Liver-Directed Therapy for Advanced Intrahepatic Cholangiocarcinoma

Liver-Directed Therapy for Advanced Intrahepatic Cholangiocarcinoma

Dr. Flavio G. Rocha

By Flavio G. Rocha, MD, FACS

Article Highlights

  • Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver cancer after hepatocellular carcinoma (HCC) with a rising incidence, and can be a challenging disease to manage.
  • Although ICC is typically not as hypervascular as HCC, the peripheral enhancement seen in ICC can be treated with TACE or TARE.
  • Phase II trials on hepatic artery infusion pump have demonstrated significant response rates in ICC.
  • For lesions not amenable to transarterial therapy external beam radiation can be an alternative local therapy for unresectable ICC.

Cholangiocarcinoma is a rare cancer of the gastrointestinal tract that can arise anywhere along the biliary tree from the liver to the pancreas. It is typically separated into extrahepatic (50% to 70%) or intrahepatic (20% to 30%) depending on the site of origin. Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver cancer after hepatocellular carcinoma (HCC) with a rising incidence.

ICC can be difficult to manage because of several factors including late diagnosis, early metastatic spread, and potential for liver dysfunction. Surgical resection with negative margins offers the only chance for long-term survival and typically involves a major hepatectomy with or without bile duct resection; however, the minority of patients are eligible. Despite curative intent surgery, both local and distant recurrence is common with as many as 60% of patients recurring at a median of 2 years following resection leading to a 5-year survival rate between 20% and 40%. 1 Adverse prognostic factors for recurrence and survival include margin status, tumor size, vascular invasion, positive lymph, and multifocal tumors. ICC can develop in the background of hepatitis C and cirrhosis where surgery may be contraindicated from a liver function perspective as opposed to tumor extent. Lastly, cancer progression in the liver leading to synthetic dysfunction, vascular occlusion, or biliary obstruction can often be the ultimate cause of death regardless of metastatic burden. Therefore, tumor control can be an effective method of prolonging survival when curative resection is not possible, regardless of stage.

Although the current standard of care for either unresectable or metastatic ICC is systemic gemcitabine and cisplatin,2 there has been recent enthusiasm for the application of locoregional therapy to the liver either as primary treatment or in combination with chemotherapy. This rationale was extrapolated from the HCC literature where liver-directed therapy has a clearly established role supported by randomized trial data. There are several strategies ranging from intra-arterial therapy to deliver chemotherapy via hepatic artery pump, drug-eluting beads, embolization particles, and selective internal radiation therapy. In addition, external beam radiation has been increasingly used in the liver. Lastly, ablative techniques using heat in the form of radiofrequency or microwaves have been used to treat ICC. Although there have not been any prospective studies to directly compare these different modalities, they have been used either alone or in combination with systemic chemotherapy in advanced ICC. The purpose of this review is to summarize the evidence behind their use with a focus on their indications and mechanism.

Intra-arterial Therapy

Transarterial chemoembolization

The rationale for arterial-based therapy arose from the observation that both primary and metastatic liver tumors derive their blood supply predominantly from the hepatic artery, whereas the portal vein preferentially provides trophic perfusion to normal hepatocytes. Since the first randomized trials in the early 2000s demonstrated a survival benefit for transcatheter arterial chemoembolization over supportive care in HCC,3 it has become the primary treatment modality for Barcelona Clinic Liver Cancer intermediate-stage tumors, as well as a downstaging strategy for resection and a bridge to transplant.

