EXPERIMENTAL AND CLINICAL TRANSPLANTATION

There are a number of factors that should be taken into consideration when predicting the outcomes and prognosis of hepatocellular carcinoma. Four of the most important factors in determining survival are the severity of the accompanying liver disease, tumor size and number, extension of the tumor into other structures, and presence of metastases. The most common and widely accepted tumor staging systems include these 4 prognostic factors. The staging systems are (1) the tumor-node-metastasis system of the American Joint Committee on cancer, (2) the Okuda system, (3) the Barcelona Clinic Liver Cancer system, (4) the Cancer of the Liver Italian Program score; (5) the recent albumin-bilirubin grading system, and finally (6) the Hong Kong Liver Cancer staging system. The successful staging of patients with hepatocellular carcinoma is difficult because of the many contributing factors to disease. Therefore, in clinical practice, there is no ideal system that can be applicable to every patient in predicting survival. As a general rule, patient and treatment modality selection criteria should be individualized for every patient and should not use an algorithmic approach because aggressive treatments in properly selected patients with hepatocellular carcinoma offer good results. A better approach for increased treatment success is one that is multidisciplinary, that is, one that considers the different variables.

Key words : Hepatocellular carcinoma, Prognostic factors, Staging

In patients with hepatocellular carcinoma (HCC), there are many variables that differ from region to region that complicate diagnosis, prognosis, and treatment recommendations. Median survival time is approximately 6 to 20 months after diagnosis. Hepato-cellular carcinoma is different from other solid tumors because of the possibility of preexisting chronic liver disease or cirrhosis condition, which have negative effects on both survival and treatment options.¹

Prognostic factors
There are 4 main groups of prognostic factors: tumor related, factors related to underlying liver status, factors related to the general health status of the patient, and factors related to the efficacy of treatment. Independent negative predictive factors of HCC outcome include α-fetoprotein levels of ≥ 20 ng/mL, serum albumin of < 3.5 g/dL, creatinine of ≥ 1 mg/dL, bilirubin of ≥ 1 mg/dL, alkaline phosphatase of ≥ 200 IU/L, presence of ascites, maximal tumor size >5 cm, multiple tumor nodules, presence of vascular invasion, presence of extrahepatic metastasis, and poor performance status (all P < .001).² Other factors that predict survival include tumor histology, living in high- versus low-incidence regions of hepatitis B and C, having antiviral therapy for hepatitis B virus-related HCC, variant estrogen receptors, diabetes mellitus, vascular endothelial growth factor, insulinlike growth factor 1, and gene expression levels of Profil.^3-8

Staging systems
Although there is a clear need for an optimal staging system that will guide practitioners to the best therapeutic approach, presently, a universally accepted prognostic system is not available. There are presently more than 15 HCC staging classifications that measure different factors and have been developed in different patient populations. Despite this, physicians are sometimes reluctant to select the optimal system. Liver function, α-fetoprotein level, performance status, tumor number and size, micro- and macrovascular invasion, and presence of metastasis are all components of the present staging systems. With these points in mind, there are several reasons why a worldwide acceptable staging system is not available.

First, HCC usually occurs in a cirrhotic background, which has some preneoplastic features; these conditions have contributing effects on patient survival. Globally, the course and prognosis of HCC differ greatly by region, with epidemiologic risk factors also varying according to geographic region. In some selected patients, the only treatment option for HCC is organ transplant. However, only 20% of all cases can be treated by surgery, which makes it difficult to use pathology-based systems such as the tumor-node-metastasis (TNM) system. Another reason for no acceptable global staging system is because molecular signatures that can predict outcomes in HCC have not been identified or incorporated in staging systems.⁹

The most commonly used staging systems are the following: the TNM system, the Okuda system, the Barcelona Clinic Liver Cancer (BCLC) system, the Cancer of the Liver Italian Program score, the albumin-bilirubin (ALBI) grading system, and the Hong Kong Liver Cancer (HKLC) staging system.

Tumor-node-metastasis staging system
In 2010, the American Joint Committee on Cancer TNM staging system was revised.¹⁰ According to this system, the most important prognostic factors are number of tumors and the extent or vascular invasion within the tumor (Table 1).¹¹

The TNM staging system was found to be the most accurate system to predict outcomes of patients undergoing liver transplant. In addition, the TNM staging system is the only one that has been confirmed for use in patients treated with hepatic resection or transplant (Table 2).¹²

The Okuda system
The Okuda prognostic scoring system uses tumor size and 3 other variables that assess cirrhosis severity: amount of ascites, serum bilirubin level, and albumin level (Table 3). In addition, unlike the TNM classification, it does not consider variables such as vascular invasion or nodal metastasis.¹³

The Cancer of the Liver Italian Program score
The Cancer of the Liver Italian Program score is a relatively new prognostic scoring system. Macroscopic tumor morphology, α-fetoprotein level, presence of portal vein thrombosis, and cirrhosis severity are evaluated to determine a prognostic score ranging from 0 to 6 (Table 4).¹⁴

In the determination of survival rates in patients undergoing nonsurgical therapies such as transarterial chemoembolization in various countries, the Cancer of the Liver Italian Program was found to be superior to the TNM, Okuda, and Child-Pugh systems.^15-17

Barcelona staging classification
The BCLC staging system is an algorithmic rather than a patient-oriented classification system. The components of this system are based on tumor volume and number, tumor invasiveness, performance status, Child-Pugh score, extrahepatic spread, and Okuda stage (Table 5).¹⁸ These components are also the main criticisms of this system, which are increasing in both Western and Asian countries.

