Mutation profiles in circulating tumor DNA (ctDNA) to predict the efficacy of sorafenib treatment in patients with advanced hepatocellular carcinoma (HCC).

Abstract

e15648 Background: To find biomarkers from blood samples through ctDNA and leucocyte genomic DNA sequencing to predict the efficacy of sorafenib. Methods: CtDNA and leucocyte genomic DNA were extracted from patients with advanced hepatocellular carcinoma for sequencing analysis before and after targeted therapy with sorafenib. Mutation profiles were obtained by deep sequencing of 620 selected genes for further bioinformatics analysis. Gene mutation frequency and status were further integrated with clinical data to dynamically monitor and predict the efficacy of sorafenib. Results: Twelve patients with advanced hepatocellular carcinoma admitted to our hospital from March 2017 to September 2018 were enrolled. A total of 27 blood samples were collected for further analysis. The results showed that there were significant differences in BRD4 and FANCG gene mutation abundance between therapeutic effect group and poor effect group (p = 0.014). The mutations of BRD4, FANCG and TP53 were associated with shorter PFS (BRD4 mutation type vs wild type: 12 weeks vs 32 weeks, p = 0.0011; FANCG mutation type vs wild type: 16 weeks vs 32 weeks, p = 0.0103; TP53 mutation type vs wild type: 8 weeks vs 32 weeks, p = 0.0095). Furthermore, we found another 19 genes (SDHA, OTUD7A, TMPRSS2, NAV3, AKT1, EIF2S2, DICER1, MAP3K1, RAD9A, TCF3, IRS1, PIK3C2G, CTCF, TSHZ2, ITPKB, KDM5C, STAG2, AMER1, CDC27) associated with therapeutic efficacy, and dynamic changes of these 19 genes mutation alleys could predict therapeutic efficacy. Conclusions: Our results showed that mutations in ctDNA could predict the efficacy and prognosis of sorafenib for advanced hepatocellular carcinoma, and that dynamic monitoring of mutation frequency in ctDNA of peripheral blood samples could be an additional method to monitor therapeutic efficacy of sorafenib. The clinical significance of the mutations of BRD4, FANCG, TP53 and another 19 genes we found needs to be confirmed on a larger scale.