Journal of Hepatology
Volume 47, Issue 4 , Pages 455-459, October 2007

Cholangiocarcinoma: Is transplantation an option? For whom?☆

Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905, USA

published online 28 July 2007.

Article Outline

Abbreviations: CT, computed tomography, DIA, digital imaging analysis, EUS, endoscopic ultrasound, FISH, fluorescent in situ hybridization, FNA, fine needle aspirate, MR, magnetic resonance, PSC, primary sclerosing cholangitis

 

Back to Article Outline

1. Introduction 

Ductular cholangiocarcinoma is a cancer with features of bile duct epithelial cell differentiation. Given this phenotype it is presumed to originate from bile epithelial cells, although it may also originate from periductular glands. It has a unique propensity to arise within the perihilar region of the liver affecting the right and/or left hepatic bile ducts and their union to form the common hepatic duct. Why this cancer affects these bile ducts and this region of the liver remains biologically enigmatic, but is important clinically. It is a desmoplastic cancer which results in bile duct stricturing and obstruction, a cardinal feature of its clinical presentation. Although bile is generally regarded as a noxious substance, this cancer has a tropism for bile and often extends along the bile ducts and resists growing tangentially or radially away from the bile ducts. This tropism likely results from the ability of bile acids to transactivate the epidermal growth factor receptor in this cell type [1]. Once a mass lesion is apparent, the cancer is quite advanced as it has developed self-sustaining growth characteristics and no longer requires the nutrient biliary stream for survival. An understanding of the composite biological features described above is critical and fundamental to its diagnosis and management. First, because of its perihilar location, this cancer encases (but infrequently invades) the portal vein and hepatic artery. Its extension directly into the liver or into the right and left secondary biliary branches with or without vascular encasement frequently precludes hepatic resection [2]. Second, its growth along rather than radially from the bile ducts due to its tropism for bile often fails to produce a mass lesion on cross-sectional imaging studies despite biliary obstruction. Third, the desmoplastic features of this cancer make it difficult to obtain a diagnosis. Endoscopic obtained brushings are often negative and only have a diagnostic sensitivity of 20–40%. Finally, this cancer predominantly recurs regionally which implies persistent micrometastatic nodal disease or disease at the margins of resection despite extended surgical resection. This regional recurrence is frequent and devastating and survival after apparent curative resection is only 20–40% at 5 years [3]. Adjuvant approaches to better control regional disease combined with operation should result in enhanced clinical outcomes.

The majority of cholangiocarcinomas arise in the absence of risk factors [4]. However, chronic bile duct inflammation is a well-recognized risk factor for this neoplasm. In particular, primary sclerosing cholangitis (PSC), an inflammatory disease of large bile ducts, clearly and unquestionably predisposes to the development of this cancer. Up to 10% of patients with cholangiocarcinoma in the United States have PSC [5], [6]. The presence of underlying PSC confounds the diagnosis and management of superimposed cholangiocarcinoma. The differentiation between a benign and malignant stricture in patients with PSC is problematic. Moreover, cholangiocarcinoma with PSC is frequently multicentric. Finally, like hepatocellular carcinoma occurring in cirrhosis, the presence of cholangiocarcinoma in PSC is frequently complicated by chronic parenchymal liver disease, cirrhosis and portal hypertension. It is not surprising therefore, that resection of cholangiocarcinoma in PSC can be considered futile [7].

These observations raise several key questions: (i) how can cholangiocarcinoma be best diagnosed definitively? (ii) If surgical resection is neither technically feasible nor rationale in patients with PSC, what is the role of liver transplantation in this setting? Our goal is to address these two issues in this mini-review.

