Info

*

Figure 6 Schematic presentation of a surface and a ¿w/fc-eroding drug delivery system. Stars represent dissolved (individual) drug molecules, whereas black circles represent undissolved drug excess (e.g., crystals and/or amorphous aggregates).

Generally, polymers containing very reactive functional groups in their backbone degrade rapidly and show surface erosion, whereas polymers containing less reactive functional groups tend to be bulk eroding. Polyanhydrides are examples for predominantly surface-eroding polymers, while poly(lactic acid) (PLA) and poly(lacticco-glycolic acid) (PLGA) are examples for predominantly bulk-eroding materials. However, it has to be kept in mind that the absolute polymer degradation rate alone not only determines the type of erosion (surface or bulk), but also the ratio "rate of polymer chain cleavage:rate of water penetration." The rate at which water enters the system strongly depends on the device dimensions. Thus, in extreme cases, drug delivery systems based on polymers with very reactive functional groups might show bulk erosion, for example, nanoparticles consisting of polyanhydrides. In contrast, devices based on polymers, which generally show bulk erosion, can exhibit surface erosion if they are large enough. Von Burkersroda et al. (42) proposed a polymer-specific, critical device dimension (Lcritical), above which the drug delivery system undergoes primarily surface erosion, and below which bulk erosion. For example, the ¿critical value is in the order of 100 |im for polyanhydrides, and in the order of 10 cm for PLGA. However, it has to be kept in mind that in the vicinity of the ¿critical values, both surface as well as bulk erosion are of importance and the overall erosion behavior of the system shows characteristics of both types of erosion.

Interesting mathematical theories have been proposed to quantify the involved physical and chemical phenomena in surface- (43) and ¿«/¿-eroding (38,40,44,45) drug delivery systems. As polymer chain cleavage is a random process, Monte Carlo simulations can effectively be used to simulate this phenomenon. The first to combine such Monte Carlo simulations with diffusional mass transport (based on Fick's second law) was A. Goepferich (46,47). This type of theories can, for instance, be applied to quantify drug release from PLGA-based microparticles (48,49). The basic idea is to divide a spherical microparticle into concentric rings of equal volume. One-quarter of a cross section of such a sphere is illustrated in Figure 7A. As it can be seen, a grid is used to divide this cross section into pixels. If this grid rotates around the z-axis, it describes half a sphere. If the drug and polymer are homogeneously distributed throughout the microparticles, there is a symmetry plane at r = 0, allowing for the calculation of the mass transport phenomena in the entire system. Each pixel shown in Figure 7A represents either drug or nondegraded polymer. This is the situation before exposure to the release medium (t = 0). Upon rotation around the z-axis, each pixel describes a ring. Importantly, the size of the two-dimensional pixels is not uniform and chosen in such as way that the volume of all three-dimensional rings is equal. This assures that the number of cleavable polymer backbone bonds in a ring representing nondegraded polymer is similar in all rings. Thus, the probability at which such a ring degrades upon first contact with water is similar. Since it is not possible to predict that at which time point which polymer backbone bond is cleaved, and as this process is random, Monte Carlo simulations can be used to describe this phenomenon. The idea is to randomly distribute "lifetime expectancies" to all pixels representing nondegraded polymer at t = 0. As soon as a particular pixel comes into contact with water, its "lifetime" starts to decrease. Once the latter expires, the pixel is assumed to be instantaneously transformed into a water-filled pore. Thus, the inner structure of the microparticles can be calculated at any time point (Fig. 7B). This information is crucial for the quantitative description of the resulting drug diffusion out of the system: The knowledge of the time- and position-dependent porosity of the device allows for the calculation of the time- and position-dependent drug-diffusion coefficients. Using Fick's second law of diffusion (22) and considering the given initial and boundary conditions, the resulting drug-release kinetics can be quantified and effects of processing and formulation parameters theoretically predicted (48,49).

