PREDICTION OF SUBOPTIMAL CYTOREDUCTION IN ADVANCED OVARIAN CANCER

Epithelial ovarian cancer is the fifth most common cause of death in women in developed countries. The 5-year survival rate is 92% for a localized process and 29% for patients with advanced disease with peritoneal metastases. The standard treatment for epithelial ovarian cancer is primary cytoreduction followed by platinum-based chemotherapy. When it is impossible, neoadjuvant chemotherapy is performed. Purpose of the study. Comparative evaluation of MRI and MSCT in identifying peritoneal distant spread in OC with determination of the most informative method for predicting suboptimal cytoreduction. Materials and methods. The study is based on the analysis of visual diagnostic results and surgical treatment of 46 women with OC. Before primary cytoreduction, all patients underwent mp-MRI (22) and MSCT (27). The local distribution and the presence of secondary spread were evaluated. For each method the diagnostic information for predicting suboptimal cytoreduction were analyzed and ROC AUC analysis was performed. Results. In 15.2% (7) of cases, stage IIIa / b was established, in 52.2% (24) – IIIc and in 32.6% (15) – IV stage of the disease (FIGO). High grade serous carcinoma was observed in 91.3% (42), and low grade – in 8.7% (4). The sensitivity, specificity and accuracy in determining secondary spread for MSCT were 0.65, 0.96, and 0.85, for MRI – 0.85, 0.89 and 0.87. The positive and negative predictive value of MSCT is 0.91 and 0.79, MRI is 0.91 and 0.87. Complete cytoreduction was performed in 34 (73.9%) cases, optimal – in 7 (15.2%), non-optimal – in 5 (10.9%) cases (including 2 with vital indications). The area under curve for MRI in predicting suboptimal cytoreduction by the sum of points ranged from 0.81 (according to ESMO / ESGO criteria) to 0.85. The ROC AUC based on the naive Bayesian classifier model was 0.88, which reflects the advantage of the weighted estimate of predictors based on their significance over the score Conclusions. MRI is superior to MSCT in the diagnostic informativeness of determining secondary spread of ovarian cancer. MRI has higher sensitivity for determining tumor metastases at the level of the most significant localizations from the point of view of cytoreductive intervention according to ESMO / ESGO data. The scale for assessing the respectability on the diagnostic criteria developed was superior to the approach based on the ESMO / ESGO criteria in its predictive accuracy.