Fayers P M, Cook P A, Machin D, Donaldson N, Whitehead J, Ritchie A, Oliver R T, Yuen P
MRC Cancer Trials Office, Cambridge, U.K.
Stat Med. 1994 Nov 15;13(21):2249-60. doi: 10.1002/sim.4780132106.
This paper describes the steps taken by the British Medical Research Council (MRC) in developing the MRC RE01 trial, a randomized clinical trial for patients with metastatic renal cancer; we discuss the reasons for adopting a triangular sequential design and the impact that this has upon the monitoring of the trial. It had been suggested to the MRC that a trial of biological agents for metastatic renal carcinoma should be initiated. The Cancer Therapy Committee (CTC) of the MRC, through its associated site specific working parties, is responsible for designing and co-ordinating randomized trials of alternative treatments in cancer in solid tumours. Since no MRC working party for renal carcinoma existed at that time, development began by the formation of an ad hoc group set up under the auspices of the CTC. They assessed, by means of a postal questionnaire, U.K. interest in the trials of, and modalities utilized for, treatment of renal cancer. The responses focused attention on the important questions to ask and indicated the level of potential collaboration. These responses and related clinical and statistical issues suggested a protocol to compare medroxy-progesterone acetate (MPA) against alpha-interferon (alpha-IFN). In view of the special problems of comparing an expensive and potentially toxic therapy with an inexpensive and non-toxic standard, a sequential design was used rather than a fixed sample size design. Statistical issues raised and solutions provided are described. The method of establishing the trial data monitoring committee and a brief review of mortality from renal carcinoma in England and Wales are also included. The trial opened to patient recruitment on 1 January 1992. The formal statements regarding statistical issues that appear in the formal trial protocol (RE01) are set out in the Appendix.
本文描述了英国医学研究理事会(MRC)开展MRC RE01试验所采取的步骤,这是一项针对转移性肾癌患者的随机临床试验;我们讨论了采用三角序贯设计的原因及其对试验监测的影响。有人向MRC建议启动一项针对转移性肾癌的生物制剂试验。MRC的癌症治疗委员会(CTC)通过其相关的特定部位工作组,负责设计和协调实体肿瘤癌症替代治疗的随机试验。由于当时不存在MRC肾癌工作组,因此在CTC的主持下成立了一个特设小组来开展相关工作。他们通过邮寄问卷的方式评估了英国对肾癌治疗试验及所采用治疗方式的兴趣。这些回复聚焦于需要询问的重要问题,并表明了潜在合作的程度。这些回复以及相关的临床和统计问题促使提出了一项比较醋酸甲羟孕酮(MPA)与α干扰素(α-IFN)的方案。鉴于将一种昂贵且可能有毒的疗法与一种廉价且无毒的标准疗法进行比较存在特殊问题,因此采用了序贯设计而非固定样本量设计。文中描述了提出的统计问题及提供的解决方案。还包括了设立试验数据监测委员会的方法以及对英格兰和威尔士肾癌死亡率的简要回顾。该试验于1992年1月1日开始招募患者。正式试验方案(RE01)中关于统计问题的正式声明列于附录中。