are waiting for the randomized data. We need a good, randomized data set in second-line. That study is ongoing. I think the challenge with T-DXd quite frankly is that it’s a harder combination strategy. So we’re not like breast cancer necessarily where we feel that in front-line, it’s going to be a stand-alone drug. I don’t know if we feel that we can replace platinum/5-FU/trastuzumab/pembro with just T-DXd. And so efforts to combine T-DXd with different chemo backbones are ongoing. But as you can imagine, as an ADC, that’s not easy to do. So finding that balance between a very high efficacy that we need in front-line which is, as I mentioned, 70% response rate, to give you a signal to lead into some big trial won’t be easy. And the efforts are ongoing. And there’s efforts ongoing also in neoadjuvant, as you mentioned. But these are going to be a little more challenging, I think, to thread that needle here.

DR LOVE: So one more thing I want to ask you about is your thoughts about supportive care of patients with upper GI cancers. And I’m sure you have the opportunity to see many patients in second opinion who have been managed by other physicians. Any, I’m not going to say errors, but maybe myths, misperceptions or things you see people doing in cases that are getting referred to you that maybe you wouldn’t do or things that you would do that they’re not doing, particularly as it relates to supportive and palliative care.

DR WAINBERG: Yeah. There’s a few like tricks that I’ve learned I guess over the years about this. I don’t — I think clinicians and general oncologists are really good at managing complex cancers including gastric cancer.

My sense in gastric cancer, and I put it like pancreas cancer the same way, we teach our fellows, is don’t doubt clinical progression in this disease. If you feel like the patient is all of a sudden having more weight loss, symptomatic disease, failure to thrive, even if you don’t see corresponding radiographic progression, it’s okay to stop therapy and make a change at that point before that performance status worsens. That’s one thing that I’ve learned over the years. And I’ve also learned the importance of nutrition. So in the beginning, when I was I guess training, I didn’t really understand how important it is to have patients with decent functional nutrition when they’re getting these chemos. And especially, in a disease like gastric cancer that a chunk of them have had a gastrectomy and have come in with weight loss. They start out with so much weight loss by the time you see them half the time. So we are trying and I think every institute tries to help these patients nutritionally whether that means consultations with nutritionists or dieticians, whether it means supportive care mechanisms to get them fed. It’s so critical. Your drugs aren’t going to work as well if your albumin is lower than 3 and you’ve lost 20 pounds in the last 2 months. And we see that very often, unfortunately.

DR LOVE: I was going to ask you, and particularly related to what you were saying about early picking up on progression, whether you integrate tumor informed ctDNA in managing these people or whether you would like to.

DR WAINBERG: I’d like to. And those are tools that I do play around with to look at ctDNA and see if changes are corresponding. I think that in gastric cancer like in pancreatic cancer, the bulk of the adjuvant ctDNA data is predictive of recurrence, quite predictive if you’re getting it in the adjuvant setting.

I do think in the adjuvant context, it is very predictive. And the lag time between positivity and the development of metastatic disease is shorter. So yes, I sometimes you know, we’re not at a place in this disease yet where we’re going to necessarily act on every positive ctDNA other than scan the patient and look for clinical symptoms. But sure, in the back of my mind, for sure.

ctDNA is a confusing place scientifically, but more importantly, for our clinicians because the clinical data of how to react to a ctDNA positive assay has not yet caught up with the science. And until those can be reconciled, sometimes we are making our patients very anxious and confused. And I do take that in mind as well.

Integration of Targeted Therapy into the Management of Advanced Biliary Tract Cancers (BTCs) — Lipika Goyal, MD, MPhil

DR GOYAL: We’re excited that we have multiple options now for HCC. So atezolizumab and bevacizumab with the IMbrave150 study and then the durvalumab plus tremelimumab regimen from the HIMALAYA study. Then, of course, we always have the tyrosine kinase inhibitors, lenvatinib and sorafenib, that have longstanding been available for patients. So given atezolizumab plus bevacizumab has a VEGF inhibitor as part of the regimen, I always think to myself, do patients have any relative contraindications to getting a VEGF inhibitor? So I find that if people have had recent significant bleeding, I think about whether or not it’s safe to give them a VEGF inhibitor. If people have at baseline relatively uncontrolled hypertension despite, for example, 3 antihypertensives and I worry about them developing further hypertension on the VEGF inhibitor, I think twice about that regimen. If people have significant proteinuria at baseline that’s hard to get on top of, I think about that. Also, of course, in our population, we see people with varices. And so if people have varices that can be treated and can move forward with atezo/bev, that’s my first choice. But if they have, let’s say, Grade 3 varices and we band them and they’re going to take some more time in order for it to be safe for them, for the patient to be treated, but the patient is having pain, for example, and more urgently needs to start on therapy, that’s the kind of patient that I would treat the varices because I want to be able to give them a VEGF inhibitor in the future, but it might make sense to start with durvalumab and tremelimumab up front.

DR LOVE: Have you used that regimen, the STRIDE regimen?

DR GOYAL: I have used it. I have used it. And certainly had success with it. I will say that as reported in the trial, there was a 20% rate of immune-related adverse events requiring steroids. When I have used it, I’ve had some patients requiring steroids for it. So certainly, patients who go on it need to be robust enough to be able to manage the adverse events that can sometimes come with the regimen. But that goes for all of our regimens including atezo/bev, lenvatinib, sorafenib. We want to make sure that if they had a complication that we’re able to certainly help them through that.

DR LOVE: I guess that is one of the big questions that oncologists have. And most of their experience with anti-CTLA-4 is ipilimumab, ipi/nivo. My understanding though is in general, I don’t know if this is just in general ipi/nivo or more in HCC or GI cancers, that the number you just brought up, the incidence of requiring steroids for autoimmune toxicity is much higher, more in the range of 50%. Clinically, do you feel like it is less — causes less immune problems than, for example, ipi/nivo?

