Marival Bermejo(1), Jessica Meulman(2), Marcelo Gomes Davanço (2,3), Patricia de Oliveira Carvalho(3), Isabel Gonzalez-Alvarez(1),Daniel Rossi de Campos (2,3.)
(1)Department of Engineering, Pharmacy and Pharmaceutical Technology Area, Facultad de Farmacia, University Miguel Hernandez de Elche, San Juan de Alicante, Alicante, Spain. (2) Sanofi Medley Farmacêutica Ltda, Industrial Development, Campinas, São Paulo, Brazil. (3) Universidade São Francisco, Postgraduate Program in Health Sciences, Bragança Paulista, São Paulo, Brazil.
Objectives: The aim of the present study was to bring additional evidence regarding the universal biopredictive dissolution medium containing 1% SLS to predict the in vivo behavior of carbamazepine (CBZ) products [1,2]
Methods: Twenty healthy volunteers have taken one IR dose of either Test and Reference formulations in a bioequivalence study (BE). Dissolution profiles were carried-out using the aforementioned medium. The dissolution media and conditions were the ones evaluated by Gonzalez-García I. et al (2). Dissolution studies were performed in USP-2 apparatus (Vankel® VK 7010) at 37 ºC and 50 r.p.m. in media containing 1% sodium lauryl sulphate (SLS). 5 ml samples were taken at 5, 10, 15, 30, 60, 120 min and sample media was replaced with 5 mL fresh warmed media. Samples were filtered and analyzed by HPLC. In vivo average plasma profiles from a previously developed BE study were used to obtain by Wagner-Nelson method the bioavailable fractions versus time profiles in Excel worksheet. A Level A In vitro–in vivo correlations (IVIVC) have been established using both a one-step and two-steps approaches as well as exploring the time-scaling approach to account for the differences in dissolution rate in vitro versus in vivo.
Two step IVIVC was developed with a previous Levy plot construction as in vivo and in vitro dissolution scales were different. A two parameter weibull function was fitted to the dissolution profiles, thus fraction dissolved Fdiss=100*{1-Exp[-(t^β)/α]}. One step approach was developed in Phoenix Winnonlin® 8.1 (Certara) using an ad-hoc ASCII library with 4 differential equations and also in Berkeley Madonna. In vivo dissolution was assumed to follow also a weibull kinetics and a 1 compartment disposition was used for the In vivo data. In vitro-in vivo data were simultaneously fitted considering a direct link between in vitro and in vivo weibull parameters through a slope factor to account for the time-scale differences.
Results: Levy plot indicated that in vitro times are approximately shorter than in vivo dissolution times. After time scaling correction, a linear IVIVC was established between fractions dissolved and bioavailable fractions. With the two-step IVIVC, the prediction errors for Cmax were lower than 5% and for AUC_inf lower than 10%. The one step IVIVC based on differential equations with a scale factor for in vivo weibull dissolution parameters showed a scale factor close to 1 for parameter β and a scale factor around 6 for parameter α times, similar to the time-scale factor obtained by the two-step procedure. In vitro and in vivo experimental values presented a correlation coefficient with predicted values higher than 0.99 (Figure 1). Prediction errors for Cmax and AUC_inf were lower in the one-step approach (less than 4% for Cmax and AUC_inf).
Conclusions: As suggested by Gonzalez-García I. et al a media containing 1% of SLS demonstrated its suitability as the universal biopredictive dissolution medium for immediate release carbamazepine formulations. A scaling adjustment by Levy plot or by scaling dissolution profile parameters in a one-step approach might be necessary due to the differences in dissolution hydrodynamics in vitro versus in vivo.
Therefore, this biopredictive medium could be a highly relevant tool in Latin-America countries to ensure and check the quality of their CBZ marketed products for which BE studies were not requested by their regulatory health authorities.
References:
[1] Lake O.A., Olling M., Barends D.M. In vitro/in vivo correlations of dissolution data of carbamazepine immediate release tablets with pharmacokinetic data obtained in healthy volunteers. European Journal of Pharmaceutics and Biopharmaceutics. 1999;48:13–19
[2] González-García, V. Mangas-Sanjuan, M. Merino-Sanjuán, C. Álvarez-Álvarez, J. Díaz-Garzón Marco, M. A. Rodríguez-Bonnín, T. Langguth, J. J. Torrado-Durán, P. Langguth, A. García-Arieta; M. Bermejo. IVIVC approach based on carbamazepine bioequivalence studies combination. Pharmazie 72: 449–455 (2017)
Reference: PAGE () Abstr 9250 [www.page-meeting.org/?abstract=9250]
Poster: Drug/Disease Modelling - Absorption & PBPK