Abstract
BACKGROUND: In recent years, immunogenic cell death (ICD) and long non-coding RNAs (lncRNAs) have been implicated in tumor invasion and growth. However, the role of ICD-associated lncRNAs in pancreatic adenocarcinoma (PAAD) remains unclear. Therefore, this study aimed to develop and validate a prognostic signature based on ICD-associated lncRNAs for PAAD patients and to explore its potential associations with immune processes. METHODS: We identified ICD-associated genes from the GeneCards database and obtained transcriptomic data and sample information for PAAD from The Cancer Genome Atlas (TCGA). Using these datasets, we primarily aimed to develop and validate a prognostic signature based on ICD-associated lncRNAs. A prognostic signature and nomogram were constructed to predict patient outcomes. To preliminarily explore potential underlying mechanisms related to immunity, we performed enrichment analysis [Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG)], tumor mutational burden (TMB) assessment, comprehensive immune correlate analysis, and pharmacosensitivity profiling. Additionally, to explore the functional role of a key lncRNA (LINC00705) in PAAD, we modulated its expression in PANC-1 and MIAPaCa-2 cell lines through knockdown and overexpression strategies. RESULTS: We identified six ICD-associated lncRNAs (ZEB2-AS1, AC099792.1, LINC01091, LINC02613, AC005304.3, and LINC00705) and constructed a prognostic signature. Survival analyses, univariate and multivariate Cox regression, and receiver operating characteristic (ROC) curve analysis and dynamic nomogram demonstrated the robust accuracy of this six-lncRNA signature in predicting PAAD patient prognosis. Pathway enrichment analysis of the low-risk cohort suggested a potential link to immune-related processes, providing preliminary insights into possible mechanisms. Patients exhibiting a higher risk score were significantly associated with elevated TMB (P=0.001). Immune-related analyses of the low-risk cohort revealed higher immune cell infiltration, improved immune function scores, and elevated expression of immune checkpoints. Pharmacosensitivity profiling identified 15 drugs with differential sensitivity. In vitro experiments showed that LINC00705 significantly affected the biological behaviors of pancreatic cancer cells. CONCLUSIONS: We developed and validated a prognostic signature based on six ICD-associated lncRNAs, demonstrating its significant accuracy in predicting outcomes for PAAD patients. This signature holds potential for providing more precise clinical guidance. Furthermore, our preliminary bioinformatic analyses suggest associations between the signature and immune processes, warranting future mechanistic investigation. Additionally, modulating LINC00705 expression may offer a promising strategy to inhibit cancer cell growth and metastasis.