DR GOYAL: Yeah, that’s exactly right. It was around 50% in the trial for ipi/nivo in HCC. And so I do find that durva plus treme, people end up having less immune-related adverse events with it. And it’s a little gentler for some patients. We’re really excited that we have both of those options. We look forward to seeing the results of the front-line ipi/nivo study to see if that becomes another front-line option. That’s the, you know, yeah. So we look forward to seeing like the front-line study results.

DR LOVE: So there’s a trial looking at ipi/nivo up front?

DR GOYAL: Yeah. It’s ipi/nivo versus dealer’s choice, sorafenib and lenvatinib.

DR LOVE: That’s really interesting. You were mentioning the issue of VEGF inhibitors and contraindications. What about just the issue of people who are older, frail and who maybe have some history of heart disease? Does that, you know, I’ve never been exactly clear whether bev actually has an adverse effect on those patients. Some people say they do. I don’t know. Does that cause you hesitation using atezo/bev?

DR GOYAL: Generally speaking, I find it to be a relatively well tolerated regimen, so age doesn’t come into play for me when giving atezo/bev. If someone has, you know, advanced heart failure, for example, it might be a patient that I’m less likely to use bev in. But even some people who have coronary artery disease, even if someone has a history of a CABG, I think it’s still reasonable to give atezo/bev. It’s such an active regimen that — and HCC is such a deadly disease that given the cost/benefit analysis, of course, with informed consent with the patient, I think it’s still reasonable to proceed.

DR LOVE: So for patients who don’t have any relative contraindications, I’m curious as most people are at this point staying with atezo/bev. But I’m just kind of curious when you think about that in terms of, again, any thoughts about why you might want to choose one versus the other in a patient who doesn’t have a contraindication? Any advantage? Anything in the data that makes you think one would be better? Or do they at this point, kind of like from the point of view of efficacy, seem very similar to you? You mentioned, you were telling me about a program you did with Dr Abou-Alfa and he’s one of the few docs who, of course, he was he PI of the HIMALAYA study, but he’s very pro STRIDE regimen. Most other people are kind of more like in your camp. But any thoughts about his thinking that maybe it’s a better regimen?

DR GOYAL: Yeah. Of course, there’s always the pitfalls of doing cross trial comparison because the populations were not exactly the same. So, for example, in the atezo/bev study, they allowed patients with main portal vein thrombosis or VP4 disease. And in the HIMALAYA study, patients with main portal vein thrombosis were not allowed. So it’s not totally the exact same population. But, of course, we do what we can with the data. And the median overall survival which was the primary endpoint in the IMbrave150 study with atezo/bev, the median survival was 19.2 months.

In the HIMALAYA study the median survival was a little over 16 months with the combination of durva and treme. And so the patients in the atezo/bev study, the sorafenib arm did better. They lived around 16 months+. And in the durva/treme study, the sorafenib arm, they lived 13.8 months. And so, again, that just shows that it’s not quite an apples to apples comparison. So if we did it head-to-head, it’s hard to know if one would win versus the other. But I think they both have their place in the treatment of HCC. And then, of course, the HIMALAYA team did the landmark analysis where they looked at what the survival was sort of long-term. And that certainly looked quite favorable for durva and treme. And we don’t know what the long-term survival is in the same way. There was not a landmark analysis for the IMbrave150 study.

DR LOVE: Yeah, I think a lot of people have come out thinking that maybe combination immunotherapy might have a longer-term. But we’re doing a program on renal cell cancer. And they have this big argument about whether to use ipi/nivo or IO/TKI. But, you know, the hope that maybe you’ll see longer-term responses with combination immunotherapy. One of the other things I want to do as we chat today also is reflect back on some of the questions that came out in the symposium from the docs, the general oncologists who presented cases because a lot of the questions were dealt with fairly briefly and there were a bunch of issues that I wanted to just bring back up. And you may remember some of these questions that came up. One was the issue of whether or not, you know, maybe these new systemic therapies should be maybe given ahead of interventional radiology procedures. One of the things you hear from a lot of oncologists is by the time they — sometimes, they don’t see the patient until they’ve already had interventional radiology. Sometimes, they’ve had a lot of it and maybe not the greatest condition for systemic therapy. Have you found yourself that you’re — of course, they don’t have the kind of interdisciplinary team that you see at tertiary centers. But in the tertiary setting, do you think there’s been less intervention or more delayed local intervention in the liver because now there are these therapies that cause responses?

DR GOYAL: I think that’s a great question and you’re hitting on a very key question that’s relevant in the field right now. And you also made the great point about multidisciplinary clinics. So now, we have so many options, both for localized disease and for systemic disease that it really makes sense to bring all of these cases to be presented with a team of interventional radiologists, radiation oncologists, transplant surgeons, hepatobiliary surgeons and medical oncologists and, of course, our radiologists and pathologists to help us make the diagnosis and think about these patients. I think having a group discussion helps us think best about the patient. And oftentimes, it’s not one or the other, it’s what now versus what later. You know? We might often start with one, for example, TACE or Y90 and then on the heels of that, end up coming in with systemic therapy. Or we might start with systemic therapy because maybe someone has a large portal vein thrombosis that we want to address on the sooner side. And then if they get a good response to atezo/bev, sometimes we might follow that or consolidate with some locoregional therapy. And so a lot of it is just trying to figure out what do we start with first.

What I will say is that for the first 10 years when I took care of HCC and we only had sorafenib, I was more likely to push localized therapy for longer because I knew sorafenib was tough, I knew the efficacy was modest. And now that we have atezo/bev, I think there is a feeling among multi-clinics that take care of HCC that we don’t want to beat up the liver too much because if people develop Child-Pugh B cirrhosis, it becomes harder to deliver systemic therapy. And so that’s why having a close relationship with the interventional radiologists and the radiation oncologists, you know, what I love about localized therapy is that it’s effective and gives people a good quality-of-life without having to come see me back and forth in clinic every 3 weeks. You know? So if one of our IR docs can do Y90 and I get 9 months without having to feel like a patient all the time, that’s a win-win for everybody. But then if we start hitting some of the normal liver along with the tumor and, of course, these patients that have liver disease in the background often have some cirrhosis or at least a component of fibrosis. If we start taking out some normal liver and then we prevent giving them effective therapies, that’s not the best option either. So it’s a conversation that we all work together to help them.

