1. DRUG RELEASE FROM MATRIX SYSTEMS
Among the great variety of CRS, matrix systems, defined as the three-dimensional network containing the drug and other substances such as solvents and excipients, can take a great advantage from the engineers, chemists, pharmacists and medical doctors cooperation. Indeed, for instance, polymers science, surface, bioadhesion and thrombogenic properties, drug-matrix interactions, drug physicochemical and therapeutic characteristics and biocompatibility merge together to determine the final reliability and effectiveness of a matrix based delivery system. Accordingly, this chapter firstly deals with the description of matrices structures and the mechanisms ruling drug release from this kind of delivery systems. Then, based on this physical frame, the development of mathematical models aimed to describe release kinetics from different matrix systems is presented including an usually briefly matched topic such as drug release from ensemble of polydisperse spherical matrices.
1.2. MATRICES
One of the most common approaches to get controlled release is to embed a drug in a hydrophobic or hydrophilic matrix in order to speed up or to reduce drug release kinetics depending on the final therapeutic target. While wax, polyethylene, polypropylene, and ethylcellulose usually constitute hydrophobic matrices, hydrophilic matrices are generally made up by hydroxipropylcellulose, hydroxipropylmethylcellulose, methylcellulose, sodium carboxymethylcellulose and, in general, polysaccharides [6,7]. In addition, poly(vinyl alcohol), acrylates, metacrylates, polylactic acid and polyglycolic acid can not be forgotten [7]. Matrices based delivery systems find a great variety of applications in the pharmaceutical field especially for what concerns oral administration. While a detailed description of all these applications is out of the aim of this chapter, the analysis of the most recent ones can be interesting. For example, in regenerative medicine, matrices (hydrogels) are used as three-dimensional support for cells growth as in the case of stem cells isolated from the human bone marrow [8] and condrocytes [9-10]. Due to a particularly soft loading process, matrices (hydrogels) are also suitable to release proteins and oligonucleotides, high molecular weight and fragile drugs which can easily undergo denaturation or degradation upon the formulative process. For example, triblock co-polymer poly(ethylene glycol) (PEG)- poly(lactic-co-glycolic acid) (PLGA)-PEG is used to accelerate the wound healing process by releasing in situ a protein able to accelerate the wound healing process [11,12]. Matrix systems are also used to deliver antimicrobial [13], antifungal [14], anti-herpes simplex type 2 (HSV-2) [15] and paclitaxel, a drug preventing in situ re-growth of the neoplastic tissue after surgical removal [16].
Matrices can be prepared by mixing the drug, in form of a thin powder, with the pre-polymer. Then, the whole mixture is poured in the polymerisation reactor. Alternatively, matrix can be structured in advance and then put in contact with a highly concentrated drug solution able to swell the matrix (solvent swelling technique). Solvent removal is achieved, for example, by means of physical treatments [17]. Another approach relies on the mechanical energy supplied to the drug-carrier (usually a polymer constituting matrix network) couple by co-grinding. In this manner, it is possible to load a drug into matrix network avoiding the use of solvents, whose elimination from the final formulation can represent a very expensive and delicate step [18]. In addition, supercritical fluids can represent a profitable tool to achieve drug loading inside matrices [19-21]. Indeed, if one hand supercritical fluids show a density approaching that of liquids (this, usually, implies a good solubility respect to drug or solvents used in the solvent swelling technique), on the other hand they are characterized by a low viscosity, typical of gases. Consequently, they easily and efficiently swell the matrix (bringing the drug inside matrix network or extracting solvents) and they can be then removed by a simple pressure decrease. Indeed, as a pressure decrease provokes the transition from the supercritical condition to the gas one, the matrix can be easily devoided by the supercritical solvent without dragging out the drug. Finally, tablets represent the simplest and the most traditional way of preparing a matrix based delivery system as their formulation requires to compress, in a proper ratio, a carrier (usually a polymer), the drug and various excipients.
1.2.1 Matrix topology
Due to the great variety of typologies and materials that can be used, it is not easy to provide a comprehensive physical picture able to comprehend all kinds of existing matrices. Nevertheless, in the attempt of giving a rigorous description without loosing in generality and for the sake of simplicity, we can think to matrices as constituted by three different phases: (a) continuous (b) shunt and (c) dispersed [22]. These phase types have to be further classified into primary, secondary, tertiary etc., on the basis of the spatial relationships of each phase to the other phases present. A primary continuous phase pervades the whole matrix and, depending on its composition relative to the compositions of associated phases, it may provide an uninterrupted diffusional path for a solute or it represents an un-accessible zone acting as a mere supporting structure. A primary shunt phase interests the whole matrix and identifies with a macro-channels structure (pores). If the shunt phase is not present, matrix is defined non-porous while it will be classified as microporous (pores diameter: 10 – 100 nm), mesoporous (pores diameter: 100 – 1000 nm), macroporous (pores diameter: 1 – 100 mm) or superporus (pores diameter: 10 – 1000 mm) in the opposite case. While in the last two cases pores diameter is much larger than the molecular size of hypothetical diffusing molecules, in the first two cases this is no longer true [23]. Finally, dispersed phases are embedded in continuous phases or shunt phases.
