Clinical characteristics of malignant ovarian germ cell tumors

This study was aimed to investigate the clinical characteristics, treatment and prognosis effect of malignant ovarian germ cell tumors (MOGCTs). A total of 39 patients with MOGCTs treated at Gynecology Obstetrics in Tianjin Central Hospital were enrolled from January 1999 to November 2009 in this study. All the patients underwent pathological examination and diagnosis was confirmed. We used univariate and multivariate logistic regression analysis to evaluate influencing factors, and Kaplan-Meier survival analysis and cox promotional hazard regression analysis were used to evaluate the survival rate. All of patients were followed up until November 30, 2014. Eight cases (20.5%) were dead. Six months were the shortest survival time and this patient with IV stage yolk sac tumor was ill-controlled after operation, progressing to death rapidly. In respect of surgical procedures, 5 cases in pregnant group underwent laparotomy and 3 cases underwent laparoscope; In unpregnancy group, 3 cases underwent laparotomy and 3 cases underwent laparoscope. There were no significant differences between two groups (P=1.000). Seven cases in pregnant group received chemotherapy and one didn't. In unpregnancy group, all the 6 cases received chemotherapy. The chemotherapy regimen was BEP (bleomycin, etoposide and cisplatine) and 4 courses were used. There were no significant differences between two groups (P=1.000). Kaplan-Meier survival analysis showed no significant differences in the impact of Federation of Gynecology and Obstetrics (FIGO) I-II staging on survival prognosis (P > 0.05), however, manifesting significant differences were found when compared to stage IV respectively (I vs IV, X-2=27.072, P < 0.001; II vs IV, X-2=6.983, P=0.008; III vs IV, X-2=10.671, P=0.001) and between stage I and stage III (X-2=4.965, P=0.026). The result of Cox regression identified only MOGCT histological type as a predictor of overall survival (OS, Wald=6.240, P=0.012). For pregnancy analysis, univariate logistic regression analysis showed that no significant difference was found between pregnancy group and non-pregnancy (Wald=2.410, P=0.121). A multivariate logistic regression analysis also showed no statistical differences between two groups. FIGO staging and MOGCT histological type are critical factors affecting the prognosis of the patients with MOGCTs. The pregnant outcome of the patients receiving chemotherapy is desirable, and there are no significant effects of age, surgical-pathological staging, MOGCT histological type, surgical procedures and tumor size on the pregnancy outcome.