DR LOVE: It’s funny when you said when we just had sorafenib alone, I was sort of flashing back on when we used to do CME programs then. It was like what are we going to talk about, you know? And all we had to talk about was sorafenib. Then, lenvatinib came along. Okay, that was a little bit better. And then, boom, everything exploded. And now, you can’t even get through it in an hour. One of the other questions that came up — it’s funny, it wasn’t that long ago that we were just talking about —

DR GOYAL: Right, right.

DR LOVE: What, 5 years ago?

DR GOYAL: Within the last 5 or 6 years that we went from having 1 drug to having like 8 or 9 regimens that are approved.

DR LOVE: I love to ask in our webinars, what fraction of you have been in oncology less than 5 years? And it’s usually like 25, 30% of people. And they have no idea how different things are nowadays. Another question that —

DR GOYAL: A barren landscape that we had for so many years. Yeah.

DR LOVE: Right, right. Another question that came out at the webinar I just wanted to bring it back up to you is there was always this sort of classic question of whether or not you need to biopsy somebody who has obvious HCC and high alpha-fetoprotein. And one of the docs, you may remember, brought up the issue of whether or not you need to biopsy just to get tissue for genomic assay and whether or not from a practical point. Yeah, sure, clinical trials, you know, that’s possible. But from a practical point of view, do you see any role of getting tissue in HCC for genomic or other assays?

DR GOYAL: Luckily, HCC is I think one of the only tumors, only solid tumors that you can diagnose radiographically. And we’re glad that’s an option. My own personal practice is in patients with early-stage HCC that I’m thinking about transplant in or resection in and they meet the criteria for radiological diagnosis including having background underlying cirrhosis plus the radiological features of early arterial enhancement and delayed venous washout and a pseudocapsule, at least 2 or those 3 if they’re greater than 2 cm, I tend to hold off on biopsy in some of those patients. Any person that I’m putting on systemic therapy, I try to biopsy 100% of those patients because every once in a while, we get surprised and a patient has cholangiocarcinoma or they have a mixed hepatocellular cholangiocarcinoma or they have some other diagnosis, you know? And so we’re not giving people Kool-Aid and these treatments can have side effects. So it’s really about the recipe in medical oncology, getting that right for whatever histology it is. And I also feel that if we’re going to make advancements in HCC, having that tissue to start to develop biomarkers to figure out when we have all these front-line regimens, atezo/bev, durva/treme, lenvatinib, sorafenib, who should we be putting toward each of these different regimens because they all have a response rate somewhere between 20 and 30%. You know? And so there’s probably certain populations of patients that respond better to one or the other.

Now there’s the age-old question about where does patient care and research, where do they all fit in together? You know? I certainly think when we’re doing clinical trials and we’re testing new drugs, I think having a biopsy at baseline and sending that tissue to the company or doing studies within the institution to try to identify biomarkers, I think everyone wins when you do that. Because ultimately, we know in HCC about 50% of patients or even maybe more are not going to make it to plan B, you know? After plan A, a lot of people don’t make it to plan B. And I’m not talking just about the trial population because those patients are obviously a bit fitter. So it really is incumbent upon us as an oncology community to try to get it right the first time as much as possible. And so if we’re able to match people with an effective first-line therapy to begin with, I think we stand to have better outcomes for patients.

DR LOVE: Actually, as I was asking that, I was just sort of flashing on, I don’t know whether this is an issue with HCC, but you hear it in almost every solid tumor which is tumor informed ctDNA. Has that been looked at? Theoretically, you wouldn’t need tissue for that. Do you see that with a future with HCC?

DR GOYAL: There are certainly studies that are coming out looking at ctDNA in diagnosis of HCC and disease monitoring for HCC and understanding resistance to HCC. There have been a series of genomic profiling studies also for HCC to figure out what kind of therapies we should be giving patients. I think it’s all a work in progress right now. There are a paucity of biomarker-directed HCC studies. Some of the studies are around using, for FGFR4 inhibitors, for example, using FGF19 overexpression or amplification as a biomarker. But most trials in HCC are biomarker unselected. I think the more we learn through ctDNA and genomic profiling and the more drug development that we do in a variety of cancers, I think we’re going to have more NASH therapies hopefully in the future for HCC. But right now, the use of ctDNA is more on the research side more so than on the mainstream clinical side.

DR LOVE: So another issue related to choice of first-line therapy in HCC, there’s been a longstanding issue and it’s been looked at so many times whether or not there’s a difference in the underlying etiologic factor, particularly viral versus nonviral and response to therapy. And or a long time, I wasn’t hearing anything about that. And then it seems now, we’re starting to hear more things, particularly with the nonviral maybe not responding as much to immunotherapy. I can’t remember who it was, I asked an investigator about HIMALAYA and she said I think about it in NASH and people with a little bit of heart disease. So a little bit of contraindication of bev and a little bit of maybe they’re not going to respond as well. What do you think about the emerging data? I guess there’s no definitive data at this point. But do you think we’re going to find out that there is a difference in response? Kind of you would think it might make sense, viral versus nonviral. Any thoughts about that whole debate?

DR GOYAL: Yeah. It’s definitely something that we talk about in the HCC community. What I would say is I think we need to be careful about subgroup analyses. I think subgroup analyses are really valuable because they generate hypotheses, but I try not to dictate my practice through subgroup analyses because they’re just not powered to show us what we’re looking at, you know? So you’re right, there have been some questions around how much patients with NASH respond to immunotherapy. But I would say there are not prospective definitive data showing efficacy of the regimen in one population versus the other.

