Philosophiae Doctor - PhD / Various formulations of the popular South African medicinal plant, Sutherlandia frutescens,are commercially available, with no documented specifications for quality assessment. With plans already underway for a clinical trial to assess its efficacy in HIV patients, there is a need for scientifically validated tests for the quality control of products of this plant. Chemical constituents of the plant are many and varied but it is still unclear which might be the most appropriate ones to monitor for activity or to describe the quality of the plant’s products. For quality control and regulatory purposes, the content and dissolution of flavonoids in the plant products can be assessed. However, these compounds are not monitored for regulation and there are as yet no HPLC or dissolution methods that can be employed for quality control of herbals like S. frutescens. Therefore, the objectives of this study were to assess the suitability of its flavonoid constituents as quality control (QC) marker compounds, and the suitability of content levels and dissolution tests of flavonoids as QC tools for S. frutescens products. To realise the afore-mentioned objectives, non-commercially available flavonoid compounds (sutherlandins) that could be used as marker compounds were isolated from S. frutescens. An HPLC assay was developed and validated for determination of flavonoid content in solution. Five S. frutescens materials viz leaf powder (LP), spray-dried aqueous extract (SDAE) and freeze-dried aqueous extracts (FDAE) were analysed for flavonoid content and dissolution. Dissolution tests were conducted for different S. frutescens materials and dissolution profiles of flavonoids in capsules containing these materials were compared using Q-release values, the similarity factor (f2) and mathematical models. To predict in vivo bioavailability of the flavonoids, in silico assessment of in vivo bioavailability of flavonoids (glycosides and aglycones) that may be contained in different S. frutescens materials was conducted. Sutherlandins A, B, C and D were successfully isolated (percentage purity approximately99 % for sutherlandins A, C and D, and 90 % for sutherlandin B) and identified, and used, along with other flavonoid compounds, for the development of a simple and robust HPLC method. Content of sutherlandins A, B, C and D, quercetin and kaempferol in different plant materials were 0.4 ± 0.3, 0.8 ± 0.2, 1.3 ± 0.2, 0.6 ± 0.1, 0.01 ± 0.02 and 0.08 ±0.1 %,respectively, and differed significantly (p < 0.001). In vitro dissolution showed faster dissolution of flavoniod glycosides compared to aglycones. The flavonoids from the LP and SDAE materials showed characteristics of immediate release with Q75 in ≤ 45 minutes, and delayed release from the FDAE material, i.e. Q75 > 45 minutes. The dissolution profiles of each flavonoid compared from different S. frutescens materials were different as signified by their f2 values which were all below 50. The mathematical models describing release were also different for each flavonoid from the different S. frutescens materials. For in vivo bioavailability modelling and prediction studies, the flavonoid aglycones met the conditions for oral bioavailability while the flavonoid glycosides did not. In conclusion, the sutherlandins isolated from S. frutescens proved to be good markers for HPLC assay and dissolution tests of S. frutescens materials. The HPLC method was suitable for assessing flavonoid levels in S. frutescens materials, and also showed differences in flavonoid content in these materials. The dissolution method was simple and reproducible, and Q-release values, the f2 and mathematical models proved to be good tools for differentiating between S. frutescens materials. In silico modelling showed that the flavonoid glycosides and aglycones differed in oral bioavailability. Although not presently required by the Medicines Control Council (MCC), quantification, release and dissolution studies and specifications may be employed as tools for routine analysis and for quality control of herbal drug formulations containing S. frutescens.

