Bethesda Handbook of Clinical Oncology, 2nd Edition

Digestive System


Primary Cancers of the Liver

Gregory D. Leonard*

William R. Jarnagin

Carmen J. Allegra

*The Royal College of Physicians, Dublin, Ireland

Department of Surgery Weill Medical College of Cornell University, New York, New York

Network for Medical Communication and Research, Atlanta, Georgia

Primary liver cancers arise predominantly from the parenchymal liver cells or hepatocytes (90%) and are called hepatocellular carcinoma(HCC). Tumors arising from the intrahepatic bile ducts (10%) are called cholangiocarcinomas and are discussed primarily in Chapter 6. HCC is the fifth leading cause of cancer in the world, with an annual incidence of one million cases (1). The incidence in the United States has increased by approximately 75% in the last decade, likely because of increases in incidence of chronic hepatitis infection (2). Research on vaccinations for hepatitis B and its consequences have prevented the development of HCC in many regions of the world. Further progress in the treatment of HCC is necessary to significantly reduce mortality from this common global disease.


  • In the United States, the incidence of clinically significant metastatic carcinoma to the liver is 20 times more common than primary liver cancer.
  • The age-adjusted incidence of HCC from 1996 to 2000 is 5 per 100,000 population ( In 2003, 17,300 (includes intrahepatic bile duct cancers) new cases of primary liver cancer and 14,400 deaths from this disease were estimated in the United States (3).
  • There is marked geographic variation in the incidence of HCC, with the highest incidences of up to 30 per 100,000 population occurring in sub-Saharan Africa and Asia.
  • Men are affected twice as often as women. The mean age at diagnosis is between 50 and 60 years.


  • Cirrhosis is present in 80% of patients with HCC. Therefore, risk factors for cirrhosis are also risk factors for HCC.
  • Hepatitis B virus (HBV) infection increases the risk of developing HCC by 100-fold. HBV causes 80% of HCC in the world. HCC develops from chronic hepatitis due to HBV at a rate of 0.5% per year.
  • Hepatitis C virus (HCV) infection accounts for 30% to 50% of HCC in the United States. HCC develops from HCV at a rate of 5% per year.
  • Alcoholic cirrhosis accounts for 15% of of HCC.
  • Hemochromatosis, hereditary tyrosinemia, and autoimmune chronic active hepatitis are other causes of cirrhosis and are associated with a significant risk for developing HCC (6).


  • There is less convincing evidence for the risk of developing HCC from:
  • Aflatoxin B1(chemical product of Aspergillus)
  • Androgenic steroids
  • Thorotrast (radiology contrast agent) and
  • Oral contraceptives.


  • The most common symptoms or signs of HCC are (7):
  • Pain (91%)
  • Weight loss (35%)
  • Vomiting (8%)
  • Hepatomegaly (89%)
  • Abdominal swelling (43%) and
  • Jaundice (7% to 41%).
  • Paraneoplastic manifestations can also occur. They include hypoglycemia, hypercalcemia, carcinoid, erythrocytosis, hypercholesterolemia, hyperthyroidism, and osteoporosis.


  • Serologic tests may reveal abnormalities in the liver profile. α-Fetoprotein (AFP) is elevated in 50% to 90% of patients with HCC but can also be elevated in other liver abnormalities. AFP values >400 µg per dL in patients without hepatitis or >4,000 µg per mL in patients positive for hepatitis B surface antigen and with radiographic abnormalities suggest the diagnosis of HCC ( If surgery is planned in these patients, an open biopsy at laparotomy is usually preferred to decrease the risk of tumor seeding, bleeding, or rupture. Des-γ-carboxy prothrombin protein, induced by the absence of vitamin K, is increased in about 91% of patients with HCC, but is also elevated in other causes of hepatitis. Carcinoembryonic antigen (CEA) is less useful in HCC and is more likely to be elevated in cholangiocarcinoma.
  • Ultrasonography is considered the best screening tool in conjunction with AFP for high-risk populations. Triphasic computerized tomography (not contrast enhanced, arterial phase, and portal phase) provides better definition, can assess for metastatic disease, and is used to guide percutaneous biopsies. Gadolinium-enhanced magnetic resonance (MR) imaging may improve characterization of small lesions. MR angiography is used to plan for surgery.
  • Laparoscopy is recommended to improve staging and to prevent unnecessary laparotomy (10).


