Gastroenterology Hepatology

Malignant tumors of the liver

Jump to Section

How can I be sure that the patient has a malignant tumor of the liver?

What primary malignancies occur in the liver?

The most common primary malignancy that occurs in the liver is hepatocellular carcinoma (HCC), which accounts for about 85% of all primary hepatic cancers. The next most frequently occurring malignancy is cholangiocarcinoma. This disease accounts for about 10% to 12% of all hepatic malignancies. The remaining liver primaries are rare and include primary lymphoma of the liver, various sarcomas, and tumors of vascular or lymphatic endothelium.

A tabular or chart listing of features and signs and symptoms

Symptoms that suggest the development of a malignancy in the liver

There are three common presentations. Patients may present with a liver mass found incidentally during investigation for unrelated symptoms or conditions. The most common focal mass in the liver is a hemangioma. This has a typical radiological appearance, so that if the radiologist diagnoses typical hemangioma, no further investigation is required. All other liver masses require aggressive investigation because the earlier a liver malignancy is treated, the better the prognosis. A delay in diagnosis may mean the difference between cure and palliation. A mass in the liver that has declared itself to be malignant by growing under observation without a firm diagnosis being made represents a failure of management.

Patients not known to have liver disease who develop a primary liver cancer may present with ascites, jaundice, or cancer symptoms, such as anorexia, weight loss, and weakness. Sometimes the first presentation is with an intrahepatic or intraperitoneal bleed due to a ruptured hepatocellular carcinoma (HCC). This would present as an acute abdomen with pain and hypotension being the major manifestations.

In patients with known chronic liver disease, particularly in those with cirrhosis, intrahepatic malignancy may present as deterioration in liver function or as complications, such as ascites, encephalopathy, renal failure, or jaundice. Thus, any deterioration in clinical condition in a patient with cirrhosis should trigger imaging of the liver.

How can I confirm the diagnosis?

Radiological diagnosis

The diagnosis of a liver mass is primarily radiological. Hepatocellular carcinoma is the most common hepatic neoplasm, and the diagnostic algorithm has been designed to diagnose HCC noninvasively if possible.

HCC has typical radiological features. These are hypervascularity in the arterial phase of a contrast-enhanced radiological study (CT scan or MRI), followed by washout or hypovascularity compared to the rest of the liver in the portal venous and/or delayed phases.

If the typical features of HCC are present on either contrast-enhanced CT scan or dynamic MRI, a biopsy is not necessary for the diagnosis. If neither study shows typical features, a biopsy of the liver mass is required. However, if the lesion requires resection, whatever it is, there is no need for a biopsy prior to resection.

Histological diagnosis

Histological diagnosis of large liver lesions (>2 cm) is not difficult. If radiology does not confirm the diagnosis, a core biopsy is recommended in most instances, although aspiration biopsies may be sufficient. However, for small lesions, aspiration biopsy is not adequate, and a core biopsy is necessary.

If the lesion is smaller than about 2 cm, the diagnosis may be difficult because the histological features of early HCC are not very different from dysplastic nodules, which are not malignant. The presence of positive staining for two of three markers confirms the diagnosis. These markers are glypican 3, HSP-70, and glutamine synthetase. Such staining should be performed in all cases where the diagnosis of HCC is not certain, such as when the pathologist questions whether the sample was from the lesion.

Comparison of a biopsy from the lesion and a biopsy from surrounding normal liver may also help make a diagnosis of HCC. There are intermediate forms between HCC and cholangiocarcinoma, such as cholangiocellular carcinoma, but not much is known about the clinical aspects of these variants.

What other diseases, conditions, or complications should I look for in patients with malignancies of the liver?


Intrahepatic and extrahepatic cholangiocarcinoma are different diseases, despite similarities in histological appearances. Extrahepatic cholangiocarcinoma is decreasing in incidence, whereas intrahepatic cholangiocarcinoma (CCA) is increasing in incidence. This section will discuss only intrahepatic cholangiocarcinoma, including hilar carcinoma.

Clinical presentation

This is usually a silent cancer until presentation with either jaundice or cancer symptoms. With todays' frequent abdominal imaging for unrelated reasons it is not uncommon for this tumor to be detected incidentally. More often it presents with jaundice, either because of obstruction to the common hepatic ducts, or with an episode of cholangitis, also related to duct obstruction.

Risk factors for HCC are well known, but seldom present. These include infection with liver flukes, Opisthorchis viverini, or Chlornorchis sinensis, Caroli's disease, or choledochal cyst. Cirrhosis also seems to be a risk factor, although the incidence of CCA in cirrhosis is much lower than for HCC. Most cases of CCA arise in patients with no known risk factors. Another predisposing condition is sclerosing cholangitis.

