Project Summary Hepatoblastoma (HB) and hepatocellular carcinoma (HCC) are the most frequently diagnosed liver tumors in children, with HBs most commonly present in young children less than 5 years of age, and HCCs are more commonly seen in adolescents. HB- and HCC-patient outcomes and treatment options vary dramatically, with 5-year overall survival rates of over 70% for HB and under 30% for HCC patients. While a combination of chemotherapy and surgery is effective for lower-risk HBs, 3-year overall survival for high-risk HBs is 50%. High- dose chemotherapy, which is often ineffective for high-risk HBs, is associated with significant morbidity. Complete surgical resection is the only chance for a cure for HCC. Molecular biomarkers can help optimize treatments for patients that will not benefit from chemotherapy or that do not require high dosage chemotherapy and identify patients that require a combination of aggressive surgery and chemotherapy. In previous work, we, and our collaborators, identified prognostic biomarkers that distinguish between low- and high-risk HBs at diagnosis. We proposed and retrospectively evaluated predictive models to classify patients based on risk—including their need for aggressive therapies. These models identify patients that do not require aggressive therapies, patients that will benefit from aggressive therapies, and patients with tumors that are more likely to metastasize and become resistant to chemotherapy. We developed and certified molecular assays to profile patients and tumors for predictive biomarker used by these models. Here, we propose to prospectively validate these biomarkers, assays, and models to produce the first validated platform for the molecular diagnosis and therapy choice for HBs and HCC. We will benefit from a close collaboration with clinical-trial produced data in the USA and EU, including AHEP1531 (USA), ChILTERN (EU) and iPC (a EU-USA collaboration).