Submitted by Jean Medeiros (jeanletras@uepb.edu.br) on 2017-02-23T15:20:42Z No. of bitstreams: 1 PDF - Amaro da Costa.pdf: 1485927 bytes, checksum: f6f5563852d161825574a80fa4aad991 (MD5) / Approved for entry into archive by Secta BC (secta.csu.bc@uepb.edu.br) on 2017-03-07T16:11:14Z (GMT) No. of bitstreams: 1 PDF - Amaro da Costa.pdf: 1485927 bytes, checksum: f6f5563852d161825574a80fa4aad991 (MD5) / Made available in DSpace on 2017-03-07T16:11:14Z (GMT). No. of bitstreams: 1 PDF - Amaro da Costa.pdf: 1485927 bytes, checksum: f6f5563852d161825574a80fa4aad991 (MD5) Previous issue date: 2015-08-21 / Fundo de Desenvolvimento do Capital Humano - Timor Leste / The absorption of a pharmaceutical product in a solid form after oral administration depends on several factors, such as release and solubilization of the drug and its permeation through the gastrointestinal tract. The aim of this study was to construct the dissolution profiles of medicinal products based on paracetamol and caffeine using partial least squares regression. The dissolution profile was determined using distilled water as the dissolution medium with apparatus 2 for 30 minutes. The collection of samples was performed in times between 5 to 30 minutes, and aliquots of the filtered and diluted with distilled water in the ratio of 1: 100 (v / v). The dissolution profiles were registered in the range 190-300 nm in molecular absorption spectrophotometer UV-Vis. To analise the data, the study used the method of partial least squares regression (PLSR) using The Unscrambler 9.8 software. The PLS models for paracetamol and caffeine had to be suitable, low value the root mean square error of calibration (RMSEC) and coefficient of determination (R2), which is closed to 1, with no significant bias and low value the relative error of prediction (REP). In comparing the dissolution profiles of the drugs, it was noted that the reference product showed a variation profile between batches much lower than generic drugs, which indicates greater uniformity in the production and quality of raw materials. When evaluated the degree of similarity between the drugs was observed that the three marks are not statistically different, since the values of difference factor (F1) to paracetamol is below the critical limit specified. Therefore, the constructed models of PLS are feasible for the use in the quality control of medicines, especially in the study of dissolution that contain acetaminophen and caffeine associated drugs. In addition, the method developed can be considered cheap and fast to achieve the desired results, without prior separation. / A absorção de um produto farmacêutico em forma sólida, após a sua administração por via oral depende de vários fatores, tais como liberação e solubilização do fármaco e sua permeação através do trato gastrointestinal. Assim, o objetivo deste trabalho foi construir os perfis de dissolução de medicamentos à base de paracetamol e cafeína, utilizando regressão por mínimos quadrados parciais. O perfil de dissolução foi realizado utilizando como meio de dissolução água destilada com aparato 2, por 30 minutos. A coleta das amostras foi realizada nos tempos entre 5 a 30 minutos, sendo as alíquotas filtradas e diluídas com água destilada na proporção de 1:100 (v/v). Os perfis de dissolução foram registrados na faixa espectral de 190 a 300 nm em espectrofotômetro de absorção molecular UV-Vis. Para a análise dos dados foi utilizada a regressão pelo método dos mínimos quadrados parciais (PLSR), utilizando o software The Unscrambler 9.8. Os modelos PLS para paracetamol e cafeína apresentaram-se adequados, com baixo valor de raiz quadrada do erro médio quadrática de calibração (RMSEC) e coeficiente de determinação (R ), que se aproxima de 1, com bias não significativos e baixo valor do erro relativo de predição (REP). Comparando os perfis de 2 dissolução dos fármacos, percebeu-se que o medicamento de referência apresentou uma variação do perfil entre lotes muito mais baixo do que os medicamentos genéricos, o que indica uma maior uniformidade na produção e qualidade das matérias prima. Quando avaliou- se o grau de similaridade entre os medicamentos, observou-se que as três marcas não são diferentes estatisticamente, pois os valores de fator diferença (f1) para paracetamol estão abaixo do limite crítico especificado. Portanto, os modelos PLS construídos são viáveis para ser utilizados no controle da qualidade de medicamentos, principalmente no estudo de dissolução que contenham paracetamol e cafeína associados em medicamento. Além disso, o método desenvolvido pode ser considerado barato e rápido para alcançar os resultados pretendidos, sem necessidade de separação prévia.

The use of computer programs to predict drug absorption in humans and to simulate dissolution profiles has become a valuable tool in the pharmaceutical area. The objective of this study was to use in silico methods through software GastroPlusTM and DDDPlusTM to simulate drug absorption curves and dissolution profiles, and to establish in vitro-in vivo correlations (IVIVCs). The work presented herein is divided into five chapters and includes the drugs ketoprofen, pyrimethamine, metronidazole, fluconazole, carvedilol and doxazosin. In Chapter 1, simulated plasma curves for ketoprofen matrix tablets are presented and IVIVC was established. The use of simulated intrinsic dissolution tests for pyrimethamine and metronidazole as a tool for biopharmaceutics classification is detailed in Chapter 2. In Chapter 3, simulation of plasma curves for fluconazole capsules with different dissolution profiles is demonstrated as a tool for biowaiver. IVIVC studies were also conducted