1. Minimally-invasive biomarkers of hepatocellular carcinoma (“Liquid Biopsy”): The presence of circulating DNA in human blood is known since the mid-1950s, but it hasn’t been until relatively recently that its potential applications in oncology has started to emerge. Different studies demonstrate how analysis of circulating DNA allows tracking tumor mutation landscape at different time points, what ultimately impacts in better tailored treatments. We expect to provide the proof-of-principle to incorporate powerful readouts of tumor biology in HCC risk stratification, using the molecular information derived from a simple blood test.

2. Liver cancer patient derived xenografts to validate oncogenic drivers in HCC and identify mechanisms of resistance to sorafenib: There is a wide consensus on the need to develop better models to reproduce with higher fidelity different aspects of human cancer, especially when trying to predict clinical efficacy of anticancer therapies. An alternative that is increasing its visibility in translational research is patient-derived xenografts (PDXs). They consist of pieces of fresh tumor tissue implanted subcutaneous or orthotopic in immunodeficient mice. Different studies (e.g., lung, colon cancer) demonstrate molecular stability between engrafted tumor and early passage expansion. Some limitations of PDX are the potential drift of stromal components from primary human to primarily mouse, as well as the lack of a fully functional human immune system. We will use this model to identify and validate mechanisms of resistance to conventional therapies.

3. Tumor heterogeneity: Phenotypic heterogeneity in liver cancer has been widely described by pathologists. However, our understanding of genetic intra-tumor heterogeneity has been limited until recently due to our incomplete structural knowledge of the cancer genome, and to a lack of technologies sensitive enough to easily detect tumor sub-clones. A Darwinian theory has been proposed to describe the evolution of tumor clones that arose and diverge from the tumor cell considered as the most recent common ancestor. Tumor heterogeneity questions the reliability of molecular classification and genetic analysis using a single tumor biopsy. This is particularly worrisome in patients with large tumors or those with multiple nodules, where the likelihood of multi-clonality increases. We will try to elucidate the role of tumor heterogeneity in HCC progression and treatment response.