An integrated model of clinical information and gene expression for prediction of survival in ovarian cancer patients

Accumulating evidence shows that clinical factors alone are not adequate for predicting the survival of patients with ovarian cancer (OvCa), and many genes have been found to be associated with OvCa prognosis. The objective of this study was to develop a model that integrates clinical information and a gene signature to predict the survival durations of patients diagnosed with OvCa. We constructed mRNA and microRNA expression profiles and gathered the corresponding clinical data of 552 OvCa patients and 8 normal controls from The Cancer Genome Atlas. Using univariate Cox regression followed by a permutation test, elastic net-regulated Cox regression, and ridge regression, we generated a prognosis index consisting of 2 clinical variables, 7 protective mRNAs, 12 risky mRNAs, and 1 protective microRNA. The area under the curve of the receiver operating characteristic of the integrated clinical-and-gene model was 0.756, larger than that of the clinical-alone model (0.686) or the gene-alone model (0.703). OvCa patients in the high-risk group had a significantly shorter overall survival time compared with patients in the low-risk group (hazard ratio = 8.374, 95% confidence interval = 4.444-15.780, P = 4.90 × 10−11, by the Wald test). The reliability of the gene signature was confirmed by a public external data set from the Gene Expression Omnibus. Our conclusions that we have identified an integrated clinical-and-gene model superior to the traditional clinical-alone model in ascertaining the survival prognosis of patients with OvCa. Our findings may prove valuable for improving the clinical management of OvCa.