ICC is typically not as hypervascular as HCC, however, the peripheral enhancement seen in ICC is still amenable to transarterial therapy. Transcatheter arterial chemoembolization is the most common liver-directed therapy for ICC and comes in two forms. Conventional transcatheter arterial chemoembolization is comprised of an emulsified chemotherapeutic agent that is injected into the hepatic artery via a catheter placed in the femoral artery followed by gelatin, polyvinyl alcohol particles, or microspheres. More recently, drug-eluting beads have been developed to provide more sustained release of chemotherapeutic agents in the particles lodged inside the tumor. The experience with transcatheter arterial chemoembolization for ICC has included several chemotherapeutic agents including gemcitabine, mitomycin C, cisplatin, oxaliplatin, and doxorubicin. Irinotecan- or doxorubicin-loaded beads have been administered in multicenter prospective study of patients with ICC who were unresectable and who experienced a median survival of 17.5 months.4 However, drug-eluting beads with transcatheter arterial chemoembolization have been associated with higher morbidity with higher toxicity and worse patient-reported post-procedure symptoms than conventional transcatheter arterial chemoembolization. A recent randomized trial of hepatic artery-directed drug-eluting beads with irinotecan (DEBIRI) in combination with systemic FOLFOX and bevacizumab for unresectable colorectal liver-only metastases was completed. Compared to the control arm with systemic therapy alone, patients who received DEBIRI experienced higher response rates (50% vs. 24%) and longer hepatic progression-free survival (PFS; 17 vs. 12 months) leading to a higher rate of conversion to resection (35% vs. 6%).5  In addition, DEBIRI was not associated with an increased rate of adverse events. DEBIRI is being evaluated in a similar phase II trial, which is currently enrolling patients, in combination with gemcitabine and cisplatin or carboplatin in unresectable metastatic ICC (NCT01648023). In the meantime, given the heterogeneity in treatments and lack of adequate prospective studies, it is difficult to recommend a standard of care transcatheter arterial chemoembolization regimen for ICC either alone or in combination with systemic chemotherapy.

Hepatic artery infusional chemotherapy

Infusion of chemotherapy alone through hepatic arterial therapy for unresectable or metastatic ICC has also been investigated. A phase II trial of 30 patients with advanced or metastatic biliary tumors using intra-arterial epirubicin and cisplatin in combination with systemic 5-fluorouracil demonstrated a partial response (PR) in 11 patients (36%) and stable disease in 12 (40%) patients, resulting in a median PFS of 7.1 months and overall survival (OS) of 13.2 months.6 This was followed by another phase II trial replacing the systemic component with capecitabine in 20 patients with advanced biliary tract cancer (12 with ICC) that had a similar PR of 31% and stable disease in 47% with a higher PFS and OS of 11.6 and 18 months, respectively.7

Regional chemotherapy to the liver has also been delivered via hepatic artery infusion pump (Fig. 1). This has been an effective modality in the treatment of colorectal liver metastases with high response rates in the unresectable and adjuvant settings. Floxuridine has been the most common agent infused given its superior hepatic first-pass metabolic profile and little systemic toxicity when combined with dexamethasone. In a phase II trial of 34 patients with unresectable primary liver tumors (26 patients with ICC and 8 patients with HCC), hepatic artery infusion pump of floxuridine without concomitant systemic therapy resulted in a disease control rate (PR plus stable disease) of 88% with a better response seen in the ICC group.8 Median disease-specific survival was 29.5 months with a hepatic PFS of 11.6 months in patients with ICC compared to 9.4 months in patients with HCC.

A follow-up study adding systemic bevacizumab to hepatic artery infusion pump of floxuridine enrolled an additional 22 patients (18 patients with ICC and 4 patients with HCC). Overall median survival was 31.1 months with a hepatic PFS of 11.2 months.9 All 18 patients with ICC had either a PR (38%) or stable disease (61%), and patients with HCC all had stable disease. However, the trial was stopped early due to biliary toxicity, a phenomenon that was also seen when bevacizumab was added to hepatic artery infusion pump with floxuridine in patients with colorectal liver metastases. The main drawback of this technology is the need for a laparotomy to implant the pump, but this can now be accomplished minimally invasively. In addition, the need for meticulous maintenance of the hepatic artery infusion pump to prevent vasculobiliary toxicity has limited this strategy to a few specialized centers. Despite those caveats, follow-up studies combining hepatic artery infusion pump and systemic chemotherapy are currently recruiting patients with unresectable ICC (NCT01525069 and NCT01862315).

