Abstract
BACKGROUND: Preoperative differentiation between benign and malignant masses can be challenging. The aim of this research was to evaluate the performance of a modified multivariate index assay (MIA) in detecting ovarian cancer and to compare the effectiveness of gynecologist assessment, cancer antigen (CA) 125, and MIA for identifying ovarian masses with high suspicion of malignancy. RESULTS: This prospective observational study included 150 women with ovarian masses who underwent surgery in the Maternity Teaching Hospital from December 2014 to May 2016. Preoperative estimation of modified MIA, assessment by a gynecologist, and CA 125 level correlated with the surgical histopathology. A modified MIA was implemented because of lack of access to the software typically used. Among 150 enrolled women there were 30 cases of malignancy, including 8 cases (26%) of early-stage ovarian cancer and 22 cases (74%) of late-stage cancer. MIA showed high specificity (96.7%) in detecting cancer and a sensitivity of 70%, with a positive predictive value of 84% and a negative predictive value of 92.8%. No significant differences were detected between the MIA results and the histopathology results (P = 0.267). For early-stage ovarian cancer, the sensitivity of MIA was 100% compared with 75% for CA 125 alone. CONCLUSION: MIA seems to be effective for evaluation of ovarian tumors with higher specificity and positive predictive value than CA 125 while maintaining high negative predictive value and with only a slightly lower overall sensitivity. For evaluation of early-stage ovarian cancer, MIA showed a much higher sensitivity that markedly outperformed CA 125 alone. This modified MIA strategy may be particularly useful in low resource setting.