Cancers of the Gallbladder and Biliary Ducts

Biliary tract cancers occur infrequently in the United States. Approximately 7,500 new cases are diagnosed each year, consisting of 5,000 gallbladder cancers and 2,500 cholangiocarcinomas.[1] Worldwide, they represent a more significant health burden, with a high frequency in Chile and parts of Asia.

The associated mortality rate in the United States is high, with approximately 4,000 to 6,000 deaths annually.[2] Few patients with unresectable cholangiocarcinomas survive longer than 1 year, and the outlook for gallbladder cancer patients is equally poor. The median survival for the latter is less than 6 months, and fewer than 5% survive 5 years.[2] The roles of chemotherapy and irradiation, particularly in the adjuvant setting, have not been well defined in these diseases.

Cholangiocarcinoma

Epidemiology and Pathogenesis

Cholangiocarcinomas are cancers of the biliary tree. Extrahepatic tumors are more common than tumors within the liver. Hilar cancers, also called Klatskin tumors, are the most common, occurring approximately two to three times more often than distal cholangiocarcinomas. Although more than 10 different histologic subtypes of extrahepatic cholangiocarcinomas have been reported, over 98% are adenocarcinomas.[3] Of the other histologic subtypes, papillary tumors are rarely invasive (and thus have a favorable prognosis), whereas the small-cell subtype metastasizes early and is nearly invariably fatal.

Little is known about the pathogenesis of adenocarcinomas except that these tumors are associated with inflammation of the biliary system. Cholangiocarcinomas express a receptor for interleukin (IL)-6, and exposure to this cytokine stimulates tumor growth in vitro.[4] They also express Fas ligand, which induces apoptosis of lymphocytes,[5] but they escape Fas-mediated apoptosis by expressing an apoptosis inhibitor (I-FLICE). Thus, cholangiocarcinomas can proliferate while avoiding the effector cells activated by the inflammation-triggered immune response.

Risk factors for the development of cholangiocarcinoma are moderately well characterized. Patients with primary sclerosing cholangitis have the highest risk, with a 7% to 10% lifetime risk of developing this cancer.[6] These tumors also occur with increased frequency in patients with hepatitis C-related cirrhosis, ulcerative colitis, and parasitic infestations.[7] Although less common than the association of hepatitis C and hepatocellular carcinoma, as many as 2% of patients with the hepatitis C virus develop cholangiocarcinoma.[8] Environmental factors have also been associated with biliary tumors. An increased incidence of cholangiocarcinomas is seen in patients exposed to the contrast agent thorium dioxide (Thorotrast).[1] Other chemical carcinogens inducing this cancer include cigarette smoke, asbestos, radon, dioxin, and nitrosamines.[1,9]

Symptoms

Cholangiocarcinomas rarely present symptomatically until they are quite large and cause obstruction of the biliary tract. Painless jaundice occurs in up to 90% of patients, with serum bilirubin often exceeding 10 mg/dL. Approximately 50% of patients experience weight loss. Obstruction caused by cholangiocarcinoma can also result in the symptoms of cholangitis, manifesting as fever and right upper quadrant abdominal pain. Unfortunately, none of these symptoms are specific for neoplasm.

Screening and Early Detection

Due to the poor outcome of patients with advanced cholangiocarcinomas as well as the fact that surgery can be curative in limited-stage disease, early detection through a dedicated screening program could conceivably decrease mortality. However, the rarity of biliary tumors makes this approach difficult. Tumor markers have been considered because of their simplicity. The carbohydrate antigen CA 19-9 is often elevated in the serum of patients with cholangiocarcinomas.

A study from the Mayo Clinic compared the CA 19-9 levels of 9 patients with primary sclerosing cholangitis and known cholangiocarcinoma to 28 control patients with primary sclerosing cholangitis and no malignancy.[6] In this setting, CA 19-9 levels exceeding 100 U/mL were 89% sensitive and 86% specific for the detection of cancer. Patients with cholangiocarcinoma without primary sclerosing cholangitis have also been studied. A CA 19-9 level exceeding 100 U/mL also correlated with a diagnosis of cholangiocarcinoma in this group.[10] However, the sensitivity decreased to 50%.

Unfortunately both of these studies suffered from an imbalance in measured serum bilirubin between cases and controls. Patients with cancer had a mean serum bilirubin of 7.0 to 8.0 mg/dL, compared to controls, with a mean of only 2.0 to 3.6 mg/dL. In nonmalignant processes, high CA 19-9 levels correlate directly with increased bilirubin.[11] Therefore, the sensitivity of CA 19-9 would likely be lower than reported if better matched controls with higher serum bilirubin levels were used.

Overall, no tumor marker studied to date has adequate sensitivity and specificity to be recommended for general screening.

Diagnostic Imaging

Cholangiocarcinoma may appear as an abdominal mass, hypoattenuated relative to the liver, on computed tomography (CT) scans. The sensitivity of CT in detecting this cancer is approximately 70%.[12,13] Although ultrasound is a less expensive test, it is also much less sensitive, suggesting a diagnosis of cancer in only 25% to 50% of cases. The primary use of ultrasound is to evaluate a patient presenting with obstructive jaundice.

Cholangiography is a potentially helpful technique, with reported sensitivity results ranging as high as 80% to 90%.[12] It is useful in both defining the location of the cholangiocarcinoma and determining whether a curative resection can be performed. However, the specificity of cholangiography is only 62% because it cannot always distinguish benign from malignant strictures.

Magnetic resonance cholangiopancreatography (MRCP) provides accurate information about tumor extent, including vascular involvement.[14,15] MRCP in conjunction with transcutaneous duplex ultrasound may soon replace other modalities in the evaluation of resectability.[16]

In the immediate preoperative period, percutaneous transhepatic cholangiography may provide additional biliary ductal detail, facilitating the placement of silastic stents that can help guide intraoperative dissection; these stents may be left in place postoperatively to provide access to the ductal system. However, prolonged biliary drainage with stents is not advantageous and increases the incidence of infectious complication.[17]

Due to their avid uptake of fluorodeoxyglucose, cholangiocarcinomas are very conspicuous on positron-emission tomography (PET).[18] Tumors as small as 1 cm can be detected by PET. Although too costly for use as a general screening tool, PET may prove useful in the evaluation of candidates being considered for surgical resection and suspected of having nonlocalized disease.

Staging

Several different staging systems have been employed for biliary tract cancers. Table 1 shows the tumor-node-metastasis (TNM) staging classification developed by the 1998 American Joint Committee on Cancer (one of the more widely used systems).[3] Stage T1 and T2 disease can be cured with surgery, with 5-year survival rates of 57% and 39%, respectively.[19] Unfortunately, the cure rate for T3 disease, with which most patients present, is significantly lower. Involvement of the regional lymph nodes also results in poor survival. No patients with N2 tumors and only 10% of patients with N1 tumors survive for 5 years.

Another commonly used staging method is the Bismuth-Corlette system,[20] which classifies tumors by the extent of biliary duct involvement. This system is preferred by many surgeons because it is said to be more clinically relevant. However, neither the TNM nor the Bismuth-Corlette staging system reliably predicts 5-year survival.

A more recently devised staging system has been proposed by Memorial Sloan-Kettering Cancer Center.[21] The T-stage classifications in this system appear to correlate well with resectability and predict the need for hepatectomy. Moreover, these clinical categories seem to be prognostic of 5-year survival. If confirmed by other centers, this schema may become the preferred clinical staging system.