PROJECT SUMMARY
The overall survival rate for children with high-risk liver cancer, including metastatic or relapse/refractory
hepatoblastoma (HB), hepatocellular carcinoma (HCC), and HB with HCC features (HBC) is below 50%. The
best odds for survival for patients with these cancers is complete surgical resection with negative margins, but
this is not achievable at diagnosis for most patients. Instead, high-risk patients are treated with combinations of
chemotherapy and surgery, where chemotherapy is used to reduce tumor burdens before surgery and to
eliminate residual cancer cells after surgery. The low survival rates for high-risk liver cancers are associated with
chemoresistance and improving these survival rates requires alternative therapies to target chemoresistant
cancer cells in combination with chemotherapy.
We propose a comprehensive approach to characterize chemoresistant cancer cells in children with high-risk
liver cancers, design technologies to detect these cancers at diagnosis, discover and establish therapies to target
these chemoresistant cancer cells, and use patient-derived models to test personalized treatment combinations
before prescribing them to high-risk patients. We propose to proceed with three aims. First, we will establish
chemoresistant models and molecularly and clinically characterize them. Second, we will use molecular data to
enable the identification of chemoresistant cancers at diagnosis utilizing prospectively collected specimens from
a major clinical trial. We will use personalized models to predict treatment response in prospectively enrolled
cancer patients and determine if chemoresistance can be detected at diagnosis. Finally, we will identify and
investigate targeted combination therapies in patient-derived models.
Conceptual innovations in our proposal include building a pediatric liver cancer atlas of molecularly and clinically
characterized cell types that could be used to predict patient responses to therapy at diagnosis, as well as
mapping out the drivers of chemoresistance in pediatric liver cancer. The study will evaluate the largest cohort
of patient-derived models for pediatric liver cancers and the utility of these models as pre-clinical tools to develop
and evaluate therapies for children with high-risk liver cancers. If successful, this study will help efforts to
transform the diagnosis and treatment of children with high-risk liver cancers leading to dramatically improved
outcomes.