Colorectal Cancer Podcast

When Is Liver-Directed Therapy for Colorectal Cancer a Good Option?

Benjamin L. Schlechter, MD; Ammar Sarwar, MD


September 06, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.

Benjamin Schlechter, MD: Hello. I'm Dr Benjamin Schlechter. Welcome to Medscape's InDiscussion series on colorectal cancer. Today we're discussing liver-directed therapy for colorectal cancer with Dr Ammar Sarwar. Dr Sarwar is an associate professor of radiology at Harvard Medical School and an interventional radiologist at Beth Israel Deaconess Medical Center in Boston.

He's also one of the clinical masters whom many of us in the Boston community turn to for help in thinking about interventional radiology interventions, not just for his willingness to do these procedures for patients, but also his willingness to say no and give us a balanced perspective on the path forward. Dr Sarwar, I like to start these conversations with a brief intro about our guest.

Can you tell me how you ended up as an interventional radiologist focused on cancer care?

Ammar Sarwar, MD: Thank you for inviting me to speak on this podcast, it's certainly a topic that I'm very passionate about. I trained in interventional radiology at a liver cancer center, which was also a liver transplant center. We did a high volume of liver-directed therapies, primarily for hepatocellular carcinoma.

As the field has evolved since I first started practice nearly 10 years ago, I identify now more as an interventional oncologist than an interventional radiologist. This changed as we started to see the benefits that some of our therapies — for example, ablation being a curative therapy — provided to our patients and started following patients longitudinally. We also developed synergies with our medical oncologists and our surgeons to try to get as many patients as we can to curative therapies. I think that's the direction that I'd like to continue seeing this field move forward toward.

Schlechter: I'm always happy to hear someone say "curative therapy." That's the goal, after all. Our main topic for today is liver-directed therapy for metastatic colorectal cancer beyond systemic therapy. I think any discussion on this topic requires a bit of an anatomy refresher. Can you walk us through some of the hepatic anatomy and how that influences your perspective as an interventional radiologist and interventional oncologist, and highlight some of the features that are important for therapeutic interventions?

Sarwar: The way that I think about the liver is primarily as segments. The reason why I'm looking at the liver in segments is to identify where the tumor is. Is it close to central hilar structures? Is it farther away? Is it closer to the capsule? How many segments are involved? That often influences the potential of that patient to undergo surgery. Finally, it helps answer the question, How selective can I be in treating this patient?

I believe the way forward for liver-directed therapies is to offer a hepatocyte-sparing approach. We want to maximize damage to the tumor but minimize damage to the normal liver. That's something that I think we can do with ablation, with chemoembolization, and with radioembolization as we offer more selective treatments.

When I'm thinking about the liver anatomy, step number one is asking, What are the segments that are involved? I think step number two, especially when it comes to transarterial therapies, is asking, What is the hepatic arterial anatomy? The hepatic arterial anatomy is very interesting.

We do about five to six hepatic arterial infusions of chemoembolizations every week. Every week, I'll see a type of anatomy that I haven't seen before. Nearly 30%-40% of our patients will have variant anatomy, and that is what keeps us on our toes when we're doing these procedures.

Schlechter: This is an important topic, right? If 80% of colorectal cancer patients have liver metastases, this is a dominant problem for these patients. Let's talk about Y90 radioembolization. This has made a lot of advances in colorectal cancer. We also have some contemporary prospective randomized trials.

The SIRFLOX study published in the Journal of Clinical Oncology by Guy van Hazel and the combined FOXFIRE trials in The Lancet by Dr Wasan in 2017. Can you give us your thoughts on these data and how they influence your practice? Both as the first-line trials, and then we'll talk about the second line trials as well.

Sarwar: I think the first-line trials were important data, as you and most of our listeners know. The data showed that there was no benefit in overall survival or progression-free survival in first line Y90 combined with systemic therapy compared to systemic therapy alone. There was some benefit in hepatic progression-free survival.

The way that I interpret that is we learned radioembolizations killed tumors inside the liver. However, the benefit wasn't sufficient to translate into a survival benefit. There are several reasons for this. First of all, just broadly as a whole, medical device trials and interventional radiology trials have a very high failure rate for the first few trials.

There are several reasons for that. Interventional radiologists typically have not been heavily involved in trial design. As an example, when we look at the protocols for the FOXFIRE and SIRFLOX trials, there's extremely detailed information on dosages for the systemic therapy drugs — how much are you giving per meter squared, how do you individualize therapy for patients? When we look at the Y90, the trial basically says, "Y90 was given," as if that's a black-or-white thing, a 0 or 1 Boolean flag. It's not black and white, and we've certainly learned this since SIRFLOX and FOXFIRE. Radioembolization is less about the delivery of radiation and much more about dose.

