Construction of a peptidase gene-based prognostic prediction model for esophageal carcinoma and identification of IMMP1L as a potential tumor biomarker

Abstract

Abstract Background Esophageal carcinoma (ESCA) is among the most prevalent malignant tumors worldwide, with a high incidence and fatality rate. However, there are presently few biomarkers connected to early diagnosis and treatment. It is essential to find the ideal biomarkers and construct reliable prognostic models. Methods and Results We identified 60 peptidase genes with differential expression in the ESCA using expression profiling data in The Cancer Genome Atlas (TCGA). Based on these genes, a prognostic risk model for ESCA was constructed by completing lasso regression analysis, ten-fold cross-validation, univariate Cox proportional hazard regression analysis, and multivariate Cox proportional hazard regression analysis. According to Kaplan-Meier survival analyses, the model demonstrated excellent performance on both the TCGA and the GEO datasets. The nomogram established by the peptidase gene and clinical variables also matched the projected and actual patient survival rates. According to the results of multivariate regression analysis, Inner Mitochondrial Membrane Peptidase Subunit 1 (IMMP1L) can be used as an independent prognostic factor for ESCA. We verified the mRNA expression level of IMMP1L in 15 esophageal cancer tissues, 12 of which were significantly increased. And we have identified the hub genes potentially targeted by IMMP1L. Conclusions we constructed and validated a prognostic risk prediction model for ESCA. And it can accurately predict survival in patients by integrating genes and tumor stage. Our results also show that IMMP1L could be used as a prospective biomarker for ESCA. These could help in the early detection and treatment of ESCA, increasing patient survival rates.