IDDF2024-ABS-0209 Bioinformatics-based analysis of PCDHB5, a relevant prognostic gene for angiogenesis in gastric cancer

Background

Angiogenesis is closely related to the prognosis of gastric cancer patients. In recent years, finding new anti-angiogenic targets for gastric cancer, developing anti-angiogenic targeted therapeutic agents for gastric cancer, and improving the prognosis of gastric cancer patients are still urgent research issues. In this study, we aimed to identify angiogenesis-related genes with prognostic value to provide more efficient anti-angiogenesis targets in gastric cancer, and to provide new ideas for the development of anti-angiogenesis-targeted therapeutic agents for gastric cancer and the improvement of the prognosis of gastric cancer clinical treatment.

Methods

The Stomach Adenocarcinoma (STAD) datasets, gene expression data, clinical information, and genomic mutation data were obtained using by The Cancer Genome Atlas (TCGA) database. The samples of adjacent tissue and tumor tissue in TCGA-STAD were examined, revealing significant separability in gene expression. The gastric cancer dataset was searched using The International Cancer Genome Consortium (ICGC) and the Gene Expression Omnibus (GEO) to obtain differential expressed genes (DEGs), and a total of 255 angiogenesis-related genes (ARGs) were identified for consistent clustering analysis using the relevant pathway information from AHGIOGENES, MSigDB, and GO databases, and the machine learning algorithm was used to establish a prognostic model based on the subtypes of the ARGs, and the gastric cancer patients were subtype classification performed. Cox regression analysis was performed on genes showing differential expression between the two subtypes to identify key genes.

Results

Ten feature genes (FGF1, KCND2, SERPINE1, PTPRD, BCHE, CYTL1, MATN3, APOD, PCDHB5, STK32A) were identified using the LASSO algorithm. Survival prognosis showed significant differences between the two groups(P<0.0001). Validation of the prognosis model was performed using a validation set based on the GEO database, and the results were consistent with the TCGA training set, indicating that this prognostic model is an independent prognostic factor for gastric cancer.

Conclusions

Ten genetic markers associated with angiogenesis were developed that predict overall survival in GC. It is considered to be a valuable prognostic model and provides new perspectives for targeted therapies.