Conclusion
The novelty of our study is the identification of genetic signatures through classification rules for SNPs in angiogenesis and ADME genes associated with sorafenib responses in HCC patients, through which it may be possible to personalize prescription. The application of GRS could allow a better stratification of patients. In addition, the network analysis conducted in this study supports the association of 8/12 analyzed genes in topological key points involved in several common biological pathways correlated to HCC and sorafenib. However, our study has some limitations. The sample size was relatively small, and further investigations in a larger sample size may be needed. The opportunity to test the classification rules on predictive biomarkers of response to sorafenib in an independent and larger validation set would give more robustness to our findings. Therefore, our findings had an exploratory aim and are intended as a “proof-of-concept” research to be further validated in a larger dataset to allow sorafenib-tailored prescriptions through predictive biomarkers of response and outcome in HCC patients. The involvement of seed genes in multiple biological pathways related to sorafenib and HCC, as well as the common interactions of ADH1A ,CYP26A1 , VEGF-A , and VEGF-C in signal transduction pathways, should allow future studies on the simultaneous targeting of different signaling pathways or common downstream proteins involved in HCC control and sorafenib response with the aim of personalizing treatment for this still uncurable disease.