Development and Validation of a Prognostic Classifier Based on Lipid Metabolism-Related Genes in Gastric Cancer

Xiao-Li Wei; Tian-Qi Luo; Jia-Ning Li; Zhi-Cheng Xue; Yun Wang; You Zhang; Ying-Bo Chen; Chuan Peng

doi:10.3389/fmolb.2021.691143

Development and Validation of a Prognostic Classifier Based on Lipid Metabolism-Related Genes in Gastric Cancer

Front Mol Biosci. 2021 Jun 30:8:691143. doi: 10.3389/fmolb.2021.691143. eCollection 2021.

Authors

Xiao-Li Wei¹, Tian-Qi Luo², Jia-Ning Li³, Zhi-Cheng Xue², Yun Wang⁴, You Zhang⁵, Ying-Bo Chen², Chuan Peng⁶

Affiliations

¹ Department of Medical Oncology, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.
² Department of Gastric Surgery, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.
³ Department of Clinical Research, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.
⁴ Department of Hematologic Oncology, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.
⁵ Zhongshan School of Medicine, Sun Yat-sen University Cancer Center, Guangzhou, China.
⁶ Department of Ultrasound, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou, China.

Abstract

Background: Dysregulation of lipid metabolism plays important roles in the tumorigenesis and progression of gastric cancer (GC). The present study aimed to establish a prognostic model based on the lipid metabolism-related genes in GC patients. Materials and Methods: Two GC datasets from the Gene Expression Atlas, GSE62254 (n = 300) and GSE26942 (n = 217), were used as training and validation cohorts to establish a risk predictive scoring model. The efficacy of this model was assessed by ROC analysis. The association of the risk predictive scores with patient characteristics and immune cell subtypes was evaluated. A nomogram was constructed based on the risk predictive score model and other prognostic factors. Results: A risk predictive score model was established based on the expression of 19 lipid metabolism-related genes (LPL, IPMK, PLCB3, CDIPT, PIK3CA, DPM2, PIGZ, GPD2, GPX3, LTC4S, CYP1A2, GALC, SGMS1, SMPD2, SMPD3, FUT6, ST3GAL1, B4GALNT1, and ACADS). The time-dependent ROC analysis revealed that the risk predictive score model was stable and robust. Patients with high risk scores had significantly unfavorable overall survival compared with those with low risk scores in both the training and validation cohorts. A higher risk score was associated with more aggressive features, including a higher tumor grade, a more advanced TNM stage, and diffuse type of Lauren classification of GC. Moreover, distinct immune cell subtypes and signaling pathways were found between the high-risk and low-risk score groups. A nomogram containing patients' age, tumor stage, adjuvant chemotherapy, and the risk predictive score could accurately predict the survival probability of patients at 1, 3, and 5 years. Conclusion: A novel 19-gene risk predictive score model was developed based on the lipid metabolism-related genes, which could be a potential prognostic indicator and therapeutic target of GC.

Keywords: gastric cancer; gene expression omnibus dataset; gene panel; lipid metabolism; nomogram; prognostic model.