Quantifying drug release from PLGA nanoparticulates
Introduction
Biodegradable polyesters composed of polylactic and or poly(lactic-co-glycolic) acids (PLGAs) have been successfully employed to design injectable drug delivery systems, such as implants and microparticulates (Vert et al., 1991, Fitzgerald and Corrigan, 1993, Beck et al., 1979). A wide range of active pharmaceutical ingredients (APIs), including conventional small drug molecules such as contraceptive steroids (Beck et al., 1979, Benita et al., 1984), anticancer agents (Wakiyama et al., 1981, Spenlehauer et al., 1988), and non-steroid anti-inflammatory drugs (NSAIDs) (Guiziou et al., 1996) as well as macromolecular biological therapeutic agents with short biological half-life (Conti et al., 1992, Sanders et al., 1984) have been delivered using these composites. The use of polyester polymers to deliver antigens, with the potential for pulsed delivery of booster dose(s), is under intensive study and there is evidence that particles in the submicron range, i.e. nanoparticulates, may give optimal particle uptake into ‘immune cells’ and improved efficacy (Florence, 1997, Köping-Höggård et al., 2005). However size reduction into the nanorange, could impact on the drug release rate profile. The mechanism(s) of API release from PLGA/PLA composites is not yet fully understood and consequently quantitative prediction of the drug release profile is problematic. Among the factors identified as impacting on drug release are, polymer/copolymer composition, processing variables, drug loading, drug physicochemical properties and possible drug–polymer interactions (Alexis, 2005). Particular attention has been paid to the size of the delivery device. Studies on microparticulates of different size and also polymer films and implants suggested that as size increases polymer degradation and mass loss accelerates and the kinetics becomes more heterogeneous (Lalla and Sapna, 1993, Dunne et al., 2000). Studies of piroxicam release from microparticles having mean sizes increasing from 13.5 to 76 μm showed a change in release profile from first order to exponential upwards with increasing size, the initial release rate decreasing with an increase in microsphere size (Lalla and Sapna, 1993). Thus it could be surmised that because of the short diffusional path, polymer degradation may play a much less important role in the release mechanism as the particle size decreases into the nanorange.
While there is considerable research on the synthesis and characterisation of PLGA nanoparticles (Astete and Sabliov, 2006), much less is published on the release mechanism from nanoparticulates. In 1992, Niwa et al. described the preparation of and release from biodegradable nanospheres (400–650 nm) containing small molecules, the low soluble indomethacin and soluble 5-fluorouracil. With indomethacin they observed a burst release phase of 50–80%, which depended on the molecular weight of the 85:15 PLGA employed. They ascribed this burst phase to drug at the surface. This phase was followed by a reduced drug release rate which they stated ‘might be controlled by the degradation speed of the polymer’. Release of 5-fluorouracil was more rapid, practically all of the drug being released in the burst phase. However these experiments lasted only for 120 h, much less than the time required for polymer degradation (Niwa et al., 1992). Other release studies using small molecules such as the marker coumarin-6 (Pietzonka et al., 2002, Eley et al., 2004, Khin and Feng, 2005), or the vitamin alpha-tocopherol (Zigoneanu et al., 2008) observed an initial rapid, but incomplete, release phase. However the experiments were not continued long enough for polymer degradation to occur. In relation to nanoparticulates containing large molecules, Blanco and Alonso (1997) examined BSA release from PLGA (50/50; RG 502, RG 503 & 503H) nanospheres, using a range of processing conditions. Common features of the release profiles observed were the burst release in the first day, followed by a slow continuous release. Experiments were conducted for 30 days, at which time incomplete release (i.e. ∼40–60%) of BSA from the RG503 type polymers had occurred. The authors concluded that not only degradation but also affinity of the BSA for polymer governed the post-burst release phase. Gutierro et al. (2002) observed similar release patterns for BSA entrapped in 200, 500 and 1000 nm PLGA (RG 506) particles, the initial rapid release phase being slightly faster the smaller the particle size. However after a month only half the loaded active was released. In contrast Panyam et al., 2003 examined BSA release from 100, 1000 and 10,000 nm PLGA (50/50, mw 120,000 Da) particulates over a 3 month period and observed protein release from the 0.1 and 1 μm particles to be greater than that from 10 μm size particles. Protein release was incomplete and they concluded that polymer degradation rates were not substantially different for different sized particles despite a 100-fold difference in surface area to volume ratio. Interestingly in none of these studies were the triphasic drug release profiles, involving an initial ‘burst’ phase, a lag phase followed by a second release phase, often reported for microparticulates (Batycky et al., 1997, Dunne et al., 2009) and implants (Gallagher and Corrigan, 2000), observed with nanoparticulates.
