EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Primary liver cancer is the fifth most common cancer overall and the second most common cause of cancer mortality worldwide. Hepatocellular carcinoma accounts for up to 90% of all primary hepatic malignancies and represents a major international health problem. It is a complex and heterogeneous malignancy, frequently occurs in the setting of a chronically diseased organ, and has multiple confounding factors. Liver transplant for hepatocellular carcinoma has been established as a standard treatment in selected patients. Liver resection and locoregional therapies could be other options for treatment. Pathologic evaluation of hepatocellular carcinoma is a complicated process that includes tumor grading and evaluation of microvascular invasion. Although macrovascular invasion can be detected with imaging techniques, microvascular invasion is diagnosed pathologically. Pathologic evaluation provides additional information about the tumor biology, using immunohistochemical and molecular methods to predict patient outcomes. Hepatocellular carcinoma requires a multidisciplinary approach to determine the most appropriate treatment, as well as requires accurate timing of various treatments for optimal outcomes.

Key words : Hepatic malignancies, Hepatocellular carcinoma, Microvascular invasion, Tumor grading

Hepatocellular carcinoma (HCC) is the most common primary malignant tumor in the liver. It is the fifth most common malignant tumor in men and the eighth most common in women worldwide. Men are affected 3 times more often than women. The incidence varies with geographic area.¹

A vast majority of HCC develop in cirrhotic livers. The following diseases increase the risk for HCC by causing cirrhosis: hepatitis B virus (HBV), hepatitis C virus (HCV), alcohol consumption, metabolic disorders, drugs, and toxins. Hepatitis B virus is the most common underlying cause of HCC worldwide. The lifetime risk of developing HCC is almost 50% in HBV-positive men and 20% in women.¹ The risk of HCC in HCV-positive patients is 17-fold compared with negative controls. Prolonged intake of alcohol can lead to cirrhosis and hence is a risk factor for HCC. The risk is higher with coexistent HBV, HCV, and diabetes.¹ Metabolic disorders such as hemochromatosis, tyrosinemia, α1-antitrypsin deficiency, and Wilson disease have the risk for HCC development. Exposure to thorotrast, aflatoxin, androgenic steroids, and progestational agents has been associated with HCC. Exposure to aflatoxin is common in areas endemic for HBV.¹

Treatment of HCC has progressed from a single option of surgical resection to a spectrum of choices, including orthotopic liver transplant, locoregional ablation and transarterial chemoembolization, local ablative therapies such as radiofrequency ablation, percutaneous alcohol injection, and microwave ablation or systemic molecular targeted therapies. Treatment modalities can be changed depending on the clinical stage of the tumor; the potential curable option for early HCC is resection and liver transplant.²

It is well known that the Milan criteria are the criterion standard for candidate selection to ensure excellent posttransplant survival for patients. However, there have been concerns about the Milan criteria as being too restrictive and far from satisfying regarding the increasing number of candidates on wait lists. An extended “up to 7 criteria” protocol has been introduced, which is aimed at improving the selection of appropriate liver transplant recipients among patients with advanced HCC.³

Even with new criteria descriptions, tumor size is always included. It is well known that solitary and small tumors are generally thought to have a better prognosis than multifocal and large tumors.³

There are several factors, including vascular invasion, satellite nodules, high differentiation grade, and underlying liver disease, that may predict outcomes after hepatic resection; in addition, tumor size is also an important prognostic factor. In a study of patients with solitary HCC, Lim and associates⁴ stratified 616 resected patients with solitary HCC according to tumor size: smaller than 5 cm, 5 to 10 cm, and larger than 10 cm. The authors concluded that microvascular invasion, satellite nodules, and a less differentiated tumor were more prevalent in patients with larger tumors. They found that patients with larger tumors were more likely to have worse overall, disease-free, and recurrence-free survival. However, there was no significant differences in overall versus recurrence-free survival between patients with tumors of 5 to 10 cm and those with tumors larger than 10 cm. Less than or larger than 5 cm is important, but the size loses its significance when the tumor is larger than 5 cm. Presence of adverse predictors influence survival.^4,5

The 5-year overall survival rate after liver transplant is about 70%; however, 27% of patients have tumor recurrence after transplant despite fulfilling the Milan criteria. Tumor recurrence is a major limitation on survival. With tumor recurrence, extrahepatic localization results in poorer outcomes.⁶

Are there any predictive factors for extrahepatic recurrence of HCC after liver transplant? In one study, Andreou and associates⁷ reviewed the clinicopathologic data of 364 consecutive liver transplant patients with HCC between 1989 and 2010. During the 78-month follow-up, 93 patients had HCC recurrence. All tumor biopsies were macroscopically evaluated for tumor burden and histologically evaluated to confirm the diagnosis and provide tumor grading. Macrovascular and microvascular invasion and DNA index were also evaluated.

