Re-estimating Sample Size While Maintaining Statistical Power

Sample size re-estimation within a confirmatory trial (Phase III) provides a mechanism for the appropriate use of the information obtained during a confirmatory study to inform and adjust the necessary sample size going forward. This process increases the confidence that an appropriate sample size has been chosen to answer the primary study questions.

The standard approach used to power a confirmatory study is first to estimate the underlying treatment effect of the primary endpoint based on available prior information. This effect denotes the true underlying difference between the treatment and control arms of the study with respect to the primary endpoint. Even though the true value of the effect is unknown, the trial investigators will usually have in mind a specific value which represents the Smallest Clinically Important Effect (SCIV) value for this trial.

Next, the trial designers will decide the sample size that can detect effect values based on prior information that exceed the SCIV with good power. The SCIV can be often pre-specified from purely clinical arguments, whereas the actual effect size is unknown. Thus it is possible in principle to design a study with a fixed sample size that will have adequate power to detect the SCIV in the absence of adequate prior information about the actual effect size of the test agent. This is what statisticians envisaged when they created the fixed-sample methodology.

However, this methodology has several drawbacks. If the actual effect size is substantially larger than the SCIV, a smaller size would have sufficed to attain adequate power. Sponsors will not often risk significant resources on trial sizes based on SCIV assumptions that would lead to larger trials than the current “best guess” about the actual effect size. Instead, a smaller trial corresponding to that best guess may be wrong; if that assumption is too optimistic, and the truth is an effect size closer to the SCIV, the trial will be underpowered and thus have a big chance of failure.

One approach to solving the problem of uncertainty about treatment effect value is to design and execute further exploratory trials (typically Phase II studies). These small Phase II studies are normally carried out to get a more precise estimate (or best guess) of the actual size effect so that the confirmatory study might be adequately powered. Each exploratory trial, although somewhat smaller than confirmatory trials, still requires significant resources to perform appropriately.

Also, the inevitable start-up time and wind-down activities between trials have to be included when deciding true program efficiency and development timelines. This might therefore not be the most efficient way to proceed from the view point of the entire clinical-trial program.

A more flexible approach to the fixed sample size methodology is needed.

Do you have any ideas? We invite you to comment below.