Document qa9RnwZdB3R4R5kvbe9rRJ5e5

The New England Journal of Medicine Re Article Medical Progress Although the numbers vary among countries and regions, about two thirds of all cases of cholangio- carcinoma are perihilar tumors, about one fourth are Biliary Tract Cancers distal extrahepatic tumors, and the remainder are in trahepatic.4 Except for embryonal rhabdomyosarco Piet C. de Groen, M.D., Gregory J. Gores, M.D., Nicholas F. LaRusso, M.D., Leonard L. Gunderson, M.D., and David M. IV)agorney, M.D. ma, the frequency of all types of biliary tract cancers increases with age. Gallbladder cancers are more fre quent in women,8 and cholangiocarcinomas are slight ly more common in men.9 These sex differences are probably related to the higher incidence of gallstones in women and of primary sclerosing cholangitis in IN the United States, an estimated 20,000 new cases of liver and biliary tract cancer are diag men. These are known risk factors for gallbladder can cer and cholangiocarcinoma, respectively. nosed annually.1 Biliafy tract cancer is the sec PATHOLOGICAL FEATURES ond most common primary hepatobiliary cancer, af As with most tumors of the digestive system, the ter hepatocellular cancer. Approximately 7500 new large majority of primary tumors are carcinomas.10-11 cases of biliary tract cancer are diagnosed per year; about 5000 of these are gallbladder cancer, and be There are several histologic types, the most common of which are adenocarcinoma, papillary carcinoma, tween 2000 and 3000 are bile-duct cancers.1 Biliary and mucinous carcinoma. The histologic grade varies tract cancers have traditionally been divided into can cers of the gallbladder, thp extrahepatic bile ducts, and the ampulla of Vater, vvhereas intrahepatic bile- from well differentiated to undifferentiated. With the exception of a rare cystadenocarcinoma, the nimors consist of clusters of cells, sometimes surrounded by duct cancers have been classified as primary liver can desmoplastic stroma (Fig. 2A). The desmoplastic re cers.2 The term "cholangiocarcinoma" was originally intended to refer only to primary tumors of the in trahepatic bile ducts and was not used for tumors of the extrahepatic bile ducts.1 Lately, however, the action of a cholangiocarcinoma can be so extensive that the specimen consists mainly of fibrous tissue with sporadic clumps of malignant cells. This feature, espe cially in patients with cholangitis, intraductal gall term has been used to include intrahepatic, perihilar, and distal extrahepatic tumors of the bile ducts.4 Perihilar tumors involving the bifurcation of the hepatic duct are also called Klatskin tumors, from Klatskin's original description in 1965.5 In this re view, the term "cholangiocarcinoma" is used for pri mary tumors of the bile ducts, including intrahepatic, perihilar, and distal extrahepatic tumors (Fig. 1). The perihilar bile-duct tumors were further classi- ; stones, or bile-duct stents, makes it very difficult to distinguish between reactive tissue and well-differen tiated cholangiocarcinoma. Because normal and malig nant bile-duct epithelial cells are not known to express a protein unique to bile-duct tissue, there is no path ognomonic immunohistochemical test to confirm the cell type of origin. However, several types of immu nohistochemical staining support the diagnosis of fied by Bismuth et al. as tumors below the conflu- j malignant biliary tract tissue. The most frequently ence of the left and right hepatic ducts (type I), ; used stains are immunohistochemical stains for cyto- tumors reaching the confluence (type II), tumors keratins, carcinoembryonic antigen, and mucins (Fig. occluding the common hepatic duct and either the 2B and 2C). Other tumor types, which occur in less right or the left hepatic ducf (types Ilia and Mb, re- j than 5 percent of cases, include squamous-cell car spectively), and tumors that are multicentric or that involve the confluence and: both the right and left j 1 cinoma, smali-cell carcinoma, and mesenchymal tu mors. In patients with the acquired immunodeficiency hepatic ducts (type IV).6 Even more detailed classi- I syndrome (AIDS), Kaposi's sarcoma and Ivmphoma fications have been proposed, but they are not used j of the biliary tract have been reported.1211 Finally, in daily practice.7 Most choljmgiocarcinomas involve ; many other types of tumor can obstruct the biliary the perihilar and distal extrahepatic bile ducts. tree by direct extension (for example, in the case of tumors of the pancreas, duodenum, stomach, or co lon), metastasis (for example, tumors of the ovary, From the Mayo Clinic, Rochester, Miryi Address reprint requests to Dr. de Groen at the Division of Gastroenterology and Hepatology, Mayo Clinic 200 First St. SW, Rochester, MN 55905, 1999, Massachusetts Medical Society. breast, or colon), or lymph-node involvement (for ex ample, lymphoma). These tumors will not be dis cussed in this review. The stage of cancers of the biliary tract is deter- 1368 October 28, 1999 V, ED1CALPROGRESS Figure 1. Classification of Cancers of the Human Biliary jract. Panel A shows the overall classification of biliary tract cancers. Panel B shows the Bismuth classification of perihilar chol;angioarcinomas. Yellow areas represent tumor, and green1 areas normal bile duct. C Figure 2. Cholangiocarcinoma. Tumor cells are shown after staining with hematoxylin and eosin (Panel A, X62), carcinoembryonic antigen (Panel B, X62), and MUC-1 (Panel C, x125). There is extensive desmoplastic stroma surrounding the tubules of the cholangiocarcinoma cells. The specimens in Panels B and C are from the same patient. Volume 341 Number 18 nnpo.'f ff 1369 The New England Journal of Medicine mined according to the tumor-node-metastasis sys within 2 years after the diagnosis of primary scleros tem of classification.14 Because the staging is slightly ing cholangitis.27-29 Patients who have ulcerative colitis different for each type of biliary tract cancer, we give in the absence of symptomatic primary sclerosing cho only a general description of the stages. Biliary tract langitis or who have long-standing intraductal gall cancers are classified from Stage 0 to stage FV. Stage stone disease also have an increased risk.2730 Other, 0 is carcinoma in situ; this: stage is not defined for rarer conditions associated with the development of intrahepatic cholangiocarcijioma. In stage I, tumor cholangiocarcinoma include bile-duct adenoma, mul invasion is limited to the mucosa, muscle layer, or tiple biliary papillomatosis, choledochal cysts, Caroli's ampulla. In stage II, local invasion of tumor is seen. disease (cystic dilatation of intrahepatic bile ducts), In stage III, tumor invasion is similar to that in stage and exposure to the radiopaque medium thorium 1 or II, but metastasis info regional and hepato dioxide (Thorotrast).30 In Southeast Asia, infestation duodenal lymph nodes or invasion of adjacent tis with the parasites Opisthorchis viverrini (in Thailand, sues has occurred. Stage IY cancer is characterized Laos, and Malaysia)31 or Clonorchis sinensis (in Japan, by extensive invasion of the liver; invasion of adja Korea, and Vietnam)32 is associated with an increase cent structures or organs; metastases in peripancre- by a factor