A predictive model based on program cell death genes for prognosis and therapeutic response in early stage hepatocellular carcinoma

The principal cause of treatment ineffectiveness in hepatocellular carcinoma (HCC) patients stems from post-surgery stagnation and treatment resistance. A comprehensive predictive model for the progression and drug response of post-surgery HCC patients remains elusive. Various programmed cell death (PCD) patterns significantly influence tumor advancement, offering potential as prognostic and drug sensitivity indicators for postsurgery HCC. The analysis in this study utilized integrated data from 12 different types of PCD, multi-omics data from TCGA-HCC and other cohorts in the International Cancer Genome Consortium, as well as clinical information of HCC patients. A PCD score was calculated using a four-gene signature determined through cox regression analysis. Validation in independent datasets revealed that HCC patients with high PCD scores had poorer prognoses post-surgery. Furthermore, an unsupervised clustering model identified two distinct molecular subtypes of HCC with unique biological processes. A nomogram exhibiting high predictive accuracy was developed by integrating a PCD signature with clinical characteristics. The association between programmed cell death, immune checkpoints genes and key components of the tumor microenvironment. was established. Patients with HCC displaying elevated PCD levels demonstrated resistance to traditional adjuvant chemotherapy and immune checkpoints inhibitor therapies. Additionally, the oncogenic function of four PCD genes was identified in an inpatient cohort. A novel scoring methodology for PCD was devised through the examination of genes linked to diverse PCD subtypes, providing valuable insights into the prognosis and drug responsiveness of HCC patients. Early-stage HCC patients may potentially derive therapeutic benefits from immune therapy directed at programmed cell death.