Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide, and its incidence will further increase, to reach a plateau in 2015-2020. The natural history of the disease is quite well known, except for its early stages, because the majority of patients at this stage are treated with radical approaches. Staging systems are key to predict the prognostics of patients with cancer, to stratify the patients according to prognostic variables in the setting of clinical trials, and to guide the therapeutic approach. The current knowledge of the disease, however, is not sufficient for recommending a staging system to be used worldwide. The conventional staging systems-Okuda stage, and TNM stage-have shown important limitations for classifying patients. Several new systems have been recently proposed, but only three of them have been validated. The Barcelona Clinic Liver Cancer (BCLC) staging classification links the stage of the disease to a specific treatment strategy. The Japan Integrated Staging (JIS) score has been proposed and used in Japan, although it needs Western validation. The Cancer of the Liver Italian Program (CLIP) score is mainly proposed for patients with advanced tumors. Early detection of HCC through surveillance programs allows the application of potentially curative therapies, such as resection, liver transplantation, and percutaneous ablation in patients with early tumors. The applicability of these treatments varies according to geographical distribution: from 50% to 70% of cases in Japan; 25% to 40% of cases in Europe and the United States; and fewer than 10% in Africa. There are no randomized controlled trials (RCTs) comparing any of the three major therapies. These studies are not feasible in the West. Therefore, there is no firm evidence to establish the optimal first-line treatment for small single HCC in patients with well-preserved liver function. Resection and transplantation achieve the best outcomes in well-selected candidates (5-year survival of 60%-70%), and compete as the first option from an intention-to-treat perspective. If surgery is precluded, local, nonsurgical therapies are applied. Percutaneous treatments provide good results (5-year survival of 40%-50%), but are unable to achieve response rates and outcomes comparable to those for surgical treatments, even when applied as the first option. Radiofrequency thermal ablation provides slightly better objective response rates than ethanol injection, but no survival advantages have been fully demonstrated. The remaining treatments have been assessed in the setting of around 70 RCTs conducted during the past 25 years. Chemoembolization has been shown to provide modest survival advantages in two RCTs and a metaanalysis, and is currently the mainstay of treatment in 10% of the whole HCC population. The ideal candidates for this option are patients with well-preserved liver function (Child-Pugh class A) and multinodular asymptomatic tumors without vascular invasion. Further RCTs are needed to assess the best chemotherapeutic agent and the ideal re-treatment schedule. There is no firstline option for patients with advanced HCC (vascular invasion, extrahepatic spread, or cancer-related symptoms). Systemic doxorubicin provides partial responses in 10% of cases, without proven survival advantages, and well-known treatment-related complications. Several other treatments, such as immunotherapy, internal radiation, tamoxifen, or anti-androgen agents, have not shown any relevant anti-tumoral effect or survival benefit. New drugs, such as tyrosine kinase inhibitors and anti-angiogenic agents, are currently being tested in the setting of clinical trials.