Abstract
High-grade serous ovarian cancer (HGSC) is an aggressive cancer with a worse clinical outcome. Therefore, studies about the prognosis of HGSC may provide therapeutic avenues to improve patient outcomes. Since genome alteration are manifested at the protein level, we integrated protein and mRNA data of ovarian cancer from The Cancer Genome Atlas (TCGA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC) and utilized the sparse overlapping group lasso (SOGL) method, a new mechanism-driven variable selection method, to select dysregulated pathways and crucial proteins related to the survival of HGSC. We found that biosynthesis of amino acids was the main biological pathway with the best predictive performance (AUC = 0.900). A panel of three proteins, namely EIF2B1, PRPS1L1 and MAPK13 were selected as potential predictive proteins and the risk score consisting of these three proteins has predictive performance for overall survival (OS) and progression free survival (PFS), with AUC of 0.976 and 0.932, respectively. Our study provides additional information for further mechanism and therapeutic avenues to improve patient outcomes in clinical practice.
Highlights
Epithelial ovarian cancer (EOC) is composed of four major histologic subtype: serous, clear cell, endometrioid, and mucinous adenocarcinomas
We integrated protein and mRNA data of ovarian cancer from The Cancer Genome Atlas (TCGA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC) and characterized High-grade serous ovarian cancer (HGSC) based on the common information from mRNA to protein
HGSC and clinical data from 169 patients were analyzed at two independent centers, JHU (n = 119) and PNNL (n = 82). 32 samples were analyzed at both centers and utilized to correct the batch effects between two sites, and merged them into a single dataset prior to analysis[12]
Summary
Epithelial ovarian cancer (EOC) is composed of four major histologic subtype: serous, clear cell, endometrioid, and mucinous adenocarcinomas. Yang et al identified nine protein markers significantly associated with progression free survival (PFS) based on the least absolute shrinkage and selection operator (lasso) and constructed a protein-driven index of ovarian cancer (PROVAR) scores to predict the recurrence time for ovarian cancer patients[11]. Univariate cox regression and lasso are effective in identifying signatures associated with the prognosis of cancer patients[13,14,15], these methods seldom combined biological information to select biomarkers, thereby it is one of the reasons that these biomarkers are not widely used in clinical practice. The sparse overlapping group lasso (SOGL) method[18], a mechanism-driven biomarker selection method, was utilized to select the main biological pathways and crucial proteins related to OS and further identified predictive proteins for OS in ovarian cancer patients. Prognosis analysis of biological pathways could provide basis for further mechanism research, and selected biomarkers of OS could provide molecule-targeted treatment and improve patient outcomes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