Back to Article Outline

2. Diagnosis of limited stage cholangiocarcinoma 

If survival following liver transplantation is expected as a viable therapeutic modality, an early diagnosis of cholangiocarcinoma is not only desirable, but is essential for cure. Therefore, early diagnostic criteria for cholangiocarcinoma need to be established and validated. In the absence of PSC, a de novo bile duct stricture involving the main right and left hepatic ducts and/or common hepatic duct is highly suspicious for cholangiocarcinoma. However, in all surgical series, 10–15% of such strictures are inflammatory. The etiopathogenesis of these strictures is unclear, but they unlikely represent early PSC because patients fare well without developing recurrent strictures after surgical resection [8]. A subset of these strictures may represent a biliary variant of autoimmune pancreatitis based on an elevation of serum IgG4 levels and resolution with corticosteroid therapy [9]. Although the presence of a hilar mass lesion with or without vascular encasement by dimensional imaging (MR being superior to CT) virtually assures the diagnosis of cholangiocarcinoma, few other modalities have proven reliable. Unfortunately, neither molecular nor proteomic markers in bile have proven useful in the diagnosis of cholangiocarcinoma. The only serum tumor marker of value is the CA19-9 determination [10]. In the presence of a malignant appearing stricture, a serum CA19-9 >130u/L has a sensitivity and specificity of 79% and 98%, respectively, for cholangiocarcinoma. Although markedly elevated serum CA19-9 values often correlate with overt regional metastases or herald distant metastases, this serum test is unlikely to be of value in screening patients for this neoplasm. Despite a multitude of dazzling advances in molecular medicine and proteomics, we are still dependent upon cytology for the diagnosis of cholangiocarcinoma.

Conventional cytology in the diagnosis of CCA is a difficult “read” and stresses the limits of what even an eminent cytopathologist can achieve with perfect vision and modern optics. Indeed, conventional cytology only has a sensitivity of 20–40% in this disease, although the specificity is perfect [11]. The difficulty with conventional cytology relates to the paucicellular nature of obtaining specimens from a highly desmoplastic cancer. To address these deficiencies, advanced, less subjective cytological methodologies have been developed to diagnose bile duct malignancies. Digital image analysis (DIA) and fluorescent in situ hybridization (FISH) for the detection of aneuploidy, a cardinal feature of cancers with chromosomal instability, have been used in the assessment of potentially malignant biliary tract strictures [11]. In particular, FISH polysomy (two or more chromosomes duplicated) identifies another 20% of patients with cholangiocarcinoma missed by conventional cytology while maintaining a perfect specificity.

An alternative approach for assessing biliary cytology in PSC is to apply a pathologic nomenclature similar to that utilized for mucosal biopsies in patients with ulcerative colitis or Barrett’s esophagus. In this approach, cytology is classified as normal, indefinite for dysplasia, low grade dysplasia and high grade dysplasia/cancer [12]. This approach is attractive because it similarly treats biliary cytology to other pre-neoplastic mucosal diseases of the GI tract. However, histological dysplasia and cytological dysplasia are not equivalent and the latter has not been widely established [13]. In fact, using this approach, approximately 25% of patients with dysplasia on cytology have neither dysplasia nor cancer in the explanted liver [12]. This percentage of false-positive results is likely too high given the high stakes game of liver transplantation.

Back to Article Outline

3. Liver transplantation alone for limited stage cholangiocarcinoma 

Liver transplantation for hepatocellular carcinoma has evolved over the last several decades. Patients undergoing liver transplantation for multicentric cancer, tumors >5cm and those cancers manifesting vascular invasion frequently develop rapidly progressive, lethal, recurrent disease. In contrast, patients with unicentric cancers <5cm or three lesions all less than <3cm, without vascular invasion, are frequently cured by liver transplantation [14]. Although the role of neoadjuvant therapy for hepatocellular carcinoma before liver transplantation remains controversial [15], limited clinical stage hepatocellular carcinoma is now a primary indication for liver transplantation worldwide. This recent era outcome data for HCC have been aided by advances in imaging technology.

