Defining the Role of Neoadjuvant Radiation in Cholangiocarcinoma

CancerNetwork® spoke with Ethan B. Ludmir, MD, associate professor in the Department of Gastrointestinal Radiation Oncology at The University of Texas MD Anderson Cancer Center, ahead of the 2026 Chicago Cholangiocarcinoma Symposium about the evolving role of radiation therapy in the management of cholangiocarcinoma.¹

Ludmir discussed the rationale for shifting radiation into the neoadjuvant setting, which patient subsets and tumor locations may benefit most, and how modern techniques such as hypofractionation, stereotactic body radiation therapy (SBRT), and proton beam therapy fit into preoperative and nonoperative care. He also addressed how radiation may be sequenced with chemoimmunotherapy regimens such as gemcitabine, cisplatin, and durvalumab (Imfinzi), the importance of overall survival (OS) as a trial end point, the ongoing randomized phase 2 HYPERION CCA trial (NCT06858735), and the September 23, 2026, FDA approval of lirafugratinib (Lyrfigtu) for FGFR2-altered cholangiocarcinoma.

CancerNetwork: What clinical rationale supports shifting radiation therapy into the neoadjuvant setting for borderline resectable or locally advanced cholangiocarcinoma, rather than reserving it for definitive or postoperative adjuvant management?

Ludmir: It speaks to a lot of the unknowns in treating cholangiocarcinoma. We have most of our data for using radiotherapy in the adjuvant setting: the SWOG S0809 trial (NCT00789958)² and several others that use radiation, for lack of another way of putting it, as a cleanup crew, particularly for margin-positive resections. Those trials included anyone with any adverse pathologic features, such as pathologic T2 disease or higher or pathologic node positivity. Most patients with extrahepatic cholangiocarcinoma, hilar cholangiocarcinoma, or gallbladder disease could qualify for adjuvant radiation if you use those trials as the paradigm.

We extrapolate from diseases like pancreatic cancer, which is a reasonable sister disease, in a way, if you think about distal extrahepatic cholangiocarcinoma, which often shows up in the head of the pancreas. Patients tend to tolerate neoadjuvant radiation a lot better than adjuvant radiation, so we tend to think that, from a patient-centric standpoint, it’s a little bit better tolerated in the neoadjuvant setting. If you believe data in the pancreas space, especially in borderline resectable pancreatic cancer, there seems to be some benefit of neoadjuvant radiation in achieving an R0 resection—getting everything out when the surgeon goes to the [operating room].

There are limitations and caveats there. Radiation can influence how things are for surgeons in the surgical field, and this gets down to individual practitioner variation and expertise. Some of our surgeons in the pancreas space, for instance, feel very strongly that neoadjuvant radiation, if delivered just so, 6 or 8 weeks before surgery, can help create edematous tissue planes that can help remove disease away from the vasculature. It might be a net positive, but there’s heterogeneity in how people practice.

Which specific patient subsets or tumor locations, intrahepatic vs perihilar or distal, stand to benefit most from neoadjuvant radiation before surgical evaluation?

We can [categorize] them in our minds as intrahepatic vs extrahepatic, writ large. For intrahepatic disease, we don’t have great data in the neoadjuvant or adjuvant settings, and more often, those patients are simply not resectable. For patients who truly are resectable, perioperative chemotherapy and chemoimmunotherapy are paradigms that are evolving and emerging with time. Radiation, for the data that we have, has not played an exceedingly strong role, and those are all very limited data based on National Cancer Database analyses and population studies, which are very weak data, as those things go. We don’t tend to believe those data are very high-quality evidence, and we don’t have excellent data to use neoadjuvant or adjuvant radiation for intrahepatic cholangiocarcinoma. That said, at a tertiary or quaternary care center that treats a lot of cholangiocarcinoma, in very select settings for intrahepatic cholangiocarcinoma, we will say, “You know what? We’re very close to this vessel. I want to be able to deliver some radiation so that when we clear that, it’s not against some very large vessel in the liver, and when we do whatever version of a hepatectomy is required, we’re more likely to get an R0 resection.”

On the flip side, to your question, distal extrahepatic cholangiocarcinoma is probably where neoadjuvant radiation shines brightest. That’s probably because it’s the closest analogy anatomically to pancreatic cancer, and we tend to think the surgical operations are very similar, so that’s where our headspace tends to live. Hilar tumors, which we bin as extrahepatic disease, are this interesting no man’s land. We often will do radiation, but we don’t do it as a segue to surgical resection. We often do it as a bridge to transplantation, which is something that’s been done by our center, by the Mayo Clinic, and by many others for many years, with different variations of what radiation looks like in that setting.

