Ovarian carcinoma (OV) is one of the most lethal gynecological malignancies globally, and the overall 5-year survival rate of OV was 47% in 2018 according to American data. To increase the survival rate of patients with OV, many researchers have sought to identify biomarkers that act as both prognosis-predictive markers and therapy targets. However, most of these have not been suitable for clinical application. The present study aimed at constructing a predictive prognostic nomogram of OV using the genes identified by combining The Cancer Genome Atlas (TCGA) dataset for OV with the immune score calculated by the Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data algorithm. Firstly, the algorithm was used to calculate the immune score of patients with OV in the TCGA-OV dataset. Secondly, differentially expressed genes (DEGs) between low and high immune score tissues were identified, and Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis was performed to predict the functions of these DEGs. Thirdly, univariate, multivariate and Lasso Cox's regression analyses were carried out step by step, and six prognosis-related DEGs were identified. Then, Kaplan-Myer survival curves were generated for these genes and validated by comparing their expression levels to further narrow the range of DEGs and to calculate the risk score. Two genes were identified, cell division cycle 20B and patatin-like phospholipase domain containing 5, which were both shown to have higher expression levels in OV tissues and to be significantly associated with the prognosis of OV. Next, a nomogram was created using these two genes and age, and using the receiver operating characteristic (ROC) curve and calibration curve, the effectiveness of the nomogram was validated. Finally, an external validation was conducted for this nomogram. The ROC showed that the areas under the curve (AUCs) of the 3- and 5-year overall survival predictions for the nomogram were 0.678 and 0.62, respectively. Moreover, the ROC of the external validation model showed that the AUCs of the 3- and 5-year were 0.699 and 0.643, respectively, demonstrating the effectiveness of the generated nomogram. In conclusion, the present study has identified two immune-related genes as biomarkers that reliably predict overall survival in OV. These biomarkers might also be potential molecular targets of immune therapy to treat patients with OV.