Expert Conversations on Cholangiocarcinoma

Management of Side Effects for Patients With Cholangiocarcinoma

Last Updated: Monday, June 27, 2022

Paula Kennedy-Newton, MSN, AGNP-C, and Natasha A. Pinheiro, MSN, AGPCNP-BC, AOCNS, discuss the management of commonly seen side effects among patients with cholangiocarcinoma. They cover side effects seen with chemotherapy, oral agents, immunotherapy, hepatic artery infusion pumps, as well as those associated with cholangiocarcinoma itself.

Meet the faculty

Paula Kennedy-Newton


Duke Cancer Center, GI Medical Oncology

Paula Kennedy-Newton, MSN, AGNP-C, is a Nurse Practitioner of GI Medical Oncology at the Duke Cancer Center in Durham, North Carolina. She specializes in the care and management of patients with hepatobiliary and neuroendocrine malignancies.

Natasha Pinheiro


Memorial Sloan Kettering Cancer Center, Solid Tumor GI Medical Oncology

Natasha A. Pinheiro, MSN, NP, AGPCNP-BC, AOCNS, is a Nurse Practitioner of GI Medical Oncology at Memorial Sloan Kettering Cancer Center. She has worked with upper GI, hepatobiliary, and colorectal malignancies. She has a distinct focus on patients with hepatic arterial infusion pumps and young-onset colorectal cancers. She spent many years as a clinical nurse specialist working on clinical guidelines and nursing educational pathways.

Paula Kennedy-Newton

Natasha, today I'd like to discuss management of the various side effects that patients with cholangiocarcinoma may encounter during their treatment course. To begin, what side effects do you most frequently see associated with chemotherapy, and what are your typical management strategies?

Natasha Pinheiro

In order to answer that, we have to talk about the standard of care for patients with cholangiocarcinoma when we're starting them on systemic chemotherapy. Usually it's a gemcitabine- or a fluorouracil-based regimen. With gemcitabine, it could be gemcitabine plus oxaliplatin, gemcitabine plus cisplatin, or gemcitabine, cisplatin, and albumin-bound paclitaxel.

Gemcitabine-based side effects typically involve a flu-like syndrome, such as low-grade fevers and fatigue on the days around treatment. We ask patients to take their temperature once a day, every day and to give us a call for anything 100.4 or higher. Patients can also use an antipyretic as needed. Nine times out of ten, fevers are self-limiting and specific to the days around treatment.

Fatigue is also very common. For patients who are experiencing fatigue—whether it's from gemcitabine or gemcitabine plus oxaliplatin and/or the cisplatin—we tell them to plan activities around the times that they're feeling well and to plan rest periods.

We ask patients to incrementally increase some sort of physical activity. For someone who's not very active, we usually just say to go out and do a daily walk and to gradually increase the distance. Although it may seem counterintuitive, many studies have shown that a way to manage fatigue is to slowly increase physical activity.1

A lot of the gemcitabine-based treatments can cause myelosuppression. The way we handle this depends on what it looks like when the patient comes back in. At my facility, if patients are symptomatic or if their platelet counts are low, we add romiplostim, an injection to help increase their platelet counts. We give romiplostim on the day of treatment and can continue to do so weekly based on weight. This enables us to not have to hold patients as often for thrombocytopenia. Obviously, we still have a parameter of what we would treat, but we're not just holding indiscriminately for anything under 100,000 per mcL, which we had in the past. We can use 75,000 platelets per mcL and up, which can allow us to keep more patients on schedule.

Platinum-based drugs add a whole other level of side effects, such as neuropathy. Depending on the extent of the neuropathy with the oxaliplatin, we're looking at cold sensitivity, which an expected side effect that is very specific to the days around treatment. We tell patients to limit their exposure to cold—to not drink or eat anything directly from the refrigerator or freezer for the first 4 days post-treatment and to try and limit touching things that are cold. It's a simple enough thing to tell people to do; however, it can be difficult in actuality, depending on where patients live and the time of year. It really can be especially problematic for patients who live in very cold areas.

What's more problematic, and this comes into play with cisplatin also, is the non–cold temperature-based neuropathy. That one is very insidious. Once we start seeing it happening, it's almost annoying to patients how often we ask if they are having any numbness and tingling or increased or decreased sensation. The reason we worry about this is because it can become long term. Whether treatment is adjuvant, neoadjuvant, or for metastatic disease, we are toeing that fine line of continuing a platinum agent where it's effective and monitoring for possible long-term neuropathy. Such long-term neuropathy can limit people's function by causing them to drop things, affect their walking, or affect mobility. A lot of the time if we're seeing that and we know that we have to continue the platinum drug, we will do a dose modification to try and offset the likelihood or the intensity of the neuropathy. If we really do have to push forward, we have the patient see neurology and can start them on a drug such as gabapentin or pregabalin. But the biggest thing in our arsenal, especially for patients who we really have to keep on treatment, is to dose reduce and get them to the point where it's a tolerable dose and is offsetting the extent of the side effects.