For their importance and their wide use in the pharmaceutical field, it is worth particularizing the above mentioned theoretical frame to polymeric matrices. In this case, usually, the shunt phase is not present (even if it is not always true as later on discussed) while a primary continuous phase (usually a liquid phase) is trapped in a swollen solid secondary continuous phase constituted by high molecular weight molecules dispersed and collocated to form a continuous three-dimensional polymeric network pervading the whole system [24]. In very low swelling matrices, however, the polymeric phase can assume the role of primary phase due to its massive prevalence on the other (liquid) continuous phase. Regardless which one is the primary phase, the ensemble of these two continuous phases can be thought as a coherent system, having mechanical characteristics in between those of solids and liquids. The presence of crosslinks (polymer–polymer junctions) between polymeric chains hinders polymer dissolution in the liquid phase that can only swell the network. This structure is roughly similar to that of sponge filled by a liquid phase. Nevertheless this is a particular sponge as, in the case of strong crosslinks (typically chemical covalent bonds), the network does not modify with time. When, on the contrary, weak crosslinks prevail (typically physical interactions such as Coulombic, van der Waals, dipole–dipole, hydrophobic and hydrogen bonding interactions), polymeric chains are not so rigidly connected to each other and the similarity with the sponge is no longer so pertinent. Indeed, while crosslink density (number of crosslinks per unit volume) is constant with time (in static conditions), brownian motion of chains and segment of chains makes the distribution of crosslinks time dependent. As a consequence, whereas average dimensions of network meshes do not modify, each mesh can modify, thus resembling a statistical network. Obviously, this kind of network can easily undergo erosion due to polymer-polymer junctions weakness. This physical frame is made more complex by the fact that the whole structure can be constituted by an ensemble of small matrix domains embedded in a continuum, usually represented by a polymer solution as it occurs for Carbopol [25,26]. In addition, it is well known that many charged polysaccharides, i.e. alginate, pectate, gellan-gum, carrageenans, can form inhomogeneous matrix structures (hydrogels) in the presence of uni- and/or multivalent cations [27,28]. Indeed, at microscopic level, they show clusters with different and high crosslink density dispersed in a less crosslinked medium [29-31]. Finally, the contemporary presence of two networks (interpenetrating structures), originated by two different polymers can further complicate the scenario. Typically, these systems are produced by an initial swelling of a monomer and reacting to form a second intermeshing network structure [32-35]. Obviously, the choice of the polymer depends on the final CRS administration route (examples include oral, ophthalmic, rectal, vaginal, subcutaneous) and on different factors such as matrix swelling degree, biodiversity, biocompatibility, interactions with drug, excipients and mechanical properties.
Polysaccharides, in particular glucans and xanthan, represent typical examples of physically crosslinked matrices [24]. Indeed, inter-chains physical interactions allow the formation of junction zones where the regular coupling of chains portion belonging to different polymeric chains takes place (Fig.6.1). The long chain segments departing from these junction zones can form, with other chains, additional junction zones so that a polymeric three-dimension network can be built up. Obviously, in general, these are weak networks (even if network strength strongly depends on polymer concentration) that can easily undergo erosion leading to structure destruction. Alginates, on the contrary, represent a typical example of strong physical networks (matrices) [27]. In this case, junction zones resemble egg boxes for their characteristics shape [36] (Fig.6.1). Indeed, bivalent cations take part to an inter-chains bond between guluronic residues that, in their planar conformation, give origin to a bond-cavity for cations. In particular, bivalent cations interact with oxygens of COO- groups and some OH groups of guluroinc residues belonging to different chains [37]. Sequences of these cavities form the above mentioned egg boxes (junction zones). Affinity series of bivalent cations is [38]:
Pb > Cu > Cd > Ba > Sr > Ca > Co, Ni, Zn > Mn
For many biological applications, Cu++ and Ca++ represent the election cations for the formation of inter-chains junction zones. Strontium and Barium are used in hydrogels designed to contain cells [39]. Crosslinked polyvinylpyrrolidone (C6H9NO) represents a particular example of strong physically crosslinked matrix. Indeed, the popcorn polymerisation used to get the polymeric matrix [40], gives origin to a very complex topological structure made up by entangled polymeric chains giving origin to a spaghetti like structure (Fig. 6.1). The fact that this matrix never dissolves in proper solvents (like water, for example) leads to the conclusion that the above mentioned spaghetti like structure is stabilized by a very small amount of chemical crosslinks [41].
