Re-engineering Precision Therapeutics Through N-of-1 Trials

该计划的目标是开发、测试和实施一个创新的技术平台，用于 改变精确疗法的N-of-1试验。目前，临床医生正在进行临床会诊 他们正试图确定针对个别患者的最佳治疗方法。这些相遇很可能是 不成功。临床医生依赖最好的可用证据(例如，来自第三阶段随机临床试验的结果； RCT)用于向患者推荐治疗方法。然而，传统的患者间随机对照试验仅提供估计。 在这些试验中，治疗对普通患者的影响。然而，个别患者往往会有反应。 与III期随机对照试验中的普通患者不同，因此，治疗反应的异质性 每天都有这些临床决定。 确定患者最佳治疗方案的最科学、最严谨、最具潜在变革性的方法是单患者试验(N-of-1)。N-of-1试验是多个交叉试验，通常是随机的，而且通常 蒙面，在单个患者内进行，客观、连续和在现实世界中收集的数据，用于 与安慰剂或其他积极疗法相比，有足够的时间来确定该疗法是否 对特定患者来说是最理想的。它们还提供有关非目标操作的信息，如副作用，因此 关于治疗对个别患者的整体益处和危害，可能会出现更复杂的图景。临床医生和患者不会常规地从事这种类型的科学努力，因为他们缺乏工具。 在这项计划中，我们将创建一个电子平台，允许临床医生和/或患者订购和 进行一项单患者试验。然后，我们将收集随机对照试验数据，以便能够估计使用 这种方法。为此，我们将使用我们的平台针对3种不同的健康状况进行3次实验，每种情况 公共卫生负担高，治疗反应异质性高，精确度优先 由先前采访的临床医生和患者确定的治疗方法。对于每个人，我们都会 随机选择60名患者，接受N-of-1试验，或接受常规护理。然后，我们将测试一个实用的设计 模拟该平台在发布日期后将如何在临床实践中实施。该应用程序将是 嵌入到临床工作流程中，临床医生能够在 通过电子健康记录直接进行N-of-1试验。在这个实验中，我们将随机选择200名患者 接受务实的N-of-1试验，或接受通常的护理。因此，我们将能够比较我们的新精度 治疗方法通常是为患者确定治疗方法的方法。 这个1中N试验平台还将促进发现治疗反应的范式转换方法 表型，因为我们将建立一个面向公众的N-of-1协议和工具注册表，以及一个未识别的存储库 N-of-1试验的数据。然后，可以挖掘这个N-of-1试验结果的国际数据库来识别表型 按治疗反应性分类。

The objective of this proposal is to develop, test, and implement an innovative technology platform for conducting N-of-1 trials that transforms precision therapeutics. Right now, clinicians are engaging in clinical encounters at which they are trying to determine the best therapy for individual patients. These encounters are likely to be unsuccessful. Clinicians rely on the best available evidence (e.g., results from phase III randomized clinical trials; RCTs) for recommending therapies to a patient. Yet, conventional, between-patient RCTs only provide estimates of the effect of therapies on the average patient in those trials. Individual patients, however, often respond differently than the average patient in the phase III RCTs, and thus, heterogeneity of therapy response plagues these clinical decisions every day. The most scientifically rigorous-- and potentially transformative--method for determining optimal therapy for apatient is a single-patient (N-of-1) trial. N-of-1 trials are multiple crossover trials, usually randomized, and often masked, conducted within a single patient, with data collected objectively, continuously, and in the real-world, for a sufficient time period to determine whether the therapy, compared to a placebo or other active therapy, is optimal for a particular patient. They also yield information on off-target actions, such as side-effects, so that a more complex picture can emerge about the overall benefits and harms of a therapy for an individual patient.Clinicians and patients do not routinely engage in this type of scientific endeavor because they lack the tools. In this proposal, we will create an electronic platform that will allow clinicians and/or patients to order and conduct a single-patient trial. We will then collect RCT data to be able to estimate the benefit (if any) from using this approach. To do so, we will conduct 3 experiments using our platform for 3 different health conditions, each of which has high public health burden, high heterogeneity of therapy response, and high priority for a precision therapeutics approach as determined by previously interviewed clinicians and patients. For each, we will randomize 60 patients to receive an N-of-1 trial, or to receive usual care. We will then test a pragmatic design that simulates how the platform will be implemented in clinical practice after its release date. The application will be embedded in the clinical workflow, with clinicians having the capability of referring and tracking their patients in N-of-1 trials directly through the electronic health record. In this experiment, we will randomize 200 patients to receive the pragmatic N-of-1 trial, or to receive usual care. We will thus be able to compare our new precision therapy approach to the way therapies are typically determined for a patient. This N-of-1 trials platform will also facilitate a paradigm-shifting approach to discovery of therapeutic response phenotypes, as we will build a public facing registry of N-of-1 protocols and tools and a repository of de-identified data from N-of-1 trials. This international database of N-of-1 trial results can then be mined to identify phenotypes categorized by treatment responsiveness.