Far more interesting is the role of radiation outside of the resectable or transplantable setting. Most patients simply do not fit into a surgical or transplant pathway. Most patients are locally advanced, or they have metastatic disease or multifocal intrahepatic disease, and those pathways look a little more systemic therapy based. The question then is, why would you do locoregional therapy? Wonderful papers from many groups—and we’re privileged to include some studies from our center in those numbers—are suggesting that even in metastatic disease, or even with enormous, what we call supermassive tumors that are not surgically resectable and are 20 cm to 25 cm across, doing definitive local therapy with radiation seems to confer meaningful benefits in terms of disease control and survival. We’re running randomized trials now at our center and across the country to answer that question more meaningfully.

How are modern targeted radiation techniques, such as hypofractionated radiation or proton therapy, enabling the delivery of biologically effective or ablative doses safely in the preoperative setting without compromising subsequent surgical planes?

I’ll spin it a few ways. Number 1, in the preoperative setting, the doses we tend to use, something like 50 Gy in 25 fractions, I wouldn’t call gentle. Patients do feel [adverse] effects from this; it is not a benign intervention. We space out the radiation over 5 weeks, and the advanced radiation techniques that we have seen emerge over the last decade or 2 are not necessarily as meaningful or as impactful in that setting. Don’t get me wrong, they’re still valuable in their own way, but the doses we’re using are not typically ablative, and that’s for a few reasons. We tend to think that in the preoperative setting, if you use exceedingly high doses of radiation, you risk creating a fair bit of fibrosis for the surgeons later, even if you go to the operating room relatively quickly. Some surgeons don’t feel this way, and some do, so there’s a lot of heterogeneity.

Certainly in the adjuvant setting, after a surgeon has done the procedure, based on everything that has to happen as a product of the surgery and the anastomoses being made, depending on what procedure is done, we typically don’t feel that hypofractionation. Proton beam radiation and many of these other tools in our toolbox are appropriate because there is increased biological effectiveness with those modalities, and we see higher risks of [adverse] effects for patients. For patients watching this, and this is very important, just because a center has access to certain advanced radiation techniques doesn’t always mean it’s the right tool for that situation. Prudence says, “What’s the right tool for the right situation?” Very often, I will not use protons for neoadjuvant or adjuvant radiation, not because there’s anything wrong with protons per se, but because we tend to believe that biologically and biophysically, there are different properties to those particles. As an institution, we’ve tended to feel that the toxicity risks are higher if you use some of those techniques in the neoadjuvant or adjuvant setting.

Now, all that said, most patients are not going to the [operating room]. What do you do for them? In those settings, there’s hypofractionated radiation, proton beam therapy, and advanced SBRT techniques with MRI-based linear accelerators…. To be clear, no one tool is better than the other. You hear a lot of commercial words, like CyberKnife, and patients hear this, and even oncologists hear this. Unless you’re knee-deep in doing liver-directed radiation all day long, the water can get muddy real fast.

Big picture, there are a few key elements here. Number 1, the data we have so far show that if radiation is your definitive local therapy, turning the dial up seems to help in terms of durable disease control. That was a seminal paper published about 10 years ago from our institution,³ and that work has only grown from there. Number 2, getting those doses in must be done very safely, especially for bigger tumors or for patients with underlying liver dysfunction, which happens quite often in patients with cholangiocarcinoma. As with any local therapy to the liver, you have to treat the cancer, but don’t break the liver. That’s a balancing act. Proton beam therapy, especially for large tumors on the right side of the liver, allows you to dial the dose very high and, most importantly, spare the uninvolved liver. That doesn’t mean one modality is always right or wrong.

All that’s by way of saying these are neat techniques that each have their right moment. For the nonoperative patients, that’s when you see proton beam therapy, high-dose radiation, and hypofractionated techniques, where you can deliver very walloping doses of radiation over 1, 2, or 3 weeks to knock out these tumors. That’s where that stuff shines.

What impact does preoperative radiation have on surgical complexity, tissue toxicity, and postoperative complication rates?

The answer is in the eye of the beholder. If you ask our surgical friends, and again we’re going to use pancreatic cancer as a basis for that answer, if you deliver, say, neoadjuvant SBRT, very high-dose radiation, there are surgeons who feel that if you wait 6, 8, 12 weeks and beyond after SBRT is delivered, that high-dose radiation causes a more exuberant fibrotic reaction and can make the surgical field more challenging. I am not a surgeon, so I don’t claim to know this firsthand. This is part of that underlying recommendation: cholangiocarcinoma is a very complex disease, so good multidisciplinary management and being at a center where we all talk to each other is very important.

