A Prognostic Model of Angiogenesis and Neutrophil Extracellular Traps Related Genes Manipulating Tumor Microenvironment in Colon Cancer

Colon adenocarcinoma (COAD) is one of the most common carcinomas worldwide. The main causes of cancer-related mortality of COAD are metastases. The fundamental processes for how angiogenesis and neutrophil extracellular traps (NETs) contributing to tumor progression and metastasis are still uncertain. In our study, The Cancer Genome Atlas (TCGA)-COAD dataset (train set) and GSE17536 (test set) were analyzed. Angiogenesis potential index (API) and NETs potential index (NPI) based on angiogenesis and NETs-related genes were respectively built using bioinformatic methods and machine learning algorithms. Subjects were split into groups with low API/NPI or high API/NPI. Survival analysis showed the high API and high NPI patients with the worst survival compared with the others. Between the high API/NPI group and the other groups, differentially expressed genes (DEGs) were found. A four-gene signature (TIMP1, FSL3, CALB2, and FABP4) was included in a risk model based on least absolute shrinkage and selection operator (LASSO) analysis. Additionally, the model displayed a significant association with many immune microenvironment characteristics. Finally, we verified the clinical significance of CALB2 expression and its role to promote the invasion and migration of colon cancer cells in vitro.

Automated summary

Colon adenocarcinoma (COAD) is a common cancer with high mortality due to metastasis. This study analyzed the Cancer Genome Atlas (TCGA)-COAD dataset and GSE17536 to explore the involvement of angiogenesis and neutrophil extracellular traps (NETs) in tumor progression and metastasis. Bioinformatic methods and machine learning algorithms were used to build angiogenesis potential index (API) and NETs potential index (NPI). Survival analysis showed that high API and high NPI patients had the worst survival. Differential gene expression analysis identified a four-gene signature (TIMP1, FSL3, CALB2, and FABP4) associated with high API/NPI group. The risk model based on these genes showed significant association with immune microenvironment characteristics.