The utility of human epididymal protein 4, cancer antigen 125, and risk for malignancy algorithm in ovarian cancer and endometriosis.

BACKGROUND

In women with pelvic mass, cancer antigen 125 (CA125) had not achieved satisfactory sensitivity and specificity in the detection of ovarian cancer, particularly in patients with underlying endometriosis. The aim of this study was to determine the diagnostic potential of human epididymal protein 4 (HE4), the combination of HE4+CA125, and the Risk of Ovarian Malignancy Algorithm (ROMA) for patients with pelvic mass, particularly in differentiating endometriosis from carcinoma.

METHODS

A prospective cross-sectional study was conducted at the Clinic for Gynecology and Obstetrics, Clinical Center of Serbia. Serum samples were obtained preoperatively from 108 women undergoing surgery for pelvic mass; 29 of them had ovarian carcinoma, and 79 had a nonmalignant ovarian disease (39 with benign tumor, 20 with endometriosis, 20 healthy controls). Sera were analyzed for the levels of HE4 and CA125 and were then compared with the final pathologic results. The diagnostic performance of HE4 and CA125 was estimated using receiver operating characteristic curve and area under the receiver operating characteristic curve.

RESULTS

The level of HE4 and CA125 was significantly higher among the patients with malignant tumors, compared with patients with nonmalignant disease. At the predefined specificity of 95%, HE4 and CA125 showed sensitivity of 65.5% and 58.6%, respectively, whereas the combination of HE4+CA125 reached 68.9% at the same specificity. Importantly, the level of HE4 did not differ significantly between the patients with endometriosis and with other nonmalignant diseases (which was not the case with CA125). Risk of Ovarian Malignancy Algorithm classified 96% of benign premenopausal cases as at low risk for ovarian cancer.

CONCLUSIONS

HE4 showed satisfactory capability of distinguishing endometriosis from ovarian cancer, which CA125 lacked. The Risk of Ovarian Malignancy Algorithm score proved to be useful in excluding malignant diagnosis in premenopausal women.