Hepatocellular Carcinoma-52nd Annual Meeting of AASLD

[Guest guest]

American Association for the Study of Liver Diseases 52nd Annual Meeting

[Medscape, 2001. © 2001 Medscape, Inc.]

Hepatocellular Carcinoma

M. Di Bisceglie, MD, FACP

Pathogenesis of Hepatocellular Carcinoma (HCC)

There are several well-established animal models of hepatocarcinogenesis,

most based on the administration of chemical carcinogens. The pathway to

cancer can be carefully traced in these animals through the appearance of

benign adenomas progressing to HCC. Oval cells have a distinct histologic

appearance within the liver of animals treated with carcinogens and seem to

be important in development of HCC. However, similar cells have not been

clearly identified in human liver tissue.

Tsamandas and colleagues[1] searched for oval cells or their equivalent in

human liver tissue of patients with chronic liver disease and HCC. Oval

cells were detected by in situ hybridization for alpha-fetoprotein mRNA and

immunostaining for CK19, CK7, LCA, and CD34 (markers that reliably identify

oval cells in animals). These investigators were able to identify oval

cells, and found that their presence correlated both with liver disease

severity and the presence of HCC.

This is an area of disease pathogenesis that warrants additional study

because it may allow risk stratification for development of HCC in

individuals with chronic viral hepatitis.

Etiologic Factors in the Development of HCC

Nonalcoholic Steatohepatitis

There is growing evidence of a role for nonalcoholic steatohepatitis (NASH)

in the pathogenesis of HCC. It has only recently been suggested that a

substantial proportion of patients with cryptogenic cirrhosis have clinical

and demographic features that match those of patients with NASH -- in

particular, diabetes and obesity. Cryptogenic cirrhosis accounts for about

10% of liver transplants in the United States.

Leone and colleagues[2] assessed risk factors for NASH in patients with HCC

in Italy. Out of a cohort of 641 patients with HCC and cirrhosis, they

identified 44 in whom there was no obvious cause for the cirrhosis. Of

these, 23 were available for their study.

The study authors then compared these patients with 115 controls, some with

cirrhosis due to chronic hepatitis C (n = 46), some with chronic hepatitis

B-related cirrhosis (n = 46), and some with alcoholic cirrhosis (n = 23).

The group with HCC and cryptogenic cirrhosis had significantly greater body

mass index and frequency of diabetes than the other groups.

Primary Biliary Cirrhosis

Although it is well known that cirrhosis predisposes to HCC, it has been

suggested that some forms of cirrhosis are more likely to lead to liver

cancer than others. In particular, cirrhosis due to 's disease and

primary biliary cirrhosis (PBC) is believed to carry a lower risk of HCC.

Investigators from a center in Japan reported on their experience with HCC

occurring in the setting of PBC.[3] Among a cohort of 396 patients with PBC,

14 (3.5%) developed HCC. All 14 had advanced liver disease at the time of

HCC appearance (stages III or IV). The median age at presentation with HCC

was 72 years. The study authors were able to calculate that the annual rate

of HCC was 0.003 per person-year of follow-up in patients with mild PBC, and

0.015 cases per person-year in advanced PBC.

Role of HIV Coinfection in Patients With Hepatitis C Cirrhosis

The role of HIV infection in modulating the risk of HCC in patients with

chronic viral hepatitis is not known. HIV infection, when advanced, is known

to predispose to several forms of malignancy, including B-cell lymphoma,

Kaposi's sarcoma, and squamous-cell carcinoma of the mouth.

Di o and coworkers[4] compared the clinical outcomes between 54

patients with chronic hepatitis C and HIV infection, and 391 patients with

hepatitis C alone. A total of 27 patients in this cohort as a whole

developed HCC. Interesting to note, however, was that only 7% of the

HIV-coinfected group developed HCC compared with 17% of those with hepatitis

C virus (HCV) infection alone, a difference that reached statistical

significance. What makes this observation even more fascinating is that the

HIV+ group had more severe liver disease, with greater risk of ascites,

portosystemic encephalopathy, and liver-related death.

Porphyria Cutanea Tarda

Porphyria cutanea tarda (PCT) is an acquired metabolic disease associated

with abnormal production of porphyrins. The conditions is characterized

clinically by photosensitivity resulting in blistering lesions on the hands,

as well as facial hirsutism. It is associated with underlying fibrotic liver

disease and iron accumulation. As many as 70% of these cases are associated

with HCV infection. It has been known for some time that PCT predisposes to

the development of HCC.

Rossini and colleagues[5] from Italy investigated whether PCT associated

with chronic viral hepatitis carried a different risk for HCC than PCT

associated with other liver diseases. They examined a cohort of 63 patients

with PCT, 10 of whom developed or had PCT at the time of presentation. All

10 (100%) of those with PCT and HCC had chronic viral hepatitis, whereas

only 33 of the 53 cases without HCC were related to chronic viral hepatitis.

