External validation of two prediction models of complete secondary cytoreductive surgery in patients with recurrent epithelial ovarian cancer

OBJECTIVE: The purpose of this study was to validate the performance of two prediction models, the AGO score and the Tian model, of complete secondary cytoreductive surgery (SCS) in patients with recurrent epithelial ovarian cancer. The predictive value of both models for survival controlled for the outcome of SCS was analyzed and known predictive factors for complete SCS were tested. METHODS: A population-based database with 408 patients, who underwent SCS between 2000 and 2013 in 38 Dutch hospitals, was used. The validation cohorts for the AGO score and the model of Tian contained 273 (66.9%) and 257 (63.0%) patients, respectively. RESULTS: The AGO score and Tian model showed a positive predictive value for complete SCS of 82.0% and 80.3% respectively, and a false negative rate of 68.5% and 55.6% respectively. A positive AGO score had no significant association with overall survival (HR 0.73; 95% CI 0.51-1.06) whereas the low risk score of the Tian model did (HR 0.62; 95% CI 0.41-0.93). A good performance status (OR 0.60; 95% CI 0.33-1.10) and the absence of ascites (OR 0.18; 95% CI 0.08-0.41) were prognostic factors for complete SCS. CONCLUSIONS: Both the AGO score and the model of Tian showed a high positive predictive value for complete SCS but also relatively high false negative rates. However, before the two prediction models can be applied in daily clinical practice the usefulness of SCS itself has to be proven first by the three ongoing randomized controlled trials: DESKTOP III trial, the GOG 213 trial and the Dutch SOCceR trial