Among the many examples of chemically crosslinked matrices, temperature responsive gels can be mentioned. Accordingly, N-alkylacrylamides with N-isopropylacrylamide (NIPA), N-N-Diethylacrylamide (DEAAm) and their copolymers where a variety of crosslinkers ranging from N,N’-methylenebis(acrylamide) and ethylene glycol dimethacrylate to N-methylolacrylamide can be remembered [42]. Peppas and co-workers [43], for example, uses poly(N-isopropylacrylamide-co-methacrylic acid) gels crosslinked by ethylene glycol dimethacrylate for the delivery of antithrombotic agents such as heparin. In addition, different polysaccharides such as methylcellulose, hydroxypropyl-methylcellulose, hydroxypropylcellulose and carboxymethylcellulose can be crosslinked with divinyl sulfone [44], adipic, sebacic, or succinic acid [45]. Finally, poly(ethylene glycol) (PEG) and poly(hydroxyethyl methacrylate) (PHEMA) are other common polymers that can undergo chemical crosslinking [28].
While the above discussed examples mainly deal with non porous structures, typical examples of porous matrices are represented by superporous hydrogels (SPH), superabsorbent polymers (SAP) [46], some kind of hydrophobic matrices and tablets. SAP and SPH are crosslinked hydrophilic polymers having the ability to absorb up to considerable amounts of water or aqueous fluids (10–1000 times of their original weight or volume) in relatively short periods of time due to their wide porous structure generated by the decomposition of a bicarbonate compound during polymerisation (Fig. 6.2). In order to prevent the formation of heterogeneous structures associated to the violent exotermic reaction involved in polymerisation, hydrophilic monomer (typically acrlylamide, salts of acrylic acid and sulfopropyl acrylate) is diluted in water by gently mixing at room temperature. Then, the neutralisation step, required for ionic monomers, is followed by crosslinker addition. Glacial acetic acid or acrylic acid are subsequently added to get a foam during polymerisation. In addition, in order to get homogeneous SPH, a surfactant (such as triblock copolymers (poly ethylene oxide - poly propylene oxide -poly ethylene oxide)) is needed to stabilise the foam. While in the case of SPA, polymerisation is obtained via both redox and thermal systems, SPH are produced only via a redox mechanism. Accordingly, oxidant and reductant (and thermal initiator) are added to monomer solution under gentle mixing. Finally, acid-dependent foaming agent (typically sodium bicarbonate) is added. SPA and SPH products are then dehydratated (heating for SPA, alcohol plus heating for SPH) to get a solid, brittle porous material, which is white in colour because of heterogeneous combination of polymer and pores. The final product can be ground into a particle shape (like superabsorbent particles), sliced into absorbent sheets, or machined into any shape and size. Obviously, pores characteristics and porosity can be varied by properly modulating the production parameters so that various grades of products can be prepared spanning from non porous up to superporous structures. Usually, swollen SPH show very poor mechanical properties and that is why their use in many practical applications turns out to be very difficult. Accordingly, new types of SPH have been proposed to overcome this serious limitation and two new generations of SPH have been realised. While first generation consists in fragile fast and high swelling ratio systems, second generation is made up by fast and medium swelling systems characterised by improved mechanical properties. Finally, their good mechanical properties make third generation suitable also for pharmaceutical and biomedical applications. In order to get second SPH generation, a crosslinked hydrophilic polymeric matrix (CHPM) able to absorb monomer solution, crosslinker, initiator and remaining components is added to monomer solution. Upon polymerization, CHPM serves as the local point of physical crosslinking (or entanglement) of the formed polymer chains. During the polymerization process, each CHPM particle acts as an isolated individual reactor in which crosslinking polymerization occurs. The final result is a structure made up by an ensemble of CHPM units connected each other by polymeric chains. Third SPH generation, on the contrary, relies on the addition of a hybrid agent (HA) that can be crosslinked after monomer polymerisation and crosslinking. Typically, HA is a water soluble or water dispersible polymer that can form a crosslinked (chemical or physical) structure pervading the primary SPH polymeric structure yielding to an interpenetrating network. Examples of hybrid agents are polysaccharides (sodium alginate, pectin, chitosan) or synthetic water-soluble hydrophilic polymers such as poly(vinyl alcohol). Examples of third SPH generation are acrylamide-based structures in the presence of sodium alginate crosslinked by calcium ions and poly(acrylamide-co-aclylic acid) in the presence of polyethyleneimine [23].