Obviously, with many systemic chemotherapy, we’re also seeing nausea, vomiting, and diarrhea. All of those we manage in the way we would manage any of the side effects from any other systemic chemotherapy. We can treat nausea with anti-nausea medications, which can be given during treatment or sent home as a prescription. Focusing on eating foods such as scrambled eggs, as tolerated, and simple carbohydrates, like crackers and toast, can be helpful for patients experiencing nausea or vomiting. Some patients may need the support of IV hydration. Diet modifications may be necessary with diarrhea. Diarrhea can also be managed with antidiarrheals. We also monitor patients’ potassium, calcium, and magnesium levels, which could then exacerbate side effects.

We're looking at labs with each treatment. We're supplementing if we see low potassium, magnesium, or calcium. And the simplest thing is just adding hydration to the treatment because that offsets some of the fatigue, nausea, and issues with diarrhea.

Paula Kennedy-Newton

With cisplatin, because of the concern for nephrotoxicity, the importance of pre- and post-hydration can't be underestimated. Is nephrotoxicity from cisplatin a problem that you frequently encounter?

Natasha Pinheiro

We are extremely diligent in telling patients that they will require hydration with treatment. There are also some patients who then get hydrated on their week off of treatment, or depending on their schedule. We really push the need for drinking at least 2 liters a day, and that isn’t only limited to water since some people find that difficult. For a person who's a big water drinker, it's great. But for the people who aren't, we tell them to try and bring it into their diet through juice, ices, salads, soups, or fruits. Some people have dietary intolerance because of long-term chemotherapy or underlying disease. For those patients, we tell them to do smoothies, shakes, and Ensures because that gives them the calories as well as the fluids. We really try to encourage them to get this amount of fluids on their own at home. But for some people who have a very difficult time, we will then organize having them come in for IV hydration.

Paula Kennedy-Newton

Absolutely. We also worry about the potential for ototoxicity with cisplatin. Do you typically have patients complete a baseline audiological evaluation?

Natasha Pinheiro

Years ago I worked with a physician who sent every single person to audiology prior to cisplatin. Before that and subsequently, I supported another physician who did not do that; they waited until there might have been some sort of sensitivity to send patients. Having seen both ends, I don't think there's any down side to getting a baseline audiology exam. It doesn't add anything really pronounced to the pretreatment regime.

On a similar note, in your practice, do you do 24-hour creatinine clearance on patients prior to starting cisplatin?

Paula Kennedy-Newton

Not routinely. We typically only obtain this if there has been some change in kidney function over the course of treatment and we're trying to get a better understanding of what is going on. We have a low threshold for stopping cisplatin if there's any concern that we're compromising kidney function.

Natasha Pinheiro

That's along the same lines of what we do here. The majority of practitioners here just check serum creatinine with their base-level labs at each treatment, although some opt for 24-hour urine creatinine clearance. We too have a low threshold to send patients to nephrology if we’re seeing elevations.

Paula Kennedy-Newton

With these types of chemotherapy agents, it is always a matter of balancing short-term side effects and the potential for longer-term side effects with the benefit that this treatment is going to be providing in terms of their cancer. I always tell patients that although we want to treat their cancer aggressively, we also have to look at the bigger picture and the rest of their life. We want to ensure that we're not causing any effects that will follow them for the rest of their life and become a problem that they'll have to deal with in the future.

Natasha Pinheiro

If we're not looking at gemcitabine-based regimens, we also use fluorouracil-based regimens with FOLFIRI and FOLFOX, although usually not in the first line. Going from FOLFOX to FOLFIRI, do you find any significant difference in patients’ side effects? In my experience, I think FOLFIRI, or the irinotecan specifically, gives more of the traditional chemotherapy-type side effects, where patients feel like they don't want to eat for a couple of days, they have taste changes, and they feel weak, tired, and nauseous. With oxaliplatin, if we can control the cold sensitivity and the neuropathy to a great extent, it is more tolerable than these traditional chemotherapy side effects. I don't know if that's been the feedback that you've gotten from patients, but I always find a big difference in tolerating FOLFOX and FOLFIRI, regardless of how it's sequenced. Usually the irinotecan is a bit more of a struggle than the oxaliplatin.

