Published: October 2013
Many cancers develop in or spread to the human liver. This chapter deals solely with hepatocellular cancer (HCC). (Figure 1)
HCC is one of the most prevalent cancers in the world (Figure 2). It is predominantly a high-risk cancer in less-developed areas of the world. Although the incidence remains relatively low in the United States, unlike many major cancers, the incidence of HCC has doubled in the past 20 years. Throughout the world, men are 2 to 3.5 times as likely as women to develop HCC. Hepatitis B or C is a major association. Aflatoxin exposure in Africa and Asia, oral contraceptive use in Europe and the United States, and alcohol use throughout the world are associated with increased risk. Tobacco use in some but not all studies has also been found to be a risk factor.
Migrants from high endemic areas bring with them a higher risk of HCC compared to those in the host country. The relative risk of HCC in a Chinese immigrant to North America is 3.3 to 11 times higher than in a person born in North America. Similarly, a West African immigrant to England or Wales carries a 32-fold increased risk of HCC compared to a native-born person. This transferable risk is largely associated with the high prevalence of chronic hepatitis B, most often acquired at birth or in early childhood, seen in high incidence in Africa and Asia.
Pathogenesis of HCC is incompletely understood. Much evidence supports the notion that DNA damage occurs, resulting in deregulation of DNA methylation, chromosomal instability, proto-oncogene activation, and tumor suppressor gene inactivation. RAS signaling pathways are observed to be activated, and this serves to activate cell proliferation.
HCC most often occurs in the setting of chronic liver disease, and many cases, particularly in economically developed countries, are found in patients with cirrhosis. Prominent risk factors for HCC include chronic viral hepatitis B or C and alcohol-related liver disease. Cirrhosis of any cause increases the risk of HCC. An emerging threat, therefore, comes from the obesity epidemic, which predisposes to nonalcoholic liver disease and cirrhosis. A recent study by our institution showed that the yearly cumulative incidence of HCC is 2.6% per year in patients with cirrhosis secondary to fatty liver disease, compared to 4.0% per year in patients with Hepatitis C cirrhosis.
In less-developed countries, particularly in tropical and subtropical climates, aflatoxin exposure is a promoter of HCC. Aflatoxins are mycotoxins produced by fungi of the genus Aspergillus, which are commonly present in soil and as contaminants of improperly stored nuts, cereals, and other produce. Other risk factors for HCC include inherited disorders such as tyrosinemia and hemochromatosis.
Precursor manifestations of HCC may be the development of dysplastic nodules. Until recently, the risk of subsequent HCC in patients in whom dysplasia was reported on liver biopsy was controversial. A recent study reports an odds ratio of 3.2 for subsequent HCC in those with dyslasia (95% confidence interval, 1.7-6.2).1
Small HCC produces no symptoms. Discovery of HCC when symptoms are present provides little value to the patient. Systemic symptoms of cancer such as anorexia, unintended weight loss, and local symptoms such as right upper quadrant pain almost guarantee the disease is untreatable. Often HCC manifests in the cirrhotic patient as worsening of cirrhotic decompensation, such as worsening ascites, hepatic encephalopathy, or gastrointestinal bleeding. Given the lack of effective therapies for advanced or recurrent HCC, early diagnosis is the goal because it allows patients to undergo treatment with good long-term survival (70% at 5 years). Unfortunately, despite earlier detection of HCC, only 20% to 30% of patients are candidates for either surgical or locoregional therapies.
Imaging study and biopsy are the most useful diagnostic tools. Serum tumor markers are occasionally quite useful. Additional details are discussed later. Discovery of small HCCs is possible through application of screening programs for those at high risk.
