Category: Formulation and Quality
Purpose: The standard extract of Ginkgo biloba extract, EGb 761, has been widely used in the treatment of neuropsychiatric disorders, such as Alzheimer’s disease. Active compounds in EGb761 consists of 22.0-27.0% Ginkgo flavonoid glycosides (glycosides of quercetin, kaempferol and isorhamnetin) and 5.0-7.0% terpene lactones (2.8-3.4% ginkgolides A, B, C, and 2.6-3.2% bilobalide). However, in addition to standardization of these marker compound content in EGb761, there is a lack of investigation on in vitro release profiles of these markers from the tablet and their correlation to the in vivo pharmacokinetics. The present study was proposed aiming to develop a clinically relevant dissolution method for prediction of the plasma pharmacokinetics of the seven active marker compounds in EGb 761 based on in vitro-in vivo correlation (IVIVC).
Methods: The dissolution tests for the seven active marker compounds in commercially available EGb 761 tablets (Dr. Willmar Schwabe, Batch No. 2371117) were carried out using a rotating paddle apparatus with paddle stirred at 75 rpm. Different dissolution medium including 0.1 M HCl, acetate buffer, H2O, fasted state simulated gastric fluid (FaSSGF), fasted state simulated intestinal fluid version 2 (FaSSIF_2) and fed state simulated intestinal fluid version 2 (FeSSIF_2) were included in the current investigations. Each EGb 761 tablet was added to the 500 ml of above-mentioned medium followed by withdrawing 3 mL sample at 5, 10, 15, 30, 45, 60, 90, 120 min and replacing with the same volume of blank solution medium. In addition to the classical dissolution method, sequential dissolution tests of EGb 761 tablets in 240 min were carried out in 0.1 M HCl /FaSSGF for 120 min followed by adjusting the pH to 7.0 by sodium hydroxide solution at 120 min. A 3 mL sample was withdrawn at 5, 10, 15, 30, 45, 60, 90, 120, 150, 180, 240 min followed by replacing with the same volume of blank solution medium. All the collected samples were immediately stored at −80 ℃ after filtration and analyzed after dilution and acid hydrolysis using the following LC/MS/MS method. To quantitatively analyse the seven active marker components, a rapid and selective LC/MS/MS method was developed. The liquid chromatographic separation was carried out on a Waters Acquity UPLC BEH C18 column (2.1×50 mm, 1.7 μm) analytical column. The gradient mobile phases consisting of (A) water-0.1% formic acid and (B) acetonitrile were eluted at a flow rate of 0.2 mL/min. The detection was performed using the positive ion electrospray ionization (ESI) in multiple reaction monitoring (MRM) mode with precursor-to-product ion transitions at m/z 303.1 → 153.0 for quercetin, m/z 287.1 → 153.0 for kaempferol, m/z 317.1 → 302.1 for isorhamnetin, m/z 327.1 → 309.1 for bilobalide, m/z 409.2 → 345.2 for ginkgolide A , m/z 425.2 → 361.1for ginkgolide B, m/z 441.1 → 325.1 for ginkgolide C and m/z 255.1 → 105.0 for internal standard, ketoprofen. Based on in vitro dissolution profiles of the three active marker compound (ginkgolides A, B, and bilobalide) from EGb 761 tablets and their corresponding in vivo plasma concentration profiles after its oral administration reported in literature [1, 2], the IVIVC equations were established for each marker compound using IVIVC toolkit in Phoenix version 64. Internal predication with the established correlation equations was carried out by predicting AUC last and Cmax, which were compared with the observed values.
Results: Except for isorhamnetin, other six active markers all demonstrated pH-dependent dissolution profiles (Figure 1). As indicated in Table 1, The IVVC model coefficients in H2O and acetate buffer were < 0.99, indicating unsatisfactory in vivo correlation for the three studied marker compounds in these two media. Both coefficients ( > 0.99) and prediction error (< 15%) for AUC last (2.087~4.183%) and C max (-7.468~9.519%) of ginkgolide A and B suggested the acceptability of our established IVIVC models in 0.1 M HCl, FaSSGF, FaSSIF, and FeSSIF. However, higher predicted AUC last for bilobalide with prediction error of 22.10-23.48% in all tested mediums for 120 min classical dissolution were found, which suggested the potential in vitro overestimation of drug release in such dissolution methods. Therefore, classical dissolution profiles in all studied medium may not correlate well with the in vivo performance of EGb761 tablets. As shown in Figure 2, our newly developed sequential dissolution method in 0.1 M HCl subsequently demonstrated a significantly decreased dissolution percentage of bilobalide when changing pH from 1.3 to 7, which was due to its degradation via ring opening in a buffer solution of pH 7.0 at 37 ℃. Therefore, such sequential dissolution method may better represent the true changes of EGb761 tablets in GI tract for better correlation of its in vivo performance.
Conclusion: Our current study recommended a more clinically relevant dissolution method of EGb761 (apparatus 2, 75 rpm, 0-120 min: 0.1 M HCl of pH 1.3; 120-240min: 0.1 M HCl of pH 7.0) for better correlation with its in vivo pharmacokinetics.
Tsun Lam Shek– The Chinese University of Hong Kong, Hong Kong, Hong Kong
Yufeng Zhang– Student, Chinese University of Hong Kong, Hong Kong, Hong Kong
Tianjing Ren– School of Pharmacy, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong, Hong Kong
Sau Wan Cheng– Hong Kong, Hong Kong
Kun Li– Shanghai, Shanghai, China (People's Republic)
Jianxin Wang– Shanghai, Shanghai, China (People's Republic)
Zhong Zuo– Director and Professor, School of Pharmacy, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong, Hong Kong