Paula Kennedy-Newton

I agree. In addition to the side effects that you mentioned, the risk of diarrhea is significant and also hard to predict. Some patients will have no problem with diarrhea whatsoever, but others will have terrible diarrhea that's very difficult to manage even with all of our usual tools. So I absolutely agree that FOLFIRI can be more difficult to tolerate than FOLFOX. Certainly patients who have had a difficult time with the cold sensitivity and neuropathy may welcome the change to FOLFIRI just to get away from that. But like you said, it can come with a whole other host of more traditional chemotherapy side effects. That really is when we rely on our triage and clinic nurses because they are the ones who hear from the patients when they call in to ask what to do once diarrhea has started. As APPs, making sure that our nurses who take those calls are well educated on the management of these side effects can be tremendously helpful since we just don't know how to predict who will experience which side effect on which regimen.

Natasha Pinheiro

The diarrhea from the irinotecan can come on the day of treatment, the day after treatment, or, more commonly for the vast majority of patients, three or four cycles in when they have totally forgotten that we mentioned it. So it's really important that we reinforce having loperamide available because it could be months before anything appears. It is also very important that we, as APPs, help our nursing colleagues feel aware and confident enough to articulate to patients that once loperamide is no longer working, we can escalate to diphenoxylate/atropine and then tincture of opium. We usually have that conversation with patients at the beginning of treatment—we say that diarrhea is a common side effect, although we don’t know when they might experience it, and this is our escalation plan for if they need it. Often there's an overlap of the agents. For example, we will have patients receive loperamide and diphenoxylate/atropine overlapping, adding in hydration again, to fit in the schedule if their diarrhea is very pronounced. Sometimes we will hold treatment or deescalate treatment to just the fluorouracil alone. This is with regard to oxaliplatin and irinotecan, if disease is stable enough with the thought of adding back the additional agent at the time of disease progression.

Now that we’ve talked about standard of care, let’s move on to more novel therapies. Which patients are you starting on these?

Paula Kennedy-Newton

For patients whose disease has progressed on the standard chemotherapy options and who have been found to have certain genetic alterations via next-generation sequencing, certain novel agents are good possibilities. Some examples that we use frequently are pemigatinib or infigratinib, which are both oral agents that we use for patients who have FGFR fusions or rearrangements.

Patients tend to like the possibility of oral agents because it saves them trips to come in for infusions. However, oral agents still have side effects and challenges. One of those is getting patients to adhere to their oral medications and take them as prescribed.

Oral FGFR-targeted agents are also associated with several GI side effects, such as nausea and vomiting, diarrhea and/or constipation, and decreased appetite.

Oral FGFR-targeted agents also tend to dry out the mucous membranes, so we see a lot of dry mouth and dry eye. There's also the potential for electrolyte derangements. Specifically with this class of drug, we see hyperphosphatemia, which can be a difficult side effect to manage. We frequently start patients on phosphate binders to try to help bring phosphate levels down. We also typically have to ask patients to adhere to a low phosphorous diet, which can be incredibly challenging because phosphorus is present in so many foods. Examples of foods that are high in phosphorus include chicken, turkey, pork, seafood, dairy, nuts, whole grains, beans and lentils; all foods that we would otherwise encourage as part of a healthy, well- balanced diet.

Other significant toxicities that we see with these agents are hand-foot syndrome and nail toxicities. Finally, there's also the possibility for retinal detachment with these agents. We typically like for patients to have a baseline ophthalmology exam before starting, and then we rely on them to let us know the first sign of any visual changes, such as blurry vision, floaters, or double vision. If any of these do occur, we have them see ophthalmology urgently. Therefore, having access to an ophthalmologist (ideally one who is familiar with these types of agents) for a baseline eye exam prior to starting therapy and then having them on standby for any possible serious ocular toxicities is best practice in the management of patients on FGFR-targeted agents.

Natasha Pinheiro

One challenge is that starting an oral agent can give patients a bit more freedom from having to come in. It's very important from a clinician standpoint that we monitor them early on and get them onto a nice cycle of when we see them. So for us, usually the recommendation is that we see them within the first 2 weeks of starting them on the oral agent. A lot of the side effects, specifically nail toxicities and hand-foot syndrome, can be seen early on, so we can try to jump in there very quickly and manage it either with a dose reduction or a dose hold and then resume the drug. Now with telehealth, we have more options for seeing patients and visualizing the side effects.

