The design of clinical trials in hepatocellular carcinoma (HCC) is complex due to the presence of concurrent liver disease, which can confound the assessment of clinical benefit. An expert panel developed guidelines to standardize trial design, emphasizing randomized phase 2 trials with time-to-event endpoints like time to progression as pivotal in HCC research. Survival remains the main endpoint in phase 3 studies, while time to recurrence is proposed for adjuvant settings. Composite endpoints like progression-free survival and disease-free survival are less reliable in HCC due to confounding factors. The Barcelona Clinic Liver Cancer (BCLC) staging system is recommended for patient selection, with new drugs tested in patients with well-preserved liver function (Child–Pugh A class). Control arms should receive standard-of-care therapy, such as chemoembolization for intermediate-stage disease and sorafenib for advanced-stage disease. Biomarkers and molecular imaging are suggested to enhance trial populations and improve cost-benefit ratios. Phase 3 trials should be conducted by centers with HCC expertise. HCC is a major global health issue, with increasing incidence in the US and Europe. Chronic hepatitis B and C infections are leading risk factors. Surgical treatments are available for 30%-40% of patients in the West, but fewer in Asia. Few HCC treatments have been thoroughly tested compared to other cancers. The panel emphasized the need for standardized diagnostic criteria, stratification factors, and appropriate endpoints. Molecular targeted therapies challenge conventional endpoints like response rate. The panel recommended using overall survival as the primary endpoint in phase 3 trials, with time to recurrence for adjuvant studies. Progression-free survival is discouraged due to confounding factors. Response rate is not recommended for phase 2 trials due to limited clinical benefit. Radiological measurements using modified RECIST criteria are suggested for assessing response and progression. Molecular imaging and new biomarkers are needed to improve trial outcomes. Trial design should consider BCLC staging, stratification, and control arms based on standard-of-care. Liver function is critical in trial design, with Child–Pugh and Model of End-Stage Liver Disease scores recommended. Safety and toxicity stopping rules should be defined by Data Safety and Monitoring Committees. The panel concluded that overall survival is the most reliable endpoint, with time to recurrence for adjuvant studies. Progression-free survival is discouraged, and response rate is not recommended for phase 2 trials. The panel emphasized the need for standardized trial design, endpoints, and patient selection to improve HCC research.The design of clinical trials in hepatocellular carcinoma (HCC) is complex due to the presence of concurrent liver disease, which can confound the assessment of clinical benefit. An expert panel developed guidelines to standardize trial design, emphasizing randomized phase 2 trials with time-to-event endpoints like time to progression as pivotal in HCC research. Survival remains the main endpoint in phase 3 studies, while time to recurrence is proposed for adjuvant settings. Composite endpoints like progression-free survival and disease-free survival are less reliable in HCC due to confounding factors. The Barcelona Clinic Liver Cancer (BCLC) staging system is recommended for patient selection, with new drugs tested in patients with well-preserved liver function (Child–Pugh A class). Control arms should receive standard-of-care therapy, such as chemoembolization for intermediate-stage disease and sorafenib for advanced-stage disease. Biomarkers and molecular imaging are suggested to enhance trial populations and improve cost-benefit ratios. Phase 3 trials should be conducted by centers with HCC expertise. HCC is a major global health issue, with increasing incidence in the US and Europe. Chronic hepatitis B and C infections are leading risk factors. Surgical treatments are available for 30%-40% of patients in the West, but fewer in Asia. Few HCC treatments have been thoroughly tested compared to other cancers. The panel emphasized the need for standardized diagnostic criteria, stratification factors, and appropriate endpoints. Molecular targeted therapies challenge conventional endpoints like response rate. The panel recommended using overall survival as the primary endpoint in phase 3 trials, with time to recurrence for adjuvant studies. Progression-free survival is discouraged due to confounding factors. Response rate is not recommended for phase 2 trials due to limited clinical benefit. Radiological measurements using modified RECIST criteria are suggested for assessing response and progression. Molecular imaging and new biomarkers are needed to improve trial outcomes. Trial design should consider BCLC staging, stratification, and control arms based on standard-of-care. Liver function is critical in trial design, with Child–Pugh and Model of End-Stage Liver Disease scores recommended. Safety and toxicity stopping rules should be defined by Data Safety and Monitoring Committees. The panel concluded that overall survival is the most reliable endpoint, with time to recurrence for adjuvant studies. Progression-free survival is discouraged, and response rate is not recommended for phase 2 trials. The panel emphasized the need for standardized trial design, endpoints, and patient selection to improve HCC research.