Background
Hepatocellular carcinoma (HCC) is a globally prevalent disease. Our article evaluates risk models based on autophagy- and HCC-related genes and their prognostic value by bioinformatics analytical
Conclusion
Our results illustrate that models, nomograms and risk scores were valuable for tumour progression. Clinical trial number: Not applicable.
Methods
Prognostic genes were identified by univariate and multivariate Cox analyses, and risk scores were calculated. The value of risk models was analysed by receiver operating characteristic curve (ROC), immune microenvironment and drug sensitivity. Prognostic gene-related regulatory mechanisms based on network database.
Results
We screened four prognosis-related genes (SQSTM1, GABARAPL1, CDKN2A, HSPB8) for model construction. The AUC for 1-, 2- and 3-year survival was higher than 0.6 in both the training and validation sets. The nomogram constructed based on risk scores, pathologic_T predicted the outcome better. There were differences in the tumour microenvironment between the high and low risk groups, as evidenced by differences in the distribution of immune cells and differences in the expression of immune checkpoints.