Background
Hepatocellular carcinoma still has a high incidence and mortality rate worldwide, and further research is needed to investigate its occurrence and development mechanisms in depth in order to identify new therapeutic targets. Ferritinophagy is a type of autophagy and a key factor in ferroptosis that could influence tumor onset and progression. Although, the potential role of ferritinophagy-related genes (FRGs) in liver hepatocellular carcinoma (LIHC) is unknown.
Methods
Single-cell RNA sequencing (scRNA-seq) data of LIHC were obtained from the Gene Expression Omnibus (GEO) dataset. In addition, transcriptome and clinical follow-up outcome data of individuals with LIHC were extracted from the The Cancer Genome Atlas (TCGA) dataset. FRGs were collected through the GeneCards database. Differential cell subpopulations were distinguished, and differentially expressed FRGs (DEFRGs) were obtained. Differential expression of FRGs and prognosis were observed according to the TCGA database. An FRG-related risk model was constructed to predict patient prognosis by absolute shrinkage and selection operator (LASSO) and COX regression analyses, and its prognosis predictive power was validated. Ultimately, the association between risk score and tumor microenvironment (TME), immune cell infiltration, immune checkpoints, drug sensitivity, and tumor mutation burden (TMB) was analyzed. We also used quantitative reverse transcription polymerase chain reaction (qRT-PCR) to validate the expression of key genes in normal liver cells and liver cancer cells.
Results
We ultimately identified 8 cell types, and 7 differentially expressed FRGs genes (ZFP36, NCOA4, FTH1, FTL, TNF, PCBP1, CYB561A3) were found among immune cells, and we found that Monocytes and Macrophages were closely related to FRGs genes. Subsequently, COX regression analysis showed that patients with high expression of FTH1, FTL, and PCBP1 had significantly worse prognosis than those with low expression, and our survival prediction model, constructed based on age, stage, and risk score, showed better prognostic prediction ability. Our risk model based on 3 FRGs genes ultimately revealed significant differences between high-risk and low-risk groups in terms of immune infiltration and immune checkpoint correlation, drug sensitivity, and somatic mutation risk. Finally, we validated the key prognostic genes FTH1, FTL, using qRT-PCR, and found that the expression of FTH1 and FTL was significantly higher in various liver cancer cells than in normal liver cells. At the same time, immunohistochemistry showed that the expression of FTH1, FTL in tumor tissues was significantly higher than that in para-tumor tissues.
Conclusion
This study identifies a considerable impact of FRGs on immunity and prognosis in individuals with LIHC. The collective findings of this research provide new ideas for personalized treatment of LIHC and a more targeted therapy approach for individuals with LIHC to improve their prognosis.