Objective To evaluate the clinical effect of GN chemotherapy protocol and GC chemotherapy protocol treatment of recurrent ovarian cancers. Methods We retrospectively analyzed the data of 60 patients with recurrent ovarian cancers in our hospital from January 2013 to June 2014. Divided into gemcitabine and nedaplatin (GN)chemotherapy group and Gemcitabine and carboplatin(GC)chemotherapy group, 30 patients in each group.Patients in GP protocol group were given Gemcitabine 1.0 g/m²,d1,8;and naphthalene(80 mg/m²,d1),21d was a period of treatment; Patients in GC protocol group were given gemcitabine 1.0 g/m²,d1,8;and carboplatin AUC=5 by calculation,21d was a period of treatment. Results For patients in GN protocol group,the short term response rate was 56.0%.For patients in GC protocol group,the short term response rate was 60.0%,the difference was not statistically significant(χ²=0.069,P=0.793). Two of the most common toxicities were myelosuppression, incidence of myelosuppression GC was slightly higher than GN, and the difference was not statistically significant. Conclusion The therapeutic effects of two chemotherapy protocols have no statistically significant difference in treatment of platinum sensitive recurrent ovarian cancer. Toxicity of GN group is light.