As ester hydrolysis is catalyzed by protons, the underlying drug-release mechanisms from PLGA-based drug delivery systems can be even more complex (50,51). As described above, water penetration into these systems is much more rapid than the subsequent polymer chain cleavage. Thus, the entire device is rapidly wetted and ester hydrolysis occurs throughout the system. Figure 8 shows exemplarily a spherical microparticle, but these phenomena can also occur in other types of drug delivery systems, exhibiting different geometries. Due to concentration gradients, the generated shorter-chain acids diffuse out of the microparticles into the surrounding bulk fluid, where they are neutralized. In addition, bases from the environment diffuse into the system (again due to concentration gradients) and neutralize the generated acids. However, diffusional mass transport is generally slow and the rate at which the acids are generated via ester hydrolysis can be higher than the acid-neutralization rate, resulting in the local accumulation of acids and, thus, microenvironmental drops in the pH (52-54). This phenomenon is often most pronounced at the center of the device, because of the longer diffusion pathways. Importantly, ester hydrolysis is catalyzed by protons (55). Thus, polymer degradation is accelerated in the regions with low local pH. Consequently, drug release can be facilitated (50,51). In case of acid-labile drugs (e.g., proteins), care needs to be taken so that the biological activity is not lost. The importance of these phenomena essentially depends on the acid generation and neutralization rate. As the latter is a

Nortdegraded polymer Drug

Nortdegraded polymer Drug

Figure 7 Principle of a Monte Carlo-based approach to mathematically model polymer degradation and drug diffusion in PLGA-based microparticles. Scheme of the inner structure of the system (one-quarter of a spherical cross section): (A) at time t = 0 (before exposure to the release medium); and (B) during drug release. Gray, dotted, and white pixels represent nondegraded polymer, drug and pores, respectively. Source: From Ref. 48.

Figure 7 Principle of a Monte Carlo-based approach to mathematically model polymer degradation and drug diffusion in PLGA-based microparticles. Scheme of the inner structure of the system (one-quarter of a spherical cross section): (A) at time t = 0 (before exposure to the release medium); and (B) during drug release. Gray, dotted, and white pixels represent nondegraded polymer, drug and pores, respectively. Source: From Ref. 48.

Figure 8 Scheme of a bulk-eroding PLGA-based microparticle exhibiting autocatalysis: Upon hydrolytic ester bond cleavage, generated shorter-chain acids diffuse out of the system, while bases from the surrounding bulk fluid diffuse in. As the relative acid-neutralization rate exceeds the acid-generation rate, the microenvironmental pH within the system drops (this is often most pronounced at the center of the microparticle because of the longer diffusion pathways). As hydrolytic ester bond cleavage is catalyzed by protons, this leads to accelerated polymer degradation (autocatalysis) and potentially to drug degradation. The stars represent drug molecules.

Figure 8 Scheme of a bulk-eroding PLGA-based microparticle exhibiting autocatalysis: Upon hydrolytic ester bond cleavage, generated shorter-chain acids diffuse out of the system, while bases from the surrounding bulk fluid diffuse in. As the relative acid-neutralization rate exceeds the acid-generation rate, the microenvironmental pH within the system drops (this is often most pronounced at the center of the microparticle because of the longer diffusion pathways). As hydrolytic ester bond cleavage is catalyzed by protons, this leads to accelerated polymer degradation (autocatalysis) and potentially to drug degradation. The stars represent drug molecules.

function of the device dimensions and mobility of the involved acids and bases, the system size and initial porosity strongly affect the extent in local acidification.

