RESUMO
Clinical trials that combine medical therapy and surgical treatment pose a number of challenges. The ANTHALYA trial investigated the efficacy and safety of adding bevacizumab to neoadjuvant carboplatin-paclitaxel chemotherapy prior to interval debulking surgery (IDS) in patients with ovarian cancer. The success of the trial depended on the efficacy and the safety of the combination and on the surgical effort. Furthermore, at the time of study design, neoadjuvant bevacizumab had not been investigated in ovarian cancer patients and it was unknown if neoadjuvant angiogenic therapy would be well tolerated or would negatively impact the success of IDS. Therefore, safety in the ANTHALYA trial was paramount. To overcome these challenges we conducted the trial in collaboration with French oncologists and surgeons who were experts in the multidisciplinary treatment of ovarian cancer. Toxicity was monitored during the neoadjuvant and IDS periods using a Bayesian approach which provided flexibility in the decision-making process regarding the continuation or rapid discontinuation of the trial in response to a safety signal. We implemented a stopping rule for safety based on a number of pre-defined bevacizumab-related complications of special interest. We also developed an electronic case report form, which greatly facilitated the collection and dissemination of safety information and other trial data. Here we describe how we used these tools to ensure the successful conduct of ANTHALYA and report how it is possible to mix a Bayesian approach for monitoring toxicity and a frequentist approach for evaluating efficacy within the same trial.