Abstract:
|
Multiplicity issues are common in vaccine clinical studies, especially in paediatric population. Common causes include co-administration with routine vaccines, multiple doses, multiple endpoints, and interim analyses. In a Phase III paediatric study, a total of 49 statistical hypotheses were identified for the primary objectives requested by authority. We have designed a sequential testing strategy with visualization using a graphical gate-keeping procedure. In the strategy, the 49 hypotheses were grouped into 10 families and distributed over 4 sequential steps, following the clinical and statistical logical relationships agreed with the clinical team. Sample size simulations were performed in SAS. With this tailored strategy the sample size was reduced by around 40%, compared with a simple Bonferroni approach. The strategy was essential to reduce the sample size and to keep the study’s feasibility within the financial budget and increase the study success. The paper includes a step-by-step example to illustrate how the simulation was done, and how the testing strategy will be performed during the analysis.
|