Abstract
Utilization of next-generation sequencing (NGS) to identify potential therapeutic targets and then prescribe matched agents provides new hope for patients with advanced cancer, such as hepatocellular carcinoma (HCC). However, intratumoral heterogeneity (ITH) challenges precise genomic profiling and may lead to target treatment failure. This study aims to evaluate whether and to what extent would genetic profiling be biased by ITH in HCC. We datamined publications focusing on the ITH of HCC and extracted the sequencing and clinicopathological information to make data reanalysis. Potential therapeutic targets and driver genes in HCC were specially pooled as reference to analyze the bias effect of ITH on genetic profiling. Five studies which analyzed ITH using NGS of multi-site samples were enrolled, with a total of 207 tumor samples from 36 HCC patients. The ITH ranged from 5.21% to 88.27% and no correlations between ITH extent and sample numbers, sequencing depth, or clinicopathological parameters were observed. In total, 72 therapeutic and 15 candidate driver genes were pooled as reference. Totally, 38.8% HCCs were found to be drugable in single-site sample, of which only 19.4% might be biased by ITH. Of the driver genes, 86% could be detected in single-site sample. HCC is a highly heterogeneous disease. While ITH indeed hinders comprehensive and precise HCC genome profiling, it has limited influences on identification of actionable and driver mutations. Single-site sampling/biopsy assayed with targeted deep sequencing might be efficient in the clinical management of HCC.