Key Takeaways

• Serial circulating tumor DNA (ctDNA) assessment and comprehensive next-generation sequencing can help oncology APPs identify actionable alterations, monitor molecular changes at progression, and support treatment decisions throughout the disease course.
• FIGHT-302 demonstrated activity with pemigatinib in FGFR2-rearranged cholangiocarcinoma, reinforcing the importance of early molecular testing to identify patients who may benefit from targeted therapy.
• Emerging HER2-directed strategies such as zanidatamab plus tislelizumab and chemotherapy may reshape treatment of HER2-positive gastroesophageal adenocarcinoma, with APPs playing a key role in monitoring and managing diarrhea, cardiac toxicity, and other treatment-related adverse events.

Tanios Bekaii-Saab, MD, chair of the Division of Hematology and Medical Oncology at Mayo Clinic in Arizona, and Jonathan Catrona, MS, RPA-C, physician assistant with New York Cancer & Blood Specialists, continue their discussion of emerging gastrointestinal cancer standards based on the latest data from the ASCO 2026 Annual Meeting. 

Bekaii-Saab and Catrona discuss the evolving role of ctDNA and next-generation sequencing at diagnosis and progression, including their potential to identify emerging resistance mechanisms and guide subsequent therapy. They also review findings from the phase 3 FIGHT-302 trial of pemigatinib in FGFR2-rearranged cholangiocarcinoma and HERIZON-GEA-01 data evaluating zanidatamab-based therapy in HER2-positive gastroesophageal adenocarcinoma. For oncology APPs, these developments underscore the growing importance of genomic literacy, appropriate molecular testing, patient education, and proactive toxicity management as biomarker-driven and bispecific therapies become increasingly integrated into gastrointestinal oncology care.

Transcript

Tanios Bekaii-Saab, MD: My name is Tanios Bekaii-Saab. I’m a GI medical oncologist and chair of the Division of Hematology and Medical Oncology at Mayo Clinic in Arizona.

Jonathan Catrona, MS, RPA-C: My name is Jonathan Catrona. I am a physician assistant at New York Cancer & Blood Specialists, which is a large community oncology practice in the Long Island metro area.

One thing, especially from an APP’s perspective for ctDNA, I think it’s important that it could be something that we need to be mindful of, especially when a patient is remaining ctDNA-negative over a long period of time. Especially over 2 to 3 years, there’s a likelihood that this patient is to be cured. But if they turn from negative to positive, it should increase your concern that recurrence is probably likely. That’s something you need to prepare your patient for, especially because more data is key. I think that’s important with informed decision-making and talking to the patients.

One big thing is especially side effect management. FOLFOX and FOLFIRI are similar, but with irinotecan, you do get less of the neuropathy compared with oxaliplatin. You don’t get any neuropathy, but you still get the diarrhea. Even though it’s a similar drug, that’s something we have to do good patient education for.

I think that is something that could be very useful going forward, no matter if you’re in the academic setting or if you’re in the community setting. Hopefully, once this study matures and once we have the data, we should know a little bit more specifically about ctDNA, especially because right now what we’ve seen so far looks promising, and a lot of us are using it. But I think we’ll know for sure once the study matures.

Bekaii-Saab: I think, to your point, just be very judicious. Certainly, in my mind, one study—the one actually I had the privilege to present, as a study I helped lead—is a study called FIGHT-302.

That’s a phase 3 international randomized trial looking at essentially pemigatinib, which is an FGFR inhibitor, versus chemotherapy in biliary tract cancer. That’s primarily the intrahepatic cholangiocarcinomas. One of the challenges, of course, with the study was the fact that FGFR fusions were less common than we initially thought. We had to screen about 4500 patients to be able to enroll about 167.

Catrona: That’s not a lot.

Bekaii-Saab: Yeah, that was a challenge. But the good news is this was first line. Our goal in this study was to see if we can get pemigatinib to replace chemotherapy, at least for those patients with FGFR2 rearrangements, and selectively so.

The study essentially was positive in the sense that the median PFS, which was the primary end point of the study, with pemigatinib was better than gemcitabine and cisplatin. The response rate was actually staggeringly better. It was about 47% versus close to 15.5%. But the other interesting aspect of this is that the median duration of response was 14.2 versus 6.3 months at the median. You know that those responses happened, and not only when they happened, they happened and they were pretty prolonged.

The median overall survival was similar in both, and that’s not surprising since there was a crossover component to this study. And 80% of the patients in the chemotherapy group ultimately received FGFR inhibitors at some point after progression, including pemigatinib in half of those because that was prespecified by the study.

In my mind, at least, the study fell short of its accrual because of the difficulties in accruing to these highly selected studies, but it was still the largest study to report out on the activity of FGFR inhibitors in the first line with pemigatinib in a randomized setting and showing that pemigatinib is a good option for those patients that have FGFR fusions.

That brings essentially the key question. Again, interested to hear how you guys think about it. For us, at this point in time, for any patient with—well, with GI cancer anyways, but biliary tract cancer—we send a tissue biopsy for next-generation sequencing and ctDNA, so liquid biopsy, to assess for these targets early in the game instead of waiting for the patients just to progress. The reason for this is a study like this. I would have a discussion with the patient now and say, “Hey, these are the results,” or pemigatinib, gemcitabine/cisplatin, plus or minus durvalumab or pembrolizumab.