DNA index is defined as an objective tool to express tumor biology. It is basically the ratio of DNA content of tumor cells. Cut-off values less than 1.5 DNA index are associated with favorable biology and tumor survival.⁷ On multivariate analysis, DNA index, beyond Milan criteria, macrovascular invasion, and tumor grading were found to be independently correlated with HCC recurrence after liver transplant. Andreou and associates analyzed for predictors of extrahepatic recurrence. On multivariate analyses, only HCC beyond Milan criteria and macrovascular invasion were independently associated with extrahepatic recurrence. In patients with HCC beyond Milan criteria who developed recurrence, only macrovascular invasion was found to be a significant predictor of extrahepatic recurrence. On the other hand, in patients with HCC within Milan criteria who had HCC recurrence after liver transplant, a DNA index of more than 1.5 was found to be the only predictor of extrahepatic recurrence on multivariate analyses. This study provides evidence that advanced HCC beyond Milan criteria and the presence of macrovascular invasion are associated with increased risks for extrahepatic recurrence, and therefore they are currently considered as relative contraindications to liver transplant.⁷ In patients with HCC within Milan criteria, the DNA index represents a valuable prognostic marker for the development of extrahepatic recurrence. DNA index as a surrogate marker for tumor biology and a prognostic factor for long-term survival has been introduced by the same group.⁸ DNA index could be used as an objective tool to predict the development of recurrences, and it is significantly associated with overall survival.⁸

When considered from the tumor biology point of view, what is the importance of assessing the HCC preoperatively for tumor grade and other variables? In another study, 153 patients were examined who were preoperatively diagnosed with HCC via needle core biopsy (NCB) between 2010 and 2012. The study’s exclusion criteria included patients who received locoregional therapies, patients with multiple nodules, and patients with insufficient sampling. Tumors were graded and categorized as well differentiated, moderately differentiated, and poorly differentiated. The authors noted whether tumors are homogeneous or heterogeneous. Ki-67 proliferation index, accuracy of NCB, and effects of the number of NCB biopsy samples were investigated. The group found that 45% of the HCCs presented intratumoral heterogeneity, meaning that the same tumor could have different grading areas. The results suggested that the risk of histologic heterogeneity increases with the size of the tumor. The rate of heterogeneity showed significant differences between HCC less than 3 cm and HCCs larger than 8 cm. The overall accuracy gradually increased along with the increased number of NCB passes. The concordance between NCBs and the surgical specimens were higher in HCCs smaller than 5 cm in which a 1-pass NCB was enough to get high accuracy. However, NCBs must be performed at least twice to get a better accuracy than only 1 pass in tumors larger than 5 cm. Increasing the number of NCB passes could improve the NCB diagnostic accuracy for HCCs larger than 8 cm.⁹

Poorly differentiated HCCs are associated with a significantly higher recurrence rate after liver transplant. In fact, tumor grade is an independent risk factor regardless of tumor number and size. Moreover, some authors support a strategy for selecting HCC patients suitable for liver transplant based on pretransplant tumor grading.¹⁰ Multivariate analyses revealed that both the size of the largest lesion and α-fetoprotein (AFP) concentration were independent risk factors. Interestingly, a combination of those 2 risk factors affected outcomes, with disease-free survival reaching only 12.5% after almost 5 years. However, if only 1 term (tumor size or AFP) was met, the disease-free survival was almost 50%. The main finding of this study is that patients with poorly differentiated tumors comprise a heterogeneous group with different prognoses. There are underlying tumor features that are linked to better outcomes after liver transplant and excellent disease-free survival even among poorly differentiated tumors. In addition, there is a simple stratification method to distinguish this group of patients based on tumor size and AFP level. If transplant centers exclude grade 3 patients from liver transplant based on the grade of pretransplant biopsy, there could be a group of patients left who could reach a satisfactory survival rate.¹⁰

