Targeted agents and immunotherapies have revolutionized cancer treatment, offering promising options for various cancer types. Unlike traditional therapies the principle of “more is better” is not always applicable to these new therapies due to their unique biomedical mechanisms. As a result, various phase I–II clinical trial designs have been proposed to identify the optimal biological dose that maximizes the therapeutic effect of targeted therapies and immunotherapies by jointly monitoring both efficacy and toxicity outcomes. This review article examines several innovative phase I–II clinical trial designs that utilize accumulated efficacy and toxicity outcomes to adaptively determine doses for subsequent patients and identify the optimal biological dose, maximizing the overall therapeutic effect. Specifically, we highlight three categories of phase I–II designs: efficacy-driven, utility-based, and designs incorporating multiple efficacy endpoints. For each design, we review the dose–outcome model, the definition of the optimal biological dose, the dose-finding algorithm, and the software for trial implementation. To illustrate the concepts, we also present two real phase I–II trial examples utilizing the EffTox and ISO designs. Finally, we provide a classification tree to summarize the designs discussed in this article.

中文翻译：

---

适应性 I-II 期临床试验设计确定靶向药物和免疫疗法的最佳生物剂量

靶向药物和免疫疗法彻底改变了癌症治疗，为各种癌症类型提供了有希望的选择。与传统疗法不同，由于其独特的生物医学机制，“越多越好”的原则并不总是适用于这些新疗法。因此，人们提出了各种I-II期临床试验设计，通过联合监测疗效和毒性结果来确定最佳生物剂量，最大限度地发挥靶向治疗和免疫疗法的治疗效果。这篇综述文章探讨了几种创新的 I-II 期临床试验设计，这些设计利用累积的疗效和毒性结果来适应性地确定后续患者的剂量并确定最佳生物剂量，从而最大化总体治疗效果。具体来说，我们重点介绍三类 I-II 期设计：功效驱动型、基于实用性以及包含多个功效终点的设计。对于每个设计，我们都会审查剂量结果模型、最佳生物剂量的定义、剂量寻找算法和试验实施软件。为了说明这些概念，我们还展示了两个利用 EffTox 和 ISO 设计的真实 I-II 期试验示例。最后，我们提供了一个分类树来总结本文中讨论的设计。