So for now, my personal practice is to treat everybody regardless of etiology with an immunotherapy regimen up-front. Specifically, with atezo/bev if they’re candidates. Otherwise, with durva/treme if they’re not candidates. And there are a lot of really smart groups trying to assess the impact of etiology on outcomes now.

DR LOVE: Any situations where you use single agent IO as first-line therapy?

DR GOYAL: Yeah, certainly. So if someone has a relative contraindication to bevacizumab and if I’m worried about the ability of that patient to tolerate immune-related adverse events, you know, the initial 300 mg dose of tremelimumab, that CTLA-4 antibody can add a bit more immune-related toxicity compared to, for example, durvalumab alone. So in patients who are a bit older, frailer, and I actually also look at what kind of social support a patient has because when patients develop immune-related adverse events, they need help. They need their family or their friends or someone who is going to be able to recognize, oh this person is having Grade 3 diarrhea or they're having Grade 3 shortness of breath, bringing them into the hospital. And then it's a long road afterwards of steroids and sort of micromanaging that care. And so how much partnership do I have with the patient's family and friends to be able to manage that patient? To be honest, that goes into my calculus as well for giving them doublet therapy up-front versus singlet immunotherapy. And then for up-front systemic therapy, we certainly have the data for durvalumab. We have data for, you know, we can consider pembrolizumab, nivolumab. All of these are reasonable options.

DR LOVE: Yeah. To me, I don't know. When I first saw HIMALAYA, I thought the durvalumab alone arm wasn't that bad. It was better than sorafenib, I think. That's what it looked like. Is that your take?

DR GOYAL: Yeah. If you look at the numbers, they look very good for durvalumab. And it was powered to look at noninferiority to sorafenib rather than superiority to sorafenib.

DR LOVE: Right. Right.

DR GOYAL: So hence, we tend to favor using durva/treme as opposed to durva alone. But as you were just asking, there's certainly a population for patients who would benefit from durva alone. And it was encouraging that the efficacy looked similar to durva/treme in many respects.

DR LOVE: So I don't know if you remember, but one of the cases that one of the docs presented was a patient who’d had a transplant and now had recurrent advanced disease. And he did not want to give the patient an IO understandably. And at that point, had not. I think the patient had already had 3 different TKIs. I think it was lenvatinib and cabo and regorafenib, I think. But I am curious, in patients with relative or really absolute contraindications to IOs, how do you sequence the TKIs?

DR GOYAL: I generally think of lenvatinib and sorafenib. I think they're both good options. Lenvatinib is generally a therapy that I often will go to because of the higher response rate and the encouraging data also that just came out of the front-line study of lenvatinib/pembro versus lenvatinib. The overall survival with lenvatinib was impressive.

DR LOVE: What's the initial dose that you use of lenvatinib?

DR GOYAL: I tend to use 12 mg in patients who are more than 60 kgs. I tend to use 8 mg in people that are less than 60 kgs. If someone is a little older and frailer, even if they're more than 60 kgs, I might start with 8.

DR LOVE: And can you talk a little bit about how you approach second and third-line therapy in patients who get either atezo/bev, I guess that's a more common first-line therapy, or treme/durva?

DR GOYAL: I'll often think about using lenvatinib or sorafenib. I think it's also very reasonable to consider using cabozantinib or using ipi/nivo in that setting. I think all 4 of those are reasonable regimens. The jury is not out yet which one of these might be the most effective after atezo/bev. My first inclination is always, do I have a clinical trial for this patient? And if I don't have a clinical trial, then the 4 of those are options.

DR LOVE: In general, how do you think through the issue of patients who have relative contraindications to IOs? This comes up in every tumor where we see people using checkpoint inhibitors. You have the issue of transplant as we were talking about, not just hepatic transplant, but some occasionally see people who've had renal transplants. And then, of course, the whole spectrum of prior autoimmune disease, everything from history of ulcerative colitis, multiple sclerosis, et cetera. Can you talk a little bit about how you think through that question of whether to use an IO in somebody who has one of these contraindications or relative contraindications?

DR GOYAL: Yeah. Given immunotherapy is so effective in, relatively speaking, in HCC, I try to do my very best to use it whenever I can. And I will phone a friend if we need help from gastroenterology or pulmonology, for example. So people who have a history of Crohn's disease or ulcerative colitis, and if they haven't had a flareup in a long period of time, certainly I will consider giving atezo/bev in that setting and then keep close contact with their Crohn's doctor. Even if they have had a recent flare, again, given the effectiveness of these regimens, we can sometimes give adjunctive therapies that can help manage the Crohn's while we're giving immunotherapy. So before I say no to immunotherapy, I will often touch base with their doctor that takes care of their autoimmune disease to see how severe is this really and what are our step-up therapies if someone were to have a flare? How dangerous would it be for this patient? So make that decision collaboratively with their other primary doctors.

When it comes to transplant, certainly we need to be quite cautious. The rejection rate post-transplant in the setting of immunotherapy is 50% or greater, so it's not my go-to. I'll certainly use the tyrosine kinase inhibitors as my first options. One of the key things to think about when you're giving TKIs after transplant is you need to make sure they do not interact with the immunosuppressants. And so people are often on tacrolimus, sirolimus, they're off and on mycophenolate mofetil. Those drugs can sometimes interact with some of the TKIs, so looking for drug-drug interactions is important. But I will often use lenvatinib or sorafenib or cabozantinib. And now recently, there was some data on ramucirumab post-transplant. So I think those are all reasonable options. I don't say no to immunotherapy outright in patients with transplant. One, I think if we have a trial and we can study it safely, I think that's the best option. But, of course, those trials are few and far between so many centers don't have them. I think it's reasonable if we've exhausted all the TKIs with a very careful discussion about the risks and benefits, and a very careful discussion with their transplant team with plans for managing the immunosuppression if they develop rejection and managing the levels of their immunosuppressants along the way and having that conversation collectively with the transplant team and them. I think if there's informed consent, I think it's not unreasonable to consider in very select patients.