for carvedilol immediate-release tablets from dissolution profiles in Chapter 4. Chapter 5 covers the application of simulated dissolution tests for development of doxazosin extended-release formulations. Simulation of plasma curves and IVIVC using the software GastroPlusTM as well as intrinsic dissolution tests and dissolution profiles using the software DDDPlusTM proved to be a tool of wide application in predicting biopharmaceutical characteristics of drugs and formulations, allowing the reduction of time and costs of experimental laboratory work. / O uso de programas de computador para prever a absorção de fármacos em humanos e simular perfis de dissolução tem se tornado uma ferramenta bastante valiosa na área farmacêutica. O objetivo deste trabalho foi utilizar métodos in silico por meio dos programas de computador GastroPlusTM e DDDPlusTM para simular curvas de absorção de fármacos, perfis de dissolução e estabelecer correlações in vitro-in vivo (CIVIVs). O material aqui apresentado é constituído por cinco capítulos incluindo os fármacos cetoprofeno, pirimetamina, metronidazol, fluconazol, carvedilol e doxazosina. No capítulo 1 são apresentadas curvas plasmáticas simuladas para comprimidos matriciais de cetoprofeno, sendo estabelecida a CIVIV. A utilização de simulações de ensaios de dissolução intrínseca para os fármacos pirimetamina e metronidazol como uma ferramenta para classificação biofarmacêutica é detalhada no capítulo 2. No capítulo 3, a simulação de curvas plasmáticas a partir de cápsulas de fluconazol contendo diferentes perfis de dissolução é demonstrada como uma ferramenta para bioisenção. Estudos de CIVIV foram também realizados para comprimidos de liberação imediata de carvedilol a partir dos perfis de dissolução no capítulo 4. Já o capítulo 5 trata da aplicação de simulações de ensaios de dissolução para o desenvolvimento de formulações de liberação prolongada de doxazosina. As simulações das curvas plasmáticas, assim como a CIVIV, obtidas com o auxílio do programa GastroPlusTM, além dos ensaios de dissolução intrínsica e os perfis de dissolução obtidos por meio do uso do programa DDDPlusTM apresentaram-se como ferramentas de grande aplicação na previsão de características biofarmacêuticas sobre os fármacos e formulações, permitindo redução de tempo e custo com trabalho experimental em laboratório.

The use of computer programs to predict drug absorption in humans and to simulate dissolution profiles has become a valuable tool in the pharmaceutical area. The objective of this study was to use in silico methods through software GastroPlusTM and DDDPlusTM to simulate drug absorption curves and dissolution profiles, and to establish in vitro-in vivo correlations (IVIVCs). The work presented herein is divided into five chapters and includes the drugs ketoprofen, pyrimethamine, metronidazole, fluconazole, carvedilol and doxazosin. In Chapter 1, simulated plasma curves for ketoprofen matrix tablets are presented and IVIVC was established. The use of simulated intrinsic dissolution tests for pyrimethamine and metronidazole as a tool for biopharmaceutics classification is detailed in Chapter 2. In Chapter 3, simulation of plasma curves for fluconazole capsules with different dissolution profiles is demonstrated as a tool for biowaiver. IVIVC studies were also conducted for carvedilol immediate-release tablets from dissolution profiles in Chapter 4. Chapter 5 covers the application of simulated dissolution tests for development of doxazosin extended-release formulations. Simulation of plasma curves and IVIVC using the software GastroPlusTM as well as intrinsic dissolution tests and dissolution profiles using the software DDDPlusTM proved to be a tool of wide application in predicting biopharmaceutical characteristics of drugs and formulations, allowing the reduction of time and costs of experimental laboratory work. / O uso de programas de computador para prever a absorção de fármacos em humanos e simular perfis de dissolução tem se tornado uma ferramenta bastante valiosa na área farmacêutica. O objetivo deste trabalho foi utilizar métodos in silico por meio dos programas de computador GastroPlusTM e DDDPlusTM para simular curvas de absorção de fármacos, perfis de dissolução e estabelecer correlações in vitro-in vivo (CIVIVs). O material aqui apresentado é constituído por cinco capítulos incluindo os fármacos cetoprofeno, pirimetamina, metronidazol, fluconazol, carvedilol e doxazosina. No capítulo 1 são apresentadas curvas plasmáticas simuladas para comprimidos matriciais de cetoprofeno, sendo estabelecida a CIVIV. A utilização de simulações de ensaios de dissolução intrínseca para os fármacos pirimetamina e metronidazol como uma ferramenta para classificação biofarmacêutica é detalhada no capítulo 2. No capítulo 3, a simulação de curvas plasmáticas a partir de cápsulas de fluconazol contendo diferentes perfis de dissolução é demonstrada como uma ferramenta para bioisenção. Estudos de CIVIV foram também realizados para comprimidos de liberação imediata de carvedilol a partir dos perfis de dissolução no capítulo 4. Já o capítulo 5 trata da aplicação de simulações de ensaios de dissolução para o desenvolvimento de formulações de liberação prolongada de doxazosina. As simulações das curvas plasmáticas, assim como a CIVIV, obtidas com o auxílio do programa GastroPlusTM, além dos ensaios de dissolução intrínsica e os perfis de dissolução obtidos por meio do uso do programa DDDPlusTM apresentaram-se como ferramentas de grande aplicação na previsão de características biofarmacêuticas sobre os fármacos e formulações, permitindo redução de tempo e custo com trabalho experimental em laboratório.

