Validity of Cancer Antigen-125 (CA-125) and Risk of Malignancy Index (RMI) in the Diagnosis of Ovarian Cancer.

OBJECTIVE: We sought to determine the validity of cancer antigen 125 (CA-125) and the risk of malignancy index (RMI) in the diagnosis of ovarian cancer in women presenting with adnexal lesions of various histopathology types.

METHODS: This retrospective cross- sectional study included all women with adnexal lesions who were evaluated at the Royal Hospital, Oman, between January 2012 and December 2014. The inclusion criteria included women who underwent surgical intervention and who had preoperative CA-125 testing and pelvic ultrasound in the work-up plan of their management. The surgical intervention was usually followed by a histopathological diagnosis of the nature of the lesion, which was used as the gold standard for the evaluation of both CA-125 and RMI.

RESULTS: The cohort included 361 women who had serum CA-125 and pelvic ultrasound prior to the surgical intervention of the adnexal lesion. Of these women, 61 (17%) had malignant ovarian lesions. Using the proposed cut-off 35 U/ml for CA-125 and 200 for RMI, the CA-125 test was more sensitive for detecting the majority of malignant ovarian tumors compared to the RMI (69% vs. 57%). Both tests were more sensitive in detecting epithelial ovarian cancer compared to other ovarian cancers. However, RMI was more specific in excluding benign ovarian lesions compared to CA-125 (81% vs. 68%). Additionally, RMI had a better area under the curve compared to CA-125 (0.771 vs. 0.745; p<0.005). Lowering the RMI cut-off to 150 resulted in a better sensitivity (62% vs. 57%) and had an acceptable specificity (78% vs. 81%) compared to a cut-off of 200.

CONCLUSION: Both CA-125 and RMI have good validity in the diagnosis of ovarian tumors. CA-125 has higher sensitivity; however, RMI has higher specificity. In combination, CA-125 might be more valid for the diagnosis of malignant ovarian cancer while RMI is more valid for excluding the diagnosis of these tumors. Differential use of these two tools will improve the triage of women with suspected ovarian tumors since both are measured in their work-up. We recommended the use of both tools in primary care to reduce referral to gynecology or oncology units.