These data suggest that it is not only the presence of underlying liver

disease in PCT that leads to HCC, but viral factors may also play a more

direct role.

Diagnosis of HCC

Imaging studies of the liver and determination of alpha-fetoprotein levels

in serum have been important tools in diagnosing HCC. However, neither

imaging nor serum alpha-fetoprotein have perfect sensitivity and

specificity. Only about 60% to 70% of patients with HCC in the developed

western world have increased serum levels of alpha-fetoprotein, and

unfortunately, levels may also be substantially increased in association

with active chronic viral hepatitis and cirrhosis. For some time now,

researchers in Japan have reported des-gamma carboxy prothrombin (DCP) to

have added predictive value over alpha-fetoprotein in diagnosing HCC.

However, this marker of HCC has been poorly studied in developed western

countries.

Marrrero and colleagues[6] from the University of Michigan compared the

utility of serum alpha-fetoprotein and DCP measurement in the diagnosis of

HCC in a cohort of 34 patients with HCC. The study authors compared the

findings in 95 controls with chronic liver disease without HCC as well as 38

normal, healthy individuals. A serum DCP level > 100 U/mL was found to have

good sensitivity and specificity for HCC as the mean value in these patients

with a 20,682 U/mL level, and only 2 of the 34 patients with HCC had levels

< 100 U/mL.

Treatment of HCC

The role of chemoembolization in the treatment of HCC is controversial. This

technique involves destruction of tumor tissue within the liver based on

occlusion of the segmental hepatic artery supplying the tumor. Usually

chemotherapeutic agents such as mitomycin C, doxorubicin, and cisplatinum

are given intra-arterially at the same time in a slurry with Lipiodol and a

water-soluble contrast agent. A randomized, controlled trial of

chemoembolization vs symptomatic treatment for unresectable HCC performed a

number of years ago showed no survival advantage, despite the fact that

chemoembolization clearly resulted in tumor shrinkage.

Thus, Llovet and colleagues[7] from Barcelona, Spain, conducted a

meta-analysis of several controlled trials of chemoembolization and found

this therapy effective overall in prolonging survival. As a part of this

meta-analysis, they included data from their own (as yet incomplete)

randomized, controlled trial, which involved 112 patients with HCC

randomized to 1 of 3 arms: (1) transarterial embolization, (2) transarterial

chemoembolization (TACE), or (3) untreated control. There were 35-40

patients in each group. An interim analysis revealed a significant survival

effect for the TACE group, resulting in this trial being discontinued.

These data are important because they will serve as an impetus for the

continued use of TACE in this setting. One crucial area for further study is

the application of TACE for tumor ablation prior to liver transplantation.

References

1.. Tsamandas AC, Gogos C, Kourelis T, et al. Oval hepatocytes express

alpha-fetoprotein and correlate with disease severity and presence of

hepatocellular carcinoma in cases of chronic hepatitis type B or C.

Hepatology. 2001;34:392A. [Abstract 879]

2.. Leone N, Bugianesi E, Carucci P, et al. Does nonalcoholic

steatohepatitis (NASH) progress to cryptogenic cirrhosis (CC) and

hepatocellular carcinoma (HCC): A case control study. Hepatology.

2001;34:251A. [Abstract 310]

3.. Shibuya A, Tanaka K, Shibata M, Miyakawa H, Morizane T. Hepatocellular

carcinoma and survival in patients with primary biliary cirrhosis.

Hepatology. 2001;34:372A. [Abstract 799]

4.. Di o V, Cavallero L, Ezenfis J, et al. Impact of HIV coinfection

on the age and the cause of death in patients with HCV cirrhosis.

Hepatology. 2001;34:446A. [Abstract 1095]

5.. Rossini A, Leali C, Contessi GB, Cariani E, Radaeli E. Hepatocellular

carcinoma in patients with porphyria cutanea tarda: role of viral

infections. Hepatology. 2001;34:182A. [Abstract 34]

6.. Marrero JA, Gosh M, Nour K, et al. Des-gamma carboxy prothrombin is a

more sensitive and reliable marker for hepatocellular carcinoma than

alpha-fetoprotein in American patients. Hepatology. 2001;34:177A. [Abstract

14]

7.. Llovet JM, Bruix J. Chemoembolization of unresectable hepatocellular

carcinoma: Meta-analysis of polled data. Hepatology. 2001;34:180A. [Abstract

27]

8.. Johlin FC, Mitros FA, Voigt MD, Wu YM, Panther MK, Jensen CS.

Surveillance cytology index ERCP for primary sclerosing cholangitis patients

detects asymptomatic cancers and pre-cancerous dysplasia. Hepatology.