For the hand-foot syndrome, we recommend that patients get a manicure and a pedicure prior to initiating treatment. We recommend this not for getting their nails colored, but more to get their skin treated. We like for them to have any possible areas of callous buffed down so that they’re starting with a clean pallet. If a patient is prone to callouses, those areas are going to be the first areas that are going to present with these agents. Sometimes patients begin treatment immediately and can't do that, but when the timing allows it, it’s usually one of our recommendations.

Let’s discuss immunotherapies. What are some of the challenges that you see with these?

Paula Kennedy-Newton

As we discussed previously, the addition of durvalumab to our standard gemcitabine–cisplatin combination has been exciting because of the potential for increased efficacy. Patients are typically enthusiastic about this prospect and they also want to understand how an immunotherapy drug is going to work since it is not a typical chemotherapy agent. I explain to them that these agents are essentially creating a powerful immune response in their body that is targeted at their cancer. Unfortunately, when that is occurring, there is also the potential for it to cause inflammation in other parts of the body that we don't necessarily want to happen. Day to day, I find immunotherapy agents to be quite well tolerated, with fairly minimal side effects such as fatigue, joint aches and mild rash or itching. However, there is the potential for rare but more serious side effects related to that immune response that's being generated in the body. We watch for things like pneumonitis, colitis, and hepatitis, which presents itself in terms of elevated liver enzymes. There is also the concern for endocrinopathies such as thyroiditis and the potential for type 1 diabetes.

Natasha, what is your standard approach for managing the various immune-related side effects that we see with checkpoint inhibitors?

Natasha Pinheiro

For endocrinopathies, we're always monitoring patients’ labs. Prior to any treatment, full blood panels should be obtained. When we're using immunotherapy, these also must include liver function tests (LFTs) and thyroid function tests (which include TSH, T3, and T4). Usually that's done with each visit or at least every other visit. At the first sign, if any, of change in a patient’s LFTs, we start them on the appropriate medication, whether it's for hyper- or hypothyroidism. What we tend to see with immunotherapies is hypothyroidism. We have a really good endocrine team here, so a lot of the time we will initiate the medication and then have the patient see the endocrine team. If the patient has a local endocrinologist, we work with them to ensure that this side effect is being managed.

Pneumonitis is a little more challenging to manage. When a patient presents with a new and acute cough that is uncomfortable and sometimes painful or acute onset dyspnea, we get a scan and assess for pneumonitis. If we do see it, we start the patient on a steroid taper.

Colitis can be a bit tricky because you have to rule out that it is not diarrhea caused by a simple GI bug or a reaction to something the patient ate. Once we've ruled out these more benign forms of diarrhea, and we can confirm to some extent that it is an immunotherapy-based colitis. In this case, along the same lines as with the pneumonitis, we would start steroids and, depending on the extent of the diarrhea, either hold the drugs while they're on the steroid or closely monitor the immunotherapy. I want to be clear in that the majority of patients, we stop the immunotherapy and have the patient on the steroid until the colitis is fully resolved. However, sometimes we’re not sure whether the diarrhea is immunotherapy based. For these patients, we start them on the steroid anyway and then once the steroid taper is complete and there is no evidence of clinical symptoms and radiologic absence of concern (if applicable), then we resume the immunotherapy.

The National Comprehensive Cancer Network and UpToDate both have great guidelines for managing immunotherapy-related adverse events.2,3

Paula Kennedy-Newton

It can be challenging to manage the changes in liver enzymes that can occur in these patients. With cholangiocarcinoma, it is not uncommon for patients to have elevations in their liver enzymes during treatment, whether it's due to biliary obstruction or declining liver function. On top of that, immunotherapy agents can also cause significant bumps in AST and ALT due to hepatitis. It can be challenging to delineate whether a bump in LFTs is related to the drug or to the cancer itself.

Natasha Pinheiro

I'm glad you brought that up. For me, this brings our conversation full circle in terms of the importance of the team approach. In this patient population specifically, patients are often referred to us from a hepatologist or a GI specialist. In those cases, it's important to keep the referring physician involved, specifically the hepatologists. As you mentioned, it can be very tricky to delineate whether a symptom is mechanical biliary dysfunction or drug related. That’s where it's important to have that GI specialist or hepatologist to help us parse through what's really happening.

Paula Kennedy-Newton

Could you talk a little about your experience managing the side effects related to the hepatic artery infusion (HAI) pumps?