  • There are many histologic types of HCC including trabecular, pseudoglandular or acinar, compact, scirrhous, clear cell, and fibrolamellar.
  • Fibrolamellar carcinoma is a histologic variant accounting for 1% of HCC. It occurs more commonly in women, is not associated with cirrhosis, and has a better prognosis than HCC.


  • The TNM staging system (see Table 7.1) has been criticized because it does not evaluate the underlying liver disease, which is clearly a major prognostic factor in patients regardless of tumor stage (11).
  • The Child-Pugh grading system has been incorporated into the management of HCC because it evaluates the status of the underlying liver function and influences treatment (12) (see Table 7.2).

TABLE 7.1. The American Joint Committee on Cancer (AJCC) 6th edition TNM stage groupings

Stage I

T1 N0 M0

T1: solitary tumor with no vascular invasion

Stage II

T2 N0 M0

T2: solitary tumor with vascular invasion or multiple tumors none >5 cm

Stage IIIA

T3 N0 M0

T3: multiple tumors >5 cm or involving a major branch of the portal or hepatic vein

Stage IIIB

T4 N0 M0

T4: tumor directly invading adjacent organs other than gallbladder or perforating visceral peritoneum

Stage IIIC

Any T N1 M0

N1: regional lymph node metastases

Stage IV

Any T any N M1

M1: distant metastases

TABLE 7.2. Child-Pugh scoring system

Chemical and biochemical parameters

Score attributed to each parameter




Class A = 5–6 points, Class B = 7–9 points, Class C = 10–15 points.









Albumin (g/dL)




Prothrombin time prolonged (s)




Bilirubin (mg/dL)








  • Surgery remains the only possiblity for cure in HCC. The treatment algorithm for HCC is determined by two factors: Tumor extent and the severity of the underlying hepatic parenchymal disease (Fig. 7.1).
  • Partial hepatectomy:Hepatic resection remains the mainstay of surgical therapy for this disease. Only 13 to 35% are surgical candidates. The majority have either disease that is beyond surgical correction or have poor hepatic reserve or both. Small tumors have the best outcomes. Recurrence is most commonly seen in the remnant liver. Repeat hepatectomy is possible in 10 to 29% of patients.

Operative mortality is <5% but is higher in the presence of cirrhosis.

Five-year survival is 30 to 40%, but is lower (12 to 37%) in patients with large tumors, tumors with vascular invasion and more advanced cirrhosis (2).

  • Total hepatectomy and liver transplantaion:Transplantation is indicated in patients with severe cirrhosis or where extensive resection leaving minimal liver reserve is required.

Criteria for transplantation are: a solitary tumor ≤5cm, 2 or 3 tumors ≤3cm in size, and the absence of vascular invasion. Based on these criteria four-year survival was reported as 75%, however, 12 of the 14 non-operative deaths were due to transplant complications as opposed to tumor recurrence (13). Patients with intrahepatic cholangiocarcinoma are not considered transplant candidates as results with transplantation in this disease have been poor.

Survival outcomes often appear more favorable compared to resection but may be due to patient selection, inclusion of incidental tumors, and lack of intention to treat analyses where survival outcomes are poorer due to disease progression while waiting for transplant (10). Survival outcomes may be further improved by living donor transplantation, although this remains controversial.



Disadvantages of transplantation are the expense, the lack of specialty centers performing operations and the lack of donor livers (currently 18,505 patients on the waiting list, 40% have been waiting for over 2 years).

Preoperative chemoembolization is often used as a temporizing modality while waiting for a liver, but the value of this approach is not yet proven.


FIG. 7.1. Treatment algorithm for the management of hepatocellular carcinoma (HCC).

Ablative Techniques

Ablative techniques have a role in patients who are not candidates for resection (12). Ablation of solitary, small lesions may be as effective as resection but a prospective comparison between surgery and one of these methods has yet to be performed. Ablation of large tumors (>5cm) has been shown to be associated with very high local recurrence rates.

