Journal of Hepatology 1997; 27: 512-516 Printed in Denmark All rights reserved ~un~sg[~urd Copenhagen
Jourual of Hepatol~ ISSN 0 168.8278
Are patients with cirrhotic stage primary sclerosing cholangitis at risk for the development of hepatocellular cancer? Denise M. Harnois’, ‘Division of’ Gastroenterology,
J. Gores’, Jurgen Ludwig2, Jeffery L. Steers3, Nicholas Russell H. Wiesner’
of Liver Transplantation, 2Departnmnt of Pathology. Mayo Clinic and Foundalion, Rochester, USA
BuckgvoumUAims: The risk of cholangiocarcinoma in primary sclerosing cholangitis is widely recognized to be III--30%,whereas the risk of acquiring hepatocellular carcinoma in primary sclerosing ~holangitis is unknown. As in other chronic liver diseases, the presence of hepatocellular carcinoma in a patient with primary sclerosing cholangitis undergoing evaluation for orthotopic liver transplantation would clearly impact on the candidacy, diagnostic evaluation, and alternative treatment options. Thus, the aim of our study was to determine the prevalence of hepatocell~ar careinoma in patients undergoing liver transplantation for primary sclerosing cholangitis. Methods: The records of the 520 patients undergoing orthotopic liver transplantation at our institution between 1985 and May 1995 were reviewed. Of the 134 patients with primary sclerosing cholangitis, three (2%) had hepato~ellular carcinoma. In the 386 patients without primary sclerosing cholangitis undergoing orthotopic liver transplantation, 22 (6%) had hepatocellular carcinoma. Results: Neither the duration of primary sclerosing
sclerosing cholangitis (PSC) is a progressive cholestatic liver disease characterized by inflammatory strictures of the large intra- and extrahepatic bile ducts. Although there is a well-documented association between cholangiocarcinoma and primary sclerosing cholangitis (l--3), the risk of developing h~patoce~lular carcinoma in PSC is less well defined. Because PSC can lead to cirrhosis, and cirrhosis is a risk factor for HCC, cirrhotic-stage PSC may be asRIMARY
Received 4 February, revised 14 April; accepted 15 April 1997
Correspondence: Gregory J. Gores, M.D., Mayo Clinic, 200 First St. SW, Rochester, MN 55905, USA.
Tel: (507) 284 0686. Fax: (507) 284 0762. email [email protected]
cholangitis (range 7-23 years) nor the presence of ulcerative colitis (two of three patients) ~stinguish~ those patients with primary sclerosing cholangitis plus hepato~ellular carcinoma from those with primary sclerosing cholangitis alone. None of the three patients with primary sclerosing cholangitis plus hepatocellular carcinoma had evidence for hepatitis B or C, alpha-1-antitrypsin deficiency, or hemochromatosis. None of the tumors was of the ~brolamellar variety of hepatocellular carcinoma. Cu~c~~si~~s~ The prevalence of hepato~ellular carcinoma in patients with primary sclerosing cholangitis undergoing orthotopic liver transplantation is 2%. These data suggest that patients with advanced cirrhotic-stage primary sclerosing cholangitis are at increased risk for developing hepat~ellular c~cinoma and should be screened for hepatocellular carcinoma as well as for ~holangio~~cinoma prior to orthotopic liver transplantation. Key words: Fibrolamellar hepatocellular carcinoma; Inflammatory bowel disease; Screening.
sociated with the development of HCC. Indeed, several case reports suggest HCC can develop in PSC, especially fibrolamelfar hepatocellular carcinoma (46). As in other chronic liver diseases, the presence of HCC in patients with PSC undergoing evaluation for orthotopic liver transplantation (OLT) would clearly impact on the patient’s candidacy, diagnostic evaluation, and alternative treatment options. Thus, the overall objective of our study was to determine the prevalence of hepatocellular carcinoma in patients undergoing liver transplantation for primary sclerosing cholangitis. Materials and Methods The records of 520 patients undergoing OLT at the Mayo Clinic between March 1985 and May 1995 were
PSC and hepatocellular
reviewed. In this population, 134 patients (26%) had PSC. PSC was diagnosed by standard clinical, biochemical, histologic and radiographic criteria, and by careful analysis of the explanted or autopsied livers. All explanted livers were sectioned and reviewed by a hepatic pathologist. The method of studying the explants has been described previously (7).
the explanted liver. Histopathology demonstrated that 90% of the lesion was necrotic, although a rim of viable HCC cells was present at the peripheral edges of the nodule. The remaining liver demonstrated a 2-cm cavernous hemangioma in the left lobe and stage 314 PSC. The patient’s AFP levels decreased to 2.2 &ml post-OLT. There is no evidence of recurrent HCC after 48 months of follow-up.
