论文部分内容阅读
目的介绍肿瘤新药Ⅱ期临床试验常用Gehan二阶段设计及Simon二阶段设计的设计原理和样本含量估计问题,提供SAS宏程序快速实现其样本含量估计。方法结合二项分布的确切概率原理说明单阶段设计、Gehan二阶段设计及两种Simon二阶段设计样本含量计算过程。利用SAS宏程序,通过设置不同参数比较其优缺点。结果 Simon极小极大设计与单阶段设计相比,不会增加最大样本含量;与Gehan二阶段设计相比,保证了在实际有效率很低时早期终止试验。结论 Simon二阶段比单阶段设计和Gehan二阶段设计更有优势,且极小极大设计在多数情况下比最优化设计受欢迎。
Objective To introduce the design principles and sample content estimation of Gehan two-phase design and Simon two-phase design of phase II clinical trial of new drug for oncology and provide SAS macro program to quickly realize the estimation of sample content. Methods Based on the exact probability theory of binomial distribution, the single-stage design, Gehan two-stage design and two Simon two-stage design samples were calculated. Using SAS macros, compare their pros and cons by setting different parameters. Results Simon’s Minimalist Design does not increase the maximum sample content compared to a single-stage design; it guarantees early termination of the test when the actual effective rate is low compared to the Gehan two-stage design. Conclusion The second phase of Simon is more advantageous than the single-phase design and the second-phase design of Gehan, and the minimax design is more popular than the optimal design in most cases.