Abstract
BACKGROUND: Characterized by its significant occurrence and high fatality, hepatocellular carcinoma (HCC) presents a challenge with treatments frequently leading to less than ideal results. The mechanism of action behind disulfidptosis, a newly identified pathway of cell death, is not well comprehended when related to HCC. This research aims to investigate a model that employs long non-coding RNA (lncRNA) associated with disulfidptosis for predicting the prognosis of liver cancer and identifying potential therapeutic measures. METHODS: The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC) provided tissue specimens from 374 and 243 cases of HCC, respectively, along with samples from 50 and 202 healthy liver tissues. By employing differential analysis and Pearson correlation, we identified lncRNAs associated with disulfidptosis. Cox and least absolute shrinkage and selection operator (LASSO) regression analyses were then utilized to assess risk and construct a prognostic model for these lncRNAs. The model's predictive performance underwent evaluation through survival analysis, receiver operating characteristic (ROC), and C-index. Furthermore, our study delved into potential therapeutic roles of disulfidptosis-related lncRNAs in HCC, scrutinizing pathways, exploring the tumor microenvironment, and investigating immune evasion mechanisms. RESULTS: The prognostic model that we developed comprises five lncRNAs associated with disulfidptosis: TMCC1-AS1, LINC01224, MKLN1-AS, MIR210HG, and DANCR, which demonstrated significant upregulation in HCC tissues. The model showed that patients in the low-risk category had superior survival rates. This model outperformed traditional predictors such as age, gender, tumor grade, and stage in accuracy, achieving an area under the ROC curve (AUC) of 0.720. It effectively forecasted survival rates at 1, 3, and 5 years, yielding AUCs of 0.778, 0.720, and 0.664, respectively. In-depth analysis, including functional pathway enrichment and studies of the tumor microenvironment and immune evasion, observed significant differences in immune cell infiltration and immune evasion mechanisms among various risk groups. This model, focused on disulfidptosis-related lncRNAs, emerges as a promising predictor for the response of HCC to immune checkpoint inhibitors as well as other prevalent anti-cancer therapies, such as Bcl-2 inhibitors, EGFR tyrosine kinase inhibitors, and PI3K inhibitors. CONCLUSIONS: A prognostic model concerning disulfidptosis-related lncRNAs was constructed to predict outcomes in HCC. This model provides insights into molecular mechanisms, characterizes the tumor microenvironment, and predicts patient responses to immunotherapy and targeted treatments.