Jelena Parojčić

The application of generalized regression neural network in the modeling and optimization of aspirin extended release tablets with Eudragit® RS PO as matrix substance

The objective of this work is to use a generalized regression neural network (GRNN) in the design of extended-release aspirin tablets. As model formulations, 10 kinds of aspirin matrix tablets were prepared. Eudragit RS PO was used as matrix substance. The amount of Eudragit RS PO and compression pressure were selected as causal factors. In-vitro dissolution-time profiles at four different sampling times, as well as coefficients n (release order) and log k (release constant) from the Peppas equation were estimated as release parameters. A set of release parameters and causal factors were used as tutorial data for the GRNN and analyzing using a computer. A GRNN model was constructed. The optimized GRNN model was used for prediction of formulation with desired in vitro drug release. For two tested formulations there was very good agreement between the GRNN predicted and observed in vitro profiles and estimated coefficients. Calculated difference (f(1)) and similarity (f(2)) factors indicate that there is no difference between predicted and experimental observed drug release profiles. This work illustrates the potential for an artificial neural network, GRNN, to assist in development of extended-release dosage forms. This method can be employed to achieve a desired in vitro dissolution profile.

Justification of biowaiver for carbamazepine, a low soluble high permeable compound, in solid dosage forms based on ivivc and gastrointestinal simulation

The aim of the present study was to use gastrointestinal simulation technology and in vitro-in vivo correlation (IVIVC) as tools to investigate a possible extension of biowaiver criteria to BCS class II drugs using carbamazepine (CBZ) as a candidate compound. Gastrointestinal simulation based on the advanced compartmental absorption and transit model implemented in GastroPlus was used. Actual in vitro and in vivo data generated in CBZ bioequivalence studies were used for correlation purposes. The simulated plasma profile, based on the CBZ physicochemical and pharmacokinetic properties, was almost identical with that observed in vivo. Parameter sensitivity analysis (PSA) indicated that the percent of drug absorbed is relatively insensitive to the variation of the input parameters. Additionally, plasma concentration-time profiles were simulated based on dissolution profiles observed under the different experimental conditions. Regardless of the differences observed in vitro, the predicted pharmacokinetic profiles were similar in the extent of drug exposure (AUC) while there were certain differences in parameters defining the drug absorption rate (C(max)t(max)). High level A IVIVC was established for the pooled data set (r = 0.9624), indicating that 1% SLS may be considered as the universal biorelevant dissolution medium for both the IR and CR CBZ tablets. The proposed methodology involving gastrointestinal simulation technology and IVIVC suggests that there is a rationale for considering CBZ biowaiver extension and introduction of the wider dissolution specifications for CBZ immediate release tablets.

Tablet disintegration and drug dissolution in viscous media: paracetamol IR tablets.

An investigation into the influence of viscous media on tablet disintegration and drug dissolution was performed with the aim to simulate the potential formulation-specific food effect for a selected highly soluble model drug. Literature data on the in vivo drug absorption in fasted and fed state have been evaluated for in vitro-in vivo correlation (IVIVC) purposes. In vitro studies were conducted in simple buffer media with or without addition of HPMC K4M as a viscosity enhancing agent. Good IVIVC correlation (r>0.95) was obtained for paracetamol dissolution in viscous media at 50rpm and fed state absorption profiles, while in vitro dissolution in simple media at lower stirring speed was predictable of drug products in vivo behaviour in the fasted state. The data obtained support the existing idea that relatively simple dissolution media and/or set of experimental conditions may be used to differentiate formulation-specific food-drug interactions. Such tests would be a useful tool in the development of formulations that would not be susceptible to the influence of co-administered meal and, furthermore, facilitate regulatory decision on the necessity to conduct food effect studies in vivo.

Artificial Neural Networks in the Modeling and Optimization of Aspirin Extended Release Tablets With Eudragit L 100 as Matrix Substance

The purpose of the present study was to model the effects of the concentration of Eudragit L 100 and compression pressure as the most important process and formulation variables on the in vitro release profile of aspirin from matrix tables formulated with Eudragit L 100 as matrix substance and to optimize the formulation by artificial neural network. As model formulations, 10 kinds of aspirin matrix tablets were prepared. The amount of Eudragit L 100 and the compression pressure were selected as causal factors. In vitro dissolution time profiles at 4 different sampling times were chosen as responses. A set of release parameters and causal factors were used as tutorial data for the generalized regression neural, network (GRNN) and analyzed using a computer. Observed results of drug release studies indicate that drug release rates vary widely between investigated formulations, with a range of 5 hours to more than 10 hours to complete dissolution. The GRNN model was optimized. The root mean square value for the trained network was 1.12%, which indicated that the optimal GRNN model was reached. Applying the generalized distance function method, the optimal tablet formulation predicted by GRNN was with 5% of Eudragit L 100 and tablet hardness 60N. Calculated difference (f1 2.465) and similarity (f2 85.61) factors indicate that there is no difference between predicted and experimentally observed drug release profiles for the optimal formulation. This work illustrates the potential for an artificial neural network, GRNN, to assist in development of extended release dosage forms.

An investigation into the factors influencing drug release from hydrophilic matrix tablets based on novel carbomer polymers.

Drug release from hydrophilic matrix tablets can be strongly influenced by the proportion of matrix forming polymer and the dimensions and geometry of the tablets. A complete two-factor, three-level factorial design, followed by multiple regression analysis and response surface methodology, was applied to investigate the influence of polymer level and tablet size on drug release kinetics from hydrophilic matrix tablets prepared with Carbopol 971P and Carbopol 71G. Tablet diameter, radius-to-height ratio, tablet surface area, and surface-area-to-volume ratio were evaluated as independent variables in terms of their applicability to characterize tablet size and geometry. The results indicate that it may be possible to control the rate of drug release by modifying the proportion of carbomer in tablets and tablet dimensions. The practical benefit of these simulations is to optimize the geometry and dimensions of a controlled release device and reduce the number of experiments involved in the development of new controlled release dosage forms.