We've made strides in the last 10 or 15 years in terms of dosimetry — maximizing dose to the tumor and minimizing dose to the normal liver. As an example, most of the therapies that were provided in the SIRFLOX and FOXFIRE trials were non-selective whole liver or lobar therapies, which were affecting not only the normal liver but also minimizing the radiation effect on tumors because you're distributing the same amount of radiation over a large area.

In most recent years, and in some of the newer trials, we start to find that if you offer more selective therapies and maximize the dose of the tumor, the effect is better. You start to see more beneficial effect in outcomes like progression-free survival.

Schlechter: That's helpful. That's such an important lesson about trial design in oncology in general. Good ideas don't make good trials and good oncology. More is not better. Better is better. This has come up a lot in this series. Let's talk about second-line data because I think that's where we've seen some more positive results.

The EPOCH study in the Journal of Clinical Oncology by Mary Mulcahy in 2021 — I'd like to hear your thoughts on this. More broadly, where do you think we should be putting Y90 in the care of colorectal cancer patients?

Sarwar: I think the EPOCH study was a powerful step forward. As you said, it all depends on the trial design. It was performed by investigators who were experienced in radioembolization at centers that had been doing radioembolization for more than 10 years. I think that's important because there's a learning curve to these transarterial localized therapies that doesn't exist for most pharmaceuticals.

Here you have experienced investigators selecting the appropriate blood vessels to treat and providing a much higher dose of radiation than if you had done non-selective therapies. I think that's one key aspect. I think the second key aspect is that this was done in second-line patients. I think that's important because, as you get toward the third line and the fourth line, the options in systemic therapy start to decline or become less appealing to patients and oncologists as well.

All these trials were done over a period of 6 to 7 years, and despite the protocol being fixed at the beginning of the trial, during that time, clinical practice changed. So even the data for EPOCH in 2023, and EPOCH that came out in 2021, the data was collected from 2008-2018 or so. The data is already old because practice has moved past what was done at that time. This is a challenge that we often face in interventional trials.

A great example of this is the stroke thrombectomy trials. The first few trials were negative, and then as the devices improved, as our ability to use the devices improved, as our selection improved, the trial started to turn positive, and now we have a great therapy. I think EPOCH is a great trial. It has certainly shown us the door in terms of how to design a trial, how to select the appropriate cohort, and how to make sure that the therapy is provided consistently between centers, especially as it relates to the technique's refinements for local-regional therapy.

Schlechter: I think it's a really helpful point. Things have moved forward. We've also learned a lot about liver vs non-hepatic disease, which we talked about in an earlier episode on the role of immune therapy with Dr Fakih.

When you're seeing a patient with hepatic disease, they've received FOLFOX, maybe they're on FOLFIRI. Where do you fit this in? Who is the right patient for this? How much extra-hepatic disease is too much? What are the lab parameters we should be looking at?

Sarwar: If I see a patient who's been referred — let's say early in their disease state, if it's not first line, maybe second line or third line — I think the key differentiator for me is really the performance status. I think performance status is number one, two, and three.

In terms of hepatic arterial therapy, let's start with ablation. Ablation will be my first choice if you have less than three tumors in the liver, and they're located in segments that can be reached with an ablation needle. For most segments of the liver, that's true.

And the reason for that is obviously the data, both for hepatocellular carcinoma as well as colorectal cancer, that shows that this is extremely similar to surgical resection but fewer complications, less morbidity, and quicker recovery times. In some instances, the patient has tumors larger than 3 cm or they're multifocal or the patient is not a candidate for ablation. If they have a normal performance status, good liver function, good renal function, and no major comorbidities, the next question is, can I potentially downstage this patient for surgical resection?

One of the things that I feel bad about, again, is that not enough patients with metastatic colorectal cancer and liver-only disease are getting a resection. The first thought that I have is how can I help my surgical oncologist downstage this patient?

There's a variety of ways that I could do that. For example, if there's bilobar disease, but the bulk of disease is in the right lobe, and there's a single tumor in the left, could I ablate the left lobe and have the patient get a hepatic resection? The second option is if the patient has predominantly right lobe disease but the future liver remnant is small, could I do techniques like portal vein embolization or hepatic venous deprivation in order to hypertrophy the future liver remnant and allow the surgical oncologist to do a curative resection?

I think resection really helps patients. Now let's say the patient is not a candidate for downstaging. The patient has received one or two lines of systemic therapy, their performance status is good. I'm going to look at their bilirubin. I think just as a rule of thumb, a bilirubin of less than 2 gives me some confidence that I can treat this patient easily without causing any damage to the liver or risk of hepatic failure.