of 25 to 50 in the risk of cholangiocar atic, periduodenal, periportal, celiac, or mesenteric cinoma. Case-control studies have shown an increased lymph nodes; and distant metastases. risk associated with smoking.33 34 However, despite all RISK FACTORS these known risk factors, many cases of cholangiocar cinoma occur in patients without obvious risk factors. Specific risk factors for the development of hepa Adenomas of the ampulla of Vater, especially when tobiliary cancer have been associated with different they are villous, are known to be premalignant le parts of the biliary tree. Gallbladder cancer is more sions. Adenomas are frequently seen in patients with frequent in patients with gallstones,15 especially if the familial adenomatous polyposis,35 who have a risk of gallstones are symptomatic1^ and large.17 Other fac ampullary adenocarcinoma that is 100 times that in tors associated with gallbladder cancer include female the normal population.36 Other possible, but not well- sex, obesity, and high carbohydrate intake, all of which established, risk factors include cholecystectomy, en are also associated with gallstone disease.16 However, doscopic sphincterotomy, and use of tobacco.33'37-3!1 the increase in the risk of gallbladder cancer in patients Finally, AIDS has been associated with cancers who have cholelithiasis but np symptoms or other risk throughout the entire biliary tract, including the factors is so low that prophylactic resection of the ampulla of Vater.1213'39 gallbladder is not recommended. For persons over 50 years of age, the rate of gallbladder cancer is about MOLECULAR ASPECTS 0.02 percent per year.18 Bacterial infection of bile, with Conversion- from normal to malignant bile-duct or without gallstone disease, occurs in up to 80 per tissue probably requires a number of successive ge cent of patients with gallbladder cancer.19 20 Studies nomic mutations similar to the sequence of events from Chile, Bolivia, and India suggest that the com proposed for other gastrointestinal cancers, although bination of chronic infection with Salmonella typhi and our knowledge of biliary tract cancers is less exten cholelithiasis is strongly associated with gallbladder sive than that of the more common gastrointestinal cancer.2122 Polyps, especially when they are more than cancers. A variety of mutations in oncogenes, as well 1 cm in diameter, and calcification of the gallbladder as tumor-suppressor genes, have been described in wall (porcelain gallbladder) are other predisposing specimens of biliary tract tumors. These include mu factors for cancer.23-24 Finally, anomalous pancrea- tations in the oncogenes K-rat, c-myc, c-neu, c-erb-b2, ticobiliary ductal junction, a rare anatomical anom and cower and the tumor-suppressor genes p53 and aly sometimes associated with a choledochal cyst and bcl-2.w43 These mutations may lead to detectable found mostly among Asians, is associated with a phenotypic changes; for instance, biliary epithelial markedly increased risk of gallbladder cancer.25 In cells switch from expressing MLIC-1 apomucin be patients with this condition, prophylactic cholecys fore birth to MUC-3 after birth.44 Malignant trans tectomy with cyst excision is recommended. formation can-reverse this process, and as mentioned Cholangiocarcinoma, both) intrahepatic and extrahepadc, is a well-known complication of primary scle before, many cholangiocarcinomas show staining with antibody to MUC-1. Similarly, core mucin carbohy rosing cholangitis. Although, lifetime risks in excess of 30 percent have been reported among patients with drate Tn and sialvl-Tn antigens were expressed in many intrahepatic bile-duct cancers.44'45 However, as primary sclerosing cholangitis, most studies mention with other tissue types, mutations and phenotypic lifetime risks of about 10 percent.2628 The time from 1 changes are also seen under nonmalignant conditions, the diagnosis of primary scferosing cholangitis to ! precluding their routine use in clinical practice. Al the development of cholangiocarcinoma ranges from 1 though there is much speculation regarding the fac 1 year to more than 25 years, although at least one j tors that induce the various mutations, such as chronic third of cases of cholangiocarcinoma are diagnosed I inflammation, ethnic background, diet, and exposure 1370 October 28, 1999 1 (0 r t." o i ^ MEDICAL PROGRESS to carcinogens, little or nothing is known about how these factors actually cause biliary tract cancer. DIAGNOSIS Table 1. Potential Tumor Markers in Gallbladder Cancer and Cholangiocarcinoma. The most common presenting symptoms of bil- j Marker Reference iary tract cancer are caused by bile-duct obstruction j In bile and include jaundice, clay-colored stools, cola-colored Tumor antigens or products urine, and pruritus.746 These symptoms tend to oc- : cur early if the tumor is located in the common Carcinoembrvonic antigen CA 19-9 CA 125 Ker et al.4* Ker et al.49 Ker et al.49 hepatic duct, the common bile duct, or the ampulla ; SiahTTn antigen Sasaki et ai.so of Vater. They develop later in perihilar disease and, i when present, are often markers of advanced disease j Fibronectin Oncogene K-raa Kbrner el a].51 Rijken et al.54: in cancer of the gallbladder and intrahepatic Cholan- 1 giocarcinoma. Pain in the right upper quadrant is Tumor-suppressor gene p53 Voravud er al.4' Suto et al.4L` the most frequent presenting symptom in gallblad Metabolic product der cancer but not in cholangiocarcinoma.- In gen Lactate eral, pain, fatigue, malaise, and weight loss occur in ; In serum Tumor antigens or products Nishijima et a advanced disease. The combination of acute right- j upper-quadrant pain, fever, and chills, in association j with cholestasis, strongly suggests cholangitis. Chol Carcinoembrvonic antigen CA 19-9 CA 50 C,A 125 Kuusela et a).s-! Kuuscla et al.,R3 Su et al.n4 Kuusela et al.53 Su et al.54 ecystitis in elderly patients is sometimes the first man ifestation of gallbladder cancer.47 Occasionally, pan creatitis is the first manifestation of a periampullary CA 195 CA 242 DU-PAN-2 Cvtokeratin 19 fragment Bhargava et al.55 Kuusela et al,53 Maeda et al.56 Kashihara et a!.57 tumor. The physical examination may reveal jajundice, right-upper-quadrant pain, hepatomegaly, and a pal Protein induced bv the absence of Nakao et al.SH vitamin K or antagonist 11 (PTVKATI) Cytokine pable gallbladder or mass, depending on the Ideation Interkukin-6 Goydos et al.fi9 and stage of the tumor. Abnormal laboratory-test results in biliary ttyct can Proteases Trvpsinogen-2 Trypsin-2-al-antitrypsin complex Hedstrom et ai.00 Hedstrom et al.60 cer can be divided into those due to interference with normal physiologic processes, resulting from inhibi Peptide Pancreatic polypeptide Bruckner et al.61 tion of normal bile flow or tumor invasion, and those due to the secretion of abnormal products. Cholesta sis and cholecystitis due to obstruction of bile flow typically result in moderate-to-marked increases in none seem to be as useful as CA 19-9.54 We also use serum levels of alkaline phosphatase, bilirubin^ y-glu- serum CA 19-9 levels to assess the effect of treat tamyltransferase, and bile acids, whereas aminotrans ment and to detect recurrence of the