Can these same principles of liver transplantation alone, for limited stage hepatocellular carcinoma, be extrapolated and applied to cholangiocarcinoma? The answer is unfortunately no! Two “recent era” publications support this resounding answer. In a collective publication from Scandinavia reporting on liver transplantation for limited stage cholangiocarcinoma in patients with PSC (no mass lesions) survival was only 30% at 5 years [16]. Likewise, in a Canadian study, the finding of cholangiocarcinoma in a liver explant (undiagnosed or incidental cholangiocarcinoma) also resulted in early cancer recurrence and death following liver transplantation [17]. Disease recurrence in these patients is local within the abdomen. Regional lymphatic extrahepatic micrometastases, which are frequent in cholangiocarcinoma but rare in hepatocellular carcinoma, likely account for the discrepant results with liver transplantation between these two primary hepatobiliary malignancies.

Neoadjuvant chemo-irradiation (or chemoradiotherapy) remarkably improves outcomes following liver transplantation for cholangiocarcinoma. Based on the premise that cholangiocarcinoma is predominantly a locoregional disease, oncology and transplant colleagues at the University of Nebraska developed a protocol of high dose intracavitary or brachytherapy plus 5-Fluorouracil (5-FU) followed by liver transplantation for this cancer [18]. Early results were promising. The Mayo Clinic oncology/transplant team refined this protocol by adding external beam radiation to obtain enhanced local control of this cancer pre-liver transplantation. Herein, we will review extensively the entry criteria, the neoadjuvant protocol and outcome data based on the Mayo protocol (Table 1).

Table 1. Entry criteria for liver transplantation in patients with cholangiocarcinoma

•Radial dimension of mass lesion ⩽3cm

•No intra- or extrahepatic metastases

•No prior abdominal radiation therapy

•No transperitoneal biopsy of the tumor

•No prior attempt at resection with violation of the bile ducts

Patients with a cholangiocarcinoma <3cm in radial diameter are eligible for the protocol. Because the extent of disease along the bile duct is virtually impossible to accurately image, there are no longitudinal limits for bile duct involvement in regard to transplant selection. There must be no evidence for intra- or extrahepatic metastases by MRI and CT imaging of the abdomen and endoscopic ultrasound (EUS) with fine needle aspiration of (FNA) regional lymph nodes. In contrast, EUS directed FNA of the primary disease is discouraged; transperitoneal biopsy frequently leads to seeding of the primary disease (personal experience) and in fact has become an exclusion criteria for this protocol. Cholangiocarcinomas often encircle and constrict the afferent hepatic blood vessels (like a boa-constrictor), a phenomenon referred to as encasement. In contrast to hepatocellular carcinoma, true vascular intraluminal invasion by cholangiocarcinoma is rare. Notably this pattern of vascular involvement by cholangiocarcinoma is not a contraindication to transplant. If present in the absence of PSC, such cholangiocarcinomas are generally deemed unresectable. Regardless, all such cancers occurring in the presence of underlying PSC are considered unresectable for the reasons described in detail above.

Following external beam radiation therapy (4500 cGy, two fractions per day×15 days) plus 5-FU via protracted venous infusion, the patients have a 2 week holiday (Table 2). At that time, they undergo brachytherapy via an endoscopic approach [19]. Within the next 2 weeks, the patients undergo an exploratory laparotomy to exclude lymph nodes and peritoneal metastases. Despite current cross-sectional imaging modalities and EUS, approximately 15% of patients will have a positive exploratory procedure, excluding further consideration for liver transplantation. Thereafter, the patients receive capecitabine 2 out of 3 weeks until liver transplantation. The demographics, complications, risk factors for disease recurrence, and clinical outcomes have been published in several serial publications. The outcomes have been extremely favorable, with 5 year survival rates of approximately 70% (Fig. 1) [20], [21], [22]. To date, these outcomes data are the best yet reported for the treatment of this otherwise tenacious locally aggressive and usually fatal neoplasm. These outstanding results have yet to be duplicated raising several interesting and evocative questions provided by the editor of this series.

