The independent demonstration of treatment effects in subgroups of special interest such as the pediatric population may not be feasible because of limited resources and patient populations. However, if there is prior knowledge available that suggest that efficacy and/or safety may be at least partly extrapolated from other populations that have already been investigated, there may be less need for independent evidence in the subgroup. To safeguard for the possibility that the extrapolation paradigm does not hold, in general, subgroup validation studies will be required. The problems of statistical inference in such settings will be discussed from a regulatory and industry perspective.

Learning Objectives:
Discuss the regulatory and industry perspectives on the problems of statistical inference in subgroups of special interest.