Develop a nomogram to predict overall survival of patients with borderline ovarian tumors

Abstract

BACKGROUND The prognosis of borderline ovarian tumors (BOTs) has been the concern of clinicians and patients. It is urgent to develop a model to predict the survival of patients with BOTs. AIM To construct a nomogram to predict the likelihood of overall survival (OS) in patients with BOTs. METHODS A total of 192 patients with histologically verified BOTs and 374 patients with epithelial ovarian cancer (EOC) were retrospectively investigated for clinical characteristics and survival outcomes. A 1:1 propensity score matching (PSM) analysis was performed to eliminate selection bias. Survival was analyzed by using the log-rank test and the restricted mean survival time (RMST). Next, univariate and multivariate Cox regression analyses were used to identify meaningful independent prognostic factors. In addition, a nomogram model was developed to predict the 1-, 3-, and 5-year overall survival of patients with BOTs. The predictive performance of the model was assessed by using the concordance index (C-index), calibration curves, and decision curve analysis (DCA). RESULTS For clinical data, there was no significant difference in body mass index, preoperative CA199 concentration, or tumor localization between the BOTs group and EOC group. Women with BOTs were significantly younger than those with EOC. There was a significant difference in menopausal status, parity, preoperative serum CA125 concentration, Federation International of gynecology and obstetrics (FIGO) stage, and whether patients accepted postoperative adjuvant therapy between the BOT and EOC group. After PSM, patients with BOTs had better overall survival than patients with EOC ( P value = 0.0067); more importantly, the 5-year RMST of BOTs was longer than that of EOC ( P value = 0.0002, 95%CI -1.137 to -0.263). Multivariate Cox regression analysis showed that diagnosed age and surgical type were independent risk factors for BOT patient OS ( P value < 0.05). A nomogram was developed based on diagnosed age, preoperative serum CA125 and CA199 Levels, surgical type, FIGO stage, and tumor size. Moreover, the c-index (0.959, 95% confidence interval 0.8708–1.0472), calibration plot of 1-, 3-, and 5-year OS, and decision curve analysis indicated the accurate predictive ability of this model. CONCLUSION Patients with BOTs had a better prognosis than patients with EOC. The nomogram we constructed might be helpful for clinicians in personalized treatment planning and patient counseling.