Optimizing Outcome Measures for Clin. Trials in Pre-Clinical Alzheimer's Disease

摘要 这笔赠款将开发统计方法，为临床试验和纵向试验得出最佳终点 对认知老化、轻度认知障碍和阿尔茨海默病的队列研究，并将发表新的 使用这些方法得出的统计上有效的临床试验结果衡量标准。待开发的方法 这笔拨款将大大提高临床试验的效率，降低成本并增加 确定有效治疗方法的概率。具体目标是： 具体目标1。推导和应用工具总分的优化计算方法。 这是我们早先使用项目反应理论找到项目的最佳得分的工作的扩展，当 计算一个乐器的总分。我们之前的工作是在横截面数据上训练重新评分算法。 扩展将以纵向数据为基础进行训练。我们有试点数据显示， 用该方法选择仪器所需的样本量。 具体目标2.推导和应用优化组合标尺结构的方法。 将认知和功能测量相结合的复合量表有望显著提高效率 这是轻度认知障碍临床试验的重要组成部分，也是积极研究的领域。我们已经得出了一个最优的 公式，并通过实际数据和模拟显示了相对于当前方法的更好的性能。 具体目标3.通过应用所开发的方法来演示从头开始的结果衡量的发展 在目标1和目标2中，针对一种不同但相关的进展性疾病，额颞痴呆(FTD)。 这项工作旨在展示我们的方法对其他疾病领域的普适性。此外， 为执行这项资助而开发的软件将在方法发布时发布，并将 适用于任何以数量性状为基础的慢性疾病的仪器开发 临床试验的终点。 这笔赠款完全是因为观察到慢性进展性疾病的临床试验是 贵得离谱。在阿尔茨海默病研究中，这限制了我们测试新疗法和 找到治疗这种疾病的方法。对于不太常见的疾病，如FTD和进行性核上性瘫痪 随着临床试验研究对象的可获得性，对更有效的终点的需求更加迫切 进一步限制了我们测试治疗的能力。每一位参加临床试验的受试者都是宝贵的资源。这 GRANT旨在改进方法，以最佳地利用从注册的受试者获得的所有信息 临床试验和增加确定有效治疗方法的可能性。

ABSTRACT This grant will develop statistical methods for deriving optimal endpoints for clinical trials and longitudinal cohort studies of cognitive aging, mild cognitive impairment, and Alzheimer's disease, and will publish new statistically efficient clinical trial outcome measures derived using these methods. Methods to be developed in this grant will substantially improve the efficiency of clinical trials, reducing the cost and increasing the probability that effective treatments will be identified. The Specific Aims are: Specific Aim 1. To derive and apply methods for optimal calculation of instrument total scores. This is an extension of our earlier work using Item Response Theory to find the optimal scoring of items when calculating an instrument total score. Our earlier work trained the rescoring algorithm on cross-sectional data. The extension will be to train on longitudinal data. We have pilot data demonstrating a 25% reduction in required sample size for select instruments by the proposed method. Specific Aim 2. To derive and apply methods for optimal construction of composite scales. Composite scales combining cognitive and functional measures promise to dramatically improve the efficiency of clinical trials of mild cognitive impairment and is an area of active research. We have derived an optimal formula and can demonstrate superior performance relative to current methods with real data and simulations. Specific Aim 3. To demonstrate de novo outcome measure development by applying the methods developed in Aims 1 and 2 to a different but related progressive disease, frontotemporal dementia (FTD). This exercise is intended to demonstrate the generalizability of our methods to other disease areas. Moreover, software developed in performance of this grant will be posted as the methods are published and will be applicable to instrument development for any chronic disease for which quantitative traits are used as endpoints for clinical trials. This grant is entirely motivated by the observation that clinical trials of chronic progressive disease are prohibitively expensive. In Alzheimer's disease research this has limited our ability to test new treatments and find a cure for the disease. For less common diseases such as FTD and progressive supranuclear palsy the need for more efficient endpoints is even more pressing, as the availability of study subjects for clinical trials further limits our ability to test treatments. Every subject enrolled in a clinical trial is a precious resource. This grant is intended to advance methods to optimally utilize all information obtained from subjects enrolled in clinical trials and increase the probability that effective treatments are identified.