This is a protocol for a Cochrane Review (diagnostic). The objectives are as follows:

To estimate the diagnostic accuracy of des-gamma-carboxy prothrombin, GALAD (Gender, Age, Lens culinaris agglutinin-reactive AFP, AFP and DCP), and alpha-foetoprotein for the diagnosis of hepatocellular carcinoma of any size, and at any stage, in adults with chronic liver disease, in either a surveillance programme or a clinical setting.

We acknowledge the possibility that theoretically, the accuracy of the tests in a surveillance programme may differ from that in a clinical setting due to variation in inclusion criteria and the prevalence of the target condition. However, we do not plan a separate analysis for surveillance and clinical settings, as they are not clearly distinct in current clinical practice (Forner 2018; Poustchi 2011).

In routine evaluation of people with chronic liver disease, index tests, as well as ultrasound, are already part of standard procedure. Given that HCC typically presents with no symptoms and is often asymptomatic, suspicion of the disease is typically based solely on the presence of advanced chronic liver disease. However, we do plan to consider the study setting as a potential source of heterogeneity.

To compare the diagnostic accuracy of des-gamma-carboxy prothrombin (DCP) alone or GALAD alone versus alpha-foetoprotein (AFP), for the diagnosis of hepatocellular carcinoma (HCC) of any size, at any stage; in adults with chronic liver disease, either in a surveillance programme or a clinical setting.

To estimate the diagnostic accuracy of DCP or GALAD versus AFP, for resectable HCC in people with chronic liver disease, in a surveillance programme and a clinical setting.

To investigate the following predefined sources of heterogeneity for each of the index tests:

• study design (case-control studies compared to cross-sectional studies);

• inclusion of participants without cirrhosis (studies including more than 10% of participants without cirrhosis compared to studies including less than 10% of participants without cirrhosis);

• study location (population differences): studies conducted in North and South America and Europe compared to Asia and Africa;

• prevalence of the target condition (studies with hepatocellular carcinoma prevalence more than 10% compared to studies with hepatocellular carcinoma prevalence less than 10%);

• participant selection (participants recruited from planned surveillance programmes compared to clinical cohorts);

• different reference standards (histology of the explanted liver compared to liver biopsy compared to another reference standard);

• different aetiology: studies including at least 90% of participants with chronic viral hepatitis compared to studies including less than 90% of participants with chronic viral hepatitis.