Case Reports Case report 1 A 43-year-old female presented to the Mayo Clinic in 1983 with a 23-year history of ulcerative colitis and recent onset of jaundice. A liver biopsy confirmed chronic hepatitis, stage 2, associated with PSC. A transhepatic cholangiogram demonstrated multiple strictures of the hepatic duct and mild dilatation, consistent with sclerosing cholangitis. Liver chemistries were as follows: total serum bilirubin (TSB)=2.2 mg/ dl (normal < 1.1 mgidl), direct serum bilirubin (DSB)= 1.3 mg/dl (normal ~0.3 mg/dl), alkaline phosphatase (AP)=687 U/l (normal <257 U/l), aspartate amino transferase (AST)=29 U/l (normal ~31 U/l). Hepatitis serologies for B and C were negative. There was no evidence of hemochromatosis by serum iron studies or iron stain of the liver biopsy specimen. The patient’s liver disease progressed and her symptoms of pruritis and fatigue worsened. A repeat liver biopsy in 1985 demonstrated stage 3 chronic hepatitis associated with PSC. Esophagogastroduodenoscopy (EGD) confirmed the presence of esophageal varices. In 1986, an ultrasound study demonstrated a 1-cm echogenic solid mass in the lateral segment of the left lobe of the liver thought to represent a cavernous hemangioma. In 1990, she underwent a total proctocolectomy with an ileoanal anastomosis for dysplasia noted on surveillance biopsies. In June of 1993, a new focal mass in the right lobe of the liver was noted on repeat ultrasound. A computed tomography (CT ) scan of the abdomen confirmed the presence of a 5-cm inhomogeneously enhancing mass in the dome of the liver, consistent with hepatocellular carcinoma. The second 2-cm lesion in the left lobe of the liver had the imaging characteristics of a cavernous hemangioma. An ultrasound-guided biopsy of the lesion at the dome of the right lobe of the liver was interpreted as grade 1 hepatocellular carcinoma. The alpha-fetoprotein (AFP) was 1410 ngiml (normal ~15 ngiml). Hepatic artery chemoembolization was performed with Ivalon particles mixed with cis-platinum, mitomycin C, and adriamycin, and she subsequently underwent OLT on November, 1993. The excised recipient liver weighed 1150 g (23 X 17X8 cm). A residual 2x2 cm necrotic nodule was present in the right lobe of
Case report 2 A 3 l-year-old male presented at the Mayo Clinic in 1987 with a 2-year history of jaundice and fatigue. A liver biopsy was nondiagnostic because of insufficient tissue. An endoscopic retrograde cholangiopancreatogram (ERCP) confirmed PSC with extensive intrahepatic involvement. An ultrasound study of the liver was negative for a mass but was consistent with cirrhosis; a CT scan of the abdomen was also negative for any evidence of a mass lesion. Liver chemistries were as follows: TSB=3.8 mgidl, DSB=2.2 mg/dl, AP=675 U/l, AST=191 U/l. Hepatitis serologies for B and C were negative, and there was no evidence of hemochromatosis by iron studies. In 1991, an AFP level was 2.1 ng/ml. There was no evidence of colitis on colonoscopic examination. Because of end-stage liver disease, he underwent OLT in July, 1992. A pre-operative alpha-fetoprotein value was normal; pre-operative CT and ultrasound studies of the liver were also negative for a mass lesion. The explanted liver weighed 1560 g (24~ 17~ 10 cm) and on the anterior surface of the right lobe a l-cm nodule was noted. Microscopic study confirmed a grade 2 hepatocellular carcinoma with focal necrosis but no evidence of vascular invasion; the remaining liver exhibited stage 4 PSC. Post transplantation, the patient developed hepatic artery thrombosis necessitating retransplantation in August 1992. Postoperatively, he developed an overwhelming fungal infection and died in November 1992. Case report 3 A 30-year-old male presented in 1966 with complaints of blood per rectum and was diagnosed after proctologic exam with chronic ulcerative colitis. In 1968, he underwent a total colectomy with establishment of a Brooke ileostomy. A liver biopsy obtained at surgery showed periportal inflammation with focal hepatic cellular necrosis and evidence of fibrosis. In 1982, a liver biopsy obtained elsewhere because of deteriorating liver chemistries, confirmed the diagnosis of PSC, stage IV. In 1991, he developed jaundice, ascites, bleeding ileal varices documented by colonoscopy, and nonbleeding esophageal varices. His liver biochemistries 513
