Choosing Interim Sample Sizes in Group Sequential Designs

This manuscript investigates sample sizes for interim analyses in group sequential designs. Traditional group sequential designs (GSD) rely on “information fraction” arguments to define the interim sample sizes. Then, interim maximum likelihood estimators (MLEs) are used to decide whether to stop early or continue the data collection until the next interim analysis. The possibility of early stopping changes the distribution of interim and final MLEs: possible interim decisions on trial stopping excludes some sample space elements. At each interim analysis the distribution of an interim MLE is a mixture of truncated and untruncated distributions. The distributional form of an MLE becomes more and more complicated with each additional interim analysis. Test statistics that are asymptotically normal without a possibly of early stopping, become mixtures of truncated normal distributions under local alternatives. Stage-specific information ratios are equivalent to sample size ratios for independent and identically distributed data. This equivalence is used to justify interim sample sizes in GSDs. Because stage-specific information ratios derived from normally distributed data differ from those derived from non-normally distributed data, the former equivalence is invalid when there is a possibility of early stopping. Tarima and Flournoy [3] have proposed a new GSD where interim sample sizes are determined by a pre-defined sequence of ordered alternative hypotheses, and the calculation of information fractions is not needed. This innovation allows researchers to prescribe interim analyses based on desired power properties. This work compares interim power properties of a classical one-sided three stage Pocock design with a one-sided three stage design driven by three ordered alternatives.

Group Sequential Designs: A Tutorial

This tutorial illustrates how to design, analyze, and report group sequential designs. In these designs, groups of observations are collected and repeatedly analyzed, while controlling error rates. Compared to a fixed sample size design, where data is analyzed only once, group sequential designs offer the possibility to stop the study at interim looks at the data either for efficacy or futility. Hence, they provide greater flexibility and are more efficient in the sense that due to early stopping the expected sample size is smaller as compared to the sample size in the design with no interim look. In this tutorial we illustrate how to use the R package 'rpact' and the associated Shiny app to design studies that control the Type I error rate when repeatedly analyzing data, even when neither the number of looks at the data, nor the exact timing of looks at the data, is specified. Specifically for *t*-tests, we illustrate how to perform an a-priori power analysis for group sequential designs, and explain how to stop the data collection for futility by rejecting the presence of an effect of interest based on a beta-spending function. Finally, we discuss how to report adjusted effect size estimates and confidence intervals. The recent availability of accessible software such as 'rpact' makes it possible for psychologists to benefit from the efficiency gains provided by group sequential designs.

Correction to: Optimality criteria for futility stopping boundaries for group sequential designs with a continuous endpoint

An amendment to this paper has been published and can be accessed via the original article.

Evaluation by simulation of clinical trial designs for evaluation of treatment during a viral haemorrhagic fever outbreak

Background Viral haemorrhagic fevers are characterized by irregular outbreaks with high mortality rate. Difficulties arise when implementing therapeutic trials in this context. The outbreak duration is hard to predict and can be short compared to delays of trial launch and number of subject needed (NSN) recruitment. Our objectives were to compare, using clinical trial simulation, different trial designs for experimental treatment evaluation in various outbreak scenarios. Methods Four type of designs were compared: fixed or group-sequential, each being single- or two-arm. The primary outcome was 14-day survival rate. For single-arm designs, results were compared to a pre-trial historical survival rate pH. Treatments efficacy was evaluated by one-sided tests of proportion (fixed designs) and Whitehead triangular tests (group-sequential designs) with type-I-error = 0.025. Both survival rates in the control arm pC and survival rate differences Δ (including 0) varied. Three specific cases were considered: “standard” (fixed pC, reaching NSN for fixed designs and maximum sample size NMax for group-sequential designs); “changing with time” (increased pC\(\text{ }\)over time); “stopping of recruitment” (epidemic ends). We calculated the proportion of simulated trials showing treatment efficacy, with K = 93,639 simulated trials to get a type-I-error PI95% of [0.024;0.026]. Results Under H0 (Δ = 0), for the “standard” case, the type-I-error was maintained regardless of trial designs. For “changing with time” case, when pC>pH, type-I-error was inflated, and when pC<pH it decreased. Wrong conclusions were more often observed for single-arm designs due to an increase of Δ over time. Under H1 (Δ=+0.2), for the “standard” case, the power was similar between single- and two-arm designs when pC=pH. For “stopping of recruitment” case, single-arm performed better than two-arm designs, and fixed designs reported higher power than group-sequential designs. A web R-Shiny application was developed. Conclusions At an outbreak beginning, group-sequential two-arm trials should be preferred, as the infected cases number increases allowing to conduct a strong randomized control trial. Group-sequential designs allow early termination of trials in cases of harmful experimental treatment. After the epidemic peak, fixed single-arm design should be preferred, as the cases number decreases but this assumes a high level of confidence on the pre-trial historical survival rate.

Optimality criteria for futility stopping boundaries for group sequential designs with a continuous endpoint

Background In clinical trials with fixed study designs, statistical inference is only made when the trial is completed. In contrast, group sequential designs allow an early stopping of the trial at interim, either for efficacy when the treatment effect is significant or for futility when the treatment effect seems too small to justify a continuation of the trial. Efficacy stopping boundaries based on alpha spending functions have been widely discussed in the statistical literature, and there is also solid work on the choice of adequate futility stopping boundaries. Still, futility boundaries are often chosen with little or completely without theoretical justification, in particular in investigator initiated trails. Some authors contributed to fill this gap. In here, we rely on an idea of Schüler et al. (2017) who discuss optimality criteria for futility boundaries for the special case of trials with (multiple) time-to-event endpoints. Their concept can be adopted to define “optimal” futility boundaries (with respect to given performance indicators) for continuous endpoints. Methods We extend Schülers’ definition for “optimal” futility boundaries to the most common study situation of a single continuous primary endpoint compared between two groups. First, we introduce the analytic algorithm to derive these futility boundaries. Second, the new concept is applied to a real clinical trial example. Finally, the performance of a study design with an “optimal” futility boundary is compared to designs with arbitrarily chosen futility boundaries. Results The presented concept of deriving futility boundaries allows to control the probability of wrongly stopping for futility, that means stopping for futility even if the treatment effect is promizing. At the same time, the loss in power is also controlled by this approach. Moreover, “optimal” futility boundaries improve the probability of correctly stopping for futility under the null hypothesis of no difference between two groups. Conclusions The choice of futility boundaries should be thoroughly investigated at the planning stage. The sometimes met, arbitrary choice of futility boundaries can lead to a substantial negative impact on performance. Applying futility boundaries with predefined optimization criteria increases efficiency of group sequential designs. Other optimization criteria than proposed in here might be incorporated.

Group sequential designs applied in psychological research

Psychological research is confronted with ever-increasing demands to save resources such as time and money while assuring high ethical standards. In medical and pharmaceutical research, group sequential designs have fundamentally changed traditional statistical testing approaches featuring only one analysis at the end of a single-stage study. They enable early stopping at an interim stage, after a group of observations, for efficacy or futility in case of an overwhelmingly large or small effect, respectively. Otherwise, the trial is continued to the next stage. On average over many studies time and money are saved and more ethical trials are facilitated by diminishing the risk of patients' exposure to inferior treatments. We provide an easy-to-use tutorial for psychological research replete with easily understandable figures highlighting the core idea of different group sequential designs, a workflow chart, an empirical real-world data set, and the annotated R code. Finally, we demonstrate the application of early stopping for efficacy.