2001;34:365A. [Abstract 771]

[Guest guest]

American Association for the Study of Liver Diseases 52nd Annual Meeting

[Medscape, 2001. © 2001 Medscape, Inc.]

Hepatocellular Carcinoma

M. Di Bisceglie, MD, FACP

Pathogenesis of Hepatocellular Carcinoma (HCC)

There are several well-established animal models of hepatocarcinogenesis,

most based on the administration of chemical carcinogens. The pathway to

cancer can be carefully traced in these animals through the appearance of

benign adenomas progressing to HCC. Oval cells have a distinct histologic

appearance within the liver of animals treated with carcinogens and seem to

be important in development of HCC. However, similar cells have not been

clearly identified in human liver tissue.

Tsamandas and colleagues[1] searched for oval cells or their equivalent in

human liver tissue of patients with chronic liver disease and HCC. Oval

cells were detected by in situ hybridization for alpha-fetoprotein mRNA and

immunostaining for CK19, CK7, LCA, and CD34 (markers that reliably identify

oval cells in animals). These investigators were able to identify oval

cells, and found that their presence correlated both with liver disease

severity and the presence of HCC.

This is an area of disease pathogenesis that warrants additional study

because it may allow risk stratification for development of HCC in

individuals with chronic viral hepatitis.

Etiologic Factors in the Development of HCC

Nonalcoholic Steatohepatitis

There is growing evidence of a role for nonalcoholic steatohepatitis (NASH)

in the pathogenesis of HCC. It has only recently been suggested that a

substantial proportion of patients with cryptogenic cirrhosis have clinical

and demographic features that match those of patients with NASH -- in

particular, diabetes and obesity. Cryptogenic cirrhosis accounts for about

10% of liver transplants in the United States.

Leone and colleagues[2] assessed risk factors for NASH in patients with HCC

in Italy. Out of a cohort of 641 patients with HCC and cirrhosis, they

identified 44 in whom there was no obvious cause for the cirrhosis. Of

these, 23 were available for their study.

The study authors then compared these patients with 115 controls, some with

cirrhosis due to chronic hepatitis C (n = 46), some with chronic hepatitis

B-related cirrhosis (n = 46), and some with alcoholic cirrhosis (n = 23).

The group with HCC and cryptogenic cirrhosis had significantly greater body

mass index and frequency of diabetes than the other groups.

Primary Biliary Cirrhosis

Although it is well known that cirrhosis predisposes to HCC, it has been

suggested that some forms of cirrhosis are more likely to lead to liver

cancer than others. In particular, cirrhosis due to 's disease and

primary biliary cirrhosis (PBC) is believed to carry a lower risk of HCC.

Investigators from a center in Japan reported on their experience with HCC

occurring in the setting of PBC.[3] Among a cohort of 396 patients with PBC,

14 (3.5%) developed HCC. All 14 had advanced liver disease at the time of

HCC appearance (stages III or IV). The median age at presentation with HCC

was 72 years. The study authors were able to calculate that the annual rate

of HCC was 0.003 per person-year of follow-up in patients with mild PBC, and

0.015 cases per person-year in advanced PBC.

Role of HIV Coinfection in Patients With Hepatitis C Cirrhosis

The role of HIV infection in modulating the risk of HCC in patients with

chronic viral hepatitis is not known. HIV infection, when advanced, is known

to predispose to several forms of malignancy, including B-cell lymphoma,

Kaposi's sarcoma, and squamous-cell carcinoma of the mouth.

Di o and coworkers[4] compared the clinical outcomes between 54

patients with chronic hepatitis C and HIV infection, and 391 patients with

hepatitis C alone. A total of 27 patients in this cohort as a whole

developed HCC. Interesting to note, however, was that only 7% of the

HIV-coinfected group developed HCC compared with 17% of those with hepatitis

C virus (HCV) infection alone, a difference that reached statistical

significance. What makes this observation even more fascinating is that the

HIV+ group had more severe liver disease, with greater risk of ascites,

portosystemic encephalopathy, and liver-related death.

Porphyria Cutanea Tarda

Porphyria cutanea tarda (PCT) is an acquired metabolic disease associated

with abnormal production of porphyrins. The conditions is characterized

clinically by photosensitivity resulting in blistering lesions on the hands,

as well as facial hirsutism. It is associated with underlying fibrotic liver

disease and iron accumulation. As many as 70% of these cases are associated

with HCV infection. It has been known for some time that PCT predisposes to

the development of HCC.

Rossini and colleagues[5] from Italy investigated whether PCT associated

with chronic viral hepatitis carried a different risk for HCC than PCT

associated with other liver diseases. They examined a cohort of 63 patients

with PCT, 10 of whom developed or had PCT at the time of presentation. All

10 (100%) of those with PCT and HCC had chronic viral hepatitis, whereas

only 33 of the 53 cases without HCC were related to chronic viral hepatitis.

