Pellets And Self-microemulsion Preparations Absorption Prediction And Formulation Optimization

There has been a concerted effort towards better prediction of pharmacokinetic properties in optimized formulation efforts. However, reliable estimation of oral bioavailability requires in vivo experimentation, and it is time-consuming and material-consuming. Instead, surrogate parameters for oral bioavailability are estimated from a battery of in silico, in vitro or in situ experiments. Hence, a systematic approach including various in vitro, in situ and in vivo approaches is required in predicting oral bioavailability and other PK properties of new formulation in screening and optimization efforts. This study was divided into two part as follows.PART ONEOBJECTIVE: The aim of the first part study was to develop an in-vitro–in-vivo correlation (IVIVC) for buspirone hydrochloride sustained release pellets formulations.METHODS: The datus of buspirone hydrochloride in vitro dissolution were obtained for each formulation using the ChP 2005(Ⅱ), method(Ⅱ), paddle stirrer at 50, 75 and 100 rpm in 0.1M HCl, pH6.8 phosphate buffer and other release conditions. The datus of buspirone hydrochloride in vivo absorption were obtained from pharmacokinetics of each formulation in beagles, rats, healthy subjects given a single oral dose.RESULTS: A sensitive and specific LC/MS/MS method for rat plasma and a GC-NPD method for beagles and human plasma for direct determination of buspirone hydrochloride have been achieved. The similarity factor f2 (<50) was used to discriminate dissolution method. The in vitro in vivo correlation was generated using pooled mean fraction of dose dissolved (FRD) and pooled mean fraction of dose absorbed (FRA) from two or more formulations. Predicted buspirone hydrochloride concentrations were obtained by derivation of equation for back calculation of Wagner–Nelson Equation. Prediction error for Cmax and AUC established the predictability of the IVIVC. The percent prediction error of the fianal formulation was 2.78% for AUC, and 17.6% for Cmax, respectively. pH6.8 phosphate buffer,50rpm was presented as the best dissolution method. We also demonstrated here that the linear correlation of fraction of oral dose absorbed of buspirone hydrochloride sustained release pellets formulations existed between human and beagle dogs(r=0.85), and also was shown between human and rats(r=0.65) which was lower. PART TWOOBJECTIVE: The aim of the second part study was to screen and investigate the effect of three novel self-microemulsifying drug delivery systems (SMEDDS) formulations (A, B, C) containing carbamazepine with poor solubility in water.METHODS: Changes in barrier properties of Caco-2 cell monolayers, including transepithelial electrical resistance (TEER) and permeability of carbamazepine, were assessed in response to three diffferent formulations. The cytotoxicity of SMEDDS on Caco-2 cells were evaluated by MTT assay. Changes in subcellular localization of the tight junction protein ZO-1, were examined by immunofluorescence. The absorption characteristics of SMEDDS in rat everted gut sac system (different intestinal sections) was studied. The pharmacokinetic behaviors of three different SMEDDS formulations were investigated in rats and beagle dogs.RESULTS: We have established a sensitive and specific LC/MS method in rat plasma/HBSS/Krebs for direct determination of carbamazepine. SMEDDS was shown to increase the permeability of carbamazepine by 2-fold to 4-fold higher than the control group. The results of MTT assay indicated that SMEDDS formulations at lower dilutions (1:100, 1:500) did not induce any toxic effect prior to 1.5h incubation. However, three formulations showed cytotoxicity with dilution of 1:10 as similar with SMEDDS without carbamazepine, but carbamazepine did not show any toxic effect itself. In immunoflourescence studies, we also found the formulation B showed no significant difference compared with the control group by detection of ZO-1 protein, they both exhibited a predominantly continuous fluorescence patterns associated with plasma membrane. However, formulation A and formulation C appeared to cause a redistribution of ZO-1protein, loss of continuous fluorescence and strong fluorescence density of cell-cell adhension point. Treatment with the SMEDDS (50μg/ml) resulted in the TEER of Caco-2 cells significantly reduction up to 40% compared with the control group. When the apical HBSS containing formulations was replaced with fresh cell culture medium after 1.5h treated by formulations, the TEER values of monolayers showed slow recovery to 70~90% after 3~6h. The effect of SMEDDS formulations on the tight junctions of Caco-2 cell monolayers is reversible. We demonstrated that the whole intestine was the good absorption segment for SMEDDS formulations. The absorption dose of formulation B at duodenum, jejunum, ileum and colon was higher than pellet significiantly. The results indicated that the absorption of carbamazepine complied with the passive transport mechanism. The bioavailability was also evaluated by the rat pharmacokinetic parameters, formulation B has the higher bioavailability. The same results were also observed in beagle dogs with high bioavailability up to 5-fold higher than control pellet.CONCLUSION: (1)In-vitro dissolution could make a good prediction of in-vivo absorption of buspirone hydrochloride sustained releas pellets. (2) The pharmacokinetic behaviors in rat could elucidate the characteristic of oral sustained release pellets, which could be used for screening of initial formulations. (3) Caco-2 cell model and rat quick-screening method could be used for prediction of new formulations in bioavailability. (4) The absorption of intestine demonstrates the regular differences in species, the datus obtained in small animals will be used for in-vivo absorption prediction of medicine in human.