Hepatocellular carcinoma is a characteristically highly vascular tumor. Therefore, it has high propensity for vascular invasion. Macrovascular invasion could be visible on gross examination; however, microvascular invasion can only be seen on microscopy and only be confirmed while evaluating the whole tumor on resection.¹¹ Microvascular invasion is generally an incidental finding, and its presence is correlated with tumor grade. In one study, the authors defined the microvascular invasion of the portal vein, hepatic vein, and bile duct. The number of invaded vessels or bile ducts and the maximum number of invading carcinoma cells were counted to determine whether there was a high degree of invasion. Patients with microvascular invasion were classified into 2 subgroups. The high microvascular invasion subgroup showed more than 50 carcinoma cells in the vasculature. Tumor size was a predictive factor for the presence of high microvascular invasion. The recurrence-free survival curves of patients without and with low microvascular invasion were similar; however, those with high microvascular invasion had significantly poorer outcomes than the other 2 groups. As a result, evaluating the degree of microvascular invasion is useful for predicting the prognosis after transplant.¹²

Although pretransplant biopsy procedures are widely performed and are defined as a safe and noninvasive method, there are several publications that the procedure has the risk of seeding and peritoneal dissemination.^13-16 A biopsy has risks, including bleeding and abdominal dissemination. To determine whether preoperative fine needle aspiration (FNA) biopsy was related to tumor recurrence, one study divided patients with HCC into 2 groups: HCC patients without FNA biopsy and HCC patients with preoperative FNA biopsy. In the group with preoperative FNA biopsy, both overall and extrahepatic recurrences were significantly higher.¹⁷

Although macrovascular invasion can be detected with imaging techniques, microvascular invasion is diagnosed pathologically. One study examined whether microvascular invasion can be accurately diagnosed preoperatively and found that patients with nonsmooth tumor margins showed higher microvascular invasion. The authors concluded that nonsmooth tumor margin on preoperative imaging studies is a significant predictive factor of microvascular invasion. The disease-free and overall survival rates were also higher in the smooth tumor margin group.¹⁸ In a study from McHugh and associates, age, tumor size greater than 3 cm, AFP more than 100, and the presence of multiple lesions were associated with increased risk of microvascular invasion. The positive predictive value of large tumor size and AFP levels greater than 100 is 88% for microvascular invasion and 55% for death.¹⁹ In another study, tumor mitotic index and histologic grade were found to be related to microvascular invasion. According to this study, preoperative liver biopsy was important because evaluating the mitotic index and histologic grade could only be possible with through-cut biopsy.²⁰

Traditionally, HCC has long been believed to transform from mature hepatocytes by a dedifferentiation process. After the importance of parenchymal cells was realized, it was recognized that HCC consisted of heterogeneous groups of subtypes. Understanding the relation between “hepatic stem cells” or “hepatic progenitor cells” and prognosis is crucial. In an evaluation of hepatic progenitor cell markers and their predictive roles in prognosis, the expression of both markers by the same tumor was related to aggressive tumor behavior. Patients with aggressive tumors have low overall survival, and the combined expression of CK19 and glypican 3 was related to microvascular invasion, regional lymph node metastasis, and distant metastasis.²¹ In another study that evaluated both glypican 3 mRNA and protein expression levels, both levels were found to be related to vascular invasion. In addition, overall survival rates were significantly lower if expression was shown.²² Several immunohistochemical markers have been studied to predict outcomes of patients with HCC. One of these studies found that p53 expression was inversely related to disease-free survival. CK7 and CK19 expression levels were associated with poor outcomes in patients with liver transplant.²³ Ki-67, which is widely used for a number of tumors and accepted as an independent marker for worse prognosis, was investigated as to whether it influenced the progression and prognosis of HCC. A meta-analysis that included 54 studies found that high Ki-67 proliferative index values were associated with poor disease-free survival, recurrence-free survival, and overall survival.²⁴

In our center, 526 liver transplants have been performed since 1988. Among these, 59 liver transplants were performed for HCC. Of the 59 patients with HCC, 14 patients (24%) had tumor recurrence. The 5- and 10-year overall survival rates were 50% and 43% in patients beyond Milan criteria.²⁵

Hepatocellular carcinoma remains a fatal malignant disease worldwide. Tumor size, serum AFP levels, microvascular invasion, and immunohistochemical profiles of the tumor can be accepted as baseline factors to estimate long-term outcomes. Future studies on management of HCC are vital to offer better outcomes and prolong survival for patients with HCC.

Volume: 15 Issue: 2 March 2017 - Supplement - 2

FULL TEXT