Regulatory requirements for average bioequivalence have been internationally harmonized (1,2), which is by no means the case for the more recent concept of individual bioequivalence (3#x2013;9). The main reason for introducing more complex replicate designs and bioequivalence criteria are the highly variable drugs, for which the setting of suitable bioequivalence ranges poses a major problem and scaling of the bioequivalence criteria by the intrasubject variability has been suggested (10#x2013;12). The shortcoming of the present two-treatment, two-period (2 #xD7; 2) crossover design to detect subject-by-formulation interaction provides a second argument in favor of the more complex replicate designs (13). A unified approach of proposed statistical procedures for the replicate design has been given by Schall (6,8). However, the availability of these methods and understanding of them seems to be limited to a small working group, so a broader international awareness of the problems and potentional solutions is desirable.
展开▼