There is a growing evidence among many doctors (for example, the International Liver Cancer Association) regarding the BCLC system that practitioners can extend the surgical resection option to patients who have intermediate or locally advanced HCC with intrahepatic venous invasion. Similarly, ablative therapies could be a logical option for patients who have tumor size of 3 to 5 cm or who have tumors even slightly larger than 5 cm, and transarterial therapy can be extended to patients with locally advanced HCC with intrahepatic venous invasion.

In summary, patients classified as having very early (stage 0) or early (stage A) disease have a chance of curative treatment modalities. Locoregional therapies rather than being curative can only control tumor growth in intermediate (stage B) patients. Patients with advanced-stage (stage C) disease according to BCLC criteria who generally have macrovascular tumor invasion and/or metastatic disease with preserved liver function as defined by the Child-Pugh cirrhosis score can benefit from sorafenib therapy. However, patients classified with terminal disease (stage D) or clearly decompensated liver cirrhosis patients with HCC have only supportive or symptomatic care options.¹⁸

Albumin-bilirubin score
Contributing factors related to tumor characteristics and liver function status are sometimes difficult to separate clinically. Johnson and associates developed the ALBI system to solve the lack of validation of Child-Pugh score in patients with advanced HCC (Table 6).^19,20 According to Johnson and associates,²⁰ the ALBI score is calculated as follows: ALBI score = (log[bilirubin in μmol/L] × 0.66) - (albumin in g/L × 0.085).

The group conducted a study comprising more than 6000 patients (predominantly with Child-Pugh A) across 4 global regions (Europe, Japan, China, and the United States) who had diverse causes of disease. After analyzing multiple HCC prognostic factors, the group tried to find an optimal liver function score among the heterogenous patient populations. Their ALBI model was developed to incorporate serum albumin and bilirubin levels alone by using multivariable Cox regression analysis on training and validation sets. The ALBI score stratifies patients into 3 categories according to functional liver status, which is comparable but much easier than using the Child-Pugh score in predicting survival.

The advantages of the ALBI score are its simplicity and its validity in large, contemporary, and internationally diverse HCC populations. Therefore, it is logical and practical to replace the Child-Pugh with the ALBI model during planning of future trials.¹⁹ Overall, the prognostic performance of the ALBI-based versus Child-Pugh-based BCLC system is similar.²¹

Hong Kong Liver Cancer staging system
A large study was conducted at Queen Mary Hospital with prospectively collected data of 3856 patients with HCC (predominantly patients with hepatitis B virus-related disease) treated from 1995 to 2008. The study used Cox regression to investigate the relative effects of 2026 variables (including demographic, clinical, laboratory, treatment, and survival data) in predicting overall survival. Classification and regression tree analyses were used to classify disparate treatment decision rules. All patients were allocated randomly into a training set or a test set in a 1:1 ratio (Table 7).²²

Comparison of the Hong Kong liver cancer and the Barcelona Clinic liver cancer staging system
Regarding prediction of overall patient survival, the HKLC system has been shown to have a better ability than BCLC to predict (area under the receiver operating characteristic curve values, approximately 0.84 vs 0.80; concordance index, 0.74 vs 0.70). The HKLC system was able to identify a subgroup of patients with BCLC intermediate and advanced-stage disease who were suitable for more aggressive treatment options, which improved their survival outcomes.^22,23 Figure 1 describes the general strategy of treatment according to the HKLC system.

However, there are some limitations related to the HKLC system. First, the cohort of the original study comprised predominantly patients with hepatitis B virus-related disease (80% of the total cohort), which makes it obligatory to be validated in a Western population as well as in those with different causes of HCC. Second, although BCLC classifies patients into 5 categories (0, A, B, C, and D), the authors subdivided the BCLC system as A1-2, A3-4, B, C, and D when assessing the area under the curve values.²⁴

Although it has some limitations, HKLC has some advantages versus the BCLC system, especially when we consider the heterogeneity of patients with BCLC-B and BCLC-C stage disease and the rigidity of BCLC in its allocation of treatment classification.

Regarding the rigidity of the BCLC system, the extended treatment options (such as resection for BCLC-B patients or transarterial chemoembolization for BCLC-C patients) should be validated in large perspective studies and in HCC patients with different causes other than hepatitis B virus infection.²⁵

However, there are some previous studies in the literature that claim that more aggressive treatment modalities could result in better outcomes than with the BCLC system approach, which may further the inspiration of other systems, such as the HKLC scoring system.^26-28

What is the best system available?
The number of staging classifications for HCC has increased in recent years, and newly developed classifications have demonstrated better results in predicting prognosis. Although improvements are still ongoing, there is no agreement on a standard and globally accepted classification.

One of the major concerns of disagreement among the available systems is related to the heterogenous characteristics of HCC. Because there are significant regional differences in terms of tumor morphology and behavior and also the presence of different comorbidities, validation studies should be completed worldwide before reaching a common consensus of the ideal classification system. Unfortunately, none of the staging classifications currently available has been validated in all patient populations, and, as such, none can be recommended for worldwide use.¹

Clinicians who treat patients with HCC should be aware of the advantages and limitations of the various systems. The understanding of tumor biology has grown; in addition, there have been advancements in imaging techniques and treatment modalities. All of these efforts will help in the development of better staging systems that can refine patient stratification, survival prediction, and treatment allocation to optimize the management of HCC.²⁵

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

FULL TEXT