These data suggest that it is not only the presence of underlying liver

disease in PCT that leads to HCC, but viral factors may also play a more

direct role.

Diagnosis of HCC

Imaging studies of the liver and determination of alpha-fetoprotein levels

in serum have been important tools in diagnosing HCC. However, neither

imaging nor serum alpha-fetoprotein have perfect sensitivity and

specificity. Only about 60% to 70% of patients with HCC in the developed

western world have increased serum levels of alpha-fetoprotein, and

unfortunately, levels may also be substantially increased in association

with active chronic viral hepatitis and cirrhosis. For some time now,

researchers in Japan have reported des-gamma carboxy prothrombin (DCP) to

have added predictive value over alpha-fetoprotein in diagnosing HCC.

However, this marker of HCC has been poorly studied in developed western

countries.

Marrrero and colleagues[6] from the University of Michigan compared the

utility of serum alpha-fetoprotein and DCP measurement in the diagnosis of

HCC in a cohort of 34 patients with HCC. The study authors compared the

findings in 95 controls with chronic liver disease without HCC as well as 38

normal, healthy individuals. A serum DCP level > 100 U/mL was found to have

good sensitivity and specificity for HCC as the mean value in these patients

with a 20,682 U/mL level, and only 2 of the 34 patients with HCC had levels

< 100 U/mL.

Treatment of HCC

The role of chemoembolization in the treatment of HCC is controversial. This

technique involves destruction of tumor tissue within the liver based on

occlusion of the segmental hepatic artery supplying the tumor. Usually

chemotherapeutic agents such as mitomycin C, doxorubicin, and cisplatinum

are given intra-arterially at the same time in a slurry with Lipiodol and a

water-soluble contrast agent. A randomized, controlled trial of

chemoembolization vs symptomatic treatment for unresectable HCC performed a

number of years ago showed no survival advantage, despite the fact that

chemoembolization clearly resulted in tumor shrinkage.

Thus, Llovet and colleagues[7] from Barcelona, Spain, conducted a

meta-analysis of several controlled trials of chemoembolization and found

this therapy effective overall in prolonging survival. As a part of this

meta-analysis, they included data from their own (as yet incomplete)

randomized, controlled trial, which involved 112 patients with HCC

randomized to 1 of 3 arms: (1) transarterial embolization, (2) transarterial

chemoembolization (TACE), or (3) untreated control. There were 35-40

patients in each group. An interim analysis revealed a significant survival

effect for the TACE group, resulting in this trial being discontinued.

These data are important because they will serve as an impetus for the

continued use of TACE in this setting. One crucial area for further study is

the application of TACE for tumor ablation prior to liver transplantation.

References

1.. Tsamandas AC, Gogos C, Kourelis T, et al. Oval hepatocytes express

alpha-fetoprotein and correlate with disease severity and presence of

hepatocellular carcinoma in cases of chronic hepatitis type B or C.

Hepatology. 2001;34:392A. [Abstract 879]

2.. Leone N, Bugianesi E, Carucci P, et al. Does nonalcoholic

steatohepatitis (NASH) progress to cryptogenic cirrhosis (CC) and

hepatocellular carcinoma (HCC): A case control study. Hepatology.

2001;34:251A. [Abstract 310]

3.. Shibuya A, Tanaka K, Shibata M, Miyakawa H, Morizane T. Hepatocellular

carcinoma and survival in patients with primary biliary cirrhosis.

Hepatology. 2001;34:372A. [Abstract 799]

4.. Di o V, Cavallero L, Ezenfis J, et al. Impact of HIV coinfection

on the age and the cause of death in patients with HCV cirrhosis.

Hepatology. 2001;34:446A. [Abstract 1095]

5.. Rossini A, Leali C, Contessi GB, Cariani E, Radaeli E. Hepatocellular

carcinoma in patients with porphyria cutanea tarda: role of viral

infections. Hepatology. 2001;34:182A. [Abstract 34]

6.. Marrero JA, Gosh M, Nour K, et al. Des-gamma carboxy prothrombin is a

more sensitive and reliable marker for hepatocellular carcinoma than

alpha-fetoprotein in American patients. Hepatology. 2001;34:177A. [Abstract

14]

7.. Llovet JM, Bruix J. Chemoembolization of unresectable hepatocellular

carcinoma: Meta-analysis of polled data. Hepatology. 2001;34:180A. [Abstract

27]

8.. Johlin FC, Mitros FA, Voigt MD, Wu YM, Panther MK, Jensen CS.

Surveillance cytology index ERCP for primary sclerosing cholangitis patients

detects asymptomatic cancers and pre-cancerous dysplasia. Hepatology.

2001;34:365A. [Abstract 771]