Serological diagnosis

CA19-9, CA 125, and carcinoembryonic antigen are frequently elevated in CCA but are nonspecific. In the presence of a liver mass, the specificity of CA19-9 is about 80%, but it is less specific in patients with sclerosing cholangitis.

Diagnosis of cholangiocarcinoma

The initial diagnosis is usually suspected on the basis of an ultrasound that shows a mass lesion. MRI is the next investigation of choice. It has the best sensitivity for evaluating the extent of disease, including extrahepatic disease. PET scanning has also been used. It is particularly useful in identifying extrahepatic disease, such as regional nodal disease.

Treatment and prognosis

The prognosis of CCA is poor. Resection is the only potentially curative therapy. However, this is only feasible in patients with peripheral lesions, and disease confined to the liver.

Patients with hilar cholangiocarcinoma may undergo resection provided the proximal extent of disease involves only one of the left or right branches of the common hepatic duct but not both. In these patients, the survival is 22% to 44%.

Patients who are not surgical candidates do not respond well to chemotherapy. Gemcitabine and 5-fluoro uracil have been used. Combinations of these agents with other agents have been used, but there is insufficient data to recommend the use of any combination therapy. The survival of these patients is mostly less than 1 year from diagnosis.

What is the right therapy for the patient with malignant tumor of the liver?

Management of patients at risk for hepatocellular carcinoma

Patients at risk for HCC should undergo regular screening. The groups of patients at risk include all patients with cirrhosis and certain noncirrhotic patients with chronic hepatitis B. However, being at risk is not by itself sufficient to enter a patient into a screening program. It depends on the magnitude of risk.

Details of at-risk groups in whom the incidence of HCC is high enough to make screening worthwhile are listed in Table I. The recommended screening protocol is performing an ultrasound every 6 months. Alpha-fetoprotein (AFP) is insufficiently sensitive or specific for routine use as a screening test. Des-gamma-carboxy prothrombin and the L3 fraction of AFP are also not recommended.

Table I. Incidence of HCC in patients at risk
Population group Incidence of HCC
Asian Men >age 40 0.4-0.6%/yr
Asian women >age 50 0.3-0.6%/yr
HBV carrier with family history of HCC Incidence unknown but higher than without family history
Cirrhotic HBV carriers 3-8%/yr
African/North American Blacks Incidence unknown but HCC occurs at a younger age
Hepatitis C cirrhosis 3-8%/yr
Stage 4 primary biliary cirrhosis 3-8%/yr
Genetic hemochromatosis and cirrhosis Unknown, but probably >1.5%/yr
Alpha 1-antitrypsin deficiency and cirrhosis Unknown, but probably >1.5%/yr
Other cirrhosis  
  Unknown, but probably 3-8%/yr

What is the most effective initial therapy?

Selecting the appropriate therapy for patients with HCC

There are two considerations: (1) the anatomical extent of the tumor and (2) the function of the underlying liver disease.

As a general statement, patients with Child's B cirrhosis are not good candidates for any form of therapy. All forms of therapy can be administered to these patients, but the procedure-related mortality and morbidity are higher than those of Child's A cirrhosis. To date, there is no outcome data for any form of treatment of HCC in Child's B cirrhosis (except liver transplantation) that shows an improvement in survival compared to no treatment.

Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.

Resection for HCC

Patients with Child A cirrhosis and insignificant portal hypertension tolerate resection well. Portal hypertension precluding resection includes a measured portal pressure gradient of more than 10 mmHg, esophageal varices, a platelet count of less than 100,000/mL, and a large spleen on imaging. Resection in these patients is associated with a poor prognosis, frequent post-resection ascites, jaundice, and slow deterioration to death.

The ideal tumor for resection is a single lesion in the periphery of the left or right lobe. Unfortunately, few tumors are ideal. The tumor must be removed with at least a 1-cm tumor-free margin. This method usually involves a right or left partial hepatectomy. Occasionally, with poorly situated tumors, a trisegmentectomy can be performed. This method requires excellent liver function. Instead of this, sometimes it is possible to perform a mesohepatectomy: removal of the medial segments of both the right and left lobes. Smaller, well-situated tumors can be removed by a segmentectomy: removal of one or two segments. Laparoscopic hepatectomy is becoming more frequent, with the attendant decrease in post-operative complications.

In the best of hands, the post-operative mortality for hepatectomy even in cirrhosis is less than 1%, as long liver function is preserved and there is no portal hypertension. The long-term survival is about 50% at 5 years. The recurrence rate is 50% to 70% at 5 years.