Then I'm starting to look at my hepatic arterial options. I think Y90 radioembolization is more likely to cause tumor necrosis, but it also needs to be delivered safely in a hepatocyte-sparing approach. If that's not an option — let's say the bilirubin's a little bit higher than 2 — then I start looking at chemoembolization. Drug-eluting beads loaded with irinotecan have certainly shown benefit in randomized trials, both in terms of [overall] survival and progression-free survival. Although that data is old at this point.

Schlechter: Excellent. What's the recovery time for patients undergoing these procedures, such as radioembolization, chemoembolization, or ablation? How much is outpatient, and how much is inpatient? This is important, especially as we think about resources.

Sarwar: Radioembolization has always been outpatient. It's two procedures: a planning procedure, which is just a diagnostic angiogram, and then a treatment procedure. Both of these are outpatient procedures 100% of the time. Almost 10-15 years ago, chemoembolization was a post-procedural admit for a day. Now, with the use of steroid regimens to ward off post-embolization syndrome and good pain medications to help with the right upper quadrant pain, all of our patients who underwent chemoembolizations in the last 3 years were done as outpatients. We looked at readmission rates and early admit rates, and it's less than 10% of patients who will need admission or readmission because of pain control or other things. In general, these therapies are really well tolerated.

There is randomized data that supports the use of a single dose of intravenous steroids at the time of the procedure. That helps with post-embolization syndrome, which is mostly related to liver inflammation. The textbook says, for example, after radioembolization or after chemoembolization, most patients will have post-embolization syndrome for up to a week. The range is 30%-80% of patients. Post-embolization syndrome occurs because of cytokine release when tumor necrosis happens and is indicated by a flu-like syndrome right upper quadrant pain, low-grade fevers, loss of appetite, and some fatigue. The textbook says 7 days. Most of my patients, because we're treating them in a selective fashion, will have it for 1 or 2 days, and then they're back to doing their activities of daily living.

Schlechter: Excellent. You talked about these interventions as a technique to downstage patients for the purpose of surgery. I think that's a fantastic point that also brings us to one of the really interesting interventions that's been pursued in colorectal cancer, going back to the 1970s, and that's placement of hepatic artery infusion pumps for the purpose of giving floxuridine (FUDR).

As opposed to the percutaneous approaches and the intravascular approaches, this is a surgical approach. It certainly would involve admissions. I want to talk about this and focus on things where we have the most mature and modern data with prospective randomized trials that are broadly available for patients and for providers at many centers. FUDR via hepatic artery infusion pumps is a little different because it's done in specialized centers.

As a reminder for our listeners, FUDR is a fluoropyrimidine like 5- fluorouracil (5-FU), and it's subject to first-pass metabolism. When patients receive it, it's quickly cleared by the liver. The idea is that if you can infuse this into hepatic artery, you can achieve high concentrations in the tumor and then quickly clear it in the liver with a reduced systemic exposure to FUDR.

And this led to tremendous interest in the 1970s and 1980s to placing hepatic artery infusion pumps and delivering FUDR. A lot of work was done to optimize how to give this in doses. Compared to what I would call oncologically ancient approaches to chemotherapy, older form of pyrimidine regimens, this seemed to be a promising regimen that improved resectability with response rates probably around 40%. When we compare that to a modern regimen like FOLFOXIRI or even FOLFOX, it's probably not as impressive, although prospective data really is lacking compared to modern regimens.

Nevertheless, this is widely used throughout the United States to downstage patients in this part of the surgical approach. I'd like to hear your thoughts on FUDR and hepatic infusion pumps. When should we reach for this? How do you think about this in the context of liver-directed therapy?

Sarwar: Sure. Full disclosure: I don't do these procedures. I think we have to look at the different paths that surgically placed intrahepatic arterial pumps took compared to percutaneous transarterial therapies. The liver has been a challenging place for systemic therapy to reach. Everybody's always been interested in treating liver tumors more effectively.

What we learned from animal studies soon after the early 1970s and 1980s when transarterial chemoembolization (TACE) was introduced, was that transarterial chemotherapy was not very effective, even though it was providing the chemotherapy directly to the tumor. The main reason for that, which I think is the main difference between TACE and hepatic arterial infusion with FUDR, is that even though you are infusing the liver with the chemotherapy, it doesn't stay there very long.

It washes out over time, and this is one of the reasons why for FUDR, your infusion times are typically about 24 hours and 48 hours, and you have to keep giving those infusions every 2 weeks for several cycles. Whereas with chemoembolization and with radioembolization, we give the therapy once. When we give that therapy once, we are combining it with an embolic agent.

