Influence of two-period cross-over design on the bioequivalence study of gefitinib tablets in beagle dogs.

Generally, two-period cross-over design is used in bioequivalence (BE) study. High intra-subject variability of gefitinib was reported in a clinical BE study, and significant changes in gefitinib exposures were observed among different periods in our previous BE study in dogs. Therefore, commercial gefitinib tablets from the same batch were used in the present study and assigned to two groups: the testing drug (GF1) group and the reference drug (GF2) group. A single-oral-dose, two-period cross-over study with a 7-day washout (approximately 24 half-lives) under fasting condition was conducted in 12 dogs to explore the factors. The results showed that the mean values of AUC(0-t), AUC(0-∞), Cmax and Tmax of these two GF1 and GF2 groups were similar. However, the GF1 and GF2 did not meet the acceptance criteria of bioequivalence with 90% confidence intervals, since the values obtained were 76.22%-117.43% for AUC(0-t) and 87.55%-131.59% for Cmax. ANOVA revealed a significant difference between the two periods (P < 0.05). Interestingly, the mean AUC(0-t) of gefitinib in the period 2 was 2.3-fold greater than that in the period 1, while Cmax in the period 2 was 1.7-fold higher than that in the period 1. However, the volume of distribution was significantly decreased, becoming 0.4-fold lower in the period 2. No statistically significant difference in the half-life and Tmax was observed between the two periods (P < 0.05). The pharmacokinetic alteration might come from the different physiological absorption and/or metabolism between periods. Since a 7-day washout interval was applied, DDI risk from P450s and/or P-gp would not play a significant role in the non-bioequivalence. As regard the variability, the intra-subject variation crossing the periods was much larger than the inter-subject variation within each period. The absorption and/or metabolism function of the gut bacteria might play an important role in the increasing exposure of gefitinib in the second period, especially with the comparison with the analysis from the high-fat-diet treatments in humans. Therefore, further studies might be needed to evaluate whether the assessment of bioequivalence could be facilitated by a much longer washout interval allowing the recovery of gut bacteria.