L'objectif de ce travail a été l'utilisation d'une méthodologie de formulation des formes à libération prolongée. Le choix de «matrices hydrophiles» a été fait en raison de l'intérêt et de l'importance des travaux qui lui sont consacrés, mais surtout en raison de la possibilité d'utilisation de la technologie simple de fabrication des comprimés par compression directe.Le principe actif choisi est le diclofenac de sodium, un anti-inflammatoire largement utilisé dont la molécule est tombée dans le domaine public. Utilisé  à raison de 100 mg par comprimé. La première partie bibliographique résume l'intérêt des formes à libération prolongée, décrit les différentes formes galéniques existantes pour la voie orale et fait le point sur les formules et les propriétés des agents matriciels hydrophiles parmi les plus utilisés, notamment les éthers de celluloses, hydroxypropymethylcellulose (METOLOSES®) (1). Dans la partie expérimentale, nous avons réalisé les différentes étapes nécessaires au développement d'une forme à libération prolongée qui correspondant aux étapes classiques  de pré-formulation, formulation et optimisation. Dans une première étape d'essais préliminaires  nous avons étudiés le comportement d'un point de vue pharmacotechnique des matières premières utilisés seuls et en mélange. Ce qui a permis de faire des orientations en ce qui concerne les différents types de METOLOSES, la nature du diluant, et leurs concentrations. Dans une deuxième étape, nous avons réalisé un premier plan d'expériences de criblage des facteurs, qui permet de déterminer le poids de chacun et leurs éventuelles interactions. Ce qui nous a permis de conclure sur l'effet des différents facteurs en formulation. Dans une troisième étape, nous avons réalisé un deuxième plan d'expériences d'optimisation de la formulation  en utilisant un plan composite centré constitué de 9 expériences ce qui a permis de sélectionner une zone de formules optimales. Enfin, la réalisation d'une formule optimale nous a permis de confirmer les résultats obtenus dans les travaux de développement et dont l'objectif été une libération sur 12 heures. / The objective of this work was the use of a formulation methodology of prolonged release dosage forms. The choice of “hydrophilic matrices” was done because of the interest and the numbered works which is devoted to it, but also for the use of a simple manufacture technology by direct compression. The selected active pharmaceutical ingredient is one of the most largely used nonsteroidial anti-inflammatory drug, the sodium diclofenac which is out of patent and used at 100 Mg per tablet. The first bibliographical part summarizes the interest of the prolonged release dosage forms, described the existing oral dosage forms and gives a progress report on the formulas and the properties of hydrophilic matrices agents among most used, in particular the cellulose ethers, hydroxypropymethylcellulose (METOLOSES®) (1). In the experimental part, we carried out the various steps necessary to the development of a prolonged release dosage form which correspond to the traditional steps of pre-formulation, formulation and optimization. In a first step of pharmaceutics preliminary tests we studied the behavior of raw materials used alone and in mixture. What made orientations with regard to the various types of METOLOSES, the nature of diluents, and their concentrations. In a second step, we carried out a first screening factors experimental design, which enabled us to conclude on the effect of the various factors in formulation. In a third step, we carried out a second optimization experimental design using a centered composite plan consisted of 9 experiments which lead us to define a space design of optimal formulas. Lastly, the manufacture of a formula from the design space enabled us to confirm the results to development work and for which the objective was a sustained release over 12 hours.

Today, orally inhaled drugs found on the market suffer from variable and discontinuous pulmonary drug release which lowers efficacy and patience compliance. This is usually a consequence of the poor understanding of the interaction and dissolution behavior of drug particles in the lung environment. Thus, the aim of this project was to investigate the effect of the dissolution medium on dissolution profiles for the well-known orally inhaled drug budesonide (BD) and fluticasone propionate (FP), in order to assess the importance of a proper selection of dissolution media for in vitro dissolution methods. In order to achieve this a modified Andersen Cascade Impactor was used to simulate deposition of particles onto filters. The dissolution was measured using a Transwell set up with polycarbonate membranes that can hold the filters with the deposited drug on it. Different media were prepared, from simple to more biorelevant. The samples taken during the dissolution experiments were analyzed quantitatively using UPLC-UV and the experimental data was processed by fitting to the Weibull function. The aim of this project was successfully achieved and the dissolution media that worked best for both BD and FP was PBS with the addition of 0.5% SDS. On the other hand, the dissolution media that performed the least for both BD and FP was the simulated lung fluid (SLF) with presence of 0.02% (w/v) DPPC. This may be due to the fact that DPPC forms liposomal aggregates which probably results in the media becoming more viscous and hence the dissolution time becomes slower.