In high-risk areas such as Hong Kong, discovery of HCC in adults with HBV (only 11% of whom had clinical cirrhosis) was nearly 7% after 5 years of screening.2 Multivariate analysis showed that cirrhosis and genotype C HBV was most often associated with the development of HCC. Translation of such yields from high-risk to lower-risk geographic regions must be done with caution. Hepatologists agree on the value of screening, even though rigorous scientific evidence in support of such as a cost-effective lifesaving measure is lacking. Limited cost-to-benefit analyses using modeling methodologies suggest screening is reasonable. Using a standard threshold of utility of less than $50,000 per quality-adjusted life-year (QALY), costs for HCC screening range from $26,000 to $55,000.3
A recent practice guideline recommends screening for hepatitis B carriers who are Asian men 40 years of age or older, Asian women 50 years of age or older, or Africans older than 20 years, or who have cirrhosis, a family history of HCC, a high HBV viral load, or current or past severe inflammatory liver disease.4 The guideline recommends screening for patients who have non-hepatitis B cirrhosis due to hepatitis C, alcohol, primary biliary cirrhosis, α1-antitrypsin deficiency, nonalcoholic steatohepatitis, autoimmune hepatitis, or iron overload.4 Evidence suggests that screening for HCC is inconsistently performed in the United States. Data from Medicare recipients suggests that among more than 3900 patients with HCC, screening had been performed on 25%, and of these, the alpha fetoprotein (AFP) alone was used most often. Patients with known hepatitis B, hepatitis C, or cirrhosis and those who were younger or female were most likely to have been screened. Patients who had been seen by a gastroenterologist were 60% more likely to have received a screening test. These data, if confirmed, suggest that knowledge about screening recommendations has been ineffectively disseminated, particularly to the general medical community.5
The mainstay of screening for HCC is liver imaging. Many modalities exist (radioisotope, ultrasound, CT, MRI, PET scanning). Considering sensitivity, cost, and repeated exposure to ionizing radiation, ultrasonography is currently the preferred screening imaging test. Optimal frequency of screening in at-risk populations is not defined by data from randomized, controlled trials. Experts recommend a frequency of no less often than yearly; some suggest intervals of every month.
Nodules are often described in cirrhotic patients without cancer. Most small nodules represent regenerative nodules, some hemangiomas, and other benign growths. Nodules smaller than 1 cm should be followed up with ultrasound at intervals of between 3 and 6 months; if after 2 years the lesion is stable, reversion to routine screening is recommended. When a lesion is 1 to 2 cm in size, further evaluation with either CT or MRI should be considered.
Serum-based tests are quite disappointing as a sole screening tool. AFP is the most commonly used serum marker. The upper limit of normal is typically 10 ng/mL. Values over 200 require further study, and values over 400 in an adult are almost certainly caused by HCC. In most cases of HCC, the AFP levels are normal or only slightly elevated. In addition, patients with chronic viral hepatitis B or C often have AFP levels of up to 100 without evidence of identifiable HCC. The sensitivity and specificity of AFP in detecting measurable HCC is low. Specificity is very high when the AFP value is greater than 400 ng/mL.
Some circulating AFP is in a glycosolated form. Measurement of the glycosolated AFP (L3 fraction) as a percentage of total AFP has been reported to add specificity to AFP. Another protein, DCP (des-γ-carboxy prothrombin), also known as Prothrombin Induced by Vitamin K Absence II (PIVKA-II), and vascular endothelial growth factor receptor might also serve as a serum marker for HCC. The clinical utility of adding these tests to an HCC screening program has not yet been clearly demonstrated and cannot be currently firmly recommended as a screening test.
In patients with chronic liver disease and portal hypertension, the most effective treatment for HCC confined to the liver in the absence of major vascular invasion is liver transplantation (LT). However, due to organ donor shortage, this approach to HCC has been limited. As a result, surgical resection and locoregional therapies such as radiofrequency ablation (RFA) and transarterial chemoembolization (TACE) have been used to treat and prevent tumor progression in selected cases.
A whole new body of knowledge is being developed regarding HCC biology. Biological tumor features that allow prediction of recurrence or response to treatment permit better patient and treatment selections. New developing molecular targeted therapies will serve as part of the neoadjuvant and adjuvant options to achieve a more comprehensive approach to this complex disease.
Multiple staging systems for HCC have been proposed. The Barcelona Clinic Liver Cancer (BCLC) staging system is the most widely accepted one. As shown in Figure 3, BCLC staging system links the prognosis and the treatment option with the stage of cancer, degree of liver dysfunction as determined by Child-Pugh score, and the patient functional status.
Liver transplantation (LT) is the treatment of choice for HCC in patients with decompensated cirrhosis, achieving 75% recurrence-free survival at 5 years in the absence of lymphovascular invasion.6-9
The early experience with LT for HCC was discouraging because only patients with extensive tumors were considered to undergo this therapy. However, patients undergoing LT for other indications who were found to have incidental HCC exhibited promising survival rates. Subsequently, in 1996 Mazzaferro introduced the Milan criteria–a single tumor of 5 cm or up to 3 tumors of 3 cm in size–and showed an 83% recurrence-free survival and 75% actuarial survival at 4 years for patients undergoing LT meeting these criteria.10 Currently, patients with HCC meeting the Milan criteria are being given priority for cadaveric donor LT. As a result, a rising number of patients with cirrhosis and early HCC undergo LT, accounting for a third of the total liver transplant operations.
