[Advanced ovarian cancer: criteria of resectability]

Residual disease is the first predictive factor for survival in women with ovarian cancer. Neoadjuvant chemotherapy still has not proved superiority in terms of overall survival compared with complete initial resection. Hence, initial surgery remains the cornerstone of management for patients with advanced ovarian cancer. Various parameters have been proposed to evaluate the ability of complete resection. Clinical evaluation of peritoneal carcinomatosis and ascite is not relevant enough while general status (ASA score) is correlated with the risk of postoperative complications. Preoperative dosage of CA-125 higher than 500 UI/L seems to be related with an increase risk of sub-optimal surgery. Recently, some authors challenged its prognostic value. For a CA-125 threshold at 500 UI/L, sensitivity, specificity, positive and negative predictive values range from 58 to 78%, 64 to 89%, 64 to 84% and 35.7 to 85.4%, respectively. Imaging criteria failed to report concordant results. Indeed, sensitivity, specificity, positive and negative predictive values vary from 52 to 100%, 75 to 100%, 49% to 100% and 50 to 100%, respectively. High-correlation has been demonstrated for carcinomatosis scores evaluating the dissemination of the disease (AUCs of ROCs curves higher than 0.6). At laparotomy, for an Eisenkop score under 6, 99% of patients could benefit of complete resection. At laparoscopy, for a Fagotti score under 4, a complete cytoreduction could be obtained in 78% of patients. This score had the best AUC (0.76). Various scoring systems are available for surgeons to evaluate the resecability of advanced ovarian cancer. Among them, peroperative scoring systems appear the best tool and should be recommended in routine especially at first laparoscopy. Finally, due to the impact on survival of complete initial resection, women with advanced ovarian cancer should be referred to specialized centres.