An Investigation into the Influence of Hydrogel Composition on Swelling Behavior and Drug Release from Poly(Acrylamide-co-Itaconic Acid) Hydrogels in Various Media

The hydrogels prepared by free radical copolymerization of acrylamide and itaconic acid were investigated with regard to their composition and crosslinking degree to find materials with satisfactory swelling and drug release properties. Samples were characterized by measuring the swelling behavior and in vitro release of paracetamol as a model drug in aqueous media with different pH values. The two-factor, three-level experimental design and response surface methodology were applied to statistically evaluate the influence of investigated factors.

Viscosity-mediated negative food effect on oral absorption of poorly-permeable drugs with an absorption window in the proximal intestine: In vitro experimental simulation and computational verification.

Concomitant food intake can diminish oral absorption of drugs with limited permeability and an absorption window in the proximal intestine, due to viscosity-mediated decrease in dosage form disintegration time and drug dissolution rate. Three poorly-permeable drugs (atenolol, metformin hydrochloride, and furosemide) exhibiting negative food effect, and one highly-soluble and highly-permeable (metoprolol tartrate), serving as a negative control, were selected for the study. In vitro and in silico tools were used to evaluate the influence of media viscosity on drug bioperformance under fasted and fed conditions. The obtained results demonstrated that increased medium viscosity in the presence of food is one of the key factors limiting oral absorption of drugs with limited permeability and absorption restricted to the upper parts of the intestine, while having negligible effect on pharmacokinetic profile of drugs with pH- and site-independent absorption. Dissolution medium pH 4.6 with the addition of hydroxypropyl methylcellulose was suggested to simulate postprandial gastric conditions for drugs whose solubility under these conditions is not the limiting factor for drug absorption. In addition, drug formulation was found to be an interfering factor in relation to the impact of medium viscosity on the rate and extent of drug absorption.

Application of dynamic neural networks in the modeling of drug release from polyethylene oxide matrix tablets.

The main objective of this study was to demonstrate the possible use of dynamic neural networks to model diclofenac sodium release from polyethylene oxide hydrophilic matrix tablets. High and low molecular weight polymers in the range of 0.9-5 x 10(6) have been used as matrix forming materials and 12 different formulations were prepared for each polymer. Matrix tablets were made by direct compression method. Fractions of polymer and compression force have been selected as most influential factors on diclofenac sodium release profile. In vitro dissolution profile has been treated as time series using dynamic neural networks. Dynamic networks are expected to be advantageous in the modeling of drug release. Networks of different topologies have been constructed in order to obtain precise prediction of release profiles for test formulations. Short-term and long-term memory structures have been included in the design of network making it possible to treat dissolution profiles as time series. The ability of network to model drug release has been assessed by the determination of correlation between predicted and experimentally obtained data. Calculated difference (f(1)) and similarity (f(2)) factors indicate that dynamic networks are capable of accurate predictions. Dynamic neural networks were compared to most frequently used static network, multi-layered perceptron, and superiority of dynamic networks has been demonstrated. The study also demonstrated differences between the used polyethylene oxide polymers in respect to drug release and suggests explanations for the obtained results.

Artificial neural networks in evaluation and optimization of modified release solid dosage forms.

Implementation of the Quality by Design (QbD) approach in pharmaceutical development has compelled researchers in the pharmaceutical industry to employ Design of Experiments (DoE) as a statistical tool, in product development. Among all DoE techniques, response surface methodology (RSM) is the one most frequently used. Progress of computer science has had an impact on pharmaceutical development as well. Simultaneous with the implementation of statistical methods, machine learning tools took an important place in drug formulation. Twenty years ago, the first papers describing application of artificial neural networks in optimization of modified release products appeared. Since then, a lot of work has been done towards implementation of new techniques, especially Artificial Neural Networks (ANN) in modeling of production, drug release and drug stability of modified release solid dosage forms. The aim of this paper is to review artificial neural networks in evaluation and optimization of modified release solid dosage forms.

In vitro – in silico – in vivo drug absorption model development based on mechanistic gastrointestinal simulation and artificial neural networks: Nifedipine osmotic release tablets case study

In vitro--in vivo correlations (IVIVC) are generally accepted as a valuable tool in modified release formulation development aimed at (i) quantifying the in vivo drug delivery profile and formulation related effects on absorption; (ii) establishing clinically relevant dissolution specifications and (iii) supporting the biowaiver claims. The aim of the present study was to develop relevant IVIVC models based on mechanistic gastrointestinal simulation (GIS) and artificial neural network (ANN) analysis and to evaluate their applicability and usefulness in biopharmaceutical drug characterisation. Nifedipine osmotic release tablets were selected as model drug product on the basis of their robustness, dissolution limited drug absorption and the availability of relevant literature data. Although the osmotic release tablets have been designed to be robust against the influence of physiological conditions in the gastrointestinal tract, notable differences in nifedipine dissolution kinetics were observed depending on the in vitro experimental conditions employed. The results obtained indicate that both GIS and ANN model developed were sensitive to input kinetics represented by the in vitro profiles obtained under various experimental conditions. Different in silico approaches may be successfully employed in the in vitro--in silico--in vivo model development. However, the results obtained may differ and relevant outcomes are sensitive to the methodology employed.