Identification of the in vivo relevant dissolution media for the three active components in EGb 761 tablet for better correlation with their pharmacokinetics in healthy subjects.

Abstract Although EGb 761, the standardized dry extract of Ginkgo biloba leaves, exhibited numerous pharmacological activities and widely used in Asia, European and North America, the quality control of its dosage forms such as tablet mainly relies on monitoring the contents of the active marker components, namely quercetin, kaempferol, isorhamnetin, bilobalide, ginkgolide A, ginkgolide B and ginkgolide C. So far, the in vitro dissolution profiles of EGb761 tablet were barely used to monitor its quality and how these dissolution profiles correlate with their in vivo pharmacokinetics was not known. Thus, the present study was proposed aiming to 1) develop the in vitro-in vivo correlations (IVIVCs) for the marker components in EGb 761 tablet; 2) identify the in vivo relevant dissolution media for the marker components in EGb 761 tablet based on the established IVIVCs. The content analyses of the marker components in EGb 761 tablet was first carried out. Then, the dissolution profiles were further obtained using paddle method of United States Pharmacopeia for bilobalide, ginkgolides A, and ginkgolide B, that have previously reported human plasma pharmacokinetics after EGb 761 tablet oral administrations. About seven different media including 0.1 M hydrochloric acid (HCl), acetate buffer, H2O, fasted state simulated gastric fluid (FaSSGF), fasted state simulated intestinal fluid version 2 (FaSSIF-V2), fed state simulated intestinal fluid version 2 (FeSSIF-V2), and sequential medium (0.1 M HCl for 2 h with pH adjusted to 7 for another 2 h) were tested in the current investigation. The obtained in vitro dissolution profiles of bilobalide, ginkgolides A and ginkgolide B from EGb 761 tablet were first fitted with four dissolution models, namely Weibull, Double Weibull, Hill and Makoid-Banakar, to obtain the best-fit model for each component in each medium. The human plasma concentration versus time profiles of the above three components were then inputted into the Phoenix WinNonlin IVIVC Toolkit to obtain their in vivo absorption profiles using numerical deconvolution. The best-fit dissolution profiles of each marker component in the seven studied media were further used to correlate with its obtained in vivo absorption profile by the linear correlation models to establish the corresponding IVIVCs in each studied medium. Finally, the best in vivo correlated medium for each investigated marker component was selected based on their adjusted correlation coefficients, Akaike Information Criterion (AIC) and Schwarz's Bayesian Criterion (SBC) values. As a result, the dissolution profiles of bilobalide, ginkgolide A, ginkgolide B from EGb 761 tablet in 0.1 M HCl, FaSSGF, FaSSIF-V2 demonstrated the best correlation with their in vivo absorption profiles, respectively. Our current studies for the first time applied the concept of IVIVC to EGb 761 tablet and successfully identified the in vivo relevant dissolution media for its three active marker components to improve its quality control.