Abstract
Targeting immunogenic cell death (ICD) may enable the response of pancreatic cancer to immune checkpoint inhibitors (ICIs). The aim of the present study was to elucidate the role of ICD-related genes in pancreatic cancer. Utilizing the k-means method, consensus clustering was employed to effectively group patients with pancreatic cancer. Subsequently, a set of differentially expressed genes was identified between the two subtypes related to ICD, facilitating the execution of a comprehensive enrichment analysis. Furthermore, the construction of an ICD-related prognostic signature (IRPS) was accomplished through LASSO Cox regression, thereby enabling the assessment of responses to both chemotherapy and immunotherapy. In addition, the biological functionality of 5'-nucleotidase ecto (NT5E) was elucidated through experimental investigations. Patients characterized as the ICD high subtype experienced a comparatively shorter overall survival. This subtype exhibited a noteworthy correlation with HLA families and immune checkpoint molecules, underscoring its immunological significance. Subsequently, patients with elevated IRPS risk scores displayed resistance towards immunotherapy interventions. Of note, synergistic downregulation of NT5E in combination with Gemcitabine was observed to significantly induce tumor cell apoptosis, emphasizing its potential therapeutic value. Leveraging ICD-related genes, a novel classification system was meticulously devised to comprehensively evaluate both the clinical outcomes and therapeutic responses of patients diagnosed with pancreatic cancer.