RESUMEN
OBJECTIVES: This study aimed to evaluate the diagnostic performances of the Copenhagen Index (CPH-I) and Risk of Ovarian Malignancy Algorithm (ROMA) in the preoperative prediction of ovarian cancer. METHODS: In a prospective cohort study, data were collected from 475 patients with ovarian masses diagnosed by gynecologic examination / ultrasound who were hospitalized at the Departments of Obstetrics and Gynecology, Hue University of Medicine and Pharmacy Hospital and Hue Central Hospital, Vietnam, between January 2018 and June 2020. ROMA and CPH-I were calculated based on measurements of serum carbohydrate antigen (CA-125) and human epididymis protein (HE4). The final diagnosis was based on clinical features, radiologic and histologic findings, and the International Federation of Gynecology and Obstetrics (FIGO) 2014 stages of ovarian cancer were recorded. Matching the values of ROMA and CPH-I to postoperative histopathology reports resulted in the preoperative prediction values. RESULTS: Among the 475 women, 408 had benign tumors, 5 had borderline tumors and 62 had malignant tumors. The two indices showed similar discriminatory performances with no significant differences (p > 0.05). At an optimal cut-off, the sensitivities/specificities of ROMA and CPH-I for ovarian cancer diagnosis were 74.2% and 91.8%, 87.1% and 78.5%, respectively. The optimal cut-off for CPH-I was 1.89%. The areas under the ROC curves (AUCs) of ROMA and CPH-I were 0.882 (95% CI: 0.849-0.909) and 0.898 (95% CI: 0.867-0.924), respectively. CONCLUSIONS: The introduction of the Copenhagen Index to help stratify the malignancy risk of ovarian tumors, irrespective of menopausal status, might be applied as a simple alternative with a similar efficacy to ROMA in clinical practice.