Optimal group‐sequential designs for simultaneous testing of superiority and non‐inferiority

Öhrn, Fredrik; Jennison, Christopher

doi:10.1002/sim.3790

Cited by 15 publications

(9 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Minimization of , which is known as the Kiefer–Weiss problem (Kiefer and Weiss, 1957), has been considered in the fully sequential case by Weiss (1962), Lai (1973) and Lorden (1976). Optimal GSTs for this and other criteria have been found by Jennison (1987), Eales and Jennison (1992, 1995), Barber and Jennison (2002) and Öhrn and Jennison (2010).…”

Section: Group Sequential Tests For Delayed Responsesmentioning

confidence: 99%

Group Sequential Tests for Delayed Responses (with discussion)

Jennison

2012

Journal of the Royal Statistical Society Series B: Statistical Methodology

Self Cite

102

View full text Add to dashboard Cite

Summary. Group sequential methods are used routinely to monitor clinical trials and to provide early stopping when there is evidence of a treatment effect, a lack of an effect or concerns about patient safety. In many studies, the response of clinical interest is measured some time after the start of treatment and there are subjects at each interim analysis who have been treated but are yet to respond. We formulate a new form of group sequential test which gives a proper treatment of these 'pipeline' subjects; these tests can be applied even when the continued accrual of data after the decision to stop the trial is unexpected. We illustrate our methods through a series of examples. We define error spending versions of these new designs which handle unpredictable group sizes and provide an information monitoring framework that can accommodate nuisance parameters, such as an unknown response variance. By studying optimal versions of our new designs, we show how the benefits of lower expected sample size that are normally achieved by a group sequential test are reduced when there is a delay in response. The loss of efficiency for larger delays can be ameliorated by incorporating data on a correlated short-term end point, fitting a joint model for the two end points but still making inferences on the original, longer-term end point. We derive p-values and confidence intervals on termination of our new tests.

show abstract

Section: Group Sequential Tests For Delayed Responsesmentioning

confidence: 99%

Group Sequential Tests for Delayed Responses (with discussion)

Jennison

2012

Journal of the Royal Statistical Society Series B: Statistical Methodology

Self Cite

102

View full text Add to dashboard Cite

show abstract

“…Adaptive sample size reestimation avoids these pitfalls and can reduce the expected sample size, and in turn the cost of the study, under a range of treatment effects. Protocols and procedures for re-specification of sample size are well described in the literature [4,[17][18][19][20][21] . This type of adaptive design can arguably reduce time and cost, but does not specifically deal with optimizing inclusion/exclusion criteria.…”

Section: Discussionmentioning

confidence: 99%

Choosing inclusion criteria that minimize the time and cost of clinical trials

Babbs¹

2014

WJM

View full text Add to dashboard Cite

AIM:To present statistical tools to model and optimize the cost of a randomized clinical trial as a function of the stringency of patient inclusion criteria. METHODS:We consider a two treatment, dichotomous outcome trial that includes a proportion of patients who are strong responders to the tested intervention. Patients are screened for inclusion using an arbitrary number of test results that are combined into an aggregate suitability score. The screening score is regarded as a diagnostic test for the responsive phenotype, having a specific cutoff value for inclusion and a particular sensitivity and specificity. The cutoff is a measure of stringency of inclusion criteria. Total cost is modeled as a function of the cutoff value, number of patients screened, the number of patients included, the case occurrence rate, response probabilities for control and experimental treatments, and the trial duration required to produce a statistically significant result with a specified power. Regression methods are developed to estimate relevant model parameters from pilot data in an adaptive trial design. RESULTS:The patient numbers and total cost are strongly related to the choice of the cutoff for inclusion. Clear cost minimums exist between 5.6 and 6.1 on a representative 10-point scale of exclusiveness. Potential cost savings for typical trial scenarios range in millions of dollars. As the response rate for controls approaches 50%, the proper choice of inclusion criteria can mean the difference between a successful trial and a failed trial. CONCLUSION:Early formal estimation of optimal inclusion criteria allows planning of clinical trials to avoid high costs, excessive delays, and moral hazards of Type II errors.© 2014 Baishideng Publishing Group Inc. All rights reserved.Key words: Adaptive trial designs; Biomarkers; Clinical trials; Device; Drug therapy; Ethics; Methodology; Optimal allocation; Personalized medicine; Sequential design Core tip: This paper presents statistical tools to model and optimize the cost of a randomized clinical trial as a function of the stringency of patient inclusion criteria. The patient numbers and total cost are strongly related to the choice of the cutoff for inclusion. Clear cost minimums exist for many realistic scenarios. Potential cost savings for typical trial scenarios range in millions of dollars. Early formal estimation of optimal inclusion criteria allows planning of clinical trials to avoid high costs, excessive delays, and moral hazards of type Ⅱ errors.Babbs CF. Choosing inclusion criteria that minimize the time and cost of clinical trials. World

show abstract

“…We then search over values of these costs to find a version of this problem for which the optimal Bayes rule has type I error rate under = 0 and so solves the problem originally stated in frequentist terms. The method of re-casting a frequentist problem as a Bayes decision problem has been used to find optimal group sequential tests; see, for example, [17][18][19][20][21]. In our problem, power depends on a vector of treatment effects, and we handle this by dealing with a one-dimensional subset of vectors at a time.…”

Section: Optimizing Power For a Family Of Configurations Ofmentioning

confidence: 99%

Optimizing the data combination rule for seamless phase II/III clinical trials

Jennison

2014

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

We consider seamless phase II/III clinical trials that compare K treatments with a common control in phase II then test the most promising treatment against control in phase III. The final hypothesis test for the selected treatment can use data from both phases, subject to controlling the familywise type I error rate. We show that the choice of method for conducting the final hypothesis test has a substantial impact on the power to demonstrate that an effective treatment is superior to control. To understand these differences in power, we derive decision rules maximizing power for particular configurations of treatment effects. A rule with such an optimal frequentist property is found as the solution to a multivariate Bayes decision problem. The optimal rules that we derive depend on the assumed configuration of treatment means. However, we are able to identify two decision rules with robust efficiency: a rule using a weighted average of the phase II and phase III data on the selected treatment and control, and a closed testing procedure using an inverse normal combination rule and a Dunnett test for intersection hypotheses. For the first of these rules, we find the optimal division of a given total sample size between phases II and III. We also assess the value of using phase II data in the final analysis and find that for many plausible scenarios, between 50% and 70% of the phase II numbers on the selected treatment and control would need to be added to the phase III sample size in order to achieve the same increase in power.

show abstract

Optimal group‐sequential designs for simultaneous testing of superiority and non‐inferiority

Cited by 15 publications

References 21 publications

Group Sequential Tests for Delayed Responses (with discussion)

Group Sequential Tests for Delayed Responses (with discussion)

Choosing inclusion criteria that minimize the time and cost of clinical trials

Optimizing the data combination rule for seamless phase II/III clinical trials

Contact Info

Product

Resources

About