Model for end-stage liver disease (MELD) exception for cholangiocarcinoma or biliary dysplasia


Cholangiocarcinoma (CCA) is a devastating disease with few treatment options. The incidence of CCA in the United States is on the rise and is currently estimated at 3,000 to 4,000 cases per year, with the incidence being especially high in patients with primary sclerosing cholangitis (PSC).1, 2 Survival after onset of symptoms of unresectable CCA is 12 to 16 months.3

There are 2 presentations of CCA: intrahepatic mass within the hepatic parenchyma, and periductal involvement of the major intrahepatic and/or extrahepatic bile ducts. Intrahepatic CCA is best treated by liver resection because liver transplantation (LT) for this condition is fraught with rapid recurrence and has been abandoned by most centers.4 Periductal CCA that is limited to the extrahepatic duct below the bifurcation of the common hepatic duct is also best treated by surgical resection. Tumors arising in the hilus are often treated by surgical resection, but many are unresectable because of vascular encasement, extension of tumor to secondary branches of the intrahepatic ducts, or underlying PSC. LT alone for these tumors is inadequate as recurrence is high, and the long-term survival rate is <25%.4 Even patients with PSC and incidental CCA discovered in the explanted liver have a poor prognosis after LT alone.5, 6 However, excellent long-term survival (78% at 5 years) has been achieved in carefully selected patients with unresectable hilar CCA and hilar CCA arising in the setting of PSC treated with neoadjuvant chemoradiotherapy before LT.7, 8

Published experiences from the University of Nebraska7 and the Mayo Clinic at Rochester, MN,8 have demonstrated long-term survival after LT for unresectable hilar CCA and hilar CCA arising in the setting of PSC. Both centers used high-dose neoadjuvant radiotherapy with chemosensitization and operative staging to exclude patients with regional lymph node metastases before LT. Efficacy of this approach is demonstrated by comparing these results with the natural history of the disease, which has a 50 to 70% mortality rate within 12 months.1, 3 Heimbach and colleagues8 demonstrated that neoadjuvant therapy and LT in patients with localized, node-negative disease achieves results similar to LT for other chronic liver diseases (i.e., hepatitis C virus infection, PSC) and hepatocellular carcinoma. Moreover, survival after LT in patients with unresectable hilar CCA or hilar CCA arising in the setting of PSC exceeded survival in patients who underwent resection.9

Region Review Board 7 members agreed to Model for End-Stage Liver Disease (MELD) score exception guidelines in September 2002. The aims of these guidelines were to provide the opportunity for deceased-donor LT to patients with unresectable CCA and CCA arising in the setting of PSC, determine the risk of disease progression while awaiting LT, and retain incentives for living-donor LT and use of extended-criteria donor livers. These guidelines recommend MELD score exceptions for patients after completion of neoadjuvant therapy and operative staging to accurately assess the risk of disease progression while awaiting LT. When the guidelines were approved, the initial MELD score of 20 was matched to the score assigned to patients with stage I hepatocellular carcinoma (which no longer receives an assigned or adjusted MELD score). The interval for additional score adjustments was 6 months, as opposed to the 3-month interval for hepatocellular carcinoma.

Rosen10 transplanted 42 patients between September 2002 and January 2006 with a median staging-to-LT interval of 114 days. After LT, 1 of 21 patients with an interval of <114 days developed recurrent disease, whereas 4 of 21 patients with an interval of >114 days developed recurrent disease (92 ± 7% vs. 56 ± 19% disease-free 2.5-year survival). Although a statistically significant difference was not achieved because of the small number of patients studied, the results suggest that prolongation of waiting time may lead to an increase in recurrent disease after LT. In addition, it was shown that changes brought about by radiation therapy and the staging operation precluded accurate reassessment of the tumor at the time of LT, and that implementation of endoscopic ultrasound-guided aspiration of regional hepatic lymph nodes before neoadjuvant therapy reduced the positive staging rate from 30 to 12%.


The excellent results achieved by protocol-driven approaches that provide neoadjuvant therapy in selected patients with localized, node-negative hilar CCA, the lack of effective alternative therapy, and untreated disease-related mortality justify a MELD score exception for patients with unresectable CCA and CCA arising in the setting of PSC.


