Background: Ovarian cancer is the leading cause of death among gynaecological cancers. Traditional serum tumour markers, like CA-125, have limited specificity, especially in premenopausal women or in conditions such as endometriosis. Human Epididymis Protein-4 (HE4) and the Risk of Ovarian Malignancy Algorithm (ROMA) have emerged as useful tools for the early and accurate differentiation of benign and malignant ovarian masses.
Objectives: To compare the diagnostic performance of HE4, CA-125, and the ROMA index in telling apart benign and malignant ovarian tumours.
Methods: A hospital-based diagnostic accuracy study involved 100 women with adnexal masses at Barasat Government Medical College over one year. Serum HE4 and CA-125 levels were measured before surgery. ROMA scores were calculated based on menopausal status. Histopathology served as the gold standard. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), ROC curves, and area under the curve (AUC) were determined.
Results: Out of 100 cases, 32 were malignant and 68 were benign according to final histopathology. Mean HE4 levels were significantly higher in malignant tumours (198.4 ± 65.2 pmol/L) compared to benign lesions (54.7 ± 22.8 pmol/L) (p<0.001). CA-125 also showed a significant increase in malignancy (231.6 ± 90.4 U/mL vs 72.5 ± 45.6 U/mL; p<0.001). HE4 had higher specificity (89.7%) than CA-125 (71.2%). The ROMA index provided the best overall performance with a sensitivity of 93.7%, specificity of 90.1%, and AUC of 0.958.
Conclusion: HE4 is better than CA-125 in specificity, while the ROMA index provides the best diagnostic accuracy for distinguishing malignant from benign ovarian masses. Including HE4 and ROMA in routine evaluations improves risk assessment and early referral.