DR LOVE: So just one more question, then we'll get on to biliary. And I was just kind of — I kind of jotted this down in the beginning when we were talking about adjuvant and you mentioned screening. I was just wondering if you have any thoughts about the issue, it's kind of an issue I hear a little bit about on and off over the years, but now I wonder with the adjuvant data whether it's going to be even more important, and that's the issue of screening for HCC. And particularly, the issue of the oncologists being aware of that. I've heard stories about oncologists seeing patients for like hypersplenism and not using that as an — they’re seeing the patient because they're thrombocytopenic, but not realizing that's an opportunity to make sure the patient's getting screened. So I don't know that oncologists are that tuned in to the whole screening thing. Can you just kind of bring us up-to-date on kind of how you think through screening and what we know about to what extent it's being effectively done?

DR GOYAL: Yeah. You're right that I don't think it's on the radar of most oncologists because if people are seeing people for hypersplenism, that's often on the hematology side or it's not usually the GI oncologists that are managing the hypersplenism. I do think you're right that the vast majority of patients that are getting screened are being screened by their primary care doctors or if they're plugged in to a hepatologist or gastroenterologist, then plugged in. For hep B, hep C patients who have cirrhosis, we have much better screening guidelines for doing ultrasound every 6 months, and sometimes we alternate ultrasound and MRI depending on how high-risk someone is. I think that the gap that we all talk about in screening is the patients with fatty liver disease and NASH who do not have cirrhosis. As we know that there's a population of patients with hepatitis B and a population of patients with NAFLD and NASH that can get to cirrhosis — I mean, can get to HCC without going through cirrhosis. But the number of patients with NAFLD and NASH without cirrhosis is so high that trying to find a cost effective way to figure out which of those patients are high-risk and screen as opposed to screening all several million patients with NAFLD and NASH, I think that's going to be a part of the holy grail. You know? I wish there were blood tests that we could do to identify who are the patients with high-risk NASH or NAFLD for conversion to HCC — not conversion, but for development of HCC. But we are not there yet. Even better than that, I wish we had treatments for NASH and NAFLD beyond what we normally do right now which is treating the hypercholesterolism and treating the diabetes and treating the obesity with diet and exercise. I think NAFLD and NASH treatments, there's a lot of research going on in this area and there are drugs that are coming out. But I think if we want to prevent HCC, I think just the way we have made great strides with treating hepatitis B and hepatitis C and that's led to a decrease in the development of HCC, I think if we can make those strides with NASH and NAFLD, I think that will be real progress.

DR LOVE: Right now, about what fraction of HCC in the United States is nonviral?

DR GOYAL: That's a great question. What I will say is the dominant causes of HCC in the US are hepatitis C, alcoholism, and NASH/NAFLD. It varies by region. So, for example, in Texas, there’s a fair number of patients with NASH and NAFLD-related HCC. In the West Coast, there are a lot of patients from Asia and you see more hepatitis B-related HCC in addition to hep c and alcohol-related HCC. So I would say that it varies by region in the US.

DR LOVE: So one other thig I wanted to ask you about is a recent press release on an adjuvant study with HCC the IMbrave 050 trial.

DR GOYAL: We were really encouraged to see that press release. I think ultimately, we’ll have to see what the data look like to get a sense of how positive the study was. But yeah, that was the IMbrave050 study. And that was patients who had high-risk hepatocellular carcinoma who underwent curative resection or ablation. They were randomized to either getting adjuvant atezolizumab plus bevacizumab or to observation. And people got treated for a year. And it looks like from the press release that it was a positive study. We’ll see in the end what the data look like. But the recurrence rate after resection for HCC is around 70 to 80% depending, of course, on how — what kind of features people have on their tumor. And even after ablation, there’s a reasonably high recurrence rate for having new HCCs or a recurrence within the ablation zone. And so there’s definitely a need to decrease rates of recurrence and increase rates of cure. So given atezo/bev is a reasonably tolerated regimen, we’re excited to see if this is going to be the new standard.

DR LOVE: So let's talk a little bit about biliary tract cancers. And before we get into first-line therapy, I just want to bring up something you mentioned earlier which is this sort of mixed tumor where it's like biliary tract and HCC. How often do you see that and how do you manage those patients? I didn't even know it existed until I heard a case and I'm like, woah, what do you do?

DR GOYAL: Yeah. Yeah, our pathologists are very smart and thorough people and they will often pick up, not often, they will rarely pick up a small component of HCC in a sea of cholangio or a component of cholangio in a sea of HCC. And sometimes it's more mixed and sometimes it's just a small component of it. I don't see it that often. I will say that in the cases that I see, I've probably seen about 10 to 15 cases in the past 10 years. So I don’t see it all that often, but it's certainly a real entity. For a long time, because we had sorafenib for HCC and we had gem/cis for cholangio, we were treating a lot of it like cholangio and giving a lot of gem/cis. Now that we have atezo/bev, we're using more atezo/bev for patients. I kind of take a cue from our pathologists to get a sense of, is it dominantly HCC or dominantly cholangio? Because we have reasonable front-line regimens for both cancers. And then I look at the CA19-9 and the AFP and the overall radiographic features and then we make an overall assessment about what's the best way to approach it. And, of course, the side effects of the different treatments that we would be giving and does anybody have any relative contraindications to immunotherapy or a VEGF inhibitor? So all of those factors go into deciding what to treat with.

DR LOVE: Can you talk a little bit about the issue of sort of the anatomy of biliary tract cancer.

And also, and I think this is one of the questions one of the oncologists had because we're going to get into this issue of first-line therapy and, of course, the TOPAZ study, whether or not you see differences in response to systemic therapy based on the systemic — on the location or the type of biliary tract cancer. You were talking about subset analysis, so I know they look at these things but then again, it's a subset analysis. Are there any sort of themes though that come through in terms of, for example, sensitivity to IOs or any other issues in terms of sort of the location?