Can preoperative radiation impact surgical complexity, fibrosis, and surgeons’ ability to cut a tumor out? Yes, absolutely, and it can cut both ways. Like we said, if you do neoadjuvant long-course chemoradiation to something like 50 Gy in 25 fractions, this 5-week course of standard radiation, there are plenty of surgeons who believe it helps with some of the tissue planes, particularly in trying to ensure that we can get an R0 resection. The data from pancreatic cancer do seem to support that we can improve our ability to obtain R0 resections in these settings. It’s a complex balancing act. Again, it speaks to the importance of making sure you’re at a place where we’re all friends with each other. That is a very nice feature of being at some of these centers, where we do talk to each other, understand the other specialties’ perspectives, and no one is siloed.

How should neoadjuvant radiation be optimally sequenced alongside modern systemic backbones, including chemoimmunotherapy regimens?

We have no idea. One of the things about cholangiocarcinoma being a relatively rare cancer, with 6000 to 7000 cases in the US per year, is that we are trying to integrate data as they change in front of our eyes. After TOPAZ-1 (NCT03875235)⁴ and KEYNOTE-966 (NCT04003636),⁵ where we now see the emergence of chemoimmunotherapy as the standard of care, it is uncertain what the right sequencing is. For hilar and distal extrahepatic cases, I do think we’re seeing more use of perioperative chemoimmunotherapy followed by or integrated with radiation in some way. There are plenty of intrahepatic cases where we know folks are doing perioperative chemoimmunotherapy without radiation, and there isn’t any clear right or wrong answer. It speaks to the fact that we’re all flying a little bit blind. We’d love to see some good, randomized data on the board.

That lets me put in a plug for the trial in locally advanced and metastatic intrahepatic cholangiocarcinoma that’s being run primarily out of Houston at MD Anderson. We’re trying to make sure it is enrolling in as many places around the country as we can. It [randomly assigns] patients who receive gemcitabine, cisplatin, and durvalumab, which is the standard chemoimmunotherapy backbone for intrahepatic cholangiocarcinoma, to local therapy with definitive radiation vs continued immunotherapy alone in a 2:1 randomization. That trial is called HYPERION CCA,⁶ and it has been enrolling very well for the last year. We’re very much hoping that we can get good randomized data to drive our clinical decision-making.

Does achieving a complete or major pathologic response following neoadjuvant radiation correlate with improved long-term disease-free and overall survival in resected patients?

It’s a great question. Insofar as I can tell, the answer seems to be yes. That’s not a shock; if you look at any solid tumor, that probably seems to be true. There are variations on that theme. Statistically, there are some interventions that we know can increase pathologic complete response or major pathologic response, but because this is the use of a surrogate end point to try to predict something longer term, what we don’t end up seeing is that it translates into a clearly seen disease-free or overall survival benefit. This is a challenge that is certainly not limited to cholangiocarcinoma. Cholangiocarcinoma is hard this way because there is an absolute dearth of data. Realistically, we don’t foresee that a randomized trial of neoadjuvant or adjuvant radiation in cholangiocarcinoma is going to be done in the US.

I do think, however, taking a bigger step back and thinking about future directions, that we can rely on the fact that biliary tract cancers are more common in other parts of the world, and other parts of the world, very candidly, have done an absolutely exceptional job of bringing cooperative group-like infrastructure together to achieve clinical trials that we should, but could not, for one reason or another, do in the States. Colleagues in India, at 2 or 3 major centers, have created these cooperative group infrastructures and run trials, several of which have been published or presented or are going to be presented, so everyone stay tuned for that. They are showing radiation leading to an OS benefit for patients with locally advanced gallbladder cancer.⁷ In the US, we could never run that kind of trial, but it is quite impressive to see a modern trial coming out of India showing that there is an OS benefit, and that is what we need to be focused on for our patients.

I do appreciate that we’re looking at pathologic response and things like that, but at the end of the day, our watchword has to be: Are we helping people live longer? Are we helping people live better? If we’re measuring quality of life and OS, that has to be at the center of what we’re thinking about. That certainly speaks to our ongoing randomized trial, HYPERION, which has OS as the primary end point,⁶ and to many other trials around the world that similarly are using nonsurrogate end points, asking very clearly: Are we helping patients with cholangiocarcinoma live longer, live better, and ideally both?

What are the key clinical trial end points and remaining questions that must be addressed to formally establish neoadjuvant radiation within national practice guidelines?

Will we see it in our time? I’m not sure. We’re going to have to rely heavily on those trials that are being done outside the US. There are caveats to those trials because in many of those countries, access to chemoimmunotherapy is more limited. For instance, they’ll have access to gemcitabine and cisplatin, but less so with the addition of durvalumab or pembrolizumab (Keytruda), and that presents a challenge. Will we use neoadjuvant radiation in a randomized trial in some way to help formally establish neoadjuvant radiation? Maybe, maybe not. A little bit of me thinks that neoadjuvant radiation in this setting will be used selectively, and it’ll be very much driven by surgical consideration, and that’s very reasonable.