disease. ferase levels are only mildly elevated or normal. The first diagnostic imaging procedure in most pa Prolonged obstruction of the common hepatic or tients with cholestasis or right-upper-quadrant pain common bile duct may lead to deficiency of fat-sol is ultrasonography of the liver and gallbladder.65 Gall uble vitamins and increased prothrombin tinje. Ma bladder cancers and intrahepatic cholangiocarcinoma lignant transformation may result in the secretion of may be detected as mass lesions. The absence of mo abnormal products into the bile or serum. Indeed, a bile filling defects within the gallbladder that atten large number of potential markers of biliary tract uate the sonographic beam with "shadow" essentially cancers have been identified. Many markers, however, excludes the possibility of cholelithiasis. Perihilar, ex- are not specific and may also be present under non- trahepatic, and periampullary cancers may not be de malignant conditions (Table l).59-62 Of these, cancer tected by ultrasonography, especially when they are antigen (CA) 19-9 is currently widely used, in partic small. Instead, indirect signs may point toward these ular for detecting cholangiocarcinoma in patients with diagnoses. The most common indirect sign is ductal primary sclerosing cholangitis (Fig. 3).6364 Serhm CA dilatation throughout the obstructed liver segments; 19-9 levels greater than 100 U per milliliter (the nor an abrupt change in ductal diameter may indicate mal level is less than 40 U per milliliter) have been the exact location of the tumor. Color Doppler im reported to have a sensitivity of 89 percent andia spec aging can detect compression, encasement, or throm ificity of 86 percent for the detection of cholangio bosis of the portal vein as well as encasement or carcinoma in these patients.63 A marker consisting of CA 19-9 in combination with carcinoembryohic an tigen (according to the formula CA 19-9 + [carcinoembryonic antigen X 40]) had an accuracy of 86 per cent.64 Other serum markers have been studied, but occlusion of the hepatic artery by tumor.66 The sensi tivity and specificity of ultrasonography vary with the type of tumor, the quality of the equipment, and the experience of the operator. Under optimal condi tions, gallbladder cancers are detected in at least 50 Volume 341 Number 18 1371 C The New England Journal of Medicine pate Figure 3. CA 19-9 and Bilirubin Levels in a 75-Year-Old Man with Primary Sclerosing Cholangitis sinfce 1972 and Cirrhosis. CA 19-9 levels gradually increased, although bilirubin levels remained near normal, and ultrasonography and computed to mography (CT) did not reveal a foca abnormality. On endoscop ic retrograde cholangiography (ERC), strictures and dilatations of bile ducts typical of primary sclerosing cholangitis were seen, with hilar strictures suggesting cholangiocarcinoma. Brush spec imens did not show atypical cells, bbt mild cytologic atypia was evident in other biopsy specimens. Repeated endoscopic retro grade cholangiography with biopsy five months later revealed a single focus of well-differentiated, invasive adenocarcinoma. percent of cases, and detection rates as high as 86 per cent have been reported for cholangiocarcinoma.6768 Computed tomographic (CjT) scanning may show an intraluminal gallbladder miss, with or without di rect invasion of the liver or other adjacent dssues (Fig. 4A).69 Intrahepatic mass lesions (Fig. 4B) and dilat ed intrahepatic ducts (Fig. 4C) are easily detected, but visualization of perihilar tumors or tumors in volving the portal venous or arterial system is best achieved by intravenous bolus-enhanced spiral or hel ical CT scanning.70 Dilatation 'of the intrahepatic bile ducts in a single, small hepatic lobe with hypertrophy j of the contralateral lobe suggests the atrophy-hyper trophy complex, as seen with tjumors chronically ob I structing a single lobe and ihvading the ipsilateral j portal vein.71 Bilobate dilated intrahepatic ducts and a normal or collapsed gallbladder and common bile duct suggest a perihilar tumor, A distended gallblad der without dilated intrahepatic or extrahepatic ducts j : I ! c Figure 4. Computed Tomographic Scans of Patients with Gallbladder Cancer and Intrahepatic and Hilar Cholangiocarcinoma. Panel A shows a large gallbladder cancer (arrow) filling part of the distended gallbladder and invading the adjoining intestine, Panel B shows a huge intrahepatic cholangiocarcinoma (arrow) limited to the right hepatic lobe. Panel C shows a hilar cholangiocarcinoma causing marked dilatation of the bile ducts in the lateral segments of the left lobe. The left medial segment is atrophied (arrow). The bile ducts in the right hepatic lobe are also dilated. 1372 October 28, 1999 f. ` t > Medical progress is seen in patients with cystic duct stones apd tu mors. On the other hand, a distended gallbladder with dilated intrahepatic and extrahepatic ducts is typical of distal extrahepatic ductal cancers, cancers of the ampulla of Vater, intraductal gallstones, or pancreatic cancers. Tumor emboli from hepatocellular Cancer, metastatic colorectal cancer, or intrahepatic cholangiocarcinoma are an unusual cause of perihilar or dis tal extrahepatic bile-duct obstruction.72 Unlike ultra sonography, CT may also show the peripamjreatic, periduodenal, periportal, celiac, and mesenteric lymph nodes. Both ultrasonography and CT permit guided fine-needle aspiration or biopsy of suspicious lesions. Magnetic resonance imaging (MBJ) permits ex cellent visualization of hepatic parenchymal abnor malities, as well as the visualization of the biliary tree and vascular structures. MRI with the use of ferrous oxide and gadolinium yields information similar to that yielded by CT, cholangiography, and angiogra phy combined.73-74 Because MRI is noninvasivle and does not involve exposure to radiation, it may replace CT and angiography for the preoperative assessment of biliary tract cancers. Without doubt, cholangiography is currently the most important radiologic procedure for assessing the resectability of a tumor. Both percutaneous cho langiography and endoscopic retrograde cholapgiogaphy are performed; the choice of procedure de pends on the suspected location of the tumor and the experience of the operators. In general, more proximal and sclerotic tumors are best assessed by percutaneous transhepatic cholangiography (Fig. 5), whereas distal extrahepatic and simple perihilar lesions are amenable to endoscopic assessment. Periampul lary tumors can be directly visualized and biopsies can be performed with a side-viewing endoscope. Both the distal and the proximal extent of tumor growth must be clearly visualized to help thq sur geon decide whether an attempt at curative resection is feasible. However, diffuse abnormalities of the bil iary system, such as those seen in primary sclerosing cholangitis, sometimes make it impossible to delin eate a tumor exactly. Because many patients with gallbladder carcinoma present with obstructive jaundice, cholangiography is also frequently used in the preoperative assess ment of this tumor; typically, a long stricture of the common hepatic duct is found. Once access to the biliary tree has been achieved, bile samples or brush cytologic or biopsy