DR GOYAL: Yeah. So for a long time, we've thought about biliary tract cancer along these anatomic subtypes, so intrahepatic which involves the bile duct involving the radicals beyond the left and right hepatic ducts, then extrahepatic cholangio which involves the left and right hepatic ducts, the perihilar region where those tumors are called Klatskin's tumors, and then the bile duct. And then sort of distal bile duct cancer also is one of the extrahepatic tumors and that involves the distal bile duct and the mid CBD. And then there's gallbladder cancer and ampullary cancer. And this is the way we used to think about biliary tract cancer. And now, with the advent of next generation sequencing, we've now sort of bucketed these tumors a little bit differently. And while we still are mindful of the anatomic subset, we also think about were there molecular features.

In terms of response to systemic therapy, really interestingly, you probably saw at GI ASCO this year that the results of the SWOG 1815 study were released by Dr Rachna Shroff, and that was the Phase III study of gemcitabine, cisplatin and nab-paclitaxel compared to gemcitabine/cisplatin. And remarkably, while that was a negative study, it did show that the patients with gallbladder cancer did quite well on the triple regimen. And I think that needs to be further explored. But it's interesting that in a gallbladder cancer, it's usually the disease site where those patients have unfortunately the worst prognosis. And the fact that they had sort of the best outcomes on that trial was certainly interesting.

I think that overall, with the immunotherapies, the different subsets are all seeing — we're seeing responses across the different subsets. I think we can do more biomarker analysis to try to understand who are going to be the responders and the non-responders to the combination of gem/cis plus durva, and gem/cis plus pembro, as I'm sure you saw the press release from Merck showing that the gem/cis plus pembro compared to gem/cis alone, that front-line study was also positive.

DR LOVE: So I guess we'll have to see how that data indirectly compares to the TOPAZ study and see how people are going to process that. Any comments though on currently the way you think through first-line therapy? And any other comments about the study that you mentioned that was presented with gem/cis/nab-paclitaxel and whether or not there's a role for that at this time? What your thoughts are about the TOPAZ study and a lot of people are — I hear people saying that it's the new standard of care, whether you agree with that or not?

DR GOYAL: Yeah. Yeah, we're encouraged that after 10 years, more than 10 years, we have finally another positive study that showed improved survival compared to gemcitabine/cisplatin. So just for your audience, the TOPAZ-1 study was a global Phase III study of durvalumab plus gemcitabine/cisplatin compared to gemcitabine/cisplatin in patients with advanced biliary tract cancers. And in that study, the patients in the durvalumab and gemcitabine/cisplatin arm had an overall survival, a median overall survival of 12.8 months compared to the placebo plus gemcitabine/cisplatin arm that had an overall, a median overall survival of 11.5 months. And this was encouraging because also the 24-month overall survival was 25% versus 10% in the doublet — in the triplet arm versus the doublet arm. Again, this landmark analysis was helpful for us to think about the benefits of immunotherapy long-term. And so those were what the data looked like. And that was really encouraging.

And do I believe that it's a new standard? I do believe that it's a new standard.

I think one of the questions about it was, was the overall survival benefit driven by the Asian population, again, looking at the subgroup analyses? And it'll be interesting to see the data from the gemcitabine/cisplatin/pembrolizumab study to see if, again, the positivity in that study was driven by the Asian population or was it seen across different subgroups. So we'll see. But now that we have 2 positive studies with the combination of chemotherapy plus immunotherapy up-front, I think overall, my practice is to give the triplet unless the patient has some reason for not getting the immunotherapy.

DR LOVE: What about the regimen you were referring to of gem/cis/nab-paclitaxel? Do you see any role for that?

DR GOYAL: Yeah. Right now, we have been using gemcitabine/cisplatin/nab-paclitaxel, especially in patients who are trying to convert from having an unresectable tumor to having a resectable tumor. We definitely see good activity with the regimen with a high response rate. And I have certainly seen success, as many of my colleagues have, in being able to convert patient's tumors. The front-line study was negative. The populations that seemed to have derived benefit in subgroup analyses were the patients with gallbladder cancer, as I mentioned, and also patients with locally advanced cancers. So I think that it's a regimen that we still can learn more about. And maybe data from the neoadjuvant setting, which there was a trial looking at neoadjuvant gemcitabine, cisplatin, and nab-paclitaxel for intrahepatic cholangiocarcinoma. I think there'll be potentially roles for this regimen. But right now, there aren't prospective data to confirm that.

I also think that Dr Shroff and her team are going to do biomarker analyses hopefully with the study because tissue was collected. And maybe that'll also help us with insight as to in whom it's worth it to give that third drug.

DR LOVE: So let's turn to I know a favorite area of you and you actually discussed this in the symposium which is targeted treatment. First, maybe you can kind of paint a little bit of a picture. Incidentally, I don't know if I'm the one who came up with this or not, but I've been saying that biliary tract cancer is the new non-small cell lung cancer. Have you heard that?

DR GOYAL: I have not heard that before, but I see the analogy.

DR LOVE: Do you agree? Or maybe not quite as — not quite?

DR GOYAL: I do believe that with intrahepatic cholangiocarcinoma. 40 to 50% of people have actionable targets and I do believe, just like in non-small lung cancer, we should be sequencing everybody's tumor to see if they are a candidate for targeted therapy. So in that way, yes, I definitely feel there's a strong analogy between the two.

DR LOVE: Yeah. And there's — that's a higher fraction than —

DR GOYAL: I would say our therapies are not as effective as the therapies in non-small cell lung cancer. So they have EGFR inhibitors, ALK inhibitors, ROS1 inhibitors and the progression free survival and the overall response rates you're seeing on — we're seeing on their drugs, we're not yet seeing with FGFR inhibitors or IDH1 inhibitors and the HER2 inhibitors. But the operative word there is yet. I am hopeful that we will get there as we develop next generation drugs.