What we are learning is that for patients who are more in the locally advanced space, radiation itself can meaningfully improve disease control and, seemingly, OS. Can it convert some of those patients to resectability? It seems that it can, or it can help bridge them to transplantation, absolutely, for which there are long-standing data using what has been classically described as the Mayo protocol for bridging radiation to transplantation for hilar tumors. Those are the questions that still linger in our minds. It’s unclear how they’re going to be answered, but it is an exciting time to be treating cholangiocarcinoma, with more molecular targeting and more genomic specificity in how we’re treating patients. That is only going to augment the nuances in both local and systemic therapies, and we’re going to see this very interesting dance play out over the next several years.

The FDA approved lirafugratinib for patients with FGFR2-altered cholangiocarcinoma on September 23, 2026. Can you comment on the impact of this approval in the space?

It’s terrific that we have another option available to us as a product of the phase 1/2 ReFocus trial (NCT04526106) for this new FGFR2 inhibitor, lirafugratinib (Lyrfigtu).^8,9 My colleagues in the medical oncology space continue to do tremendous work to move the needle forward for targeted therapy. We know that FGFR alterations are common, particularly among patients with intrahepatic cholangiocarcinoma, so it’s outstanding. It’s also very hard for radiation oncologists like me, who aren’t medical oncologists, to keep up with the FGFR inhibitors—and how to pronounce them correctly. The more we get these effective targeted therapies, the more we’re going to see patients, quite candidly, live longer and live better.

The other element there is that as systemic control gets better, the importance of local control suddenly comes into focus. That becomes an issue that we need to work on, because when there’s gross residual disease, there’s a high risk of resistant clonogens, and that’s where resistance seems to emerge. You see this stepwise approach, where better systemic therapy translates to the need for better local therapy, and that’s how I think we are moving the needle forward. It’s terrific that we have another approval in the space. Hopefully, it heralds more approvals ahead, not just for FGFR-targeting agents but for others as well. It is an exciting time.

References

1. Ludmir E. Shaping the future: role of neoadjuvant radiation therapy before surgical intervention in cholangiocarcinoma. Presented at the 2026 Cholangiocarcinoma Symposium. September 25, 2026.
2. Ben-Josef E, Guthrie KA, El-Khoueiry AB, et al. SWOG S0809: a phase II intergroup trial of adjuvant capecitabine and gemcitabine followed by radiotherapy and concurrent capecitabine in extrahepatic cholangiocarcinoma and gallbladder carcinoma. J Clin Oncol. 2015;33(24):2617-2622. doi:10.1200/JCO.2014.60.221
3. Tao R, Krishnan S, Bhosale PR, et al. Ablative radiotherapy doses lead to a substantial prolongation of survival in patients with inoperable intrahepatic cholangiocarcinoma: a retrospective dose response analysis. J Clin Oncol. 2016;34(3):219-226. doi:10.1200/JCO.2015.61.3778
4. Oh DY, He AR, Qin S, et al. Durvalumab plus gemcitabine and cisplatin in advanced biliary tract cancer. NEJM Evid. 2022;1(8):EVIDoa2200015. doi:10.1056/EVIDoa2200015
5. Kelley RK, Ueno M, Yoo C, et al. Pembrolizumab in combination with gemcitabine and cisplatin compared with gemcitabine and cisplatin alone for patients with advanced biliary tract cancer (KEYNOTE-966): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2023;401(10391):1853-1865. doi:10.1016/S0140-6736(23)00727-4
6. Koay EJ, Ludmir EB, Lee S, et al. HYPERION CCA: gemcitabine/cisplatin/durvalumab with or without adaptive liver-directed radiotherapy for cholangiocarcinoma. Oncologist. 2026;31(suppl 1):oyag205.060. doi:10.1093/oncolo/oyag205.060
7. Engineer R, et al. A phase III randomized clinical trial evaluating perioperative therapy (neoadjuvant chemotherapy versus chemoradiotherapy) in locally advanced gallbladder cancers (POLCAGB). J Clin Oncol. 2025;43(16 suppl):4007. doi:10.1200/JCO.2025.43.16_suppl.4007
8. FDA approves lirafugratinib for previously treated, unresectable, locally advanced or metastatic cholangiocarcinoma. News release. FDA. September 23, 2026. Accessed September 25, 2026. https://tinyurl.com/4hf56udk
9. Hollebecque A, et al. Efficacy and safety of lirafugratinib in FGFRi-naive cholangiocarcinoma (CCA) patients harboring FGFR2 fusions/rearrangements (FGFR2 f/r). J Clin Oncol. 2026;44(suppl 2):476. doi:10.1200/JCO.2026.44.2_suppl.476