Circulating tumor DNA profiling reveals clonal evolution and real-time disease progression in advanced hepatocellular carcinoma

Circulating tumor DNA (ctDNA) provides a potential non-invasive biomarker for cancer diagnosis and prognosis, but whether it could reflect tumor heterogeneity and monitor therapeutic responses in hepatocellular carcinoma (HCC) is unclear. Focusing on 574 cancer genes known to harbor actionable mutations, we identified the mutation repertoire of HCC tissues, and monitored the corresponding ctDNA features in blood samples to evaluate its clinical significance. Analysis of 3 HCC patients' mutation profiles revealed that ctDNA could overcome tumor heterogeneity and provide information of tumor burden and prognosis. Further analysis was conducted on the 4th HCC case with multiple lesion samples and sequential plasma samples. We identified 160 subclonal SNVs in tumor tissues as well as matched peritumor tissues with PBMC as control. 96.9% of this patient's tissue mutations could be also detected in plasma samples. These subclonal SNVs were grouped into 9 clusters according to their trends of cellular prevalence shift in tumor tissues. Two clusters constituted of tumor stem somatic mutations showed circulating levels relating with cancer progression. Analysis of tumor somatic mutations revealed that circulating level of such tumor stem somatic mutations could reflect tumor burden and even predict prognosis earlier than traditional strategies. Furthermore, HCK (p.V174M), identified as a recurrent/metastatic related mutation site, could promote migration and invasion of HCC cells. Taken together, study of mutation profiles in biopsy and plasma samples in HCC patients showed that ctDNA could overcome tumor heterogeneity and real-time track the therapeutic responses in the longitudinal monitoring.