DR LOVE: Yeah, and I thought — I can tell you, oncologists are excited about it because they love targeted therapy in general, particularly in lung cancer. And when they see it — and when they saw it here, they were all over it. But maybe you can paint a little bit of a picture of what we know about, you know, they're also telling me all the time that they can't find any of these patients when they look for it. So I'm curious what the yield is when you look at tissue and also liquid biopsy, and what the spectrum is in frequency globally, not exactly, but just generally of the targetable lesions that you see in biliary tract.

DR GOYAL: Yeah. So each of the different subtypes have different rates of these different mutations. FGFR2 fusions are seen in about 10 to 15% of intrahepatic cholangiocarcinoma, rarely seen in extrahepatic cholangio, and almost never seen in gallbladder. IDH1 mutations are also seen in about 10 to 15% of intrahepatic cholangiocarcinoma, maybe 1 to 2% of extrahepatic cholangiocarcinoma, and almost never in gallbladder cancer. Then there are BRAF V600E mutations which are seen in about 3 to 5% of biliary tract cancers. We see HER2 amplification, more common in extrahepatic cholangio and gallbladder cancer, somewhere up to 15% even in gallbladder cancer, and probably closer to 3 to 5% in intrahepatic cholangiocarcinoma with ECC being in between. Then we have more rare targets such as NTRK fusions, RET fusions. In my experience, I have actually never picked up one in the years that I've been taking care of patients with cholangiocarcinoma biliary tract cancers. I would put it down around less than 1%. But the RET inhibitors and the NTRK inhibitors are very effective molecules that have high response rates and good durations of response. So definitely worth sequencing for those fusions because when we see it, we have effective therapy. Then we also have microsatellite instability where it's seen in about 1 to 2% of biliary tract cancers, and we certainly have pembrolizumab and different trials of immunotherapy agents for these patients that can be quite effective.

There is also tumor mutational burden. In the study where tumor mutational burden was studied as a biomarker and that led to the approval of pembro, there really weren't patients with biliary tract cancer that had high tumor mutational burden in that study and showing benefit with pembrolizumab. So in the absence of other options, if we have a patient with high tumor mutational burden in biliary, it's very reasonable to try pembrolizumab or immunotherapy, but I think it's not totally validated yet in biliary cancer. And then there is smattering of other targets such as KRAS G12C and MET amplification. And every once in a while, even an R-spondin fusion. So it's definitely worth doing broad NGS to look for these rare targets because there are certainly basket trials for some of these rare targets that are histology agnostic that our patients can certainly benefit from. That's how for RET and NTRK, there were patients with cholangio in those basket trials.

DR LOVE: And you were mentioning TMB. And I don't know if you remember the case, I can't even remember whether it was in the HCC section or biliary, I think it was biliary, but where one of the docs said it looked — a lot of times I'm seeing high TMBs on liquid biopsy. Is there something not right about that? And I think the faculty — I hadn't heard that before, and I think the faculty said yes. Is that right, that you see higher values and maybe can't depend upon 10 being the level to look at?

DR GOYAL: Yeah. The use of 10 mutations per megabase, that was based on tissue-based testing. And I agree, there's published data showing that on ctDNA, the rates of the mutational burden is certainly higher than what you see on tissue. And so I generally do not use liquid biopsy for my cutoff to figure out if I should give someone immunotherapy.

DR LOVE: So let's talk a little bit about each one of these starting out with FGFR fusions. First of all, do you always pick those up on NGS or do you need to use an RNA assay?

DR GOYAL: No, the assay definitely matters. If you use some of the big commercial assays, they certainly cover the FGFR fusions. And sometimes even if they don't identify the partner, they'll identify an FGFR rearrangement. I think the sensitivity and specificity of the large commercial assays is quite good. There are some commercial assays that do RNA sequencing along with DNA sequencing, and those also are good at picking up the FGFR fusions. I think these panels that are limited panels that mainly look at point mutations, I think we're going to miss fusions on some of those panels and certainly on some of the liquid biopsy panels also. It's not easy picking up fusions in — on blood biopsies.

And we just published a study looking at patients getting the Guardant360 assay. And we found that 18% of patients who had effusion, their effusion was picked up on Guardant when we looked at them matched with patients who had the FGFR2 fusion in tissue. Now then we also recently published data from the futibatinib study, the FOENIX-CCA2 study. That was a Phase II single arm study looking at futibatinib in patients with FGFR2 fusion and rearrangement positive cholangiocarcinoma. And that study when we looked at blood and tissue, the positive percent agreement between tissue and blood, if I remember exactly, it was 88%, but it was certainly much higher than 18%. And that was using the Illumina TruSeq 500 assay. And that's a research assay right now. And so I think the assay definitely matters. If you pick up a fusion on ctDNA, great. The specificity is high so I think it's real. If you don't pick it up, I think certainly still doing tissue-based testing is relevant because we don't want to miss an FGFR2 fusion for our patient.

DR LOVE: So one of the — it seems strange to me that there's so many of these drugs that are being tested. It's incredible for something so uncommon. But first of all, is there — can you kind of just — I assume, it seems like in general they're similar. You can track out the differences. But globally, what do you see in terms of efficacy in this situation? Is it a typical targeted kind of thing where you see the vast majority of people responding and then progressing? What kind of model globally does this fit into?

