Details

Autor(en) / Beteiligte

• Wang, Ying
• Huang, Wenting
• Zheng, Shanshan
• Wang, Liming
• Zhang, Lili
• Pei, Xiaojuan

Titel

Construction of an immune-related risk score signature for gastric cancer based on multi-omics data

Ist Teil von

• Scientific reports, 2024-01, Vol.14 (1), p.1422-1422, Article 1422

Ort / Verlag

England: Nature Publishing Group

Erscheinungsjahr

2024

Link zum Volltext

Quelle

EZB-FREE-00999 freely available EZB journals

Beschreibungen/Notizen

• Early identification of gastric cancer (GC) is associated with a superior survival rate compared to advanced GC. However, the poor specificity and sensitivity of traditional biomarkers suggest the importance of identifying more effective biomarkers. This study aimed to identify novel biomarkers for the prognosis of GC and construct a risk score (RS) signature based on these biomarkers, with to validation of its predictive performance. We used multi-omics data from The Cancer Genome Atlas to analyze the significance of differences in each omics data and combined the data using Fisher's method. Hub genes were subsequently subjected to univariate Cox and LASSO regression analyses and used to construct the RS signature. The RS of each patient was calculated, and the patients were divided into two subgroups according to the RS. The RS signature was validated in two independent datasets from the Gene Expression Omnibus and subsequent analyses were subsequently conducted. Five immune-related genes strongly linked to the prognosis of GC patients were obtained, namely CGB5, SLC10A2, THPO, PDGFRB, and APOD. The results revealed significant differences in overall survival between the two subgroups (p < 0.001) and indicated the high accuracy of the RS signature. When validated in two independent datasets, the results were consistent with those in the training dataset (p = 0.003 and p = 0.001). Subsequent analyses revealed that the RS signature is independent and has broad applicability among various GC subtypes. In conclusion, we used multi-omics data to obtain five immune-related genes comprising the RS signature, which can independently and effectively predict the prognosis of GC patients with high accuracy.