Transplantation for Primary Hepatic Malignancy

Epidemiology

The majority of the malignant lesions of the liver in the United States are metastatic in origin. HCC accounts for 80% to 90% of the primary malignancies, whereas CCA constitutes most of the rest. Some of the tumors exhibit mixed features of HCC and CCA. Most patients with HCC in the Western Hemisphere (70% to 90%) have some degree of cirrhosis. In Asia and West Africa, HCC occurs frequently in normal liver tissue. The incidence rate in these areas is 30 cases per 100,000 population. HCC is the fifth most common malignancy worldwide, with more than a million new cases a year. It ranks as the third most fatal tumor worldwide. The incidence of HCC more than doubled in the United States in the last decades : it rose from 1.6 cases per 100,000 population in 1975 to 4.9 per 100,000 in 2005. The case-fatality ratio is 0.8, and the median survival is less than 6 months in most cases.

In the United States, HCC is more prevalent in Asians and American Indians, followed by African Americans, and whites. HCC is three times more frequent in men compared to women. Patients with established cirrhosis have an annual HCC incidence of 3% to 10% depending on the underlying liver disease. Chronic inflammation in the liver leads to necrosis followed by increased regeneration, which can lead to a higher rate of cell mutation and hence carcinogenesis. The hepatitis B virus can also act directly with disruption of the hepatocyte DNA followed by faulty repair. HCC can also occur via the pathway of adenomatous hyperplasia, which can lead to malignant transformation. These changes are associated with marked alteration of the apoptosis-to-mitosis ratio.

In Asia the incidence of HCC correlates with that of hepatitis B, whereas the most frequent cause of HCC in the West is hepatitis C (HCV). The risk of HCC seems to be maximal 30 years after infection with HCV. There is an increase in the incidence of HCC parallel to the rise of HCV, which is expected to peak in 2025. In Western countries, the United States in particular, the obesity epidemic leads to an increase in the incidence of nonalcoholic steatohepatitis, which if uncontrolled, will be the leading cause of cirrhosis and HCC. Hemochromatosis, although relatively rare in the population, carries a 45% risk of HCC. Genetic abnormalities, such as chromosome aberrations or gene mutations, have been described with HCC, but a unifying genetic/molecular mechanism was not characterized. Other risk factors for HCC include aflatoxin, azo dyes, aromatic amines, N -nitroso compounds, chlorinated hydrocarbons, hydrosol compounds, pesticides, radiation, thorotrast, smoking, porphyria, Budd-Chiari syndrome, oral contraceptives, anabolic steroids, and α ₁ -antitripsin deficiency. HCC is less common in patients with autoimmune hepatitis, Wilson's disease, and alcohol-induced cirrhosis.

Clinical Presentation and Diagnosis

In patients with known chronic hepatitis, most cases of HCC are found by screening imaging. Unfortunately, HCC tumors become symptomatic when already in an advanced stage. Some patients may present with diffuse or localized abdominal pain, sometimes radiating to the right shoulder, postprandial fullness, or jaundice. Others present with a palpable mass or deterioration of liver function. Sudden abdominal enlargement is an ominous sign, as it can be related to portal vein thrombosis or tumor bleeding. Patients typically display symptoms of chronic liver disease, with splenomegaly, ascites, and jaundice. Some patients have paraneoplastic syndromes, with polycythemia, polymyositis, deep vein thrombosis, hypoglycemia, and chronic diarrhea.

Serum α-fetoprotein (AFP) level is a useful marker when elevated, especially if above 400 ng/mL, and 70% of patients have elevated AFP level. Des-gamma-carboxy prothrombin level is similarly useful when elevated ; however, it has not gained widespread use. Further diagnosis is made with imaging. Ultrasonography is mostly a screening tool for the diagnosis of HCC. Contrast-enhanced sonograms (with liquid microspheres) increase the sensitivity for HCC ; the procedure is not available in most countries. A more complete assessment of tumor size, characteristics, and tumor burden is made with multiphasic contrast CT scanning or MRI scan. Most tumors have the characteristic arterial blush. In this situation the suspicion of HCC is very strong. CT or MRI provides complete imaging of the liver. Both CT and MRI scans have good sensitivity and specificity. In a systematic review of studies where imaging was compared with pathological characteristics of explants, the sensitivity of imaging increased from 60.5% for sonograms, to 67.5% for CT, and 80.6% for MRI, whereas the specificity decreased from 96.9% (sonogram), to 92.5% (CT), and 84.8% (MRI). Although PET scans may predict microvascular invasion, they are rarely used in practice.

