Identification of metabolism-related genes for predicting peritoneal metastasis in patients with gastric cancer.

OBJECTIVE

The reprogramming of metabolism is an important factor in the metastatic process of cancer. In our study, we intended to investigate the predictive value of metabolism-related genes (MRGs) in recurrent gastric cancer (GC) patients with peritoneal metastasis.

METHODS

The sequencing data of mRNA of GC patients were obtained from Asian Cancer Research Group (ACRG) and the GEO databases (GSE53276). The differentially expressed MRGs (DE-MRGs) between a cell line without peritoneal metastasis (HSC60) and one with peritoneal metastasis (60As6) were analyzed with the Limma package. According to the LASSO regression, eight MRGs were identified as crucially related to peritoneal seeding recurrence in patients. Then, disease free survival related genes were screened using Cox regression, and a promising prognostic model was constructed based on 8 MRGs. We trained and verified it in two independent cohort.

RESULTS

We confirmed 713 DE-MRGs and the enriched pathways. Pathway analysis found that the MRG-related pathways were related to tumor metabolism development. With the help of Kaplan-Meier analysis, we found that the group with higher risk scores had worse rates of peritoneal seeding recurrence than the group with lower scores in the cohorts.

CONCLUSIONS

This study developed an eight-gene signature correlated with metabolism that could predict peritoneal seeding recurrence for GC patients. This signature could be a promising prognostic model, providing better strategy in treatment.