  • Percutaneous ethanol injection into tumors causes cellular dehydration, coagulative necrosis and localized tissue ischemia. It is frequently used for up to three localized tumors of <5cm that are not surgical candidates usually due to cirrhosis. It is relatively inexpensive and well tolerated.
  • Radiofrequency ablation is performed percutaneously using ultrasound guidance and causes focal coagulative necrosis of tumors via thermal energy. It is most efficacious for tumors <3cm where complete necrosis can occur in up to 90% of tumors.
  • Cryotherapy is also safe and more effective than RFA for larger tumors but is less suited to a percutaneous approach.
  • Hepatic artery chemoebolization is based on the principal that 80% of the blood supply to tumors is from the hepatic artery which supplies only 20–30% of normal liver parenchyma. Ligation or embolization of the hepatic artery can induce temporary tumor responses but when combined with chemotherapy can be more efficacious. Further discussion on this topic is in the chemotherapy section.




  • External beam radiotherapy has a limited role in HCC owing to the poor tolerance of radiation by the liver. However, safe and effective doses can be given to palliate the pain.
  • Radioactive isotopes have demonstrated efficacy in the adjuvant treatment of HCC.

Iodine-131 combined with lipiodol (a dye used for lymphangiography that is known to concentrate in tumors when given by hepatic arterial infusion) given by the hepatic artery has been compared to the observation in 43 patients in Hong Kong who had curative resections for HCC (12). At interim analysis, the study was closed owing to an improved disease-free survival from 13.6 months to 57.2 months with the radioactive isotope. This provocative study has yet to be validated in a multicenter trial.


  • Single-agent chemotherapy has demonstrated response rates of approximately 15% to 30%, which increases to 20% to 35% with combination therapy. Cisplatin and anthracycline combinations have been studied most extensively, but there is no reference regimen for this disease (14).
  • Although most chemotherapy trials are in phase II, some trials have demonstrated a statistically significant survival advantage compared to resection alone (15) (see Table 7.3). Chemotherapy is usually used for palliation in patients with unresectable disease and should only be used for adjuvant therapy as part of a clinical trial.
  • The PIAF regimen (intravenous cisplatin, recombinant interferon alfa 2b, doxorubicin, and 5-fluorouracil) demonstrated a 50% objective response rate in 50 patients with unresectable disease from Hong Kong (16). Nine of the 13 responding patients were operated on and four were found to have complete pathological remission. In addition to highlighting a potentially active regimen, this trial also demonstrates that radiologic assessment may underestimate tumor cell kill.

TABLE 7.3. Selection of trials demonstrating a survival advantage with the use of chemotherapy





1-, 2-, and 3-yr survival


HAI, hepatic arterial infusion; 5-FU, 5-fluorouracil.
Median overall survival.

Lai et al. (17)




11 wk





7 wka

Nakashima et al. (18)



HAI cisplatin or HAI 5-FU, doxorubicin and mitomycin C

63%, 50%, 7%





39%, 22%,12%

Lai et al. (19)



Epirubicin, HAI iodized oil with cisplatin

69%, 53%, 48%



50%, 36%, 18%



Regional Chemotherapy

  • Hepatic arterial infusion chemotherapy acts on the premise that tumor blood supply is by the hepatic artery. Response rates are often higher with this therapy. Trials comparing efficacy of various systemic chemotherapies are contradictory. At the University of California, less than 10% of patients with HCC were candidates for surgical placement of infusion pumps. Pump insertion requires a laparotomy, is expensive, and is performed only in certain specialized centers.
  • Chemoembolization involves regional administration of chemotherapy followed by embolization of specific arteries using gelatin sponges, collagen, alcohol, or microspheres. Embolization can cause tumor ischemia and increases the potential for improved chemotherapeutic exposure to the tumor. Patients with portal vein thrombosis or poor liver reserve are usually excluded because normal liver parenchyma relies on the hepatic arterial blood supply. Lipiodol is commonly used for embolization because it can transport lipophilic chemotherapeutic agents and can be radiolabeled. No trials have compared the outcomes for different chemotherapeutic administration techniques in HCC.