specimens can be obtained. Bile samples, obtained through a percutaneous stent, Con tain cancerous cells in 30 to 40 percent of cases of cholangiocarcinoma.75'76 The use of brush biopsy and cytologic examination may increase the yield to 40 70 percent. Unfortunately, even percutaneous or -ndoscopic biopsy not infrequently yields nonqiagnostic tissue because of the desmoplastic nature of the lesion. The highest diagnostic yield may cpme Figure 5. Cholangiogram of a Patient with Cholangiocarcinoma. RA denotes the right anterior ductal system, RP the right pos terior ductal system, L the left ductal system, and G the gall bladder. The broad, open arrow points to part of the duct that is narrowed by tumor; the broad, solid arrow to the track of the percutaneous needle used to inject contrast material; the ar rowhead to the cystic duct; and the curved arrow to the com mon hepatic duct. from ductal shave biopsies with an atherectomy cath eter and percutaneous biopsies at the location of the suspected tumor immediately adjacent to a previously placed biliary stent.77 Placement of percutaneous or endoscopic stents relieves symptoms, improves hepat ic function, and allows palpation of the ductal struc tures at the time of exploration. Angiography accurately documents vascular en casement and thrombosis of the portal vein and hepat ic artery, but in most cases it is not necessary before surgery. When combined with cholangiography, it correctly predicts resectability in the majority of cas es.78'79 However, as mentioned before, MR] may re place angiography for the assessment of vascular en casement and patency. Several new and promising imaging techniques have recently become available. Endoscopic ultrasonogra- Voiume 341 Number 18 1373 The New England Journal of Medicine phy can be used to visualize the distal extrahepatic bil tively treated with laparoscopic cholecystectomy. Al iary tree, the gallbladder, and the regional lymph nodes though curative resection of stage II disease may in great detail.80 The use of an endoscope equipped also be achieved by laparoscopic cholecystectomy, the with linear ultrasonography allows ultrasound-guided survival rates are higher when more extensive, open fine-needle aspiration of suspected tumors and ab resections are performed.90 For most stage II, III, normal or enlarged lymph nodes.81 Positron-emission and IV gallbladder cancers, extended or radical chol tomography (PET) permits, the metabolism of bile- ecystectomy is preferred.91 In addition to the gall duct epithelial cells to be assessed in vivo by means of bladder, the adjacent liver tissue and regional lymph a glucose analogue, [ l8F jtluoro-2-deoxy- D-glucose,82-83 nodes are removed. Depending on the extent of liver Cholangiocarcinoma cells halve a high glucose uptake. invasion, subsegmental, bisegmental, lobar, or extend Both glucose and [18F]fluorp-2-deoxy-D-glucose are ed lobar resection is performed. Japanese investiga phosphorylated, but [ 18F ]fluoro-2-deoxy-D-glucose is tors found a five-year survival rate of 75 to 80 per not further metabolized. As a result, cholangiocarci cent in patients with disease limited to the mucosa, noma cells accumulate [18F]lluoro-2-deoxy-D-glucose, muscularis, or subserosa who were treated by ex causing "hot spots" on scanning. In addition, hepato- tended cholecystectomy.92 Even more aggressive sur cytes have high glucose-6-phQsphatase activity and rap gery has been performed in patients with extension of idly turn over [18F]fluoro-2-deoxy-D-glucose, thereby tumor into the duodenum, pancreas, colon, or kid further increasing the signal - to-background ratio. Sev ney fossa.93 As expected, mortality and morbidity have eral small studies have documented the ability of PET been high, and the outcome in general has been dis to detect cholangiocarcinomas as small as 1 cm in di appointing. The overall survival, except for patients ameter.82-83 Other, less frequently used new diagnos with stage 0 or I tumors, is dismal. Most patients tic methods include intradfictal ultrasonography,84 present with advanced disease, and the combined five- endoscopic or percutaneous flexible cholangioscopy,86 year survival for all stages of gallbladder cancer is be and radiolabeled antibody or ligand imaging.86-87 tween 5 and 10 percent. Cancers of the gallbladder and of the perihilar and Intrahepatic cholangiocarcinoma is generally treat distal extrahepatic bile ducts may spread directly into ed by hepatic resection alone.94-95 The surgical treat adjacent organs or the abdominal cavity, where they ment of perihilar cholangiocarcinoma depends on can be detected by ultrasonography, CT, MRI, or the Bismuth class.6 Most authors report that about endoscopic ultrasonography. The exact extent of in one third of patients can undergo curative resection, vasion, however, may be difficult to discern. In gall but some suggest that up to two thirds of patients bladder cancer, metastatic disease is rather common should undergo resection with curative intent.4 En and occurs early. In perihilar and distal extrahepatic bloc resection of the extrahepatic bile duets and gall bile-duct cancers, distant mejastases are relatively in bladder, regional lymphadenectomy, and Roux-en-Y frequent and occur late in the course of the disease. hepaticojejunostomy are recommended for type I and The lungs and bones are most commonly involved. II tumors, and that treatment plus hepatic lobecto The metastases can be detected by chest radiography, my is recommended for type III tumors.88-96-97 The CT, or bone scanning. intent is to achieve a tumor-tree proximal margin of TREATMENT ANJ3 SURVIVAL Surgery at least 5 mm.98 Because type II and III tumors often involve the ducts of the caudate lobe, caudate lobec tomy is recommended to improve local control and At present, only surgical excision of alt detectable ; survival for patients with type II or III tumors.88-99 1011 tumor is associated with imjprovement in five-year , In distal extrahepatic tumors and cancers of the am survival.4-8-9-46 Multiple factors related to both the pulla of Vater, pancreatoduodenectomy is the thera patient and the tumor need tp be evaluated when as py of choice. Commonly, a pylorus-preserving Whip sessing resectability.7 Poor performance status, major : ple procedure is performed. Survival is directly related cardiopulmonary disease, and preexisting cirrhosis are to the stage of disease. The median survival for pa the most common patient-related factors precluding tients with intrahepatic cholangiocarcinoma without surgical exploration. Poor nutritional status, sepsis, | involvement of the hilum varies among centers from and severe cholestasis also predict a poor outcome, but , 18 to 30 months. The median survival for patients these factors can sometimes be reversed preoper- with perihilar cholangiocarcinoma is slightly less, vary atively. Distant metastases, extensive regional lym- ing from 12 to 24 months.96-101 The five-year survival phadenopathy, and regional vascular encasement or i for both groups of patients varies between 10 percent invasion preclude resection. Although laparoscopic and 45 percent, with the best results reported from ultrasonography in some cases may alter either the Japan.88-100-102 