Bioinformatics analysis to screen the key prognostic genes in ovarian cancer

Abstract

BackgroundOvarian cancer (OC) is a gynecological oncology that has a poor prognosis and high mortality. This study is conducted to identify the key genes implicated in the prognosis of OC by bioinformatic analysis.MethodsGene expression data (including 568 primary OC tissues, 17 recurrent OC tissues, and 8 adjacent normal tissues) and the relevant clinical information of OC patients were downloaded from The Cancer Genome Atlas database. After data preprocessing, cluster analysis was conducted using the ConsensusClusterPlus package in R. Using the limma package in R, differential analysis was performed to identify feature genes. Based on Kaplan-Meier (KM) survival analysis, prognostic seed genes were selected from the feature genes. After key prognostic genes were further screened by cluster analysis and KM survival analysis, they were performed functional enrichment analysis and multivariate survival analysis. Using the survival package in R, cox regression analysis was conducted for the microarray data of GSE17260 to validate the key prognostic genes.ResultsA total of 3668 feature genes were obtained, among which 75 genes were identified as prognostic seed genes. Then, 25 key prognostic genes were screened, including AXL, FOS, KLF6, WDR77, DUSP1, GADD45B, and SLIT3. Especially, AXL and SLIT3 were enriched in ovulation cycle. Multivariate survival analysis showed that the key prognostic genes could effectively differentiate the samples and were significantly associated with prognosis. Additionally, GSE17260 confirmed that the key prognostic genes were associated with the prognosis of OC.Conclusion AXL, FOS, KLF6, WDR77, DUSP1, GADD45B, and SLIT3 might affect the prognosis of OC.

Coexpression network analysis identified Krüppel‐like factor 6 (KLF6) association with chemosensitivity in ovarian cancer

“…The results indicated that relatively high KLF6 expression in ovarian cancer leads to poor OS and PFS. This is consistent with a previous report of Li et al…”

“…The results indicated that relatively high KLF6 expression in ovarian cancer leads to poor OS and PFS. This is consistent with a previous report of Li et al 33 Previous studies have showed that KLF6 is associated with cell proliferation, angiogenesis, invasion, and tumorigenicity. 34 In addition, downregulation of KLF6 also can promote hepatocyte regeneration by affecting autophagy.…”

GADD45B as a Prognostic and Predictive Biomarker in Stage II Colorectal Cancer

“…GADD45B eta is a stress-activated protein that plays a vital role in regulating apoptosis, proliferation, and DNA repair. It has been demonstrated to be an indicator for predicting clinical outcomes of gastric cancer [ 18 ], ovarian cancer [ 19 ], and glioma [ 20 ] according to the previous studies. Verzella et al found that elevated GADD45B expression correlated with rapid disease progression in 13 of the top 15 solid cancers for mortality and the patient cohorts expressing high GADD45B levels exhibited significantly shorter recurrence-free survival and OS than the corresponding cohorts, which expressed low GADD45B messenger-RNA (mRNA) levels [ 16 ].…”

Role of microRNAs as Clinical Cancer Biomarkers for Ovarian Cancer: A Short Overview

“…miRNA expression patterns were identified and correlated with ovarian cancer samples through PCA [119][120][121], arriving at a highly differentiated profile among ovarian tumoral cell lines (e.g., OVCAR3, OVCAR420 etc), ovarian cancer tissues, and non-tumorigenic cells (HOSE-B). Another approach for determining differential expression between groups is constituted by the t-statistic method linear models for microarray data (LIMMA) [122]. The agglomerative hierarchical clustering method, as employed by Dahiya et al, used a complete linkage method to test the natural tumor samples grouping based on gene-expression profiles correlations [119].…”

“…The model that was employed was a multilayer perceptron with the number of hidden layers being optimized to avoid overfitting [127]. Differences in false-positive and false-negative assignment were compared while using Fisher's exact test, and by the use of 14 miRNAs from pre-and post-operative original signature, provided very good discriminatory power on the testing set (AUC 0.93, 95% CI 0.81-1.00), with an overall sensitivity of 75% and specificity of 100% [122].…”