Background: Ovarian cancer (OVC) is a devastating disease worldwide; therefore the identification of prognostic biomarkers is urgently needed. We aimed to determine a robust microRNA signature-based risk score system that could predict the overall survival (OS) of patients with OVC.
Methods: We extracted the microRNA expression profiles and corresponding clinical data of 467 OVC patients from The Cancer Genome Atlas (TCGA) database and further divided this data into training, validation and complete cohorts. The key prognostic microRNAs for OVC were identified and evaluated by robust likelihood-based survival analysis (RLSA) and multivariable Cox regression. Time-dependent receiver operating characteristic (ROC) curves were then constructed to evaluate the prognostic performance of these microRNAs. A total of 172 ovarian cancer samples and 162 normal ovarian tissues were used to verify the credibility and accuracy of the selected markers of the TCGA cohort by quantitative real-time polymerase chain reaction (PCR).
Results: We successfully established a risk score system based on a six-microRNA signature (hsa-miR-3074-5p, hsa-miR-758-3p, hsa-miR-877-5p, hsa-miR-760, hsa-miR-342-5p, and hsa-miR-6509-5p). This microRNA based system is able to characterize patients as either high or low risk. The OS of OVC patients, with either high or low risk, was significantly different when compared in the training cohort (p < 0.001), the validation cohort (p < 0.001) and the complete cohort (p < 0.001). Analysis of clinical samples further demonstrated that these microRNAs were aberrantly expressed in OVC tissues. The six-miRNA-based signature was correlated with the prognosis of OVC patients (p < 0.001).
Conclusions: The study established a novel risk score system that is predictive of patient prognosis and is a potentially useful guide for the personalized treatment of OVC patients.
Keywords: Bioinformatics analysis; Ovarian cancer; Prognosis; TCGA; microRNA signature.
© 2022. The Author(s).