Finally, five-year survival rates of 15 to diagnosis or the staging of the tumor while obviat- i 25 percent in patients with distal extrahepatic cho ing the need for open laparotomy,88-89 its value is not langiocarcinoma and of 50 to 60 percent in patients fully known. with periampullary cancers have been reported.103105 Stage 0 and stage I gallbladder cancers are etfec- i Patients with unresectable intrahepatic or perihi- 1374 October 28, 1999 o 0(c;. - tp MEDICAL PROGRESS hr cholangiocarcinoma in the absence of extfahepat- life in patients with biliary tract cancers. A large ic disease are potentially treatable with total resection number of agents, including fluorouracil, mitomycin, of the liver, bile ducts, and hilar lymph nodes, followed methotrexate, etoposide, doxorubicin, I-(2-chloro- by orthotopic liver transplantation.106 107 The results of I ethyl)-3-(4-methylcydohexyI)T-nitrosourea, and cis- resection of cholangiocarcinoma in patients With in- platin, have been tested as single or combination trahepatic or perihilar cholangiocarcinoma asf a com therapies without appreciable effects. Partial responses jplication of primary sclerosing cholangitis are dismal.29 lasting from weeks to several months have been ob- These patients can also potentially be treated by or ! served in approximately 10 to 20 percent of cases.119 thotopic liver transplantation. However, the early re I sults of orthotopic liver transplantation in both groups Other Palliative Treatments of patients have been disappointing, w'ith rapid recur In most patients wfio are not candidates for sur rence of disease.108 109 At the Mayo Clinic, patients un | gery, endoscopically or percutaneously placed plastic dergoing orthotopic liver transplantation for cholan I or metal stents relieve the symptoms associated with giocarcinoma are selected according to very strict i cholestasis.120 Plastic stents tend to become occlud- criteria and undergo preoperative external-beam and 1 ed and require replacement approximately every' three internal transcatheter radiation, continuous intrave j months. Metal stents tend to stay open longer be- nous chemotherapy, and pretransplantation explor I cause of their larger diameter. They' rarely migrate ative laparotomy. According to preliminary data, all and are currently widely used.121 According to a re patients treated so far have had prolonged tumor- cent report, the use of a hematoporphyrin derivative free survival. as a sensitizer, followed two day's later by intralumi Palliative surgery is used selectively.96'110 Removal nal photoactivation, resulted in prolonged biliary de of the gallbladder may prevent acute cholecystitis in compression and an improved quality of life.122 Pain patients with gallbladder cancer. In all forms of bil is managed by oral and percutaneous narcotics and, iary tract cancer, a gastrojejunal bypass for the treat if needed, by a celiac-plexus block. Survival is related ment or prevention of obstruction of the'gastric to the stage of the disease. The median survival var outlet is selectively indicated; intraoperatively, a neu ies from 6 to 12 months, with most centers report rolytic celiac-plexus block can be performed fpr pain ing no patients surviving at 5 years.46'9t>'104-110 control. Some centers have reported good palliation and quality' of life after a segment III or V hepjiticoje- FUTURE DIRECTIONS junostomy.111'112 finally, adequate relief of cholestasis Major improvement in the survival of patients w'ith due to distal extrahepatic tumors can be obtained cancers of the biliary tree will probably not result with choledochojejunostomy or hepaticojejunostomy. from more aggressive or advanced surgical techniques Radiation or oncologic radiation therapy. Instead, efforts should be directed at prevendon, early detection, and novel Biliary tract cancers are difficult to control [locally treatments derived from basic research. The universal by external-beam radiation therapy alone. However, benefits of smoking cessation and weight reduction external irradiation, alone or in combination with are obvious. Populadons at risk -- that is, patients fluorouracil, may relieve pain and contribute to bil with primary sclerosing cholangids, intraductal stones, iary decompression. cystic diseases of the biliary system, cholelithiasis When external irradiation is used in combination combined with S. typhi infection, liver flukes, or fa with total resection in patients w ith microscopically milial adenomatosis poly'posis -- need to be identi involved margins, or in combination w'ith flubrour- fied so they can be offered prevendve strategies and acil and supplemental transcatheter brachytfjerapy, prophylacdc or early treatment.123 For instance, pa survival appears to be prolonged, and in a few cases tients with primary sclerosing cholangitis who are long-term survival has been reported.113115 The me found to have cellular atvpia on initial brush cyto dian survival increases from 6 to 8 months among logic examination may benefit from programs of an patients treated with surgery or palliative stent iplace- nual surveillance consisting of a serum CA 19-9 ment and chemotherapy to 12 to 19 months among those wfio receive similar treatment combined with measurement and endoscopic retrograde cholangi ography. The use of brush cytologic examination or external-beam irradiation.114 116 Escalation of the ra biopsy would be analogous to current surveillance diation dose may increase survival.113116 Other inves strategies for inflammatory bowel diseases.124 tigators have not found a significant survival benefit There is a need for new, cost-effective screening resulting from the addition of irradiation to sdrgical resection or stenting.117118 methods, including simple assays for tumor markers in the serum, bile,125 or stool.126 A noninvasive radiolog Jhemotherapy ic technique that can detect disease at a curable stage may already exist, if the early data from studies of PET Preoperative or postoperative chemotherapy does scanning are confirmed in larger series. Several prom not significantly improve survival or the quality of I ising chemopreventive agents, currently under inves- Volume 341 Number 18 1375 t The New England Journal of Medicine X tigation for other types of cancer, may also be bene j 22. Dutta U. Garg PK, Kumar R, Tandon RK. Chronic salmonella typhi ficial to patients with primary sclerosing cholangitis. carrier state and cholelithiasis: cofactors in the pathogenesis of carcinoma of the gallbladder. Digestion I998;59:,$uppl 3:576. abstract. For patients with nonresedtable disease, new drugs 23. Chijiiwa K, Tanaka M, Polypoid lesion of the gallbladder: indications that selectively target malignantly transformed cells of carcinoma and outcome after surgery for malignant polypoid lesion. Int without damaging normal tissue are required. In I Surg 1994,79:106-9. 