Identification of ATXN3 and UBE2S as prognostic markers for osteosarcoma by a regression model for integrating multiomics
WCRJ 2021;
8
: e2051
DOI: 10.32113/wcrj_20217_2051
Topic: Cancer biology
Category: Original article
Abstract
Objective: Osteosarcoma (OS) is a malignant tumor in children, which seriously endangers children's health. However,few studies have been conducted to identify potential targets of OS. The purpose of this study was to identify the protein markers in plasma that are associated with the occurrence, metastasis, and prognosis of OS by combining multinomial analysis and constructing regression models.
Materials and Methods: The miRNA expression profile GSE65071 dataset and protein expression (GSE78192) dataset were downloaded from the Gene Expression Omnibus (GEO) database. The differentially expressed proteins (DEPs) and miRNAs (DEMIs) were identified from primary OS vs. normal control, and metastatic OS vs. primary OS, respectively. A miRNA-protein network was constructed based on miRNA-protein pairs. Functional analysis of DEPs involved in miRNA-protein network was utilized to obtain insights into the functions of the network. Subsequently, we screened out the genes differentially expressed similarly in both plasma and osteosarcoma tissues of OS patients. Eventually, genes in miRNA-protein network closely related to survival were identified by the intersection of lasso regression and K-M survival analysis, and Cox regression model was constructed for further analysis.
Results: A total of 277 DEMIs and 50 DEPs, 3 DEMIs and 254 DEPs were identified with contact of OS occurs and metastasis. Furthermore, 11 proteins involved in miRNA-protein network were confirmed to exhibit the same dysregulation in both primary OS plasma and tissue and 3 proteins exhibit dysregulation in metastatic patients. Moreover, for patients without metastasis, UBE2S and ATXN3 not only have the same differential expression trend in both plasma and tumor tissues of OS patients, but also are closely related to the survival time of patients.
Conclusions: In OS plasma, 11 proteins, including UBE2S, ATXN3 and LARP6 are potential plasma markers for the diagnosis of OS. Totally 3 proteins (UGT3A1, IGF1R and SLC7A1) were the potential plasma markers to predict metastasis of OS. Furthermore, the C-index and the AUC at 4-year of Cox regression model were 0.79 and 0.97, suggesting good reliability and potential of UBE2S and ATXN3 for prognostic prediction and targeted therapy of primary OS.
Materials and Methods: The miRNA expression profile GSE65071 dataset and protein expression (GSE78192) dataset were downloaded from the Gene Expression Omnibus (GEO) database. The differentially expressed proteins (DEPs) and miRNAs (DEMIs) were identified from primary OS vs. normal control, and metastatic OS vs. primary OS, respectively. A miRNA-protein network was constructed based on miRNA-protein pairs. Functional analysis of DEPs involved in miRNA-protein network was utilized to obtain insights into the functions of the network. Subsequently, we screened out the genes differentially expressed similarly in both plasma and osteosarcoma tissues of OS patients. Eventually, genes in miRNA-protein network closely related to survival were identified by the intersection of lasso regression and K-M survival analysis, and Cox regression model was constructed for further analysis.
Results: A total of 277 DEMIs and 50 DEPs, 3 DEMIs and 254 DEPs were identified with contact of OS occurs and metastasis. Furthermore, 11 proteins involved in miRNA-protein network were confirmed to exhibit the same dysregulation in both primary OS plasma and tissue and 3 proteins exhibit dysregulation in metastatic patients. Moreover, for patients without metastasis, UBE2S and ATXN3 not only have the same differential expression trend in both plasma and tumor tissues of OS patients, but also are closely related to the survival time of patients.
Conclusions: In OS plasma, 11 proteins, including UBE2S, ATXN3 and LARP6 are potential plasma markers for the diagnosis of OS. Totally 3 proteins (UGT3A1, IGF1R and SLC7A1) were the potential plasma markers to predict metastasis of OS. Furthermore, the C-index and the AUC at 4-year of Cox regression model were 0.79 and 0.97, suggesting good reliability and potential of UBE2S and ATXN3 for prognostic prediction and targeted therapy of primary OS.
To cite this article
Identification of ATXN3 and UBE2S as prognostic markers for osteosarcoma by a regression model for integrating multiomics
WCRJ 2021;
8
: e2051
DOI: 10.32113/wcrj_20217_2051
Publication History
Submission date: 30 Mar 2021
Revised on: 11 May 2021
Accepted on: 30 Jun 2021
Published online: 12 Jul 2021
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.