Optimal adaptive group sequential design with flexible timing of sample size determination

Zhang, Lanju; Yang, Bo

doi:10.1016/j.cct.2017.04.005

Cited by 12 publications

(14 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We believed that it is necessary to improve the quality of research and to the registered clinical research programmes in strict accordance with the guidelines for clinical trials 32- 35 . In addition, current clinical trials by different hospitals conducted spontaneously are not effectively organized and coordinated, so more scattered and disorderly.…”

Section: Discussionmentioning

confidence: 99%

Systematic Review of the Registered Clinical Trials of Coronavirus Disease 2019 (COVID-19)

Zhu

Gao

Robert

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: Since the outbreak of coronavirus diseases 2019 (COVID-19), many researchers in China have immediately carried out clinical research scheme of the COVID-19. But, there is still a lack of systematic review of registered clinical trials. Therefore, we made the first systematic review of the clinical trials of COVID-19 in order to provide evidence for the control of the COVID-19. Methods: The database from the Chinese Clinical Registration Center and the ClinicalTrials.gov were searched to collect the registered clinical trials of COVID-19. The retrieval inception date is February 9, 2020. Two evaluators independently selected literature, extracted data and evaluated the risk of bias. This study is based on the recommendations of PRISMA in Cochrane handbook. Results: A total of 75 COVID-19 registered clinical trials (63 interventional studies and 12 observational studies) were obtained. 97.3% of clinical trials were initiated by Chinese organizations. Only 11 trials have begun to recruit patients, and all registered clinical trials have not been completed. Most of the trials are early clinical exploratory trials or in pre-experiment stage (only two trials of Remdesivir in Ⅲ stage), and the sample size of subjects recruited is small. The main intervention methods include traditional Chinese medicine treatment, western medicine treatment and integrated Chinese and Western medicine treatment. The subjects were mainly non severe adult patients (≥ 18 years old). The main outcomes were clinical observation and examination. The duration of most trials was more than 5 months, and the median of the intervention study was 180 d (95% CI: 146.3 -328.9 d); the median of the observation period was 334 d (95% CI: 166.6 -363.4 d). Overall, both the methodology quality of intervention register trials and observational trials are low. Conclusions: Disorderly and intensive clinical trialsof COVID-19 using traditional Chinese medicine and Western medicine are ongoing or will be carried out in China. However, based on the poor quality and small sample size and long completion period, we will not be able to obtain reliable, high-quality clinical evidence about COVID-19 treatment for quite a long time in the future. Improving the quality of study design, prioritizing promising drugs, and using different designs and statistical methods are worth advocating and recommending for the clinical trials of COVID-19 in China.

show abstract

Section: Discussionmentioning

confidence: 99%

Systematic Review of the Registered Clinical Trials of Coronavirus Disease 2019 (COVID-19)

Zhu

Gao

Robert

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The CHW design is an adaptive GS design with flexible timing of sample size determination. 16 It becomes a sample size reestimation design only if the targeted sample size is initially determined at the beginning of a trial. With the CHW design, a clinical trial can start without an initially targeted trial sample size.…”

Section: Discussionmentioning

confidence: 99%

“…This can be achieved if the trial sample size can be determined in an interim analysis using more relevant information from the cashed in trial data and the partially realized value of the test statistic. The reformulated CHW design as an adaptive GS design with flexible timing for sample size determination is for the purpose …”

Section: Adaptive Gs Design With Flexible Timing Of Sample Size Determentioning

confidence: 99%

“…The reformulated CHW design as an adaptive GS design with flexible timing for sample size determination is for the purpose. 16 Suppose that a total of K analyses are planned in the two-arm clinical trial and that hypothesis (1) is to be repeatedly tested in K − 1 interim analyses and in the final analysis. For the purpose, on the scale of [0, 1], the information time is freely chosen upfront as 0 = r 0 < r 1 < · · · < r l < · · · < r K − 1 < r K = 1.…”

Section: Adaptive Gs Design With Flexible Timing Of Sample Size Determentioning

confidence: 99%

See 1 more Smart Citation

On the efficiency of adaptive sample size design

Zhang

2018

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

Adaptive sample size designs, including group sequential designs, have been used as alternatives to fixed sample size designs to achieve more robust statistical power and better trial efficiency. This work investigates the efficiency of adaptive sample size designs as compared to group sequential designs. We show that given a group sequential design, a uniformly more efficient adaptive sample size design based on the same maximum sample size and rejection boundary can be constructed. While maintaining stable statistical power at the required level, the expected sample size of the obtained adaptive sample size design is uniformly smaller than that of the group sequential design with respect to a range of the true treatment difference. The finding provides further insights into the efficiency of adaptive sample size designs and challenges the popular belief of better efficiency associated with group sequential designs. Good adaptive performance plus easy implementation and other desirable operational features make adaptive sample size designs more attractive and applicable to modern clinical trials.

show abstract

“…The VALOR trial in oncology is a recent example of such strategy to obtain additional funding for promising interim outcome (Ravandi et al., ). Cui, Zhang, and Yang () reformulate their SSR method as an adaptive group sequential method with flexible timing for sample size determination so that the method may be viewed not as sample size reestimation but as a delayed upfront sample size determination before the final analysis. For unblinded SSR, since the final sample size depends on the interim data, statistical adjustment is usually needed to account for the potential inflation of type I error rate (Proschan & Hunsberger, ; Wittes & Brittain, ).…”

Section: Introductionmentioning

confidence: 99%

An efficient sample size adaptation strategy with adjustment of randomization ratio

Wang

Zhong

et al. 2019

Biometrical J

Self Cite

View full text Add to dashboard Cite

In clinical trials, sample size reestimation is a useful strategy for mitigating the risk of uncertainty in design assumptions and ensuring sufficient power for the final analysis. In particular, sample size reestimation based on unblinded interim effect size can often lead to sample size increase, and statistical adjustment is usually needed for the final analysis to ensure that type I error rate is appropriately controlled. In current literature, sample size reestimation and corresponding type I error control are discussed in the context of maintaining the original randomization ratio across treatment groups, which we refer to as "proportional increase." In practice, not all studies are designed based on an optimal randomization ratio due to practical reasons. In such cases, when sample size is to be increased, it is more efficient to allocate the additional subjects such that the randomization ratio is brought closer to an optimal ratio.In this research, we propose an adaptive randomization ratio change when sample size increase is warranted. We refer to this strategy as "nonproportional increase," as the number of subjects increased in each treatment group is no longer proportional to the original randomization ratio. The proposed method boosts power not only through the increase of the sample size, but also via efficient allocation of the additional subjects. The control of type I error rate is shown analytically. Simulations are performed to illustrate the theoretical results. K E Y W O R D Sadaptive design, conditional power, nonproportional sample size increase, randomization ratio adjustment, sample size reestimation, type I error control

show abstract

Optimal adaptive group sequential design with flexible timing of sample size determination

Cited by 12 publications

References 24 publications

Systematic Review of the Registered Clinical Trials of Coronavirus Disease 2019 (COVID-19)

Systematic Review of the Registered Clinical Trials of Coronavirus Disease 2019 (COVID-19)

On the efficiency of adaptive sample size design

An efficient sample size adaptation strategy with adjustment of randomization ratio

Contact Info

Product

Resources

About