24. Polk HC Jr. Carcinoma and the calcified gall bladder. Gastroenterolo deed, several promising novel therapeutic agents, in gy 1966,50:582-5. cluding farnesyl transferase and angiogenesis inhibi tors, as well as immunotherapeutic methods directed 25. Chijiiwa 1C, Kimura H, Tanaka M. Malignant potential ot'the gallbtaddcr in patients with anomalous pancrcaticobiliary ductal junction: the dif ference in risk between patients with and without choledochal cyst. Inc against tumor-specific antigens, may soon be tested j Surg 1995;80:61-4. in clinical studies. Finally, qnce we understand the 26. Kocnfcld D, Ekbom A, Ihre T. Survival and risk of cholangiocarcino ma in patients with primary sclerosing cholangitis: a population-based complex molecular pathobiqlogy of the transforma- study. Seand J Gastroenterol L997;32:1042-5. don from normal to malignant biliary tract dssue, ge 27. Broome U, Olsson R, Loof L, et al. Natural history and prognostic factors in 305 Swedish patients with primary sclerosing cholangitis. Gut netic repair of the mutations responsible for the ma ; 1996;38:610 5. lignant phenotype may become possible. 28. Farges O, Malassagnc B, Sebagh M, Bismuth H. Primary sclerosing cholangitis: liver transplantation or biliary surgery. Surgery 1995;! 17:146- Supported by the Mayo Foundation. 55. 29. Rosen CB, Nagornev DM, Wicsner RH, Coffey RJ Jr, LaRusso NF. Cholangiocarcinoma complicating primary sclerosing cholangitis. Ann REFERENCES i Surg 1991;213:21-5. 30. Cailea F, Sergj C, Fabbretti G, Brisigotti M, Cozzutto C, Medicina D. 1. Landis SH, Murray T, Bolden S, Wingjo PA. Cancer statistics, 1998. CA Prccancerous lesions of the biliary tree. J Surg Oncol Suppt 1993;3:131-3. Cancer J Clin 1998;48:6-29. [Errata, CAGancer J Clin 1998;48:192, 329.] 31. Watanapa P. Cholangiocarcinoma in patients with opisthorchiasis. Br J 2. Percy C, Van Holten V, Muir C, eds. International classification of dis eases for oncology. 2nd ed. Geneva: Wofld Health Organization, L990. Surg 1996;83:1062-4. 32. Shin HR, Lee CU, Park HJ, et al, Hepatitis B and C virus, Clonorchis 3. AJborcs-Saavcdra J, Henson DE, Sobjn LH. Histological typing of tu sinensis for the risk of liver cancer: a case-control study in Pusan, Korea. mours of the gallbladder and extrahepatic bile ducts. 2nd ed. Berlin, Ger Int J Epidemiol 1996;25:933-40. many: Springer-Verlag, 1991. 33. Chow WH, McLaughlin JK, Menck HR, Mack TM. Risk factors for 4. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma: a spectrum extrahepatic bile duct cancers: Los Angeles County, California (USA). of incrahepatic, pcrihilar, and distal tumors. Ann Surg 1996;224:463-75. Cancer Causes Control 1994;5:267-72. 5. Klatskin G. Adenocarcinoma of the hepatic duct at its bifurcation with 34. Bergquist A, Glaumann H, Persson B, Broome U. Risk factors and in the porta hepatis: an unusual tumor With distinctive clinical and patho clinical presentation of hepatobiliary carcinoma in patients with primary logical features. Am J Med L965;38:24Io6. sclerosing cholangitis: a case-control study. Hepatology 1998;27:311- 6. Bismuth H, Nakache R, Diamond T.: Management strategies in resec 6. tion for hilar cholangiocarcinoma. Ann Surg 1992;2L5:31 -8. ! 35. Sanabria JR, Croxford R, Berk TC, Cohen Z, Bapat BV, GalUnger S. 7. Pitt HA, Dooley WC, Yeo CJ, Camerdn JL. Malignancies of the biliary Familial segregation in the occurrence and severity of periampullary neo- jtree. Curr Probl Surg L995;32:l-90. | plasms in familiai adenomatous polyposis. Am J Surg 1996;171:136-41. 8. Henson DE, Albores-Saavedra J, Cor^e D. Carcinoma of the gallblad 36. OfFerhaus GJ, Giardiello FM, Krush AJ, et al. The risk of upper gas- der: histologic tvpes, stage of disease, grade, and survival races. Cancer i trointestinal cancer in familial adenomatous polyposis. Gastroenterology 1992;70:1493-7' ! 1992;102:1980-2. 9. Idem. Carcinoma of the extrahepatic tyiie ducts: histologic types, stage 37. Ekbom A, YueaJ-, Karlsson BM, McLaughlin JK, Adami HO. Risk of of disease, grade, and survival rates. Cancer 1992;70:1498-501. pancreatic and periampullar cancer following cholecystectomy: a popula 10. Tumors of the gallbladder and extraJjcpadc bile ducts, [n: Alborcs- tion based cohort study. Dig Dis Sci 1996;41:387-91. Saavedra J, Henson DE. Atlas of tumor pathology. 2nd series. Fascicle 22. 38. Hakamada K, Itoh T, Endoh M, Sasaki M. Late bile duct cancer after Washington, D C.: Armed Forces Institute of Pathology, 1986. the manipulation of the sphincter of Oddi in the treatment of benign pan- 11. Tumors of the liver and incrahepatic bile ducts. In: Craig JR, Peters creatobiliary diseases. Digestion 1998;59:Suppl 3:577 abstract. Rl., Edmondson HA. Atlas of tumor pathology. 2nd series. Fasaclc 26. 39. Seitz JF, Giovannini M, Wartelle C, Monges G, Dhiver C, Gastaut JA. Washington, D.C.: Armed Forces Institutje of Pathology, 1989. .Association d'un sarcome de Kaposi de 1'ampoule de Vater et d'une cho- 12. Segarra P, Abril V, Gil M, Ortega E, Ballester JE, Traves V. Sarcoma langite sclerosante a Cryptosporidium chez un patient porteur d'un SIDA. de Kaposi de la via biliar sin afectacion cutanea en paciente con SIDA. Rev 1 Gastroenterol CLm Biol 1990;14:889-91. Esp Enfcrm Dig 1996;88:637-9. ! 40. Suto T, Sugai T, Nakamura S, et al. Assessment of the expression of 13. O'Boyle MK. Gallbladder wail mass on sonography representing large- I p53, MIB-l (Ki-67 antigen), and urgyrophilic nucleolar organizer regions cel! non-Hodgkin's Ivmphoma in an AfDjS patient. J Ultrasound Med 1994;13:67-8. 14. Fleming ID, Cooper JS, Henson DE, et al., eds. AJCC cancer staging in carcinoma of the extrahepatic bile duct. Cancer 1998:82.86-95. 41. Terada T, Nakanuma Y, Sirica A. Lmmunohistochemical demonstra tion of MET overexpression in human intrahepatic cholangiocarcinoma manual, 5th ed. Philadelphia: Lippincott-Raven, 1997. and in hepatoiithiasis. Hum Pathol 1998;29:175-80. 15. Adson MA. Carcinoma of the gallbladder. Surg Clin North Am 1973; 42. Rijken AM, van GuJik TM, Polak MM, Sturm PD, Gounu DJ. Ofter- 53:1203-16. haus GJ. Diagnostic and prognostic value of incidence of K ras codon 12 16. Zatonski WA, Lowenfels AB, Boyle P, et al. Epidemiologic aspects of mutations in resected distal bile duct carcinoma. | Surg Oncol 1998;68: gallbladder cancer; a case-control study of the SEARCH Program of the 187-92. International Agencv tor Research on Cancer. J Natl Cancer Inst 1997;89: 1132-8. 1 17. Lowenfels AB, Walker AM, Althaus pP, Townsend G, Domehot L. 43. Voravud N, Foster CS, Gilbertson JA, Sikora K, Waxnun J. Oncogene expression in cholangiocarcinoma and in normal hepatic development. Hum Pathol 1989;20:1163-8. Gallstone growth, size, and risk of gallblaqkier cancer: an interracial study. ; 44. Sasaki M, Nakanuma Y, Terada T, Kim YS. Biliary epithelial expression Int J Epidemiol 1989;18:50-4. 18. RansohotFDF, Grade WA. Treatment of gallstones. Ann Intern Med 1993;119:606-19. , ol ML'Cl, MUC.2, MUC3 and MUC5/6 apomucins during intrahepatic bile duct development and maturation: an immunohistochemicai studv -Am J Pathol 1995;147:574-9. 19. Csendes A, Becerra M, Burdilcs P, De|mian I, Bancalari K.Xscndes P. 45. Yamashita K, Yonezawa S, Tanaka S, et al. lmmunohistochemical studv Bacteriological studies ol bile from the gallbladder in patients with carci ' of mucin carbohydrates and core proteins in hepatoiithiasis and cholangi noma of the gallbladder, cholelithiasis, common bile duct stones and no ocarcinoma. Int J Cancer 1993;55:82-91 gallstones disease. Eur ) Surg 1994;i60:3<p3-7. , 46. Farley DR, Weaver .AL, Nagornev DM. "Natural history" of unrcsect 20. Nath G, Singh H, ShukJa VK. Chronic tvphoid carriage and carcino- \ ed cholangiocarcinoma: patient outcome after uoncurative intervention. ma of the gallbladder. Eur J Cancer Prev J997;6;557-9. 