Statistical Planning of Translational Research

该项目将是大学医学中心Göttingen医学统计系和慈善机构Universitätsmedizin柏林生物计量学和临床流行病学研究所的合资企业。它致力于改善基础研究转化为临床科学的统计方面，以满足3R动物实验原则：替代，减少，改进和动物福利当局制定的标准。任何试验都应该从仔细的计划开始，尤其是样本量的计算，特别是样本量和功率方面的考虑。实验的计划阶段是关键，因为统计计划中的错误会对结果和从数据中得出的结论产生严重后果。在转化研究（临床前和早期临床）中，错误结论严重影响后续试验，因此，错误激增，这是一个相当不道德的结果。在统计实践中，大多数研究都是基于t检验和wald型统计（包括方差分析）来规划的，并做出一些严格的分布假设。样本量通常很小，如果不满足计划假设，试验要么动力不足，要么规模过大，从而导致错误的结论，浪费资源，甚至可能具有误导性。另一方面，非参数排序方法（如Wilcoxon-Mann-Whitney检验、Brunner-Munzel检验、多重对比检验及其推广）是这些参数方法的优秀替代方法。然而，样本量公式以及详细的功率分析尚未用于此类测试的广泛类别。此外，包括样本量重新估计在内的群体顺序设计和自适应设计提供了一个灵活的研究框架，因此在翻译研究的所有阶段都是可取的。提前停止试验不仅从节省成本的角度来看是值得的，动物福利也将其提升到了一个不同的水平。在本项目中，将开发使用创新重采样算法的近似解决方案，因为样本量通常很小且通常有限。除了开发非参数组序列测试，我们将探索现有方法在小样本设置中的应用，以及它们在（验证性）动物试验总体工作流程中的实施。总而言之，该项目不仅将开发新的统计程序和算法，还将改善工作流程，并最终改善临床前研究的道德标准。

This project will be a joint venture of the Department of Medical Statistics of the University Medical Center Göttingen and the Institute of Biometry and Clinical Epidemiology of the Charité Universitätsmedizin Berlin. It aspires to improve the statistical aspects in the translation of basic research to clinical sciences to meet standards set by the 3R’s principle of animal experiments: Replacement, Reduction, Refinement and by animal welfare authorities. Any trial should start with its careful planning and especially with sample size calculations, in particular with regard to sample size and power considerations. The planning phase of an experiment is key, since errors in the statistical planning can have severe consequences on both the results and conclusions drawn from the data. In translational research (preclinical and early clinical), false conclusions highly affect subsequent trials and thus, mistakes proliferate, a rather unethical outcome. In statistical practice, most studies are planned based on t-tests and Wald-type statistics (including ANOVA) and make some strict distributional assumptions. Sample sizes are typically small and if planning assumptions are not met, the trials are either underpowered or too large and thus result in wrong conclusions and waste resources and might even be misleading. On the other hand, nonparametric ranking methods (such as Wilcoxon-Mann-Whitney test, Brunner-Munzel test, multiple contrast tests and their generalizations) are excellent alternatives to such parametric approaches. However, sample size formulas as well as detailed power analyses are not implemented yet for broad classes of such tests. Furthermore, group sequential designs and adaptive designs including sample size re-estimation provide a flexible research framework and are therefore desirable in all phases of translational research. Early stopping of trials is not only worthwhile from cost saving perspectives, animal welfare rises it to a different level. In this project, approximate solutions using innovative resampling algorithms will be developed, since sample sizes are typically small and often limited. Beyond developing nonparametric group sequential tests, we will explore the application of existing methods in small sample settings and their implementation in the general work flow of (confirmatory) animal trials overall. All in all, this project will not only develop new statistical procedures and algorithms, but will also improve work flows and ultimately the ethical standards of preclinical studies.