Table 2. Pre-operative chemo-irradiation prior to and immunosuppression following liver transplantation

•External beam radiation therapy (4500cGy in two fractions per day over 15 days)

•5-Fluorouracil, 225mg/m2/day by protracted venous infusion during external beam radiation therapy

•Brachy therapy with iridium, 2000cGy to a 1cm radius

•Capecitabene after brachy therapy, 1000mg/m2/day in two divided doses/day, 2 out of every 3 weeks until transplant

•Immunosuppression following liver transplantation consists of prednisone, mycophenolic acid and tacrolimus the first four months, and then monotherapy with tacrolimus thereafter

  • View full-size image.
  • Fig. 1. 

    Overall survival following liver transplantation for cholangiocarcinoma. Overall survival is shown for patients (n=78) undergoing liver transplantation for cholangiocarcinoma at Mayo Clinic Rochester, MN from 1994 to October 2006. One year survival was 90% and 5-year survival 74%.

Back to Article Outline

4. When should liver transplantation be performed for cholangiocarcinoma? 

First and foremost, the authors want to affirm an exceedingly conservative approach to an aggressive protocol. Like a legal document, text and phraseology is often open to interpretation. We emphasize that in all instances we have taken a conservative interpretation to our protocol declining participation for patients with criteria in the ‘gray area’. Consequently, it may be difficult to generalize this protocol to all transplant centers. Considerable resources, experience and effort have been provided for this patient population in our Center. Therefore, only Centers with a multidisciplinary team focused in the care of these patients (Hepatologists, Oncologists, Radiation Oncologists, Endoscopists, Transplant Surgeons) may wish to incorporate this protocol into their transplant armamentarium.

Center resources aside, the question persists, when should patients with cholangiocarcinoma undergo liver transplantation? Based on an evidence-based approach, patients with PSC meeting the entry criteria for the Mayo protocol should undergo liver transplantation. There is little rationale for resection of cholangiocarcinoma in the setting of PSC [7]. For patients with unresectable disease meeting the Mayo criteria for transplantation, the approach is simple: their only hope for a cure is liver transplantation! What is more provocative is the patient with resectable disease. Here we encourage dialogue and data-driven discussions. Advocates of resection will note that surgery for this disease continues to improve with the recognition the caudate lobe requires removal in hilar cancers, vascular reconstruction can be performed safely, and the approach of en bloc tumor removal [23]. In this context, resection avoids the arduous process of chemo-irradiation, necessity of identifying a deceased or living liver donor and post-operative immunosuppression. On the other hand, liver transplantation continues to improve in efficacy, and immunosuppression regiments have become less intense. Thus, a legitimate question is whether patients with limited stage disease should preferentially undergo liver transplantation. In an initial attempt to address this question, we examined survival in similar cohorts of patients undergoing liver transplantation and hepatic resection for hilar cholangiocarcinoma [22] and survival was significantly greater in patients undergoing liver transplantation. Given this was a retrospective, post hoc, cohort analysis, there were differences in the patient populations. The patients undergoing liver transplantation were younger, more likely to have PSC, and in the absence of PSC were Bismuth type IV cholangiocarcinomas. In contrast, the patients undergoing resection had Bismuth type I to III cholangiocarcinomas. In addition to differences in selection criteria, age and Bismuth stages, these data have been criticized inasmuch as patients undergoing liver transplantation waited for transplant following neoadjuvant chemo-irradiation. The waiting time has been proposed to select for more indolent disease. However, drop-out on the waiting list for liver transplantation due to disease progression (patients followed with CT scans of the chest and abdomen at 3 month intervals) is less than 5% in the Mayo experience disputing this putative postulate, despite waiting times for cadaveric donors of 9–12 months for blood type A, 12–15 months for blood type O, 6–9 months for blood type B, and 3–6 months for blood type AB. Also, only 30% of our cohort of patients underwent living donor transplantation, which would shorten these waiting times. Although the role of neoadjuvant chemo-irradiation for patients undergoing hepatic resection has not yet been explored, most surgeons are unwilling to address the technical challenges of liver resection, biliary and vascular anastomoses in an irradiated field. Therefore, a randomized trial comparing hepatic resection and liver transplantation is unlikely. The question of whether patients with potentially resectable disease should be considered for liver transplantation is disputable. Recent outcomes for hepatic resection with en bloc portal vein reconstruction suggest significant improvement in survival compared to historical controls with standard hepatic resection and 5-year survival has approached 60% [23]. Nonetheless, the question is legitimate and should not be categorically dismissed.

