探寻两种制剂混合而成的药物的最大耐受剂量（MTD）

发布时间：2019-02-13 12:21

【摘要】：本文研究的是在第一阶段临床试验中,寻找两种制剂混合而成药物的最大耐受剂量的方法。对于一种制剂的最大耐受剂量,目前已有比较成熟的方法,比如,CRM方法和Up and Down。对于两种及其多种制剂的(MTD)研究,还在探索。美国Peter F.Thall等人提出一种复杂模型和算法,研究这个问题。本文本着简化模型和算法的想法,对此进行改进探索。我们提出一个自适应的两阶段贝叶斯设计,用于混合使用两种细胞霉素制剂于第一阶段临床试验以找出一个或多个可接受的(AD)剂量对。此方法要求每一种制剂在混合使用前已分别研究过,并且此制剂在使用时稳定。我们设定一个关于两个制剂量的函数的中毒概率参数模型。每一种制剂的中毒概率参数的信息先验或者由设计试验的医生引出,或者由以前的历史数据引出,对于两种制剂之间的相互作用的参数先设定一个模糊先验。先给出两种制剂各自的参数先验的引出。我们这种设计应用于GEM和CTX的临床试验,并给出模拟研究。
[Abstract]:In the first phase of the clinical trial, we studied the method of finding the maximum tolerated dose of the two preparations. For the maximum tolerated dose of a preparation, there are currently more mature methods, such as CRM and Up and Down. (MTD) studies on two or more preparations are still being explored. Peter F.Thall et al. Proposed a complex model and algorithm to study this problem. Based on the idea of simplified model and algorithm, this paper explores how to improve the model. We propose an adaptive two-stage Bayesian design for the use of a mixture of two kinds of cytomycin in the first phase clinical trial to find one or more acceptable (AD) dosages. This method requires that each preparation has been studied separately before mixed use and that the preparation is stable at the time of use. We set up a poisoning probability parameter model for the function of two formulations. A priori of the poisoning probability parameters of each preparation is obtained either by the doctor who designed the experiment or by previous historical data to set a fuzzy priori for the parameters of the interaction between the two preparations. A priori introduction of the parameters of the two preparations was given. Our design was applied to clinical trials of GEM and CTX, and a simulation study was given.
【学位授予单位】：东北师范大学
【学位级别】：硕士
【学位授予年份】：2006
【分类号】：R311

【共引文献】