A range of models, involving contributions from either diffusion and/or polymer degradation, have been proposed in an attempt to quantify and predict drug release from biodegradable implants and microparticulates (Arifin et al., 2006, Faisant et al., 2002, Siepmann and Goepferich, 2001, Shah et al., 1992, Batycky et al., 1997, Leelarasamee et al., 1986, Jalil and Nixon, 1990, Pradhan and Vasavada, 1994).
Other compounds show three phases in the release versus time profile; an initial burst phase, a lag phase and a drug release following the lag phase, linked to polymer mass loss rather than the decrease in polymer molecular weight (Batycky et al., 1997, Fitzgerald and Corrigan, 1993, Fitzgerald and Corrigan, 1996, Rothstein et al., 2008).
Previously we described the degradation kinetics of pure PLGA micro- and nanoparticulates (Dunne et al., 2000) and showed that the kinetics of polymer mass loss may be described by an equation, reflecting a bulk degradation process, of the form:where x is the fractional polymer mass loss, t is the elapsed time, m = −ktmax where k and tmax are the constants reflecting the bulk degradation process (Prout and Tompkins, 1944, Jacobs, 1997). In the case of microparticulates and macroscopic discs (Fitzgerald and Corrigan, 1993, Fitzgerald and Corrigan, 1996, Ramtoola et al., 1992), not only the polymer mass loss, but also the concomitant drug release followed the same equation, consistent with polymer degradation as the controlling step dictating drug release.
Subsequently we proposed that the release of drugs from PLGA/PLA composites, at low loadings below the percolation threshold, could be considered in two phases, where the first phase of the release reflects diffusion controlled dissolution of drug accessible to the solid/dissolution medium interface and the second phase reflects release of drug entrapped in the polymer, the release of which is dependent on the bulk degradation of the polymer. In this simple mathematical model, if drug is dispersed as a separate phase in the polymer, then the initial diffusion or ‘burst’ release phase (FB) can be described by Eq. (2):where FBIN is the burst fraction at time infinity and kb is the first-order rate constant associated with the ‘burst’ release. The rate constant kb is equal to where D and Cs are the diffusion coefficient and solubility of the drug, respectively, A is the surface area of drug available for dissolution and h1 is the apparent aqueous diffusion boundary layer thickness. Thus kb is expected to increase with increased drug solubility and surface area.
The second release phase (FDeg), describing release of drug trapped in the polymer (1 − FBIN) is considered dependent on polymer erosion and may be described by bulk degradation kinetics Eq. (3):where tmax and k are the time to the maximum rate and the rate constant respectively of the polymer degradation release phase (Gallagher and Corrigan, 2000).
Therefore Ftot, the total fraction of drug released at a given time t is given by:Eq. (4) describes the fractional drug release versus time profile and enables estimation of the kinetic parameters k, kb and tmax by non-linear least squares fitting. The amount of active released (At) will be given by (FtotL) where L is the drug loading. This equation has been successfully used to describe drug release from biodegradable unit dosage forms (Gallagher and Corrigan, 2000, Milallos et al., 2008) and microparticulates (Dunne et al., 2009) and modified for cases where drug is molecularly dispersed in the polymer (He et al., 2005).