The question is further confounded by the lack of definitive staging at the time of decision. Despite EUS aspiration of regional hepatic lymph nodes, 10–15% of the patients will be found to have regional lymph node metastases at the time of operative staging which is done after completion of neoadjuvant radiotherapy. The unexpected finding of regional lymph node metastases excludes these patients from liver transplantation. These patients would have been candidates for resection since limited regional lymph node involvement may not preclude success [2]. Unfortunately, neoadjuvant radiotherapy at the Mayo Clinic protocol does often lead to hilar bile duct necrosis which prohibits resection and biliary reconstruction. Many of these patients would also go on to develop liver failure, even if a potentially curative resection were possible. Indeed, we are currently contemplating operative staging prior to neoadjuvant therapy for patients with potentially resectable disease. Those with regional lymph node metastases could be treated with resection and those found to have unresectable disease in the absence of lymph node metastases could then receive neoadjuvant therapy and transplantation. We have transplanted three such patients, but one developed recurrent disease. More experience is necessary to determine whether staging before neoadjuvant therapy increases the risk for tumor dissemination and/or recurrence after transplantation. Patients who meet criteria for either approach should be informed of the two surgical options, in our opinion, and provided necessary guidance to make an individual, informed decision.

In summary, the Mayo protocol has identified a regiment and patient selection criteria which enable our astounding success rate for patients undergoing liver transplantation with cholangiocarcinoma. Based on this center-specific experience, the Organ Procurement Transplant Network will allow regions to prioritize patients with cholangiocarcinoma meeting specific diagnostic criteria and neoadjuvant therapy [24]. Further data will emanate from this experience.

Back to Article Outline

Acknowledgement 

The authors thank Erin Bungum for the excellent secretarial assistance.