Liver transplantation for HCC

Liver transplantation has become an accepted form of therapy for HCC. Post-transplant survival in properly selected patients who actually receive a transplant is about 75% to 80% at 5 years. However, if all patients listed for transplantation are counted, the 5-year survival is closer to 65% because of drop-out on the waiting list.

Not all HCCs are suitable for transplantation. Two major issues remain to be resolved. These are 1) criteria for listing for transplantation and 2) downstaging (i.e., treating a tumor that exceeds listing criteria to bring it within criteria).

Listing criteria for liver transplantation for HCC

The classical criteria are the so-called "Milan Criteria." The Milan criteria state that patients whose tumors are smaller than 5 cm, if single, or smaller than 3 cm and no more than 3 in number can be transplanted with a survival rate equivalent to other indications (i.e., about 80% at 5 years).

The Milan criteria were based on pretransplant radiological assessment of tumor extent. However, it is clear that some patients whose tumors exceed Milan criteria can be transplanted with good survival. This has led to the development of several additional criteria. None have yet gained wide acceptance. These include the San Francisco Criteria, the "Metroticket concept," and criteria based on tumor volume, rather than cross-sectional diameter. Most programs still use the Milan Criteria.


Downstaging is the process of treating an HCC that exceeds listing criteria to bring it within criteria. The literature on this topic is very confusing. There is no uniformity in maximum extent of tumor that can be treated, nor in the target size that is acceptable after downstaging so that transplantation has a reasonable survival.

Assessment of downstaging relies largely on the anatomical extent of the tumor, rather than on its biology. Because larger tumors generally have worse prognoses, it is not clear if simply making the tumor smaller reduces the risk of recurrence to the same level as that for tumors within Milan criteria, to start with.

Some clinicians have used the response to chemoembolization as a criterion to assess suitability for transplant. Tumors that respond by shrinkage and show no further growth for 3 months after chemoembolization have a much better prognosis than those tumors that do not respond to treatment.

Contraindications to liver transplantation for HCC

Vascular invasion on pretransplant radiology carries a very high risk of post-transplant recurrence, and is usually a contraindication to transplantation. Poorly differentiated tumor morphology is also an adverse prognostic sign. However, the importance of this criterion in patients whose tumors are within the Milan criteria is not clear. A high AFP (>400 ng/mL) also carries a high risk of post-transplant recurrence. However, it is not clear whether this should be an absolute contraindication to transplantation.

In addition to tumor-specific contraindications, there also are general medical contraindications to major surgery, such as significant heart or lung disease. There are also social contraindications. Patients have to be able to comply with the post-transplant follow-up and medical regimen. In patients who developed liver failure on the basis of alcoholic cirrhosis, most programs require a 6-month period of abstinence before considering transplant.

Local ablation

There are two commonly used forms of local ablation: 1) percutaneous ethanol injection (PEI) and 2) radiofrequency ablation (RFA). Radiofrequency ablation is the preferred option because tumor destruction is more complete, and it takes fewer sessions to achieve this. In randomized controlled trials, RFA is associated with better survival than PEI. Both can be performed as percutaneous outpatient procedures.

The commonly used radiofrequency probes can successfully ablate a lesion of up to 4 cm in diameter. In lesions smaller than 2 cm, complete eradication with recurrence rates of less than 1% at 5 years can be achieved. As the tumor size increases, the ability to achieve complete eradication decreases, so that at about 3 cm and larger, the "cure" rate is about 58%.

Larger lesions can be ablated using multiple probes, but the success rate for these probes is unknown, and there is also no data on survival. It is not certain that achieving a large ablation zone for a large tumor improves survival because the larger the tumor, the more likely that there is at least microvascular invasion, with the risk of metastases.

RFA destroys a rim of normal liver surrounding the tumor, as well as the tumor itself. The larger the tumor, the greater the amount of liver destroyed. The amount of damaged liver increases exponentially, so that for a 2-cm tumor, about 10 cc of normal liver is destroyed; for a 5-cm tumor, about 65 cc of normal liver is destroyed.

Therefore, in patients with marginal liver function, ablation of larger tumors carries a risk of deterioration of liver function. Thus, although RFA can be used in patients with Child's B cirrhosis, it is usually reserved for patients with small tumors (<3 cm) and those who have a Child-Pugh score of 7 to 8, rather than 9 to 10.