D. M. Harnois et al. TABLE
Characteristics of patients with primary sclerosing cholangitis plus hepatocellular carcinoma (HCC) and PSC alone
Age, years (mean/range) Gender Duration of PSC, years (mean/range) Inflammatory bowel disease (%)
PSC plus HCC
50 2 14 2
45 15 11 95
(3655) M/l F (7-23) (66%)
(15-71) Ml51 F (l-39) (83%)
Clinical information on the three patients with primary cholangitis (PSC) plus hepatocellular carcinoma (HCC)
Duration of PSC (yr)
Tumor size (cm)
Pl P2 P3
55 53 36
M F M
23 10 7
yes yes no
1.5 5.0 1.0
4.2 1410 2.1
were as follows: TSB=29.0 mg/dl, DSB=16.2 mg/dl, AP=673 U/l, and AST= 118 U/l. He had negative hepatitis serologies for B and C, and no evidence of hemochromatosis by serum iron studies. A serum alpha-fetoprotein was 4.2 @ml. Preoperative ultrasound and magnetic resonance imaging (MRI) studies of the liver showed no evidence of a hepatic mass. Because of end-stage liver disease, he underwent OLT in April, 199 1. Pathologic examination of the explanted liver revealed biliary cirrhosis with PSC and grade 2 hepatocellular carcinoma, 1.5 cm in diameter without evidence of vascular invasion. The weight of the explanted liver was 1800 g (24~18XlOcm). He is now 5 years post-transplant and has had no evidence of recurrent HCC by ultrasound or CT scan; his serum AFP remains normal.
angitis can predispose to the development of hepatocellular carcinoma. This is the first paper, to our knowledge, to review the risk of developing hepatocellular carcinoma in a large number of patients with end-stage primary sclerosing cholangitis. Because the whole liver was available for pathologic examination, we can determine a true prevalence of hepatocellular carcinoma in this patient population, patients with end-stage liver disease from PSC. The prevalence of HCC in this PSC population was 2%, compared to a prevalence of HCC of 6% (22 of 386 patients) in patients undergoing OLT for all other indications, including hepatitis B and C, alpha-1-antitrypsin ZZ phenotype, alcoholic liver disease, cryptogenic cirrhosis and hemochromatosis (Fig. 1). No patients with PSC were turned down at our institution during this time period because of a known HCC. Thus, the 2% prevalence of HCC in this patient population should be accurate. In contrast, the prevalence of HCC in the non-PSC population is not representative, as it reflects a selection bias of preferentially transplanting patients with small tumors following chemoembolization and turning down patients for OLT who have advanced cancer. Further categorization of the non-PSC population with HCC demon-
‘3 Patients in each group with hepatohiliary neoplasia (%)
Results/Discussion The average age, gender, duration of PSC, and the presence or absence of inflammatory bowel disease did not distinguish those three patients with PSC plus HCC from the 134 patients with PSC alone (Tables 1 and 2). None of the three patients with PSC plus HCC had evidence of hepatitis B, hepatitis C, alpha-l-antitrypsin (MZ or ZZ phenotype), alcohol-related liver disease, or hemochromatosis to explain the development of hepatocellular carcinoma. None of these patients had received ursodeoxycholic acid therapy. All of the patients had well-established cirrhosis with diffuse regenerative nodules. These cases suggest that advanced biliary cirrhosis from primary sclerosing chol514
Non-PSC HCC n=22/386
PSC HCC n=3/134
Non-PSC CCA n=3/386
PSC CCA n=3/134
Fig. 1. The prevalence of hepatocellular carcinoma (HCC) and cholangiocarcinoma (CCA) in patients with or without primary sclerosing cholangitis (PSC) undergoing orthotopic liver transplantation. Using a Fisher’s exact test, there was no signtj?cant difference between the prevalence of HCC in patients with vs. those without PSC. Likewise, there was no signiJicant difference between the prevalence of CCA in patients with vs. those without PSC. It should be noted, however, that these are prevalence figures for patients selectedfor liver transplantation at our institution and may not reflect the prevalence of these complications in unselected patients.