21. Strom BL, Soloway RD, Rios-Daletiz JL, et al. Risk factors tor gall- j Mayo Clin Proc 1995;70:425-9. 47. Liu KJ, Richter HM, Cho MJ, Jarad J, NadimpalH V, Donahue PE. bladder cancer: an international collaborative case control studv Cancer I Carcinoma involving the gallbladder in elderly patients presenting with 1995;76:1747-56. ' j acute cholecystitis. Surgery 1997;122:748-56! 1376 * October 28, 1999 I I*^ .j st MEDICAL PROGRESS -i-- ------------------- 48. Ker CG, Wu CC, Chen JS, Hou MF, Lee KT, Sheen PC. A study of CEA, CA 19-9 and CA 125 in biliary tract diseases. Kao HsiuHg 1 Hsueh Ko Hsueh Tsa Chih 1989;5:107-13.' 49. Ker CG, Chen JS, Lee KT, Sheen PC, Wu CC. Assessment of serum and bile levels of CA19-9 and CA125 in cholangitis and bile duct carcino ma. j Gastroenterol Hepatol 1991;6:505-8. 50. Sasaki M, Morita T, Hoso M, Nakanuina V, Tanimura H. Carcinoem- brvonic antigen and blood group-related carbohydrate antigen^ in glyco proteins in human bile in hepatoiithiasis. Hepatology 1996,23:258-63. 51. Korner T, Kropf J, Taspersen D, Schorr W, Hammar C-H, Gressner AM. Fibronectin in menschiicher Galle -- ein neuer Parameter zur Diagnostik maligner Gallengangsprozesse? -- Eine Pilotstudie. Z Gastroenterol 1994;32:87-90. 52. Nishijima T, Nishina M, Fujiwara K. Measurement of lactase levels in serum and bile using proton nuclear magnetic resonance in parents with hepatobiliary diseases, its utility in detection of malignancies Jpn J Clin Oncol 1997;27:13 7. 53. Kuusela P. Haglund C, Roberts PJ. Comparison of a new Rumour marker CA 242 with CA 19-9, CA 50 and carcinoembryonic afitigen (CEA) in digestive tract diseases. Br ) Cancer 1991;63:636 40. 54. Su WC. Chan KK, Lin XZ, et al. A clinical study of 130 parents with biliary tract cancers and periampullary tumors. Oncology 1996,$3:488-93. 55. Bhargava AK, Petrciii NJ, Kama A, et al. Serum levels of cancer-asso ciated antigen CA-195 in gastrointestinal cancers and its comparison with CA19-9. } Clin Lab Anal 1989;3:370-7. 56. Maeda I, Ku Y, Iwasaki T, et al. A case of advanced cholangioceliular carcinoma treated successfully by percutaneous isolated liver perfusion with cisplatin. Gan To Kagaku Rvoho 1996;23:1607-9. (In Japanese.) 57. Kashihara T, Ohki A, Kobavashi 1, et ai. intrahepatic cholalngiocarci- noma with increased serum CYFRA 21-1 level. J Gastroenterol 1998;33 447-53. 58. Nakao A, Taniguchi K, Inoue S, et ai. Clinical application of a new monoclonal antibody (19B7) against PIVKA-II in the diagnosis^ of hepa tocellular carcinoma and pancreatobiliarv malignancies. Am J Gastroenterol 1997;92:1031-4. 59. Goydos JS, Brumfield AM, Frezza E, Booth A, Lotze MT, Carry SE. Marked elevation of serum interleukin-6 in patients with cholangiocarcinoma: validation of utility as a clinical marker. Ann Surg 1998,227:398 404. 60. Hedstrom J, Haglund C, Haapiainen R, Stenman UH. Serpm trypsinogen-2 and trypsin-2-alpha(l)-antitrypsin complex in malignant and be nign digestive-tract diseases: preferential elevation in patients with cholan giocarcinomas. Int } Cancer 1996;66:326-31. 61. Bruckner FIW, Chesser MR, Mandeli J, Farber LA, DiGiovanni G. Circulating pancreatic polypeptide in patients with adenocarcinoma of the bile duct. Acta Oncol 1993;32:627-9. 62. Nakeeb A, Lipsett PA, Lillemoe KD, et al. Biliary carcinoembryonic antigen levels arc a marker for cholangiocarcinoma. Am J Surg J996;171: 147-53. 63. Nichols JC, Gores GJ, LaRnsso NF, Wjesner RH, Nagorncy E?M, Ritts RE Jr. Diagnostic role ofserum CA 19 9 for cholangiocarcinoma in patients with primary sclerosing cholangitis. Mayo Clin Proc 1993;68:874-9. 64. Ramage JK, Donaghy A, Farrant JM, iorns R, Williams R. Scrum tu mor markers for the diagnosis of cholangiocarcinoma in primary sclerosing cholangitis. Gastroenterology 1995;108:865-9, 65. Saim S. imaging of the hepatobiliary tract. N Engl J Med 1997;336: 1889-94. 66. Hann LE, Schwartz LH, Panicek DM, Bach AM, Fong Y, ^lumgart LH. Tumor involvement in hepatic veins: comparison of MR imaging and US for preoperative assessment. Radiology' 1998;206:651-6, 67. Bach AM, Loring LA, Hann LE, Iilescas FF, Fong Y, Blumgart LH Gallbladder cancer: can ultrasonography evaluate extent of disease? J Ul trasound Med 1998;17:303-9. 68. Frezza EE, Mczghebe H. Gallbladder carcinoma: a 28 year experience Int Surg 1997;82:295-300. 69. Furukawa H, Kosuge T, Shimada K, ct al. Small polypoid lemons of the gallbladder: differential diagnosis and surgical indications by helical computed tomography. Arch Surg 1998,133:735-9. 70. Tillich M, Mischingcr HI, Preiseggcr KH, Rabl H, Szolar DfT. Mu! tiphasic helical C i in diagnosis and staging of hilar cholangiocarcinoma. AIR Am J Roentgenol 1998;171:651-8. 71. Hann LE, Getrajdman GI, Brown KT, et al. Hepatic lobar atrophy: as sociation with ipsilatcral portal vein obstruction. AJR Am ) Roentgenol 1996;167:1017 21. 72. Prinz RA, Ko TC, Maltz SB, Reynes Cl, Marsan RE, frecarlj. RJ Common bile duct obstruction by free floating tumor. HPB Sure 1993-6 319-23. 73. Fulcher AS, Turner MA. HASTE MR cholangiography in the eval uation of hilar cholangiocarcinoma AJR Am J Roentgenol 1997-169 1501-5. 74. Schwartz LH, Coakley FV, Sun Y, Blumgart LH, Fong Y, Panicek DM. Neoplastic pancrcaticobiliary duct obstruction: evaluation with breath-hold MR cholangiopancreatography. AJR zVm I Roentgenol 1998: 170:1491 5. 75. Dcsa LA, Akosa AB, Lazzara S, Domizio P, Krausz T, Benjamin IS. Cvtodiagnosis in the management of extrahepatic biliarv stricture Gut 1991;32:1188-91, 76. Mansfield JC. Griffin SM, Wadebra V, Matthcwson K. A prospective evaluation of cytology from biliary strictures. Gut 1997;40:671-7 77. Scheducr MS, Doemeny JM, Johnson JO. Biliary ductal shave biopsv with use of the Simpson atherectomy catheter. J Vase Lnterv Radiol 1993; 4:819-24 78. Dooley WC, Cameron JL, Pitt HA, Lillemoe KD, Yue NC, Vcnbrux AC. Is preoperativc angiography useful in patients with periampullary ru mors? Ann Surg 1990,211:649-55. 79. Kuszvk BS, Sever P, Bluemke DA, Fishman EK. Intrahepatic cholan giocarcinoma: the role of imaging in detection and staging. Cm Rev Diagn Imaging 1997;38:59-88. 80. Tio TL. Sic LH, Kaliimams G, et al. Staging of ampuilary and pancre atic carcinoma: comparison between endosonography and surgery. Gas tromtest Endosc 1996;44:706-13. 81. Wierscma MJ. Vilmann P, Giovannini M, Chang KJ. Wierscma LM. Endosonography-guided fine-needle aspiration biopsy: diagnostic accuracy and complication assessment Gastroenterology 1997;112:1087-95. 82. Delbeke D, Martin WH, Sandier MP, Chapman WC, Wright JK Jr, Pinson CW. Evaluation of benign vs malignant hepatic lesions with posi tron emission tomography. Arch Surg 1998;133:510-6 83. Keiding S, Hansen SB, Rasmussen HH, et aJ. Detection of cholangio carcinoma in primary sclerosing cholangitis bv positron emission tomogra phy. Hepatology 1998;28:700-6. 84. Itoh A, Goto H, Naitoh Y, Hirooka Y, Furukawa T, Havakawa T. In traductal ultrasonography in diagnosing tumor extension of cancer of the papilla of Vater. Gastrointest Endosc 1997;45:251-60. 85. Nimura Y, Kamiya J. Cholangioscopy. Endoscopy 1998;30:182-8. 