Characterization of tumor microenvironment and programmed death-related genes to identify molecular subtypes and drug resistance in pancreatic cancer
Background: Immunotherapy has been a key option for the treatment of many types of cancer. A positive response to immunotherapy is heavily dependent on tumor microenvironment (TME) interaction. However, in pancreatic adenocarcinoma (PAAD), the association between TME mode of action and immune cell infiltration and immunotherapy, clinical outcome remained unknown. Methods: We systematically evaluated 29 TME genes in PAAD signature. Molecular subtypes of distinct TME signatures in PAAD were characterized by consensus clustering. After this, we comprehensively analyzed their clinical features, prognosis, and immunotherapy/chemotherapy response using correlation analysis, Kaplan-Meier curves analysis, ssGSEA analysis. 12 programmed cell death (PCD) patterns were acquired from previous study. Differentially expressed genes (DEGs) were acquired based on differential analysis. Key genes affecting overall survival (OS) of PAAD were screened by COX regression analysis and used to develop a RiskScore evaluation model. Finally, we assessed the value of RiskScore in predicting prognosis and treatment response in PAAD. Results: We identified 3 patterns of TME-associated molecular subtypes (C1, C2, C3), and observed that clinicopathological characteristics, prognosis, pathway features and immune features, immunotherapy/chemosensitivity of patients were correlated with the TME related subtypes. C1 subtype was more sensitive to the four chemotherapeutic drugs. PCD patterns were more likely to occur at C2 or C3. At the same time, we also detected 6 key genes that could affect the prognosis of PAAD, and 5 genes expressions were closely associated to methylation level. Low-risk patients with high immunocompetence had favorable prognostic results and high immunotherapy benefit. Patients in the high-risk group were more sensitive to chemotherapeutic drugs. RiskScore related to TME was an independent prognostic factor for PAAD. Conclusion: Collectively, we identified a prognostic signature of TME in PAAD patients, which could help elucidate the specific mechanism of action of TME in tumors and help to explore more effective immunotherapy strategies.