DR GOYAL: Yeah. I think FGFR2 fusions are unique in that you do see responses to the FGFR inhibitors, so for example, with pemigatinib, the first drug to ever be approved in cholangiocarcinoma and it was an, it is an FGFR inhibitor. It's a reversible ATP competitive inhibitor and the response rate was 36% in patients with FGFR2 fusion and rearrangement positive cholangiocarcinoma. And then with futibatinib, the response rate is 42% in patients, a similar population but restricted to intrahepatic cholangiocarcinoma. And that's a covalently binding irreversible inhibitor. Having a response rate of 36 to 42% in a fusion positive population, that's good, not great. Like what we see with ALK inhibitors, ROS1 inhibitors is we can see response rates of 70 to 80%. So, there's a lot of research going on right now about primary resistance. And for example, with pemigatinib, the team for that study showed that if patients have a TP53 mutation concurrent with their FGFR2 fusion, in that study there were no responses. And then in the futibatinib study, they showed that whether someone had a TP53 mutation or not, the response rates were similar with futibatinib. So the numbers were small. There were about 100 patients in both studies. The number of patients with TP53 mutations was roughly around 10 plus or minus in both studies, so it's a little hard to make big definitive statements. But I think it sort of remains to be seen, what are the drivers of primary resistance? Because, we certainly want to get that 40% closer to 80% and, ideally, up to 100% with our FGFR inhibitors.

DR LOVE: At this point, can you distinguish them either based on efficacy or tolerability? Including erdafitinib which is approved in bladder cancer, I know it's been studied here as well, but when you kind of look at these various agents, indirectly comparisons with fairly small numbers of patients, anything you can separate out? Do you have theoretically a preference about what you think might be best? And any thoughts, you know, we were talking about lung cancer. Lung cancer, most of the targeted agents are given up-front because, like you say, they have response rates of 60%+. Any discussion about trying to bring these molecules up front?

DR GOYAL: So with pemigatinib and futibatinib, I would say the data are similar. Just to tell you the numbers, for pemigatinib as I mentioned, response rate was 36%, futibatinib 42%. The progression free survival for pemigatinib was 7 months, and the progression free survival for futibatinib was around 9 months. But very similar numbers, very similar populations, and I think both are reasonable options in the refractory setting.

I think that in terms of bringing them to the front-line, there have been efforts to try to look at pemigatinib versus gemcitabine/cisplatin in the first-line, infigratinib versus gemcitabine in the first-line, and futibatinib versus gemcitabine/cisplatin in the first-line. It's been difficult to enroll to those studies to be honest. One, because biomarker testing right now takes time. It can take up to 2 to 4 weeks to identify if someone has an FGFR2 fusion. And understandably, when patients are diagnosed, both patients and oncologists are eager to start treatment. And when you have a 10 to 15% chance at having the target and you can still get the FGFR inhibitor in the second-line, I think the enrollment has been difficult.

Do I think that they could outcompete gemcitabine/cisplatin, on overall survival? I think it's possible but ultimately, we would have to do the study to see. There are newer next generation inhibitors that are coming out, covalent inhibitors, reversible inhibitors, that have been designed to overcome many of the mutations that develop at the time of resistance to these drugs. So what we know about FGFR resistance is that in a percentage of patients, they're, it's mediated by kinase domain mutations, so secondary mutations in the FGFR gene in the kinase domain. And so a lot of these drugs have been designed to be able to address those clones. And ideally, these drugs will have higher response rates and longer progression free survival. I think the higher response rates are more related to the potency against FGFR2. And the duration of response and progression free survival potentially are more related to being able to stomp out some of these resistant clones and having the tolerability for patients to be able to stay on drug longer.

And so some of the other data that we talked about during the symposium at GI ASCO were, for example, the RLY-4008 study, R-L-Y-4008. That is an FGFR2 specific tyrosine kinase inhibitor. It's covalently binding, hits FGFR2 quite hard. And their preliminary data showed in 17 patients that the overall response rate was 82% in patients who had FGFR2 fusion positive cholangiocarcinoma that were FGFR inhibitor naïve. And the duration of response is still maturing for that study so we'll see what that looks like, but that was certainly quite encouraging. And then we also presented data from tinengotinib. It's another drug that is a multi-kinase inhibitor that's there to address many of the kinase domain mutations. And then there are other companies that are also looking in this space. Futibatinib is another drug that we presented at the symposium. So luckily, there's a lot of drug development in this area. I think time will tell in the end what the response rates and the durations of response are for all of these different drugs.

DR LOVE: Any comments on tolerability of these agents? You may remember one of the cases, the patient had a great response to pemigatinib but had problems with muscle pains or arthralgias. In general, of course, we know from bladder cancer, I don't know how exactly how comparable it is or issues with phosphate and other ophthalmic issues. What do you see with these agents?

DR GOYAL: Yeah, I would say the 5 or 6 most common things we see are the hyperphosphatemia certainly across drugs and the stomatitis or mouth sores, the hand foot syndrome, or palmar-plantar erythrodysesthesia, diarrhea, and the nail toxicity that we see where the nails become bridle, sometimes lift off. Sometimes their nails actually fall off unfortunately. And then we rarely luckily see some of these ophthalmic toxicities that you mentioned where we can see retinal detachment or some fluid underneath at the central serous retinopathy that we see. In all cases that I've seen, the CSR has been reversible. We stopped the drug, the floaters go away, and we follow them closely with ophthalmology. But certainly, a concern that we need to take seriously and follow closely for our patients.

DR LOVE: You mentioned HER2, and I believe there have been data with T-DXd, the antibody drug conjugate in biliary. Maybe I'm not remembering that correctly. But any, any efforts to use anti-HER therapy that have been successful with biliary tract cancers?

DR GOYAL: Yeah, there have been several agents tried in biliary tract cancers. There was the MyPathway study of trastuzumab and pertuzumab, and the overall response rate with that combination and HER2 overexpressing or amplified biliary tract cancer was 23%. That's now on the NCCN guidelines as a combination that we can use. You're right, there was trastuzumab/deruxtecan as well, and that study, there was a 36% response rate and a 7.4-month duration of response. And so that's another one that's not yet on the NCCN guidelines but a drug that's being studied in biliary. And the other one is zanidatamab which is basically a bispecific antibody that's kind of like a combination of trastuzumab and pertuzumab. And that one in a small study had an overall response rate of 40%, but that's now being studied primarily in a biliary tract cancer population to see if we can move that drug forward for biliary.