Hepatocellular carcinoma (HCC) is an increasingly common and highly morbid malignancy. Appropriate tumor staging is crucial to inform prognosis and guide treatment decisions in clinical practice; however, staging for HCC can be complex because of many patient-level factors that can impact prognosis and treatment eligibility. HCC is a unique malignancy in that it typically occurs in the setting of underlying organ dysfunction (i.e., cirrhosis), so most staging systems take both tumor burden and the degree of underlying liver dysfunction into account for prognostication. The most commonly used staging systems for HCC include the Barcelona Clinic Liver Cancer (BCLC) system [1], the Hong Kong Liver Cancer (HKLC) system [2], the Cancer of the Liver Italian Program (CLIP) system [3], the Japan Integrated Staging (JIS) system [4], and the Model to Estimate Survival in Ambulatory HCC patients (MESIAH) (Table 1) [5]. However, there has been a lack of consensus regarding the optimal staging system, and none is universally accepted. Most existing systems were derived from single-center data, lack prospective or external validation, and lack granularity in intermediate- and advanced-stage patients. In this issue of PLOS Medicine, Alessandro Vitale and colleagues detail the derivation and validation of a novel staging system for HCC, the ITA.LI.CA system [6]. The ITA.LI.CA system attempts to address many deficiencies of prior staging systems and demonstrates better discriminant ability for predicting survival than existing HCC staging systems in both internal and external validation cohorts.

The ITA.LI.CA system was derived from a prospective multicenter database of over 5,000 HCC patients from Italy. The majority of patients in the cohort had hepatitis C infection, nearly all (97%) had good performance status, and three-fourths had well-compensated cirrhosis. External validation was performed using data from a Taiwanese cohort of over 2,600 patients, with the primary etiology of liver disease for patients in this cohort being chronic hepatitis B infection. Using a priori variable selection based on prior staging systems and a literature review, the authors derived a model that uses a prognostic score based upon tumor burden (categories of 0, A, B1–3, and C), functional status, Child-Pugh score, and alpha fetoprotein (AFP) concentration (≤1,000 or >1,000 ng/ml). The model had better discriminant ability than any of the existing staging systems in the training, internal validation, and external validation cohorts (c-statistic values being 0.72, 0.71, and 0.78, respectively).

The BCLC staging system is currently the most widely accepted staging system and has been endorsed by the American Association for the Study of Liver Disease (AASLD) and European Association for the Study of the Liver (EASL) [7,8]. Though some aspects of the ITA.LI.CA system are rooted in the BCLC, it is distinct in several important ways: first, in subclassifying BCLC stage B patients into B1, B2, and B3 categories based on degree of intrahepatic tumor burden; second, in differentiating patients with intrahepatic and extrahepatic metastases; and finally, by incorporating the serum biomarker AFP. In the BCLC system, all patients with liver-isolated disease, without metastases or vascular invasion, are grouped together as BCLC stage B [1]. However, differential survival and locoregional treatment allocation for BCLC stage B patients has been demonstrated in several studies [9,10]. For example, distinguishing whether BCLC stage B patients are within (B2) or beyond (B3) Milan criteria is important when considering liver transplantation. Similarly, recent data suggest prognosis in patients with extrahepatic metastases is worse than those with intrahepatic metastases, so the differentiation in the ITA.LI.CA system, essentially subclassifying the BCLC stage C patients, adds further granularity to estimating prognosis [11]. Finally, AFP is not part of the BCLC staging system but can serve as a surrogate for occult vascular invasion, distant metastases, or aggressive tumor biology. Patients with an AFP > 500 ng/ml have a higher risk of recurrence post-transplant as well as a lower likelihood of response to locoregional therapy [12]. These three important distinctions as compared to the BCLC system likely explain, in part, the higher prognostic accuracy of the ITA.LI.CA staging system in derivation and validation cohorts.

Although the model demonstrated good prognostic discrimination among study patients, it should be noted that most patients in both cohorts had good performance status, compensated cirrhosis, and early or intermediate stage tumors. It is unclear if the ITA.LI.CA staging system would perform as well in cohorts with high rates of hepatic decompensation, poor performance status, and/or advanced tumor stage—subgroups that currently account for the majority of HCC patients in several countries, including the United States. Further, very few patients in this study—less than 2% in the derivation cohort and none in the external validation cohort—underwent liver transplantation, a curative therapy for both the tumor and underlying cirrhosis that plays a crucial role in the management of HCC patients.

Potential future steps in further refinement and validation of the ITA.LI.CA staging system include prospectively assigning treatment allocation recommendations to patients in different stages and validation in more contemporary cohorts, in which transplantation or systemic therapies are utilized to a greater extent.