No Difference in CEA vs CA125/CEA Ratio for Identifying Ovarian Masses
No significant difference was identified in CA125/CEA ratio vs CEA alone in distinguishing ovarian neoplasms, including benign and malignant, from metastases.
No significant difference was found between cancer antigen 125/carcinoembryonic antigen (CA125/CEA) ratio vs CEA alone in differentiating benign and malignant ovarian neoplasms from ovarian metastases in masses with both multilocular and multilocular-solid morphology, according to a study published in Ultrasound Obstetrics and Gynecology.
A team of investigators from Italy identified 396 patients with multilocular (≥5 locules) or multilocular solid ovarian masses from 3 ultrasound centers (Milan, n=195; Rome, 143; Prague, 58). Ultrasound examinations were performed following International Ovarian Tumor Analysis (IOTA) protocols, and both transvaginal and transabdominal scans were performed on each patient. Mass definition was described using IOTA terminology, including size and specific characteristics. The primary end point was histologic diagnosis of the ovarian mass.
Of all patients included in the study, 350 (88.4%) were found to have an ovarian neoplasm and 46 (11.6%) had an ovarian metastasis. Compared with ovarian metastases, ovarian neoplasms were smaller in diameter (largest diameter: 97 mm vs 146 mm); patients diagnosed with neoplasms were also generally younger than patients with metastases (median age, 55 vs median age, 62).
Although most primary epithelial ovarian carcinomas (191/197, 97%) and ovarian metastases (34/46, 73.9%) were multilocular-solid, more carcinomas than metastases were described as multilocular (6/197 vs 12/46; 3% vs 26.1%). Ovarian neoplasms were associated with fewer masses containing ≥10 locules compared with ovarian metastases (18.9% vs 54.3%).
The optimal cut-off value of CA125 for differentiating ovarian neoplasms from ovarian metastases was 265.9 U/mL. The predictive performance of the CA125 cut-off value was as follows: AUC 0.559 (95% CI, 0.478-0.639), accuracy 41.2%, sensitivity 35.7%, specificity 82.6%, positive predictive value 94%, and negative predictive value 14.4%. The optimal cut-off value of CEA for differentiating ovarian neoplasms from ovarian metastases was 2.33 ng/mL. The predictive performance of the CEA cut-off value was as follows: AUC 0.791 (95% CI, 0.711-0.870), accuracy 73.7%, sensitivity 73.1%, specificity 78.3%, positive predictive value 96.2%, and negative predictive value 27.7%. For CA125/CEA ratio, the optimal cut-off value for predicting ovarian neoplasms vs ovarian metastases was 11.92. The predictive performance of the CA125/CEA ratio cut-off value was as follows: AUC of 0.758 (95% CI, 0.683-0.833), accuracy 79.8%, sensitivity 82.3%, specificity 60.9%, positive predictive value 94.1%, and negative predictive value 31.1%.
“CA125/CEA ratio and CEA alone did not show any significant difference in distinguishing ovarian neoplasms (including benign and malignant) from ovarian metastases in described masses with multilocular and multilocular-solid morphology,” stated the authors. “Therefore, in this morphological subgroup of ovarian masses, CEA alone is enough to use for differentiating between the ovarian neoplasms (including benign and primary malignant) and ovarian metastases.”
ReferenceMoro F, Pasciuto T, Djokovic D, et al. Role of CA125/CEA ratio and ultrasound parameters in identifying metastases to the ovaries in patients with multilocular and multilocular-solid ovarian masses [published online July 5, 2018]. Ultrasound Obstet Gynecol. doi: 10.1002/uog.19174