Back to Article Outline

References 

  1. Werneburg NW, Yoon JH, Higuchi H, Gores GJ. Bile acids activate EGF receptor via a TGF-alpha-dependent mechanism in human cholangiocyte cell lines. Am J Physiol Gastrointest Liver Physiol. 2003;285:G31–G36
  2. Jarnagin WR. Cholangiocarcinoma of the extrahepatic bile ducts. Semin Surg Oncol. 2000;19:156–176
  3. Rea DJ, Munoz-Juarez M, Farnell MB, Donohue JH, Que FG, Crownhart B, et al. Major hepatic resection for hilar cholangiocarcinoma. Arch Surg. 2004;139:514–525
  4. Malhi H, Gores GJ. Review article: the modern diagnosis and therapy of cholangiocarcinoma. Aliment Pharmacol Ther. 2006;23:1287–1296
  5. Lazaridis KN, Gores GJ. Primary sclerosing cholangitis and cholangiocarcinoma. Semin Liver Dis. 2006;26:42–51
  6. LaRusso NF, Shneider BL, Black D, Gores GJ, James SP, Doo E, et al. Primary sclerosing cholangitis: summary of a workshop. Hepatology. 2006;44:746–764
  7. Rosen CB, Nagorney DM, Wiesner RH, Coffey RJ. LaRusso NF. Cholangiocarcinoma complicating primary sclerosing cholangitis. Ann Surg. 1991;213:21–25
  8. Are C, Gonen M, D’Angelica M, DeMatteo RP, Fong Y, Blumgart LH, et al. Differential diagnosis of proximal biliary obstruction. Surgery. 2006;140:756–763
  9. Kamisawa T, Tu Y, Egawa N, Nakajima H, Tsuruta K, Okamoto A. Involvement of pancreatic and bile ducts in autoimmune pancreatitis. World J Gastroenterol. 2006;12:612–614
  10. Levy C, Lymp J, Angulo P, Gores GJ, Larusso N, Lindor KD. The value of serum CA 19-9 in predicting cholangiocarcinomas in patients with primary sclerosing cholangitis. Dig Dis Sci. 2005;50:1734–1740
  11. Moreno Luna LE, Kipp B, Halling KC, Sebo TJ, Kremers WK, Roberts LR, et al. Advanced cytologic techniques for the detection of malignant pancreatobiliary strictures. Gastroenterology. 2006;131:1064–1072
  12. Boberg KM, Jebsen P, Clausen OP, Foss A, Aabakken L, Schrumpf E. Diagnostic benefit of biliary brush cytology in cholangiocarcinoma in primary sclerosing cholangitis. J Hepatol. 2006;45:568–574
  13. Baskin-Bey ES, Moreno Luna LE, Gores GJ. Diagnosis of cholangiocarcinoma in patients with PSC: a sight on cytology. J Hepatol. 2006;45:476–479
  14. Varela M, Sanchez W, Bruix J, Gores GJ. Hepatocellular carcinoma in the setting of liver transplantation. Liver Transpl. 2006;12:1028–1036
  15. Lesurtel M, Mullhaupt B, Pestalozzi BC, Pfammatter T, Clavien PA. Transarterial chemoembolization as a bridge to liver transplantation for hepatocellular carcinoma: an evidence-based analysis. Am J Transplant. 2006;6:2644–2650
  16. Brandsaeter B, Isoniemi H, Broome U, Olausson M, Backman L, Hansen B, et al. Liver transplantation for primary sclerosing cholangitis; predictors and consequences of hepatobiliary malignancy. J Hepatol. 2004;40:815–822
  17. Ghali P, Marotta PJ, Yoshida EM, Bain VG, Marleau D, Peltekian K, et al. Liver transplantation for incidental cholangiocarcinoma: analysis of the Canadian experience. Liver Transpl. 2005;11:1412–1416
  18. Sudan D, DeRoover A, Chinnakotla S, Fox I, Shaw B, McCashland T, et al. Radiochemotherapy and transplantation allow long-term survival for nonresectable hilar cholangiocarcinoma. Am J Transplant. 2002;2:774–779
  19. Simmons DT, Baron TH, Petersen BT, Gostout CJ, Haddock MG, Gores GJ, et al. A novel endoscopic approach to brachytherapy in the management of Hilar cholangiocarcinoma. Am J Gastroenterol. 2006;101:1792–1796
  20. Heimbach JK, Haddock MG, Alberts SR, Nyberg SL, Ishitani MB, Rosen CB, et al. Transplantation for hilar cholangiocarcinoma. Liver Transpl. 2004;10:S65–S68
  21. Heimbach JK, Gores GJ, Nagorney DM, Rosen CB. Liver transplantation for perihilar cholangiocarcinoma after aggressive neoadjuvant therapy: a new paradigm for liver and biliary malignancies?. Surgery. 2006;140:331–334
  22. Rea DJ, Heimbach JK, Rosen CB, Haddock MG, Alberts SR, Kremers WK, et al. Liver transplantation with neoadjuvant chemoradiation is more effective than resection for hilar cholangiocarcinoma. Ann Surg. 2005;242:451–458discussion 458–461
  23. Neuhaus P, Jonas S, Settmacher U, Thelen A, Benckert C, Lopez-Hanninen E, et al. Surgical management of proximal bile duct cancer: extended right lobe resection increases resectability and radicality. Langenbecks Arch Surg. 2003;388:194–200
  24. Gores GJ, Gish RG, Sudan D, Rosen CB. Model for end-stage liver disease (MELD) exception for cholangiocarcinoma or biliary dysplasia. Liver Transpl. 2006;12:S95–S97

☆ The authors declare that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.

PII: S0168-8278(07)00398-4

doi:10.1016/j.jhep.2007.07.003

Journal of Hepatology
Volume 47, Issue 4 , Pages 455-459, October 2007

Article Tools

* Image Usage & Resolution