Two stage sequential testing
At the end of the second stage, the results from both groups combined are used in the final analysis. If bioequivalence is not proven at the end of the first stage, the second stage is conducted employing an appropriate number of additional subjects as determined based on the variance estimates and point estimate calculated from the stage 1 data. If bioequivalence is proven at this point the study can be terminated. The analysis undertaken at the end of the first stage is treated as an interim analysis. Several regulatory agencies such as Japan, EU, Canada accept these studies, and so does WHO.Īs per WHO Annex 6 – the number of subjects employed in the first stage is generally based on the most likely intra-subject variance estimate with some added subjects to compensate for drop-outs.
#TWO STAGE SEQUENTIAL TESTING TRIAL#
Such studies come under the umbrella of adaptive trial designs which are defined by the USFDA as a BE study design that allows for prospectively planned modifications to one or more aspects of the design based on accumulating data from subjects in the trial.”Īdaptive designs are well known and utilized for conduct of patient based efficacy and safety studies often, though their use in BE studies is limited.Ī two stage sequential design is a kind of adaptive design based on use of comparative data (interim) valuable tool that can be employed for BE studies.
In such situations a two- stage sequential study design can be employed such that an accurate estimate of the variability can be determined in the first stage of the study. If the basis of assumptions is unreliable or unknown the assumed variance can be either too small or too large, leading, respectively, to studies that are underpowered or overly large. We have to remember that the sample size is based on a set of assumptions we make. The planning of bioequivalence (BE) studies, requires a prior specification of an effect size for the determination of power and an assumption about the variance and therefore the calculation of the sample size. For a two-stage group sequential robust procedure with a single interim analysis, two critical values for the maximum tests are provided based on a given alpha spending function to control the desired overall type I error.Adaptive Designs are defined “as a clinical trial design that allows for prospectively planned modifications to one or more aspects of the design based on accumulating data from subjects in the trial.” We propose and compare several approaches of controlling type I error rates when group sequential analysis is conducted with the maximum test for family-based candidate-gene association studies. When the group sequential method is applied, type I error should be controlled. The group sequential approach allows interim analyses and has been applied to many test statistics, but not to the maximum statistic.
Recently, cost-effective group sequential approaches have been introduced to genetic studies. This robust test is preferred over a single optimal test.
When a family of genetic models is scientifically plausible, the maximum of several tests, each optimal for a specific genetic model, is robust against the model mis-specification. Using a single test statistic optimal for one genetic model may lose substantial power when the model is mis-specified. In practice, however, genetic models for many complex diseases are usually unknown. In family-based association studies, an optimal test statistic with asymptotic normal distribution is available when the underlying genetic model is known (e.g., recessive, additive, multiplicative, or dominant).