earticle

영어

In this study, a new diagnosis model based on multiple serum biomarkers and meno-pausal information was developed for the early detection of ovarian cancer. A total of 254 serum samples from 202 patients with benign tumors and 52 patients with ovarian cancer were used. The concentrations of 22 ovarian-cancer-specific serum biomarkers were de-termined using Luminex, and premenopausal and postmenopausal status was binary mapped as 0 and 1. To identify the best biomarker combinations effectively differentiating ovarian cancer from benign tumors, all combinations with 2 and 3 biomarkers were eval-uated according to the area under the receiver operating characteristic curve (AUC). Menopausal information was included in the combinations of 2 and 3 biomarkers that showed the best classification performance, with AUC values of 84.53 and 87.01, respec-tively. The diagnosis model for ovarian cancer developed in this study based on the best biomarker combinations proved that combinations of biomarkers together with menopau-sal information show more accurate performance than mere combinations of biomarkers alone.