Catrona: We do something similar. In New York, we actually have a practice policy for anybody who’s stage III and IV, especially in the GI cancers. We must do next-generation sequencing on tissue and on blood because you’re looking for targets like EGFR, PD-L1, or HER2, because it could be practice-changing.

Especially with the TKIs, some of the TKIs you have to use before PD-L1 unless you get these crazy side effects. Especially when you look for EGFR, that is something that, even though you won’t see a lot of it, once you do, you have something you can use that has significant benefit over what’s normal standard of therapy.

I think for anyone, especially in academic or in a private practice, doing NGS testing on tissue and blood at the time of diagnosis, especially for later-stage cancers, is key. Even now with other trials we’re seeing, even in the adjuvant setting, that’s going to be more important. Even now, we’re looking at possibly using PD-L1 therapies in stage II, stage III adjuvant setting. That’s something you need NGS for to screen for.

Normally, they might reflex it when they do the biopsy, but usually NGS can give you the clue. It also gives you the percentage of PD-L1. That’s important. But also, it may give you other markers, God forbid, if the patient does recur.

One other thing. One question I had for you is, do you guys ever, let’s say you do the next-generation sequencing at time of diagnosis, do you ever repeat it? I know that’s always been a contentious point in our practice. Some of our clinicians don’t. Some of us do. I usually do, but it’s always a misnomer because some of us will say you won’t see anything different, but if it’s progressed, you might see something different. One of the more obvious things, like breast cancer, you see PR, ER, HER2 change all the time. Why don’t we see that with—not those markers—but different markers change as well? I want your perspective on that.

Bekaii-Saab: I can see the challenge with this. Universally, to start this, everyone gets it at point zero before we start any treatment. They get a liquid and a tissue. But yeah, I actually think that it’s very reasonable to, and I do that in my clinic routinely, at every point of progression, draw another ctDNA.

Why? Because do things change dramatically? No. But I’ll give you the example of this FGFR, that FIGHT-302 study. What we found, for example, is that about a third of the patients that progress on pemigatinib, whether it was first line or at progression, develop FGFR2 mutations and others, so activating mutations.

Well, it turns out that there’s an agent called futibatinib that has essentially coverage for these mutations. Many times what we do in practice, when we see those arising, then the patient will get futibatinib, and many of those patients—and that has been shown in other studies—will actually respond.

Now, we’ve extended the life of FGFR inhibitors. Other instances of that are with EGFR inhibitors in colorectal cancer with EGFR rechallenges. If you persistently continue to show absence of RAS mutations or even BRAF mutations, so MAPK mutations or alterations, then you have a 20% to 30% chance to see a response again with rechallenging the patients with EGFR.

HER2, same thing. You don’t lose HER2 per se, but with colorectal cancer, you have large heterogeneity of tumors and clones. Some clones may not be as expressive as others. You can see the decrease or absence of clonality.

Yes, I think with a prespecified intent, meaning that, one, I know that that’s going to benefit the patient, and two, I may have agents that help salvage, I think it’s actually very reasonable to do serial circulating tumor DNA assessment.

You can’t biopsy. You can’t follow the patient with a needle. Two reasons for that. One, of course, it’s disruptive to quality of life and it certainly poses its risk. But the second is, because of the heterogeneity of the tumors, of the cancers we treat, a needle going into a single point does not necessarily represent the totality of the cancer, which is something that the liquid biopsies do a little bit better.

Catrona: Speaking of HER2, there was one big trial that came out in HER2 in the gastroesophageal adenocarcinoma space, which was the zanidatamab.

Bekaii-Saab: Zanidatamab.

Catrona: Zanidatamab. Looking at the new trial from HERIZON-GEA-01, the triplet, which is zanidatamab, tislelizumab, and chemotherapy, beat trastuzumab and chemotherapy. Overall survival, 26.4 months versus 19.2 months, which is surprising because trastuzumab has been the HER2 go-to for a while. Especially since it was first approved in the HER2 setting for gastroesophageal cancer, it was probably the go-to. Now it looks like it’s being usurped, which is good for the field.

One thing that’s interesting is diarrhea was the most common side effect, which we see with trastuzumab. Especially in my practice, trastuzumab-induced diarrhea is something we see constantly. You get treatment with loperamide or Lomotil or even the medieval opium—I mean, tincture of opium. It’s something they use routinely. I think that’s something that’s probably going to continue.

One thing I wish they would have touched on is cardiotoxicity. I know with trastuzumab, you do see some cardiotoxicity. You don’t see it as much. You usually see more in breast with the doxorubicin because you get the augmented effect with both. Trastuzumab usually is reversible. It’s not like doxorubicin, where it’s irreversible. But that is something I am wondering, once we get the full trial, what exactly we’re looking for in cardiotoxicity, because that is a big important thing for trastuzumab that usually prevents us from using it.