In the case of chemoembolization, it was originally lipiodol. For colorectal metastases, we use drug-eluting beads. These beads get stuck in the arterials, and they have the drug attached to them. They basically diffuse the drug over the next several weeks. You don't need to have a pump implanted in the patients.

The same is true with radioembolization. It's a brachytherapy approach. The beads get lodged into the arterioles next to the tumor, and they attack the tumor with radiation over the next several weeks until the half-life causes a decay, and the radiation goes down to baseline. I think that's the main reason why, first of all, when you combine the transarterial infusion of chemotherapy with embolization, you're able to give more drug directly to the tumor. There's less systemic seepage that happens, and there are animal studies that prove that. This not only improves the safety of the mechanism by which TACE occurs compared to hepatic arterial infusion but also the efficacy.

I look forward to a time when these two therapies start to merge. We're always challenged by funding for these novel ideas. I would love to see a therapy in which even if you placed a pump and you're giving the hepatic arterial infusion, that you combine it with an embolic so that you can use a smaller dose of chemotherapy, and it stays in the tumor longer.

I look forward to having some kind of TACE approach with FUDR, because right now we're only using irinotecan when it comes to colorectal metastases.

Schlechter: That's a great thought that I hadn't thought of before. As we think about trial design for FUDR, one of the challenges is the lack of modern prospective data, because I don't think we can look at publications from 1989 or 1993 and say that this stands up to modern rigor. At the same time, you hear reports of patients who've done very well. Of course, we hear reports of patients who haven't. It would be an interesting way to randomize patients. You could randomize by who would receive standard FUDR reports plus or minus an embolization agent to try and improve penetration and reduce systemic exposure.

That would be, I think, appealing to the surgeons who are placing in the pumps and feel committed to them, and the patients are committed to the idea of pursuing it. It would also allow us to look at the risk-benefit ratio for patients. I love that idea. Today we've been hearing from Dr Ammar Sarwar discussing liver-directed therapy for colorectal cancer. We had a refresher on hepatic anatomy, how we should think about segments, the arterial infusion of drugs into the liver, and how we can use embolic agents to slow down transfer of these drugs through the liver and improve anticancer effect as opposed to systemic effect and quick washout. We talked about Y90 radioembolization and how it really is a preferred agent in the later lines of therapy, and it doesn't really improve outcomes in the first line. Maybe that's a biological difference. Maybe that's just because we've gotten better at it.

The later studies had better results. We talked about ablative techniques and using percutaneous needles to treat small metastases, which can have excellent outcomes, both as part of surgical management and palliative management. We also talked about FUDR and infusions via hepatic artery and some of the limitations of that but also some potential innovations going forward.

Thank you for joining us. This is Dr Benjamin Schlechter for InDiscussion.


Colon Cancer

Locoregional Liver-directed Therapies to Treat Unresectable Colorectal Liver Metastases: A Review

History and Evolution of Yttrium-90 Radioembolization for Hepatocellular Carcinoma

SIRFLOX: Randomized Phase III Trial Comparing First-line mFOLFOX6 (Plus or Minus Bevacizumab) Versus mFOLFOX6 (Plus or Minus Bevacizumab) Plus Selective Internal Radiation Therapy in Patients With Metastatic Colorectal Cancer

First-line Selective Internal Radiotherapy Plus Chemotherapy Versus Chemotherapy Alone in Patients With Liver Metastases From Colorectal Cancer (FOXFIRE, SIRFLOX, and FOXFIRE-Global): A Combined Analysis of Three Multicentre, Randomised, Phase 3 Trials

Radioembolization With Chemotherapy for Colorectal Liver Metastases: A Randomized, Open-label, International, Multicenter, Phase III Trial

FOLFOX Regimen

FOLFIRI (Folinic Acid, Fluorouracil, and Irinotecan) Increases not Efficacy but Toxicity Compared With Single-agent Irinotecan as a Second-line Treatment in Metastatic Colorectal Cancer Patients: A Randomized Clinical Trial

Hepatic Chemoembolization

Post-embolization Syndrome as an Early Predictor of Overall Survival After Transarterial Chemoembolization for Hepatocellular Carcinoma

When Should You Consider Immunotherapy in Colorectal Cancer

FOLFOXIRI Reintroduction in Metastatic Colorectal Cancer

A Review of Applying Transarterial Chemoembolization (TACE) Method for Management of Hepatocellular Carcinoma

Lipiodol Transarterial Chemoembolization for Hepatocellular Carcinoma: A Systematic Review of Efficacy and Safety Data

Follow Medscape on Facebook, X (formerly known as Twitter), Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.