Development of a hypoxia-responsive macrophage prognostic model using single-cell and bulk RNA sequencing in pancreatic cancer.

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is characterized by a low survival rate and limited responsiveness to current therapies. The role of hypoxia in the tumor microenvironment is critical, influencing tumor progression and therapy resistance. The aim of this study was to implement the complex dynamics of the hypoxic tumor microenvironment in PDAC in a hypoxia-related prognosis model. We utilized single-cell RNA sequencing (scRNA-seq) data and integrated it with TCGA-PAAD database to identify hypoxia-responsive macrophage subsets and related genes. Kaplan-Meier survival analysis, Cox regression, and Lasso regression methods were employed to construct and validate a hypoxia-related prognostic model. The model's effectiveness was evaluated through its predictive capabilities regarding chemotherapy sensitivity and overall survival. Our research integrated data from scRNA-seq and the TCGA-PAAD database to construct a hypoxia-related prognostic model that encompassed 13 critical genes. This hypoxia model independently predicted chemotherapy response and poor outcomes, outperforming traditional clinicopathologic features. Additionally, a pan-cancer analysis affirmed the relevance of our hypoxia-related genes across multiple malignancies, particularly highlighting KRTCAP2 as a pivotal biomarker associated with worse prognosis and reduced immune infiltration. Our findings underscored the prognostic potential of hypoxia-related model and offered a novel avenue for therapeutic targeting, aiming to ameliorate outcomes in pancreatic cancer.