Chemoembolization involves inserting a catheter into a branch of the hepatic artery feeding the HCC via the femoral artery and injecting a chemotherapy agent mixed with lipiodol, an oily radiographic contrast agent. This procedure is usually followed by embolizing the artery feeding the tumor. The chemotherapy used is most commonly doxorubicin, cisplatin, or mitomycin C.

Chemoembolization has been used for all but the most advanced stages of disease. However, data on survival from randomized controlled trials comes only in patients with good liver function (Child's class A cirrhosis) and no radiographic evidence of invasion of any of the hepatic veins, portal veins, or bile ducts. Chemoembolization improves survival but at about 12 to 24 months.  

Complications of chemoembolization

The most common complication is the post-embolization syndrome of fever, pain, and nausea. This lasts about 24 to 28 hours. The chemotherapy agent causes bone marrow depression, with a risk of infection during the neutropenic phase.

Patients with biliary-enteric anastomoses are at particularly high risk for infection and should not be treated by chemoembolization. Patients usually develop alopecia. Peripheral biliary strictures are possible, as evidenced by the presence of dilated bile ducts on imaging follow-up.

Patients with chronic hepatitis B may develop a recrudescence of viral replication and, subsequently, of acute or chronic hepatitis. Such patients need to be covered with an antiviral for the duration of the treatment and for up to 3 months subsequently.

Chemobolization in patients with Child B cirrhosis

In a randomized controlled trial that included patients with more advanced liver disease, chemoembolization failed to show any survival advantage. The post-procedure mortality in this study was high, suggesting that patients with more advanced liver disease do not tolerate the procedure well. Therefore, chemoembolization is not recommended for patients with Child B cirrhosis. These patients may do better with other forms of therapy, such as liver transplantation. 

Chemoembolization in patients with vascular invasion

The randomized controlled trials showing a survival advantage of chemoembolization excluded patients with vascular invasion. Cohort studies have shown that patients with vascular invasion have a much shortened survival compared to those without vascular invasion. Therefore, it is not known whether chemoembolization in these patients confers a survival advantage. Nonetheless, chemoembolization has been given in such circumstances.

Most clinicians would avoid chemoembolization in patients with main portal vein obstruction, for fear of infarcting the liver by embolizing the hepatic artery. However, many continue to give chemoembolization to patients with second-order branch portal vein invasion. This is not recommended in the guidelines.

Systemic therapy for HCC

Only one systemic agent has ever been shown to enhance survival in patients with HCC, and that is sorafenib. The improvement in survival is about 3 months compared with untreated patients. The randomized controlled trials that demonstrated this outcome included only patients who had good liver function (Childs' A) and good performance status (WHO stage 1-2). Sorafenib is a multikinase inhibitor that is thought to work by inhibiting angiogenesis and blocking signal transduction in a number of important intracellular pathways.

Conventional chemotherapeutic agents have not been shown to enhance survival significantly and are associated with significant toxicity. Some trials have even demonstrated a decrease in survival in the treated group. This therapy is not recommended.

Infusion chemotherapy, where drug is infused into the hepatic artery via a subcutaneous injection port, is popular in Japan. However, the results are not good. There are no randomized controlled trials of sufficient power to demonstrate a benefit. In some studies, the survival of the treated group is no different from that expected for that stage of disease. This therapy is not recommended.

A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies

Other forms of therapy for HCC

There are several additional forms of treatment that have been devised for HCC. None, however, have been adequately tested.

Radioembolization involves injecting radiolabelled particles (glass beads or resin particles) into the branch of the hepatic artery that feeds the tumor. This treatment has been associated with substantial tumor responses, but it remains to be demonstrated that radioembolization is superior to chemoembolization or other forms of treatment.

Hepatic artery ligation now is seldom used. Bland embolization continues to be used, but there is no satisfactory evidence of efficacy in improving survival. Chemoembolization can be delivered using drug-eluting beads. These have been shown in a single trial to be approximately equivalent to standard chemoembolization but with fewer side effects.

Listing of these, including any guidelines for monitoring side effects.


How should I monitor the patient with hepatocellular carcinoma?

Post-treatment monitoring of patients with HCC

The post-treatment management of patients with HCC is directed at evaluating the response to treatment, presence or absence of recurrence, and progression of liver disease. Monitoring for response and recurrence is usually performed by imaging, usually with the same imaging method that was used in the initial diagnosis. The same diagnostic criteria are used: namely, arterial hypervascularity and venous phase washout. Recurrence is shown by a focus exhibiting these characteristics. In patients who had AFP-secreting tumors, the AFP also can be used to monitor for recurrence. However, because evaluation of the recurrence requires imaging even if AFP is rising, unless the lesion is visible on imaging, nothing can be done. It is therefore not clear that monitoring AFP adds anything.