A practical example for a drug treatment with a biodegradable-controlled drug delivery system is illustrated in Figure 9. The delivery system is a disk-shaped wafer (flat cylinder, commercialized under the trade name "Gliadel") and contains 3.85% of the anticancer drug BCNU [l,3-bis(2-chloroethyl)-l-nitrosourea; carmustine] (56-60). Despite its lipophilicity and low molecular weight (and, thus, ability to cross the blood-brain barrier to a certain extent), a systemic treatment with BCNU is not feasible because of the severe, dose-limiting side effects (in particular bone marrow suppression and pulmonary fibrosis) combined with a relatively short half-life (<15 min) (61). The biodegradable matrix former in Gliadel is a polyanhydride: poly[bis(p-carboxyphenoxy) propanensebacic acid] [p(CPP:SA)]. This advanced drug delivery system has been developed by Brem and coworkers and presented a major breakthrough in the field of controlled local brain delivery, which is highly challenging due to the blood-brain barrier. Gliadel got approved by the Food and Drug Administration (FDA) in 1996 for the treatment of recurrent glioblastoma multiforme. Figure 9A illustrates the basic principle of this treatment method: schematic cross sections of a human brain are shown. The black circle represents the tumor, the surrounding tissue being infiltrated by tumor cells. As the surgeon cannot remove large quantities of the surrounding tissue (due to the risk to affect vital brain functions), the probability that tumor cells remain within the brain is considerable. Consequently, many patients die due to local tumor recurrence in the direct vicinity of the primary tumor. To reduce this risk, up to eight BCNU-loaded wafers are placed into the resection cavity of the tumor, during the same operation, when the cranium is still open (Fig. 9A). The anticancer drug is then released in a time-controlled manner into the resection cavity and penetrates into the surrounding tissue. As the matrix-forming polymer is biodegradable, there is no need to remove empty remnants. In addition, the drug is released in the direct vicinity of the site of action, thus, systemic side effects are

Tumor

Surgical resection

Wafer implantation

Tumor

Surgical resection

40 80 120 160 200 Time (weeks)

Wafer implantation

ITT population

ITT population

0 6 12 18 24 30 36 42 Months from implant surgery

Figure 9 Treatment of operable brain tumors with BCNU-loaded, p(CPP:SA)-based wafers. (A) Schematic cross sections through a brain: The tumor is illustrated as a black circle; the surrounding tissue is infiltrated by cancer cells. Upon tumor resection, up to eight drug-loaded wafers (cylindrical disks) are placed into the resection cavity to minimize the risk of local tumor recurrence. (B) Overall survival of 222 patients with recurrent brain tumors (phase III clinical trial, after adjustment for prognostic factors). (C) Overall survival of 240 patients with newly diagnosed brain tumors (phase HI clinical trial, including results from the long-term follow-up). Abbreviation: ITT, intent-to-treat. Source: From Refs. 58 and 63 (B and C).

0 6 12 18 24 30 36 42 Months from implant surgery

Figure 9 Treatment of operable brain tumors with BCNU-loaded, p(CPP:SA)-based wafers. (A) Schematic cross sections through a brain: The tumor is illustrated as a black circle; the surrounding tissue is infiltrated by cancer cells. Upon tumor resection, up to eight drug-loaded wafers (cylindrical disks) are placed into the resection cavity to minimize the risk of local tumor recurrence. (B) Overall survival of 222 patients with recurrent brain tumors (phase III clinical trial, after adjustment for prognostic factors). (C) Overall survival of 240 patients with newly diagnosed brain tumors (phase HI clinical trial, including results from the long-term follow-up). Abbreviation: ITT, intent-to-treat. Source: From Refs. 58 and 63 (B and C).