In 2001, Yao introduced the University of California at San Francisco (UCSF) criteria showing that patients with a single lesion of 6.5 cm or up to 3 lesions with none larger than 4.5 cm and a total tumor burden less than 8 cm exhibited patient survival of 90% and 75.2% at 1 and 5 years, respectively.11 Patients with tumors beyond these criteria had 50% 1-year survival.
A major shortcoming of the Milan and UCSF criteria is that they tend to exclude patients with HCC from the LT candidacy based solely on the size of the individual tumor and/or the number of tumors. Certainly, among patients with tumor burdens beyond the currently adopted Milan or UCSF criteria, a subset of patients with favorable tumor biology exists that would have good outcome if transplanted. On the other hand, even with the implementation of the Milan criteria, recurrence rates after LT have been shown to be 8%-15%. This likely represents a subgroup of patients with aggressive HCC behavior, however morphologically they meet the Milan criteria. To overcome these shortcomings of Milan and UCSF criteria, we have recently proposed the "rate of tumor growth" as a useful tool for assessing patients with HCC for consideration of LT. We suggested a cut-off value of tumor growth of 1.61 cm3 per month. Patients whose tumor burden exceeds the Milan criteria, but with tumor growth <1.61 cm3 per month, experienced less recurrence of HCC after LT than those beyond the Milan criteria with tumor growth ≥1.61 cm3 per month. The criteria of tumor growth have to be further validated in larger multicenter trials.
Living donor liver transplantation (LDLT) is an alternative source of liver grafts that has been used to treat patients with end-stage liver disease and HCC. LDLT reduces waiting time and dropout rates on the liver transplant list. However, there is a controversy on whether LDLT is associated with an increased risk of tumor recurrence after LT, compared to the deceased donor. Some studies showed that LDLT for patients with HCC increases the risk of tumor recurrence following transplantation. This is thought to be secondary to the decreased waiting time on the liver transplant list, so that patients with rapidly growing tumors could be rushed to transplantation before the identification of the aggressive behavior of their tumors.
Locoregional therapies such as RFA and TACE have been implemented to control tumor progression and prevent patient drop-out from the waiting transplant list. Among patients within Milan criteria who are eligible for transplantation, the role of these therapies is particularly important and cost effective in patients with expected waiting time longer than 6 months as a bridge to transplantation. In patients with tumor burden beyond the Milan criteria, locoregional therapies have been recently used to downstage the tumors so that they can eligible for transplantation. Herein the strategy of "treat and wait" has been introduced; in which patients beyond Milan who were traditionally excluded from the liver transplant candidacy, could be treated with locoregional therapy to downstage the tumor. Patients should be monitored for the progression of tumor for about three months after the application of locoregional treatment, and considered for transplantation if no evidence of tumor progression.
Tumor recurrence rates after LT in patients within Milan and UCSF criteria ranges between 10% and 15%. Vascular invasion and poor tumor differentiation have been identified as the main risk factors for HCC recurrence. Particularly in the absence of vascular invasion, it seems that the number and size of tumors should not affect patient survival. Vascular invasion is a consequence of aggressive tumor biology. Unfortunately, pretransplant indicators of microvascular invasion remain undiscovered.
Several tumor biological factors such as DNA aneuploidy, high tumor cell proliferation index, high telomerase activity, and mutation of the p53 gene have been associated with increased risk of postoperative HCC recurrence. However, the prognostic impact of these factors remains uncertain due to a limited number of studies and conflicting data.
There has been growing interest in the study of gene expression profile in relation to prognosis of HCC patients. The copy DNA microarray is a modern technology that allows analysis of thousands of genes simultaneously using gene chips. Several studies have identified gene expression profiles predictive of the recurrence or prognosis of HCC patients.
In a study from the University of Pittsburgh, Marsh and colleagues found that patients with low HCC rates of mutation in critical tumor suppressor genes had lower tumor recurrence rate after LT, even in the presence of microvascular invasion. Larger studies are needed to confirm these associations, but given the absence of serum markers that can predict tumor behavior, pretransplant HCC biopsy might become a routine practice to assess HCC biology.