Liver disease should be monitored by regular blood tests. In addition, patients with cirrhosis should have a gastroscopy to look for esophageal varices and, if present, these should be treated by band ligation.

If the underlying liver disease can be treated, the risk of recurrent disease may be reduced. This applies mainly to hepatitis B and hepatitis C.

What's the evidence?

Yao, FY, Mehta, N, Flemming, J. " Downstaging of hepatocellular cancer before liver transplant: long-term outcome compared to tumors within Milan criteria". Hepatology. 2015. vol. 61. Jun. pp. 1968-77.

(Review of downstaging HCC before transplant, long-term outcomes.)

Venkatesh, SK, Chandan, V, Roberts, LR. "Liver masses: A Clinical, Radiographical and Pathological Perspective". Clin Gastroenterol Hepatol. vol. 12. 2014. pp. 1414-1429.

Marrero, JA, Ahn, J, Rajender Reddy, K. "American College of Gastroenterology. ACG Clinical Guideline: The diagnosis and management of focal liver lesions". Am J Gastroenterol. vol. 109. 2014. pp. 1328-1347.

(ACG guidelines regarding management of focal liver lesions.)

Sangiovannia, A, Manini, MA, Iavaron, M. "The diagnostic and economic impact of contrast imaging techniques in the diagnosis of small hepatocellular carcinoma in cirrhosis". Gut. vol. 59. 2010. pp. 638-44.

(This article is on the radiologic diagnosis of HCC.)

Di Tommaso, L, Destro, A, Seok, JY. "The application of markers (HSP70 GPC3 and GS) in liver biopsies is useful for detection of hepatocellular carcinoma". J Hepatol. vol. 50. 2009. pp. 746-54.

(On histological markers of HCC.)

Zhang, BH, Yang, BH, Tang, ZY. "Randomized controlled trial of screening for hepatocellular carcinoma". J Cancer Res Clin Oncol. vol. 130. 2004. pp. 417-22.

(Evidence of the benefit of HCC screening.)

Forner, A, Reig, ME, de Lope, CR, Bruix, J. "Current strategy for staging and treatment: the BCC update and future prospects". Semin Liver Dis. vol. 30. 2010. pp. 61-74.

(Description of the BCLC staging system that formalizes the approach to management of patients with HCC.)

Mazzaferro, V, Regalia, E, Doci, R. "Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis". N Engl J Med. vol. 334. 1996. pp. 693-9.

(On the Milan criteria.)

Shiina, S, Teratani, T, Obi, S. "A randomized controlled trial of radiofrequency ablation with ethanol injection for small hepatocellular carcinoma". Gastroenterology. vol. 129. 2005. pp. 122-30.

(Evidence that RFA is superior to PEI. This was a randomized controlled trial.)

Llovet, JM, Real, MI, Montana, X. "Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial". Lancet. vol. 359. 2002. pp. 1734-9.

(Evidence that chemoembolization improves survival. This was a randomized trial.)

Llovet, JM, Ricci, S, Mazzaferro, V. "Sorafenib in advanced hepatocellular carcinoma". N Engl J Med. vol. 359. 2008. pp. 378-90.

(Evidence that sorafenib improves survival. This was a randomized controlled trial.)

Ribero, D, Nuzzo, G, Amisano, M. "Comparison of the prognostic accuracy of the sixth and seventh editions of the TNM classification for intrahepatic cholangiocarcinoma". HPB (Oxford). vol. 13. 2011. pp. 198-205.

(Survival in cholangiocarcinoma, treated surgically.)

Rea, DJ, Heimbach, JK, Rosen, CB. "Liver transplantation with neoadjuvant chemoradiation is more effective than resection for hilar cholangiocarcinoma". Ann Surg. vol. 242. 2005. pp. 458-61.

(On liver transplantation for cholangiocarcinoma.)
You must be a registered member of ONA to post a comment.

Sign Up for Free e-newsletters

Regimen and Drug Listings


Bone Cancer Regimens Drugs
Brain Cancer Regimens Drugs
Breast Cancer Regimens Drugs
Endocrine Cancer Regimens Drugs
Gastrointestinal Cancer Regimens Drugs
Genitourinary Cancer Regimens Drugs
Gynecologic Cancer Regimens Drugs
Head and Neck Cancer Regimens Drugs
Hematologic Cancer Regimens Drugs
Lung Cancer Regimens Drugs
Other Cancers Regimens
Rare Cancers Regimens
Skin Cancer Regimens Drugs