Abstract
Purpose: To construct a prognostic model for pancreatic cancer based on glycolytic and hypoxic metabolic subtypes. To analyze the biological characteristics of these subtypes and explore potential therapeutic options. Methods: We obtained mRNA, simple nucleotide variation (SNP), and clinical data for pancreatic cancer from The Cancer Genome Atlas (TCGA). Patients were classified into four metabolic subtypes. We focused on glycolysis and hypoxia subtypes. Single-sample gene set enrichment analysis (ssGSEA) assessed immune cell infiltration. We evaluated the effects of immunotherapy and chemotherapy on these subtypes. Cox regression and random survival forest algorithms were used to build a prognostic model. Validation was performed using data from the International Cancer Genome Consortium (ICGC) and ArrayExpress database. Results: We identified four subtypes. Kaplan-Meier survival analysis showed the glycolytic subtype had the longest survival, while the hypoxic subtype had the shortest. The glycolytic subtype exhibited higher immune cell infiltration. Immunotherapy and chemotherapy appeared more beneficial for the glycolytic subtype. KRAS mutations were more frequent in the hypoxic subtype. Our prognostic model indicated a worse prognosis for high-risk groups, validated by external data. Conclusion: The glycolytic metabolic subtype of pancreatic cancer is associated with longer survival and better response to chemotherapy and immunotherapy compared to the hypoxic subtype.