Immunotherapy is the most promising new cancer treatment for various pediatric tumors and has resulted in an unprecedented surge in the number of novel immunotherapeutic treatments that need to be evaluated in clinical trials. Most phase I/II trial designs have been developed for evaluating only one candidate treatment at a time, and are thus not optimal for this task. To address these issues, we propose a Bayesian phase I/II platform trial design, which accounts for the unique features of immunotherapy, thereby allowing investigators to continuously screen a large number of immunotherapeutic treatments in an efficient and seamless manner. The elicited numerical utility is adopted to account for the risk‐benefit trade‐off and to quantify the desirability of the dose. During the trial, inefficacious or overly toxic treatments are adaptively dropped from the trial and the promising treatments are graduated from the trial to the next stage of development. Once an experimental treatment is dropped or graduated, the next available new treatment can be immediately added and tested. Extensive simulation studies have demonstrated the desirable operating characteristics of the proposed design.

中文翻译：

---

用于儿童免疫治疗试验的贝叶斯适应性I / II期平台试验设计

免疫疗法是针对各种小儿肿瘤的最有希望的新型癌症治疗方法，已导致需要在临床试验中评估的新型免疫疗法的数量空前增加。大多数I / II期试验设计已经开发出来，一次只能评估一种候选疗法，因此对于该任务而言并非最佳。为了解决这些问题，我们提出了一种贝叶斯I / II期平台试验设计，该设计考虑了免疫疗法的独特功能，从而使研究人员能够以有效且无缝的方式连续筛选大量的免疫疗法。采用引起的数值效用来说明风险与收益之间的权衡，并量化剂量的可取性。在审判期间 从试验中适应性地放弃无效或过度毒性的治疗，有希望的治疗从试验中逐步发展到下一阶段。一旦实验性治疗方法被放弃或逐渐消失，就可以立即添加和测试下一个可用的新治疗方法。大量的仿真研究证明了所提出设计的理想工作特性。