Liver resection for HCC is the treatment of choice in noncirrhotic patients. However, unfortunately this option is feasible in only 5% of the cases in Western countries and in 40% of the cases in Asia. Surgical resection is primarily indicated for patients with unifocal HCC not associated with underlying chronic liver dysfunction or for early-stage disease with well-compensated cirrhosis. Determination of tumor extension and liver function are the cornerstone steps to define tumor respectability.
Although noncirrhotic patients tolerate major resections with low morbidity, patients with advanced cirrhosis and portal hypertension are at risk of developing hepatic failure, bleeding, ascites and infection.
Clinical features such as Child A cirrhosis along with platelet count of 100,000/mm3 and hepatic vein pressure gradient less than 10 mm Hg have been associated with good tolerance to resection, with 5-year survival rate of 70%. However, because of underlying chronic disease in most cases, between 60% and 100% of patients develop HCC recurrence after resection and require additional treatment such as re-resection, ablation, or liver transplantation.
Improvement in surgical technique and patient selection made the perioperative mortality for hepatic resection drop from 10% to 30% to less than 1%. With good surgical technique, blood transfusion during surgical resection may be required in less than 10% of patients. The overall morbidity is between 20% and 50% (pleural effusion, perihepatic abscess, ileus, bile leak, wound infection, deep vein thrombosis, urinary tract infection), and its reduction depends on implementing parenchyma-sparing resections and decreased intraoperative blood loss.
In a large study with case matching by tumor stage and degree of cirrhosis and liver dysfunction, overall 5-year survival rates were similar for patients with HCC treated by LT or resection, but transplantation provided a clear advantage over resection in terms of recurrence-free survival at 5 years (54% vs 14%, P < 0.0001). Long-term outcomes support the use of primary resection in patients with Child A cirrhosis with small unifocal tumors. Multifocal tumors are associated with higher risk of recurrence after resection and are best approached by primary LT.
Patients with nonresectable HCC have exhibited survival benefit when treated with RFA and TACE. Recent studies demonstrated that RFA and TACE had similar efficacy compared to surgery when used to treat small tumors. Refined technique and improvements in technology will continue to expand the indications of these modalities for unresectable HCC, patients waiting for LT, in need of tumor downstaging, or needing control of progression. The potential combination options of such therapies with resection, or systemic antineoplastic agents such as sorafenib and others, are expected to improve patient outcome, although time is needed to draw this conclusion.
By generating temperatures higher than 60°C with alternating current (±500 kHz) between a radiofrequency probe and grounding pads on the patient's lower extremities, tumor coagulative necrosis is induced when inserting the probe into the tumor substance.
The efficacy of RFA depends on appropriate technical use and size of tumor. Lesions of 3 cm can be effectively treated, achieving complete tumor necrosis in 80% to 90% of cases. However, for lesions of 5 cm, complete tumor necrosis is unlikely, and tumor recurrence is expected in more than 50% of cases.
Tumor location is another important factor to determine treatment efficacy and safety. Lesions adjacent to large vessels might have a worse treatment response due to the thermal protection provided by the contiguous blood flow. Lesions located near the liver hilum should be avoided due to the risk of vascular and bile duct injuries. Peripheral lesions should also be avoided due to the risk of needle tract tumor dissemination. Overall, the morbidity risk associated with RFA is about 5% to 10%, which includes injury to adjacent anatomic structures, liver failure, hemorrhage, abscess formation, intercostal nerve injury, tumor lysis syndrome, and pneumothorax. The mortality risk has been reported as less than 1%. To reduce the morbidity risk, laparoscopic RFA is a suitable approach when indicated.
Patient survival with unresectable HCC treated with RFA has been reported as 75% to 92% at 1 year, 37% to 59% at 3 years, and 28% at 5 years. In the case of resectable small tumors, RFA has been associated with similar survival rates when compared with resection in selected cases.12
Due to technologic improvement and potential association with other locoregional therapies such as TACE, RFA will continue to be a promising tool to approach HCC.
