Identification of Context-Specific Fitness Genes Associated With Metabolic Rearrangements for Prognosis and Potential Treatment Targets for Liver Cancer

Abstract

Liver cancer is the most frequent fatal malignancy. Furthermore, there is a lack of effective therapeutics for this cancer type. To construct a prognostic model for potential beneficiary screens and identify novel treatment targets, we used an adaptive daisy model (ADaM) to identify context-specific fitness genes from the CRISPR-Cas9 screens database, DepMap. Functional analysis and prognostic significance were assessed using data from TCGA and ICGC cohorts, while drug sensitivity analysis was performed using data from the Liver Cancer Model Repository (LIMORE). Finally, a 25-gene prognostic model was established. Patients were then divided into high- and low-risk groups; the high-risk group had a higher stemness index and shorter overall survival time than the low-risk group. The C-index, time-dependent ROC curves, and multivariate Cox regression analysis confirmed the excellent prognostic ability of this model. Functional enrichment analysis revealed the importance of metabolic rearrangements and serine/threonine kinase activity, which could be targeted by trametinib and is the key pathway in regulating liver cancer cell viability. In conclusion, the present study provides a prognostic model for patients with liver cancer and might help in the exploration of novel therapeutic targets to ultimately improve patient outcomes.