At times it is difficult to differentiate benign nodules in the cirrhotic liver from HCC, or focal nodular hyperplasia can mimic malignancy. In this situation a tumor biopsy is warranted. Tumor biopsy had excellent sensitivity and specificity, except for small nodules ; 2% to 11% of specimens are inadequate. Routine biopsy of lesions suspicious for HCC has been discouraged. Because HCC is a well-vascularized tumor, there is a higher risk of bleeding than with liver biopsy. There is also a 1% to 5.1% risk of parietal tumor seeding, which results in metastasis, with a median occurrence of 17 months, and most metastases are hard to treat. Currently the diagnosis of HCC can be made using two different imaging modalities or one imaging modality and elevated AFP level. Diagnosing HCC without a biopsy comes at the expense of false-positive results (e.g., no tumor in the pathological examination of the explant).

When performed, a tumor biopsy can diagnose poorly differentiated tumors or vascular invasion, in which case a liver transplant is contraindicated. However, tumors, especially large ones, are heterogenous, and at times tumor differentiation can be overestimated, and vascular invasion is missed. Certain tumors require differentiation from CCA or metastatic tumors, typically with additional staining for AFP, carcinoembryonic antigen, CD10, and CD15.

Imaging for extrahepatic staging of HCC include CT scan of chest, abdomen, and pelvis, and bone scan ( Fig. 16-7 ).

Staging

Multiple staging systems were adopted for HCC, whether strictly oncological or prognostic or combined. The TNM system, based on tumor size and number (T), lymph node involvement (N), and the presence or absence of metastasis (M) has undergone a few revisions; the most recent version is shown in Table 16-1 . Liver transplantation mostly benefits patients with stage I or II tumors, although some transplants are performed for some stage IIIA patients using expanded tumor inclusion criteria. The TNM system has limitations, because the data used in decision making are based on imaging and not actual histological results, and imaging can underestimate tumor burden in stage II patients by 27% to 33%. The severity of the background liver disease, an important part of prognosis, is not part of TNM. TNM staging does not correlate with posttransplant survival.

To better evaluate prognosis, several groups have proposed staging systems that include both pathological and functional criteria. The Okuda staging system ( Table 16-2 ) combines tumor size with serum bilirubin level, albumin level, and ascites. The Cancer of the Liver Italian Program (CLIP) score combines tumor morphological characteristics, Child-Turcotte-Pugh (CTP) score, portal vein thrombosis, and AFP levels, with severity ranging from 0 to 6 ( Table 16-3 ).

Although considered a staging system, the Barcelona Clinic Liver Cancer (BCLC) classification is a treatment algorithm and not a pure staging tool. It combines the Okuda staging with the CTP score, number and size of nodules, portal vein thrombosis, and functional status to recommend treatment modalities ( Fig. 16-8 ).

Prognostic Factors and Risk Factors for Poor Tumor Biological Properties

Because most of the HCC tumors arise in the setting of chronic liver disease with or without cirrhosis, the prognosis of HCC is dictated not only by the tumor biological properties and progression, but also by the severity and evolution of the underlying liver disease. Adverse tumor factors have been studied extensively. Adverse tumor factors include tumor size, number of nodules, poor tumor differentiation, vascular invasion, nodal or extrahepatic involvement, and portal vein thrombosis. Portal vein thrombosis due to HCC is a contraindication and needs to be differentiated from bland portal vein thrombi. Poor tumor differentiation and vascular invasion are strong determinants of poor oncological outcome. Unfortunately, vascular invasion cannot be diagnosed with the current imaging or laboratory tests available. Pathological data are seldom available before transplant, and a tumor biopsy sample can miss vascular invasion. Vascular invasion is the strongest predictor of poor outcome, and it correlates with tumor size and number of nodules. Tumors greater than 5 cm in diameter, a larger number of nodules, and high AFP level are associated with poorer differentiation and a higher likelihood of vascular invasion. AFP values greater than 400 ng/mL and 1000 ng/mL are associated with poor outcome in group comparison, but not for every patient. High levels of des-gamma-carboxy prothrombin (protein induced by vitamin K antagonist II [PIVKA-II]) are associated with worse posttransplant outcomes. Fast tumor progression typically denotes poor tumor biological properties and is associated with high recurrence rate and mortality after transplantation.