Selective HCC arterial catheterization followed by the sequential infusion of chemotherapy and embolization particles is the principle of TACE. To improve treatment tolerance, drug-eluting beads (DEB) delivered by selective transcatheter arterial chemoembolization (DEB-TACE) have been recently introduced. This new modality is as effective as the traditional TACE in the management of HCC, however with fewer side effects, possibly because the patients have less systemic exposure to the chemo drug since the drug is released at slow rate. Several studies established the efficacy of TACE in treating unresectable HCC. A randomized, controlled trial showed that the 1, 2, and 3-year survival rates were 57%, 31%, and 26% in the treatment group versus 32%, 11%, and 3% in the control group (P = 0.002). In a systematic review of randomized trials for unresectable HCC, TACE was shown to improve patient 2-year survival compared to control groups (P = 0.017). In addition, a meta-analysis concluded that TACE reduced the 2-year mortality rate compared with control groups (P = 0.015).13
The role of TACE as a bridging therapy to transplantation has been reported, and conflicting data were observed. The efficacy of TACE in controlling tumor progression and down-staging in patients on the waiting list is accepted, but no conclusive survival benefit has been identified. The current consensus opinion recognizes the benefit of TACE in liver transplant candidates with waiting time longer than 6 months.
The risks associated with TACE for unresectable HCC include death (5%-10% in Child C cirrhotic patients), liver failure (5%-10% in Child C patients), upper gastrointestinal bleeding, abscess formation (30%-80% when the sphincter of Oddi is compromised), encephalopathy, nontarget embolization, tumor lysis syndrome, pulmonary embolism, and acute renal failure. The postchemoembolization syndrome is the most common side effect of TACE. Patients develop abdominal pain, nausea, and fever, which are self-limited in most cases and well controlled with hydration and with pain and anti-nausea medications.
TACE can be safely used in the absence of portal vein thrombosis and severe liver dysfunction.
Recent studies have reported improved patient survival when TACE and RFA were implemented in a combined approach to treat HCC larger than 3cm compared with either RFA or TACE alone. In a randomized, controlled trial including 291 patients with HCC larger than 3 cm, the 1-, 3-, and 5-year survival rates for lesions of 5 cm were superior for TACE followed by RFA over either treatment alone. Further experience is needed to confirm these observations.14
Selective transarterial radiotherapy able to deliver up to 150 Gy of beta-emitting radiation has been introduced to treat unresectable HCC showing encouraging tumor response. Approximately 50% of patients experience 50% tumor reduction, with a median time to maximum response of 120 days.
Yttrium-90 microspheres (TheraSphere) have been used as a bridge to transplantation to control tumor progression and for downstaging. Also, the capability of this technique to induce tumor burden shrinkage allowed subsequent treatments with surgical resection and RFA.
An important advantage of this technique is that because hepatic artery occlusion is not induced, patients with portal vein thrombosis can be treated safely and in several courses. On the other hand, this treatment should not be used if total bilirubin level is greater than 2 mg/dL, and in the presence of large lung shunt to avoid radiation induced liver and lung injuries.
A randomized control trial to compare the efficacy of RFA, TACE, and radioembolization is underway (NCT00956930, available at: www.clinicaltrials.gov).
Systemic therapy for advanced HCC is evolving. Sorafenib is an oral multikinase inhibitor of the vascular endothelial growth factor receptor, the platelet-derived growth factor receptor, and Raf. Sorafenib is the first systemic therapy that has showed a survival advantage in patients with advanced HCC. In a recent multicenter, randomized control trial of patients with advanced HCC, patients treated with Sorafenib (400 mg oral daily) had an overall median survival of 10.7 months, compared with 7.9 months for the placebo group. Most patients were of compensated liver disease classified as Child-Pugh class A. The most common side effects of Sorafenib are weight loss, hand or foot skin reactions, and diarrhea. Significant adverse events can occur; these include severe hypertension, myocardial infarction, gastrointestinal bleeding, bowel perforation, and hepatic decompensation.
Several targeted therapies are being evaluated in phase 1 to 3 trials, as first- or second-line alternative therapy to Sorafenib, or in combination with Sorafenib.
Sorafenib is also currently being studied as adjuvant therapy in combination with surgical resection or with locoregional therapy.
Future studies are needed to assess the efficacy and safety of these agents in preventing HCC recurrence after LT.
The most effective strategies to reduce HCC disease burden are to eliminate the precursor disease states. Immunization against hepatitis B has resulted in a sharp decline in HCC. Treatment of chronic hepatitis B and C have similarly been shown to decrease HCC. Effective public health measures related to environmental toxins and control of aflatoxin in food supplies reduce HCC rates. Reduction of high-risk behavior resulting in cirrhosis can also be expected to reduce HCC.