Natasha Pinheiro

With HAI pumps, we're always looking at patients’ liver function and trying to delineate whether results are from the drug that we're giving directly into the liver pump, a symptom of the underlying disease, or a result of the possible adjunct systemic chemotherapy. We use the pumps to treat with floxuridine (FUDR), a fluorouracil-based drug, which we administer directly into the liver at significantly higher doses than we can give systemically because the liver can regenerate. We're always going to assume that if we see an elevation in a patient’s LFT, it is a result of the drug because assuming it's not and dosing at the same dose can put patients at risk for long-term liver damage.

When we start treating with the HAI pump depends on whether there was a resection prior to pump placement. If there was no resection when the pump was placed, then we can start treating with FUDR within 2 weeks. If there was a resection, just like with systemic chemotherapy, we have to wait for the patient’s body to heal before administering the FUDR. In this case, the patient would first receive two flushes with heparin and normal saline, and then at 4 weeks post-op, they would come in for the FUDR.

We look at patients’ AST, ALT, alkaline phosphatase, and bilirubin at every visit. The schedule is to alternate flushes and FUDR every two weeks, so patients get FUDR once a month. It is done this way so that the liver can recover after getting that such high dose of chemotherapy; the flush helps bathe the liver to aid in recovery. We look at patients’ labs on the first day of treatment with FUDR. If their AST, ALT, alkaline phosphatase, and bilirubin are all within normal limits we will treat them. Those levels are what we will compare to 1 month from now, and it’s how we dose their FUDR. There's a whole formula as to how the dose is calculated that is specifically based on the flow rate of the pump and the patient’s body weight.

When a patient comes in 2 weeks after their initial dose of FUDR for their flush, we expect to see some mild elevation in their LFTs because their liver was just bathed in chemotherapy. We're not going to do anything at that point unless their alkaline phosphatase or their bilirubin are significantly increased, in which case we will add dexamethasone to the flush as an anti-inflammatory. So just keep in mind, we are not always adding dexamethasone to the flush week, but dexamethasone is always in when the patient gets the drug. So they get heparin, normal saline, dexamethasone, and FUDR. On their flush week, if their numbers are within range and normal for them, they just get heparin and normal saline. At any time, if we see an elevation during a flush, we will add the steroid as an anti-inflammatory.

When they come back 2 weeks later, if their numbers have gone back to baseline or are within range of their baseline, we're using the alkaline phosphatase as the parameter for treatment. On day 1 of treatment, we multiply it by 1.2 and then 1.5. If it is under 1.2 above that range, they stay at the dose. If it's between 1.2 and 1.5 upper limits of that previous dosing, then they are dose reduced. If it is outside of that 1.5, then they are held, flushed, and reevaluated in a month and will always be dose reduced because it had gone up to the point of being held.

There's a true algorithm to this. If a local oncology team is comfortable with flushing a hepatic pump, they can do that. However, the majority of local oncologists do not order floxuridine if they don’t have a pump program and are not always comfortable with the algorithm used for dose reducing and dose modifying. Because of this, all of our patients will come to us at least once a month for the FUDR. It is important to make patients aware in advance that it is possible at any point for them to come in and either their numbers will be such that they will have to be held or they will have to be dose reduced. Some patients travel a significant amount of time to get this treatment, only to come in and be told that we can’t give it because of their labs. It is, therefore, important to monitor expectations and tell them that this is a possibility from the outset.

Any time people come here to talk about the placement of the hepatic pump, we tell them about the logistics of managing it. We make sure they understand that they will have to commit to going to a pump center at least once a month to manage the FUDR versus coming here every 2 weeks versus not placing it. Because there is a lot of travel involved, it is important for patients to be able to look at logistical and financial concerns in advance of pump placement. We have some patients who fly here every month or every 2 weeks. These are not always wealthy people, but they make it happen. All they need is one mishap in their life or one day of bad weather to throw it all off.

The other thing that's important is what sits in the pump and how long. We try to tell people that they have to adhere to the schedule, but when it's a flush in the pump, we have some leeway by a couple of days or even up to a week. However, when there's FUDR in the pump, we do not like that to go over even a day because at that point it risks causing a patient’s numbers to become even more elevated.


  1. Mustian KM, Alfano CM, Heckler C, et al. Comparison of Pharmaceutical, Psychological, and Exercise Treatments for Cancer-Related Fatigue: A Meta-Analysis. JAMA Oncol. 2017;3(7):961-968.
  2. NCCN Guidelines for Management of Immunotherapy-Related Adverse Events. Accessed June 25, 2022.
  3. Postow M. UpToDate. Toxicities associated with checkpoint inhibitor immunotherapy. Accessed June 25, 2022.