Transarterial radioembolization

Selective internal radiation therapy with Yttrium-90 (Y-90) is the most common mechanism of transarterial radioembolization. Y-90 is a beta-emitting isotope that is loaded into either glass or resin microspheres and delivered by hepatic artery infusion by interventional radiologists. Unlike transcatheter arterial chemoembolization, selective internal radiation therapy is nonocclusive as its antitumoral effects rely on the availability of oxygen to provide the free radicals to destroy the cancer cells with a tissue penetration up to 2.5 mm from (Fig. 2). Therefore, this modality can be used in patients who may have ipsilateral portal vein thrombosis from either bland clot or tumor invasion.

The largest retrospective study of Y-90 for 46 patients with unresectable ICC reported an OS of 15.6 months for those with mass-forming tumors versus 6.1 months with periductal infiltrating lesions.10 Another proposed indication of selective internal radiation therapy is the potential for radiation “segmentectomy” or “lobectomy,” whereby higher doses of Y-90 are used to treat large, unresectable tumors. This can be associated with an increased rate of tumor necrosis and lead to contralateral increase of the future liver remnant similar to that seen with portal vein embolization. In select cases, this has allowed for conversion to resectability, although hepatectomy following selective internal radiation therapy can be quite challenging, secondary to liver fibrosis and scarring. This observation, coupled with the suggestion of prolonged hepatic PFS in the SIRFLOX trial (FOLFOX + bevacizumab + selective internal radiation therapy) for unresectable colorectal liver metastases, have propagated transarterial radioembolization as a potential first-line therapy option for unresectable liver tumors. 11 Recently, Salem et al also demonstrated a longer time to progression of 26 months for transarterial radioembolization compared to 6.8 months for transarterial radioembolization in patients with Barcelona Clinic Liver Cancer stage A or B HCC.12 To date, there are no randomized trials comparing transcatheter arterial chemoembolization to transarterial radioembolization in ICC, and the sequencing with systemic chemotherapy still needs to be determined.

Local Therapy

External beam radiation

Given the limitations of decreased hypervascularity, large size, and anatomic distribution of ICC precluding effective transarterial embolization, external beam radiation has been proposed as an alternative local therapy for unresectable ICC.

Stereotactic body radiation therapy and intensity-modulated radiotherapy techniques and doses had previously been established to treat small liver tumors. More recently, doses up to 100 Gy (median 58.05 Gy) with ablative effects have been delivered to larger tumors (median size 8 cm) with effective tumor control in inoperable ICC resulting in a median OS of 30 months.13 Proton therapy has also been used with a similar efficacy of 2-year local control rate of 94% and 2-year OS of 46% (Fig. 3).14 The majority of patients received systemic chemotherapy and most progressions occurred at extrahepatic sites. An ongoing multicenter phase III trial to evaluate this high-dose conformal radiation strategy after induction chemotherapy is underway (NCT02200042).


Thermal ablation with either radiofrequency or microwaves has a definitive role in the curative treatment of HCC supported by several randomized trials. It is typically reserved for lesions that are under 3 cm located away from major vascular or biliary structures such as the hepatic vein confluence or portal bifurcation. Ablation can be performed percutaneously, laparoscopically, or at the time of laparotomy in patients who may have inconveniently located tumors that would require a major resection precluding an adequate liver remnant, multiple tumors, or in those patients with poor liver function who could not tolerate surgery. Although the experience is limited for ICC compared to HCC, the same principles are thought to be applicable. The existing literature is comprised of mostly small, retrospective series. However, a systematic review of radiofrequency for ICC pooling results of 7 studies with 84 patients suggested 1-year, 3-year, and 5-year survival rates of 82%, 47%, and 24%, respectively.15 More recently, the application of irreversible electroporation may avoid the limitations of radiofrequency and microwaves by allowing for nonthermal destruction of tumors near critical vascular structures.


In summary, the experience learned from the application of locoregional therapy to the liver for HCC and metastatic colon cancer combined with the development of new technologies has led to an expansion of the therapeutic armamentarium for ICC. Further evaluation in the form of rigorous clinical trials, though, is needed to integrate these components with the established modalities of surgery and systemic chemotherapy to continue to improve outcomes for this rare but fatal disease.  

About the Author: Dr. Rocha is a general, thoracic, and vascular surgeon at the Virginia Mason Medical Center.