The degradation, swelling and erosion properties of biodegradable implants prepared by extrusion or compression moulding of poly(lactide-co-glycolide) and ABA triblock copolymers

Abstract

In the design of parenteral delivery systems the modulation of the biodegradation of a polymer matrix represents a promising strategy to control drug release. We have investigated the degradation of ABA triblock copolymers, consisting of poly(lactide-co-glycolide) A-blocks and poly(oxyethylene) B-blocks, and PLG, poly(lactide-co-glycolide), with respect to swelling behaviour, molecular weight loss and polymer erosion. Implants were prepared by either compression moulding or extrusion using a laboratory ram extruder. Insertion of an elastoplastic B-block did not lower the processing temperature, but the entanglement of the polymer chains was significantly reduced as can be seen from the diameters of the extruded rods. The swelling of the rods showed a volume extension of 130% for an ABA containing 50% PEO and 20% for an ABA containing 20% PEO. Using ${}^1\text{H-NMR}$ it was found that protons in the B-blocks of the swollen ABA copolymers were mobile, while the A-blocks remained rigid during incubation. The analysis of the pH inside ABA rods using electron paramagnetic resonance, EPR, gave a pH of 5.2 after incubation with a subsequent increase to pH 6.0 during the first day, approaching the pH of the medium after nearly 33 d. Acidic degradation products did not accumulate inside the ABA rods. Degradation and erosion started immediately upon incubation. By contrast, PLG rods showed the typical profile of degradation and erosion. In this case, the influence of the geometry of the device was insignificant. Consequently, ABA triblock copolymers may widen the spectrum of parenteral drug delivery with regard to release of pH-sensitive drugs as well as erosion-controlled release kinetics.

Introduction

Biodegradable polymers have attracted considerable interest as parenteral depot systems to control the release of peptides, proteins and DNA [1], [2], [3]. The most widely investigated polymers with regard to toxicological and clinical data are copolyesters of poly(lactide-co-glycolide), PLG [4]. Although these polymers have been commercially used for peptide delivery, such as Zoladex^™ or Decapeptyl^™, their utility for proteins is limited due to their degradation behaviour and release properties. On one hand, the degradation of PLG, which is defined as decrease in weight and number average of molecular weights [5], leads to a drastic pH drop inside the matrix as low as pH 2 in vivo [6]. This may lead to an autocatalytic acceleration of the polymer degradation [7], as well as to an acidic microenvironment, which can compromise the stability of the encapsulated drugs, such as peptides or proteins [8]. The coencapsulation of buffer salts may neutralise the internal pH, but affects the erosion profile negatively [9]. On the other hand, a quantitative modelling of degradation and erosion of PLG is complicated due to a large number of parameters [9], [10]. This `bulk erosion’ [11] leads to striking differences from the desired infusion-like profile of protein release.

ABA triblock copolymers, consisting of poly(lactide-co-glycolide) A-blocks and poly(oxyethylene) B-blocks, have been developed to modulate both release and degradation properties of PLG. The insertion of a PEO B-block into PLG chains can yield star-shaped or linear ABA triblock copolymers [12]. Since the polymer mass loss or erosion of the ABA in vivo was substantially faster than PLG [13], we investigated the influence of compositional variations of linear ABA copolymers on the in vitro erosion. The release profile from ABA microspheres was influenced by polymer erosion, but the relation between water uptake or swelling and erosion has not been studied in detail [14], [15]. The pH of the microenvironment of a spin-labelled human serum albumin decreased after incubation [8]. For ABA films containing a semi-crystalline l-poly(lactic acid) A-block the preferential cleavage between A-block and B-block led to a biphasic degradation, while for ABA films with a PLG A-block the compatibility of hydrophilic and hydrophobic polymer phases was improved. As the degradation and erosion progressed linearly with the increase in block compatibility [16], the variation of the B-block and the shape of the matrix was investigated in this study.

Furthermore, PEO is advantageous for a melt processing, because PEO is a widely used plasticiser [17]. Since the copolymers consisting of lactic acid and glycolic acid require processing temperatures in the range of 70–90°C [18], the incorporation of a drug or an internal plasticiser like PEO may lead to heating conditions that are acceptable even for macromolecules like proteins.

From a mechanical point of view, common extruders either use shear forces and pressure, e.g. screw extruders, or high pressure in the range of a hundred bars requiring a hydraulic press, e.g. ram extruders [18], [19]. Since both extrusion techniques, which are widely used in processing of plastics, may lead to drug degradation, a laboratory ram extruder was constructed especially adapted to the requirements of thermoplastic materials. In order to evaluate this extrusion process, rods of PLG and ABA triblock copolymers were compared to tablets, prepared by compression moulding.

This study compares two melt-processing techniques for manufacturing implants from biodegradable polymers. Moreover, the decomposition under in vitro conditions of these tablets and rods was investigated with respect to the influence of geometry as well as of polymer properties of PLG and ABA triblock copolymers on degradation, erosion and swelling.