Prevention and Novel Therapies

  • Tamoxifen, antiandrogen therapy, and octreotide have been extensively investigated and do not have a beneficial effect on patients with HCC. HMG-CoA reductase inhibitors such as pravastatin have had more encouraging results (20).
  • Interferon-α reduces the onset of liver damage and its progression to cirrhosis in 10% to 30% of patients with chronic hepatitis B.
  • Refrigerated storage of food grains and transportation of grains in refrigerated vehicles should help reduce the risk of ingesting aflatoxin.
  • Acyclic retinoid, polyprenoic acid, reduces the incidence of second primary of HCC after initial resection and requires further investigation.
  • Screening of high-risk populations with with α-fetoprotein at 4-month intervals and with ultrasound at yearly intervals has been shown to identify patients with earlier stages of HCC and may improve survival in high-risk groups (21).


  1. Jemal A, Murray T, Samuels A, et al. Cancer Statistics, 2003. CA Cancer J Clin2003;53:5–26.
  2. Cha CH, Ruo L, Fong Y, et al. Resection of hepatocellular carcinoma in patients otherwise eligible for transplantation. Ann Surg2003;238:315–323.
  3. Schafer DF, Sorrell MF. Hepatocellular carcinoma. Lancet1999;353:1253–1257.
  4. Valea FA. Liver and hepatic duct cancer. Clin Obstet Gynecol2002;45:939–951.
  5. El S, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med1999;340:745–750.
  6. Okuda K, Obata H, Nakajima Y, et al. Prognosis of primary hepatocellular carcinoma. Hepatology1984;4S:3S.
  7. American Joint Committee on Cancer. Cancer staging manual, 6th ed. New York: Springer-Verlag, 2002:131–138.
  8. Pugh RN, Murray L, Dawson JL, et al. Transection of the esophagus for bleeding esophageal varices. Br J Surg1973;60:646–649.
  9. Bismuth H, Chiche L, Adam R, et al. Liver resection versus transplantation for hepatocellular carcinoma in cirrhotic patients. Ann Surg1993;218:145–151.
  10. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment of early hepatocellular carcinoma: resection versus transplantation. Hepatology1999;30:1434–1440.
  11. Lau W, Leung W, Ho S, et al. Adjuvant intraarterial iodine-131-labelled lipiodol for respectable hepatocellular carcinoma. A prospective randomized trial. Lancet1999;353:797–801.
  12. Venook AP. Regional strategies for managing hepatocellular carcinoma. Oncology2000;14:347–354.



  1. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med1996;334:693–699.
  2. Johnson PJ. Are there indications for chemotherapy in hepatocellular carcinoma. Surg Oncol Clin North Am2003;12:127–134.
  3. Haskell CM. Liver: chemotherapy. In: Haskell CM, ed. Cancer treatment, 5th ed. Philadelphia, PH: W. B. Saunders, 2001:778–785.
  4. Leung TWT, Patt YZ, Lau WY, et al. Complete pathological remission is possible with systemic combination chemotherapy for inoperable hepatocellular carcinoma. Clin Cancer Res1999;5:1676–1681.
  5. Lai CL, Wu PC, Chan GC, et al. Doxorubicin versus no antitumor therapy in inoperable hepatocellular carcinoma. A prospective randomized trial. Cancer1988;62:479–483.
  6. Nakashima K, Kitano S, Kim YI, et al. Postoperative adjuvant arterial infusion chemotherapy for patients with hepatocellular carcinoma. Hepatogastroenterology1996;43:1410–1414.
  7. Lai EC, Lo CM, Fan ST, et al. Postoperative adjuvant chemotherapy after curative resection of hepatocellular carcinoma: a randomized controlled trial. Arch Surg1998;133:183–188.
  8. Kawata S, Yamasaki E, Nagase T, et al. Effect of pravastatin on survival in patients with advanced hepatocellular carcinoma. A randomized controlled trial. Br J Cancer2001;84:886–891.
  9. Yang B, Zhang B, Xu Y, et al. Prospective study of early detection for primary liver cancer. J Cancer Res Clin Oncol1997;123:357–360.