139 Development of a drug release methodology for carbamazepine cr tablets based on bioequivalence evaluation I. Homšek 1 , J. Parojčić 2 , M. Mitić 1 , S. Simić 1 , N. Cvetković 1 , Z. Ðurić 2 1 Galenika a.d., R&D Institute, Batajnički drum bb, 11080 Belgrade, Serbia 2 Department of Pharmaceutical Technology, Faculty of Pharmacy, University of Belgrade, Vojvode Stepe 450, 11221 Belgrade, Serbia *Correspondence: homsek@yahoo.com The aim of this study was to investigate the infuence of experimental conditions on carbamazepine (CBZ) release from the two controlled-re- lease (CR) tablet formulations with proven bioequivalence and to propose the universal release method which would be biorelevant. In vitro data were obtained in various release media using the USP apparatus II in order to assess their infuence on CBZ release from CR tablets. An empirical correlation between in vivo plasma concentration data expressed as a fraction drug absorbed and the cumulative amount of in vitro release was established. The most complete release and highest level A correlation was observed by using the half change method (HCM) compared to other release methodologies tested. The results obtained indicate that the release test developed offers a promising in vitro tool for predicting the in vivo performance of CBZ CR tablets, ensuring batch to batch bioequivalence and verifcation of certain postapproval changes without the need for additional in vivo studies. Key words: Carbamazepine – Controlled-release – Bioequivalence – Release – IVIVC. J. DRUG DEL. SCI. TECH., 18 (2) 139-144 2008 Controlled-release pharmaceutical formulations may offer one or more advantages over the conventional (immediate release, IR) dosage forms of the same drug, including a reduced dosing frequency, a decreased incidence and/or intensity of adverse effects, a greater selectivity of pharmacological activity and a reduction in drug plasma fuctuation resulting in a more constant or prolonged therapeutic effect [1]. The drug evaluated in this study was CBZ, a well-established drug, used mainly in the treatment of different forms of epilepsy. It is generally classifed as a drug with low solubility and high permeability throughout the intestinal tract, meeting the criteria of class II substan- ces according to the Biopharmaceutics Classifcation System [2]. It is characterized by a slow and irregular gastro-intestinal absorption due to its low water solubility [3]. Controlled-release formulations of CBZ produce steady-state blood concentrations of the drug similar to immediate-release formulations [4]. It may improve compliance and patient convenience and cause a reduction in the concentration-related side effects, leading to better antiepileptic therapy [5-7]. In vitro dissolution testing is important for providing process control and quality assurance. In certain cases, especially for oral controlled-release formulations, drug release test can also serve as an indicator of how the formulation will perform in vivo [1]. The infuence of media composition on drug release kinetics is well recognized and reported in the literature [8-12]. With an attempt to reconstruct the gastrointestinal environment, “physiologically based dissolution media” matching the pH value, buffer capacity, bile salt and lecithin levels and osmolarity of the luminal content were designed. Four media were established, namely: fasted state simulated gastric fuid (FaSSGF), fed state simulated gastric fuid (FeSSGF), fasted state simulated intestinal fuid (FaSSIF) and fed state simulated intestinal fuid (FeSSIF) [13-16]. Extensive research work has been performed and the in vivo predictability of the proposed media was stressed out for poorly soluble drugs danazol, ketoconazole [13], albendazole [17], atovaquone and troglitazone [14]. The obtained data were in rank order with the in vivo observations [18, 19]. However, besides certain advantages, there are also problems associated with the use of these media since, with the increasing complexity of media, time for setting up and conducting dissolution test increases and analytical procedures and validation become more complicated. Shah et al. [8] studied the release of CBZ IR tablets in various media and 1% SLS was selected as the most suitable. The USP XXII (1990) incorporated the release test for CBZ tablets using the appa- ratus II at 75 rpm and 900 mL of 1% SLS. Since 2001, USP 25 has specifed release tests for CBZ extended release tablets, using 900 and 1,800 mL of water as a medium, for 200 and 400 mg doses, respectively. Although there is a debate on the rational use of surfactants in drug release media and their in vivo relevance, Ikinci et al. [20, 21] used 1% SLS aqueous solution to predict the in vivo behavior of CBZ CR tablets. Katzhendler et al. [22] proposed 700 mL of 1% SLS aqueous solution for release testing of CBZ sustained release hydrophilic matrix tablets. There are reports on the comparative release profles of generic CBZ extended release tablets vs. Tegretol XR in different release media: pH 1.2, pH 4.5, pH 7.5 as well as in water containing 0.2% SLS, at 50 and 75 rpm [23]. The best correlation between the in vitro profles and in vivo data was obtained in 0.2% SLS aqueous solution at a stirring speed of 75 rpm. Giunchedi et al. [24] developed the in vitro release test for CBZ modifed release dosage forms, by performing the change of release media during the test (media pH 1.2 from 0 to 2 h, followed by the phosphate buffer pH 6.8 from 2 to 18 h). The authors showed that the determining factors controlling drug release were the type and amount of enteric polymer constituting the granules and the amount of HPMC mixed with them. Similar results, refecting highly signifcant correlation were obtained with theophyl- line modifed release formulations tested in pH 1.2 for 1 h followed by pH 6.8 phosphate buffer [25]. Another approach to overcome the low solubility problems and maintain sink conditions in vitro was refected by the use of high volumes of the release media. Giunchedi et al. [26] used a cylindrical vessel with 5,000 mL volume instead of 1,000 mL and a stirring rate of 100 rpm for release testing of CBZ from extended-release erodible matrices. Ertan et al. [27] developed a new in vitro/in vivo kinetic correlation method for nitrofurantoin matrix tablet formulations by using the USP paddle and half-change