TABLE 2

Summary of the dynamic and the static mechanistic model–based predictions of repaglinide-rifampicin interactions

Dosage Regimen^{^a}	Observed AUCR	PBPK Model Prediction
Dosage Regimen^{^a}	Observed AUCR	CL_int,h/CL′_int,h	Fg′/Fg	AUCR	[I_u,max,in] (µM) ^{^b}	[I_u,gut] (µM)^{^b}	R-Value ^{^c}	CL_int,h/CL′_int,h	Fg′/Fg	AUCR
0 hours	0.52 (Bidstrup et al., 2004)	1.70	0.40	0.68	1.76	3.75	3.1	1.95	0.30	0.59
1 hour	0.69 (Hatorp et al., 2003)	1.83	0.41	0.75	1.76	3.75	3.1	1.95	0.30	0.59
12.5 hours	0.43 (Niemi et al., 2000)	1.15	0.33	0.38	0.27	0	1.3	0.91	0.26	0.24
24 hours	0.20 (Bidstrup et al., 2004)	0.88	0.33	0.29	0.02	0	1.0	0.73	0.26	0.19
PPE				24%						19%
RMSE				0.10						0.11
AFE				1.24						1.26

AFE, average-fold error model; PPE, percentage prediction error model; Fg′, fraction of drug escaping gut-wall metabolism in the presence of rifampicin; Fg, fraction of drug escaping gut-wall metabolism in the absence of rifampicin; RMSE, root mean square error model.
↵^a Time of repaglinide oral dose after the last dose of rifampicin.
↵^b For the static model, CYP3A4 induction was assumed constant for 24 hours after the rifampicin treatment and was calculated using 1.76 µM for [I_u,max,in] or 3.75µM for [I_u,gut].
↵^c R-value for OATP1B1-mediated interaction (Giacomini et al., 2010; US Food and Drug Administration, 2012) was calculated as 1+(I_u,max,in/Ki).