But the data from ASCO are reassuring. Even the progression-free survival is sitting up any longer. I believe it was 12.4 months versus 8.1 months, which is reassuring.

Bekaii-Saab: Yes, on all these fronts. Zanidatamab is a bispecific and targets domains both pertaining to trastuzumab and to pertuzumab.

Many years ago, when trastuzumab became the standard, there was a study called JACOB, which essentially looked at trastuzumab, pertuzumab versus trastuzumab. Two antibodies versus one. The study was numerically positive but statistically negative. It stopped the development of further dual therapy.

Zanidatamab is a bispecific, and the bispecifics have an advantage. We think we’re giving dual antibodies just because of focalization, and they actually attach with more potency to the individual sites. Zanidatamab, essentially, not surprisingly, is better than trastuzumab.

Where it becomes very interesting, I think, is when we remember KEYNOTE-811, where trastuzumab, pembrolizumab, and chemotherapy was better than trastuzumab and chemotherapy in the PD-L1-positive group only. In fact, in the negative, it was detrimental. With zanidatamab—and that’s what we’ve seen at ASCO—with zanidatamab, whether it’s PD-L1-positive or PD-L1-negative, you are doing better with zanidatamab and tislelizumab. It’s a different outcome. It’s a game changer in that sense.

Catrona: One minor thing with zanidatamab that always concerns us in the community setting is access, because usually we have a community partner. In New York, this space is pretty competitive, so it is a little difficult finding a partner. But especially in a hospital-based setting, this is something I know that colleagues who work at practices at Memorial Sloan Kettering and New York Presbyterian use routinely. This is something, especially in the GI space, that’s pretty new.

Bekaii-Saab: We’ll see what the FDA does with it. It’s probably likely they’ll approve it. If it does, then hopefully that will resolve that issue.

I want to go back to the issue with diarrhea, and you’re absolutely right. That was something that was very intriguing and certainly problematic. But the chemotherapy that was used in this study was primarily CAPOX.

Catrona: Oh, yeah.

Bekaii-Saab: And as you know, CAPOX is just pure misery for many of these patients.

Catrona: Yeah, that’s the full-dose CAPOX.

Bekaii-Saab: I think that actually drove a lot of the toxicities that have been seen. I suspect that, as we in the United States are more likely going to use FOLFOX, that we may see less. Now, that doesn’t mean that you’re not going to see diarrhea, but it’s much, much less significant.

Also, when you mentioned the left ventricular dysfunction, there was a little bit more left ventricular dysfunction, but mostly reversible, again, like we’ve seen with trastuzumab. But you do see it more. You still have to keep an eye on the potential for cardiac toxicity in those patients. It seems like it’s just a HER2 class effect, which is the cardiac—less so the diarrhea. Diarrhea is part of it as well. Diarrhea was a bit unusual. But you do see more diarrhea with zanidatamab, but not to that level. I think the CAPOX really did not help. If anything, probably made it much worse.

Catrona: Especially for APPs, bispecifics have kind of become a new frontier. That’s something that we need to be aware of, especially something that’s probably going to redefine our field in the next 5 to 10 years to come. We’re seeing it now with multiple myeloma, with CLL, with these new agents coming out, even with small cell lung cancer with the new bispecific approved. Now we’re going to hopefully probably see it in GI cancers if this wins FDA approval.* This is something that we need to keep an eye on.

One thing, especially going back with toxicity, as APPs, this is our bread and butter. We need to be mindful of the diarrhea. We need to treat it. Especially the same agents you use with CAPOX and FOLFOX are the same agents you’re probably going to use for this regimen as well. That’s something that we need to keep in mind, and we need to stay vigilant.

I know for my patients, especially at the beginning, if I know they’re going to be on FOLFOX, FOLFIRI, or even CAPOX, I give them a prescription of Lomotil. I say, “Take it if you need it. I hope you never need it.” Usually they do. And if you start using both loperamide and Lomotil, let me know so I can call in something else.

One other thing you can use is also short-term Sandostatin. It’s something you can use. They can come in once a day for it, which is something you could use. That works once in a while. But patients are usually pretty reluctant to come to the office every single day for the next injection. But it does help. It’s good for when you have patients who come into the office, let’s say they’re having a day already needing hydration. That’s something you could give them. This is something that’s probably going to reinforce our practice, which is something that we need to keep mindful of. 

Going back to zanidatamab, even though it’s a bispecific, you do not see the CRS because of the lack of a T-cell engager. That’s important. You don’t see much of the other side effects that we have unfortunately become accustomed to seeing in lung cancer and in multiple myeloma. This is something we’re not happy seeing. Thankfully, because of this, we won’t see the step-up dosing with zanidatamab compared with the other agents, which is reassuring.

I’m kind of hoping we’re also going to see this in the community setting sooner rather than later, especially because if we can avoid the hospital, I think all patients would like that in the community setting. But especially going forward, starting these new agents, we need to be mindful of side effects going forward.

*Editor’s note: The speakers discuss zanidatamab as awaiting FDA action, but the FDA subsequently approved zanidatamab-hrii with chemotherapy, with or without tislelizumab-jsgr depending on HER2 testing criteria, on August 25, 2026.