The release studies from nanoparticulates reviewed above suggest that in many of these studies experiments were not conducted long enough to observe a polymer degradation controlled drug release. The objective of this work was to examine the release of a range of model actives from PLGA nanoparticulates over a sufficient experimental period to encompass the polymer degradation/mass loss phase and evaluate quantitatively the release profile. To evaluate the use of Eq. (4) to describe and factor out the relative contributions from burst diffusion and polymer degradation to the release profile and thus extend our knowledge on the likely contribution of polymer degradation to the drug release process from biodegradable nanoparticulates. A range of model actives, both small molecules such as indomethacin, ketoprofen, 6-coumarin and larger molecules such as HSA and ovalbumin were employed encapsulated in PLGA nanoparticulates having mean diameters in the range 420–750 nm.
Section snippets
Measurement of molecular weight
Gel permeation chromatography (GPC) was used to determine polymer molecular weights; Mw the weight average molecular weight, Mn the number average molecular weight and polydispersity index. The system consisted of a Waters Styragel HPLC column, a Waters 510 HPLC pump and a Waters refractive index detector 410. Samples were prepared in tetrahydrafuran (THF). The mobile phase was THF and the flow rate was 1 ml/min. The internal and external temperatures of the detector were 30 and 25 °C,
Indomethacin and ketoprofen loaded PLGA nanoparticulates
The molecular weight characteristics determined for the PLGA sample employed were Mw 55.4 ± 6.3 kDa, Mn 53.8 ± 6.1 kDa and polydispersity 1.03 ± 0.01. The Mw obtained agrees with the value of 54.12 kDa reported by Mollo and Corrigan (2002) for the same batch of product.
Indomethacin (molecular weight; 357.79 Da) and ketoprofen (molecular weight; 254.28 Da) were encapsulated in PLGA to form nanoparticles. Their physical properties are summarised in Table 1. Compared to the drug free PLGA nanoparticles,
General discussion
The results obtained confirm the suggestions of others (Panyam et al., 2003), that the release of APIs from PLGA nanoparticulates (size range 400–700 nm) is similar in time frame/scale to that observed from micro- and macrodelivery systems. The drug release profiles obtained at low loadings, below the percolation threshold (Bonny and Leuenberger, 1991), indicate that the mechanism of release is similar to that obtained with microspheres and larger implants, the release profiles having a slow
References (63)
- et al.
Mathematical modelling and simulation of drug release from microspheres: Implications to drug delivery systems
Advanced Drug Delivery Reviews
(2006) - et al.
A Theoretical model of erosion and macromolecular drug release from biodegrading microspheres
Journal of Pharmaceutical Sciences
(1997) - et al.
A new long-acting injectable microcapsule system for the administration of progesterone
Fertility and Sterility
(1979) - et al.
Characterisation of drug loaded poly(d,l-lactide) microspheres
Journal of Pharmaceutical Sciences
(1984) - et al.
Development and characterization of protein-loaded poly(lactide-co-glycolide) nanospheres
European Journal of Pharmaceutics & Biopharmaceutics
(1997) - et al.
Differential thermal, solubility, and aging studies on various sources of digoxin and digitoxin powder: biopharmaceutical implications
Journal of Pharmaceutical Sciences
(1979) - et al.
The control of protein release from PLGA microparticles by variation of the external aqueous phase surfactant in the water-in oil-in water method
Journal of Controlled Release
(1998) - et al.
Influence of particle size and dissolution conditions on the degradation properties of polylactide-co-glycolide particles
Biomaterials
(2000) - et al.
First order release from porous PLA microspheres with limited exit holes on the exterior surface
Journal of Controlled Release
(2000) - et al.
PLGA-based microparticles: elucidation of mechanisms and a new, simple mathematical model quantifying drug release
European Journal of Pharmaceutical Sciences
(2002)