86. Jewkes A), Macdonald F, Downing R, Drolc Z, Allum WH. Labelled antibody imaging m pancreatic cancer, cholangiocarcinoma, chronic pan creatitis and sclerosing cholangitis. Eur J Surg Oncol 1991;17:354-7. 87. Tan CK, Podila PV, Taylor JE, et al. Human cholangiocarcinomas ex press somatostatin receptors and respond to somatostatin with growth in hibition. Gastroenterology' 1995;108:1908-16. 88. Washburn WK, Lewis WD, Jenkins RL. Aggressive surgical resection for cholangiocarcinoma. Arch Surg 1995;130:270-6. 89. van Delden OM, de Wit LT, Nieveen van Dijkum EJ, Smits NJ, Gouma DJ, Reeders JW. Value of laparoscopic ultrasonography in staging of proximal bile duct tumors. J Ultrasound Med 1997;16:7-12. 90. Fong \r, Heffernan N, Blumgart LH. Gallbladder carcinoma discovered during laparoscopic cholecystectomy: aggressive rcrcscction is beneficial. Cancer 1998;83:423-7. 91. Donohue JH, Nagorney DM, Grant CS, Tsushima K, Ilstrup DM, Adson MA. Carcinoma of the gallbladder: does radical resection improve outcome? Arch Surg 1990;125:237-41. 92. Ogura Y, Mizumoto R, Isaji S, Kusuda T, Matsuda S, Tabata M. Rad ical operations for carcinoma of the gallbladder: present status in Japan. World J Surg 1991;15:337-43. 93. Nakamura S, Nishiy'ama R, Yokoi Y, et al. Hepatopancreatoduodenectomy for advanced gallbladder carcinoma. Arch Surg 1994;129:625-9. 94. Lieser MJ, Barry' MK, Rowland C, Ilstrup DM, Nagorncy DM. Sur gical management of intrahepatic cholangiocarcinoma: a 31-year experience. J Hepatobiliary Pancreat Surg 1998;5:41-7, 95. Sasaki A, Aramaki M, Kawano K, et al. Intrahepatic peripheral cho langiocarcinoma: mode of spread and choice of surgical treatment. Br J Surg 1998;85:1206-9. 96. Kiempnaur J, Ridder GJ, Werner M, Weimann A, Pichlmayr R. What constitutes long-term survival after surgery for hilar cholangiocarcinoma? Cancer 1997;79:26-34. 97. Nimura Y, Kamiya J, Nagino M, et al. Aggressive surgical treatment of hilar cholangiocarcinoma J Hepatobiliary Pancreat Surg 1998;5:52 61. 98. Sakamoto E, Nimura Y, Hayakawa N, et al. The pattern of infiltration at the proximal border of hilar bile duct carcinoma: a histologic analysis of 62 resected cases. Ann Surg 1998;227;405-U. 99. Bartlett D, Fong Y, Blumgart LH. Complete resection of the caudate lobe of the liver: technique and results. Br J Surg 1996;83:1076-81, 100. Nagino M, Nimura Y, Kamiya J, et al. Segmental liver resections for hilar cholangiocarcinoma. Hepatogastroemerology 1998;45:7-13. 101. Madariaga JR, Iwatsuki S, Todo S, Lee RG, Irish VI', Starzl TE. Liver resection for hilar and peripheral cholangiocarcinomas: a studv of 62 cases Ann Surg 1998;227:70-9. 102. Reding R, Buard JL, Lebcau G, Launch B. Surgical management of 552 carcinomas of the extrahepatic bile ducts (gallbladder and periampul Volume 341 Number 18 * 1377 The New England Journal of Medicine lary tumors excluded): results of the French Surgical Association Survey. 115. Aldcn ME, Mohiuddin \1. The impact of radiation Jose in combined Ann Surg 1991;213:236-41. external beam and intraluminal ir-192 brachythcrapy for bile duct cancer, 103. Monson JR, Donohue JH, McEhtec GP, ct al. Radical resection for [nt J Radiat Oncol Biol Phys 1994;28:945-51. carcinoma of die ampulla of Vater. Arcji Surg 1991;126:353-7. 11$. Kuv.shinoff BW, Armstrong IG, Fong V, ct al. Palliation of irresect* 104. Fong Y, Bluingart LH, Lin E, Foftncr JG, Brennan ME Outcome of able hilar cholangiocarcinoma with biliary drainage and radiotherapy. 8r J treatment for distal bile duct cancer. Bt J Surg 1996;83:1712-5. Surg 1995,82:1522-5. 105. Yeo CJ, Sohn TA, Cameron JL, Fjruban RH, Lillcmoc KX>, Pitt HA. 117. Pitt HA, Nakeeb A, Abrams RA, et al. Perihilar cholangiocarcinoma: Periampullary adenocarcinoma: analvsi? of 5-vear survivors. Ann Surg postoperative radiotherapy does not improve survival. Ann .Surg 199S;221: 1998;227:821-31. 788-98, 106. (watsuki S, Todo S, Marsh JW, et al. Treatment of hilar cholangio- 118, Bowling TE, Galbraith SM, Hartield AR, Solano J, Spittle MF. A ret- carcinoma (Klatskin tumors) with hepatic resection or transplantation. 1 rospective comparison of endoscopic stenting alone with stenting and ra J Am Coll Surg 1998;187:358-64. , diotherapy in non-resectable cholangiocarcinoma. Gut 1996;39:852-5. 107. Urcgo M, Flickinger JC, Carr BlJ Radiotherapy and multimodality 113. Sanz-Altamira PM, Ferrante K, Jenkins RJL, Lewis WD, Huberman management of cholangiocarcinoma. hit J Radiat Oncol Biol Phvs 1999; MS, Stuart KE. A phase U trial of 5-tluorouracil, leucovorin, and carbo- 44:121-6. platin in patients with unresectablc biliary tree carcinoma. Cancer 1998;82: 108. Goldstein RM, Stone M, Tillery GW, ec al. Is liver transplantation 2321-5. indicated for cholangiocarcinoma? Am J Surg 1993;166:768-72. 1 120. Prat F, Chapac O, Ducot B, et al. Predictive factors for survival of pa 109. Goss JA, Shackleton CR, Farmer5 DG, ct al. Orthotopic liver trans tients with inoperable malignant distal biliary .strictures: a practical man plantation for primary sclerosing cholangitis: a 12-year single center expe agement guideline. Gut 1998;42:76-80. rience. Ann Surg 1997;225:472-83. 121. Schima W, Prokesch R, Osterrcichcr C, ct al. Biliary Wallstenr en 110. Nordback 1H, Pitt HA, Coleman J, ct al. Unresectablc hilar cholan doprosthesis in malignant hilar obstruction: long-term results with regard giocarcinoma: percutaneous versus operative palliation. Surgery 1994;il5 to the type of obstruction. Clin Radiol 1997;52:213-9. 597-603. 122. Ortner MA, Liebetruth j, SchreiberS, et al. Photodynamic therapy 111. Taschieri AM, Elli M, Danelli PQ, Crisuldi M, Montecamozzo G, of nonresec table cholangiocarcinoma. Gastroenterology 1998;114:536-42. Porretta T. Third segment cholangiojejunostomy in the treatment of 123. de Groen PC, Barry JA, Sehaller WJ. Applying World Wide Web unresectablc Klatskin tumors. Hepatogastrocnterology 1995;42:597- technology to the study of patients with rare diseases. Ann Intern Med 600. 1998;129:107-13. 112. Jarnagin WR, Burke E, Powers C, Fong Y, Biumgart LH. Intrahe- 124. Johlin FC, Voigt M, Wu YM. Surveillance cytology (SC) in the de patic biliary enteric bypass provides effective palliation in selected patients tection of asymptomatic progression to cholangiocarcinoma (CCC) in pa with malignant obstruction at the hepatic duct continence. .Am J Surg tients with primary sclerosing cholangitis. Hepatology 1998;28:Suppl: 1998;175:453-60. 393A. abstract. 113. Foo ML, Gunderson LL, Bender CE> Buskirk SJ. External radiation 125. Ito R, Tamura K, Ashida FI, et al. Usefulness of K ras gene mutation therapy and transcatheter iridium in the treatment of extrahepatic bile duct at codon 12 in bile for diagnosing biliary strictures. Int J Oncol 1998;12: carcinoma, hit J Radiat Oncol Biol Phy$ 1997;39:929-35. 1019-23. 114. Gonzalez Gonzalez D, Gerard JP, Maners AW, et al Results of radi 126. Coldas C, Hahn SA, Hruban RH, Redston MS, Yeo CJ, Kern SE. De ation therapy in carcinoma of the proximal bile duct (Klatskin tumor). tection of K-ras mutations in the stool of patients with pancreatic adenocar- Semin Liver Dis 1990;10:131-41. i cinoma and pancreatic ductal hyperplasia. Cancer Res 1994;54:3568-73. 1378 October 28, 1999 f ;i M