CCA, cholangiocarcinoma; PSC, primary sclerosing cholangitis; LT, liver transplantation; MELD, Model for End-Stage Liver Disease.


We propose that LT candidates who meet the criteria listed in Table 1 be granted a MELD score exception equivalent to 10% mortality at 3 months and that the score be increased by a 10% mortality equivalent at 3-month intervals. Required data collection fields will be determined by the United Network for Organ Sharing Liver and Intestinal Organ Transplantation Committee for future reevaluation of CCA as an exception for MELD score adjustment and priority for LT.

Table 1. Criteria for MELD Exception for Liver Transplant Candidates With Cholangiocarcinoma
  1. Abbreviations: MELD, Model for End-Stage Liver Disease; UNOS, United Network for Organ Sharing; CCA, cholangiocarcinoma; PSC, primary sclerosing cholangitis; CT, computed tomographic; MRI, magnetic resonance imaging; LT, liver transplantation.

• Centers must submit a written protocol for patient care to the UNOS Liver and Intestinal Committee before requesting a MELD score exception for a candidate with CCA. This protocol should include selection criteria, administration of neoadjuvant therapy before transplantation, and operative staging to exclude patients with regional hepatic lymph node metastases, intrahepatic metastases, and/or extrahepatic disease. The protocol should include data collection as deemed necessary by UNOS.
• Candidates must satisfy diagnostic criteria for hilar CCA: malignant-appearing stricture on cholangiography and biopsy or cytology results demonstrating malignancy, carbohydrate antigen 19-9 >100 U/mL, or aneuploidy. The tumor should be considered unresectable on the basis of technical considerations or underlying liver disease (e.g., PSC).
• If cross-sectional imaging studies (CT scan, ultrasound, MRI) demonstrate a mass, the mass should be <3 cm.
• Intra- and extrahepatic metastases should be excluded by cross-sectional imaging studies of the chest and abdomen at the time of initial exception and every 3 months before score increases.
• Regional hepatic lymph node involvement and peritoneal metastases should be assessed by operative staging after completion of neoadjuvant therapy and before LT. Endoscopic ultrasound-guided aspiration of regional hepatic lymph nodes may be advisable to exclude patients with obvious metastases before neoadjuvant therapy is initiated.
• Transperitoneal aspiration or biopsy of the primary tumor (either by endoscopic ultrasound, operative, or percutaneous approaches) should be avoided because of the high risk of tumor seeding associated with these procedures.

Although there are no data to permit extrapolation of this recommendation to children, there is little reason to believe that CCA behaves differently in pediatric patients, especially those with PSC. The MELD score exception guidelines should be applied to children whose priority is determined by MELD score.



The natural history of biliary dysplasia in PSC is unknown. There are few data to describe the risk of CCA in patients with biliary dysplasia or the progression of dysplasia to CCA. There are no data describing the progression of dysplasia to CCA beyond LT criteria. Although CCA clearly arises in patients with PSC, it may now be treated effectively with neoadjuvant therapy and LT. There are no data suggesting efficacy for prophylactic LT in patients with PSC, and the overall risk of CCA developing in patients with PSC approximates the risk of death with LT. The risk of development of CCA is approximately 10% during the first 10 years after the initial diagnosis of PSC, which is similar to the mortality rate observed during the first year after LT. The risk of CCA in patients with dysplasia is likely even higher, but these patients can be observed with close follow-up and effective intervention (neoadjuvant therapy and LT) should they develop definitive evidence of CCA. Even if the presence of dysplasia were shown to be a surrogate for CCA, results with LT alone for incidental CCA are poor. Additional investigation and/or observation is necessary to establish a definitive diagnosis of CCA, which would then warrant a MELD score exception and neoadjuvant therapy before LT.

Synthesis of Available Data

Although the presence of biliary dysplasia may identify patients at higher risk of concurrent or later development of CCA compared with other patients with PSC without dysplasia, it is not known whether this risk warrants treatment with LT. Thus, there is no justification at this time to prioritize patients with biliary dysplasia or to recommend a standard MELD score exemption.