PSC and hepatocellular carcinoma
Patients in each group with HCC (%)
Metabolic Other Viral hepatitis liver disease 1x=9/208 n=10/54 n=2125
Fig. 2. The prevalence ofhepatocellular carcinoma (HCC) in patients undergoing orthotopic liver transplantation for primary biliary cirrhosis (PBC), viral hepatitis (hepatitis B and C), metabolic liver diseases (amyloid, hemochromatosis, and Wilson s disease), other categories of liver disease (ethanol-related, Budd-Chiari syndrome, non-alcoholic steatohepatitis, aIpha I antitrysin ZZ or MZ phenotype, fulminant hepatic failure, cryptogenic cirrhosis) and primary sclerosing cholangitis [PSC). Using a contingency table analysis, the prevalence of HCC was statistically greater in patients with viral hepatitis as compared to the other groups (pG.01). It should be noted, however, that these are prevalence figures for patients selected for liver transplantation at our institution and may not reflect the prevalence of these complications in unselected patients.
strates that the largest number of patients undergoing transplantation for end-stage liver disease who were found to have HCC had viral hepatitis (19%) (Fig. 2). These data are consistent with previous reports that HCC is a frequent complication of advanced viral liver disease (8). Using a Fisher’s exact test, the 2% prevalence of HCC in PSC was not statistically different from the 1% prevalence of HCC in patients with primary biliary cirrhosis (PBC) (Fig. 2). Our prevalence of HCC in PBC is similar to that reported by others (9,lO). Thus, although the risk for developing HCC in cholestatic liver diseases is lower than that for viral hepatitis, the prevalence of HCC in patients with these diseases undergoing OLT is high enough to warrant screening. We have also confirmed previously reported data that cholangiocarcinoma occurs commonly in patients transplanted with PSC (2%) (Fig. I) (1,2,8). However, these data underestimate the true prevalence of cholangiocarcinoma in PSC, since many patients with cholangiocarcinoma were excluded or never evaluated for OLT.
The practical conclusion of this study relates to the evaluation of patients with PSC for OLT Although patients with PSC are screened for cholangiocarcinoma prior to being considered for OLT, the risk of HCC in these patients may not be appreciated. In addition to being screened for cholangiocarcinoma, patients with end-stage PSC should also have a serum alpha-fetoprotein determination and ultrasound and/or CT scan during their evaluation for OLT. Finding an HCC may impact on their candidacy for OLT or the treatment of the patient prior to OLT. For example, Case 1 of this paper underwent chemoembolization of her HCC before undergoing OLT These issues become increasingly important with the longer waiting times world-wide for OLT. Finally, we note that our data regarding the risk for developing HCC in PSC relate only to patients with advanced PSC associated with extensive fibrosis and diffuse nodular regeneration of the liver. Our data may not pertain to patients with noncirrhotic or even early cirrhotic stage (portal to portal biliary fibrosis in the absence of regenerative nodules) disease. Therefore, the question of screening patients with PSC for HCC in the absence of well-established advanced cirrhosis cannot be addressed by our data.
Acknowledgements This work was supported by NIH Grants DK 41876 (GJG) and DK 24031 (NFL), NC1 grant CA 15083-23F3.2 (GJG), the Gainey Foundation, St. Paul, MN (GJG), and the Mayo Foundation, Rochester, MN, USA.
References 1. Farges 0, Malassagne B, Sebagh M, Bismuth H. Primary sclerosingcholangitis: liver transplatation or biliary surgery. Surgery 1995;117: 146-55. 2. Rosen CB, Nagorney DM, Wiesner RH, Coffey RJ, LaRusso NE Cholangiocarcinoma complicating primary sclerosing cholangitis. Ann Surg 1991; 213: 21-5. 3. Broome U, Olsson R, Loof L, Bodemar G, Hultcrantz R, Danielsson A, et al. Natural history and prognostic factors in 305 Swedish patients with primary sclerosing cholangitis. Gut 1996; 38: 610-5. 4. Snook JA, Kelly P Chapman RW, Jewel1 DP Fibrolamellar hepatocellular carcinoma complicating ulcerative colitis with primary sclerosing cholangitis. Gut 1989; 30: 243-5. 5. Klompmaker IJ, de Bruijn KM, Gouw AH, Barns JL, Slooff MJH. Recurrence of hepatocellular carcinoma after liver retransplantation. Br Med J 1988; 296: 1445. 6. Ismail T, Angrisani L, Hubscher S, McMaster I? Hepatocellular carcinoma complicating primary sclerosing cholangitis. Br J Surg 1991; 78: 360-I. 7. Ludwig J, Ottman DM, Eichmann TJ. The preparation of native livers for morphological studies. Mod Path01 1994; 7: 790-3.
D. M. Harnois et al. 8. Riegler JL. Preneoplastic conditions of the liver. Semin Gastrointest Dis 1996; 7: 7487. 9. Farinati F, Floreani A, De Maria N. Fagiuoli S, Naccarato R, Chiarmonte M. Hepatocellular carcinoma in primary biliary cirrhosis. J Hepatol 1994; 21: 315-6.
10. Loof L, Adami HO, Sparen P, Danielsson A, Eriksson LS, Hultcrantz R, et al. Cancer risk in primary biliary cirrhosis: a population-based study from Sweden. Hepatology 1994; 20: 1014.