reduced. In 1995, a multicentered, randomized, double-blinded, and placebo-controlled phase m clinical trial was conducted with 222 patients with recurrent malignant brain tumors (58). Importantly, the median survival time of the group of 110 patients who were treated with drug-loaded wafers was 31 weeks compared to only 23 weeks in the case of the 112 patients who received placebo disks. Figure 9B shows the overall survival of the patients (after adjustment for the examined prognostic factors). Furthermore, there were no clinically important side effects caused by the BCNU-loaded, p(CPP:SA)-based disks—neither locally within the brain, nor systemically. Later on, Gliadel was also used for the initial treatment of malignant gliomas (59,62). Upon surgical tumor resection, 240 patients received either BCNU-loaded or drug-free wafers. Both groups were postoperatively treated with external beam radiation. The median survival time in the intent-to-treat (ITT) group was 13.9 months for the Gliadel-treated patients and 11.6 months for the placebo group. The one-year survival rates were 59.2% and 49.6%, respectively. A long-term follow-up of this phase m clinical trial with 59 patients showed that 11 were alive at 56 months: 9 of them had been treated with Gliadel and only 2 with placebo wafers (63). The extended Kaplan-Meier curves for all 240 patients are shown in Figure 9C. Clearly, the survival advantage of the Gliadel-treated group was maintained even after one, two, and three years.

Another interesting example of the major benefits that a biodegradable, time-controlled drug delivery systems can offer is the delivery of growth factors and/or cytokines to proliferating/differentiating cells. This type of approach can be very useful for cell therapies: Up to now, the success of such advanced treatment methods, for instance, for neurodegenerative disease (e.g., Parkinson's disease) is limited because (i) the survival rate of transplanted cells within the brain tissue is low, and (if) the integration of the cells in their new environment is poor. Generally, about 90% of the transplanted cells die within the first two weeks after administration. One strategy to overcome these restrictions is to combine controlled-release microparticles and cell transplantation (64). The microparticles can, for example, release specific growth factors and/or cytokines at a predetermined rate and, thus, help to reduce cell death and to improve cell integration into the brain tissue. An even more sophisticated strategy is the use of microparticles not only as time-controlled drug delivery systems, but at the same time as microcarriers for the transplanted cells (64-68). This type of devices are also called "pharmacologically active microcarriers (PAM)." Figure 10A illustrates the concept of this approach. The microparticles exhibit two characteristic features: (i) a coating with cell adhesion or extracellular matrix molecules, and (if) time-controlled release of suitable biologically active agents at predetermined rates. Figure 10B, C show optical and scanning electron micrographs of such devices with PC12 cells adhering on their surfaces. Furthermore, the microparticles can additionally release drugs that are able to modify the microenvironment,

-Cell support (or culture -PAM degrade first from the inside After complete degradation ol PAM.

and transplantation -Enhancement ot survival and cells can integrate the parenchyma differentiation of transplanted ceils -Modulation of the microenvironrneni

-Cell support (or culture -PAM degrade first from the inside After complete degradation ol PAM.

and transplantation -Enhancement ot survival and cells can integrate the parenchyma differentiation of transplanted ceils -Modulation of the microenvironrneni

Figure 10 PAM used to improve the efficacy of cell therapies: (A) Schematic illustration of the principle of the approach. (B and C) Optical and scanning electron micrographs of cells adhering onto PAM. Abbreviations'. GF, growth factor and/or cytokine; PAM, pharmacologically active microcarriers. Source: From Ref. 65.

Figure 10 PAM used to improve the efficacy of cell therapies: (A) Schematic illustration of the principle of the approach. (B and C) Optical and scanning electron micrographs of cells adhering onto PAM. Abbreviations'. GF, growth factor and/or cytokine; PAM, pharmacologically active microcarriers. Source: From Ref. 65.

for example, favor angiogenesis or local immunodepression. Upon complete microparticle degradation, the cells can integrate the parenchyma. Recently, nerve growth factor-releasing PAM conveying PC 12 cells were transplanted into "Parkinsonian rats" (64). The idea is as follows: When PC12 cells expressing tyrosine hydroxylase (TH) are exposed to nerve growth factor, they stop cell division, extend long neuritis, become excitable, and after depolarization, can release significant amounts of dopamine (the neurotransmitter that is missing in Parkinson's Disease). Interestingly, first results showed that these nerve growth factor-releasing PAM can indeed reduce cell death and improve the amphetamine-induced rotational behavior of the rats.

0 0

Post a comment