On enhancement of bone formation using local drug delivery systems

Sammanfattning: Introduction: Despite that many reports have confirmed the long-term clinical success rates associated with implant treatment, implant failure to achieve and maintain osseointegration still occurs in many cases. Local and sustained drug release at the bone-implant interface is one of the strategies that have been suggested to improve the osseointegration. The local drug release could help avoiding the risks usually associated with systemic administration, such as high drug dose or the loss of drug bioavailability. Aims: To map out the most commonly used chemical compounds and drug delivery systems used in animal experiments for implant research (Study I). Furthermore, to develop a new surface coating designed for medical devices and implants, and to examine the drug release mechanism from the coating using near infrared light (NIR) as an external stimulus (Study II). In addition, we examined the release of clarithromycin from PLGA microspheres within beta-tricalcium phosphate (β-TCP), and evaluated their osteogenic effect in a calvaria defect model in vivo (Study III). In Study IV, we evaluated the local release of strontium ranelate (Sr-ranelate) from implant surface coated with mesoporous titania films, and investigated if the local release of Sr-ranelate could improve bone formation around implants in an animal model. Materials and Methods: The articles included in the present thesis consist of four different studies. For Study I, an electronic search was done in three databases (PubMed, Scopus, Embase) to map out the most commonly used methods for local drug and chemical compound delivery to implant sites, and to assess their influence on bone response. Meta-analyses were performed for the outcome of bone-to-implant contact (BIC). In Study II, PNIPAAm-AAm polymers were synthesized at different compositions. The polymers were then incorporated with gold nanorods (GNRs) since these rods at predetermined aspect ratio can absorb NIR light to generate heat within the polymeric layer to initiate a drug release. The volume-phase transition behavior for the polymers was analyzed using differential scanning calorimetry (DSC). The GNRs-incorporated PNIPAAm was characterized using scanning electron microscopy (SEM) and quartz crystal microbalance with dissipation monitoring (QCM-D). The release behavior using phenol as drug model was investigated upon NIR irradiation using UV/VIS spectroscopy. In addition, the antibacterial behavior of polymer layers loaded with vancomycin was examined against Staphylococcus epidermidis. In Study III, four bone defects (5 mm of diameter) were created in the calvaria of New Zealand White rabbits (n = 21, n= 7/time point). The defects were randomly designated to four groups. Group 1: no augmentation (sham), Group 2: β-TCP, Group 3: β-TCP with 0.12 mg clarithromycin, and Group 4: β-TCP with 6.12 mg PLGA microspheres loaded with 0.12 mg clarithromycin. After 2, 4, and 12 weeks of healing, bone regeneration was evaluated using micro-computed tomography (µCT) and histology. In Study IV, mini-screw titanium implants were coated with mesoporous TiO2 films using Pluronic (P123) with or without poly propylene glycol (PPG) to create films with two different pore sizes. The implants were then incorporated with Sr-ranelate. SEM evaluation was performed to visualize the mesoporous TiO2 films and determine the pore size. The absorption and release kinetics of Sr-ranelate from mesoporous TiO2 films were evaluated by QCM-D. For the in vivo experiment, mini-screw titanium implants with or without Sr-ranelate were inserted in rats’ tibia bone to evaluate bone formation after 2 and 6 weeks. Results: In the systematic review (Study I), sixty-one studies met the inclusion criteria. Calcium phosphate (CaP), bisphosphonates (BPs), and bone morphogenetic proteins (BMPs) were the most commonly used chemical compounds. There were two main methods for local drug delivery at the bone-implant interface: (1) directly from an implant surface by coating or immobilizing techniques, and (2) the local application of drugs to the implant site, using carriers. There was a statistically significant increase in BIC for both local drug delivery methods (p= .02 and p < .0001, respectively) when compared to controls. There was a statistically significant increase in BIC when CaP (p= .0001) and BMPs (p= .02) were either coated into implants or delivered to the implant site, in comparison to when drugs were not used. The difference was not significant for the use of BPs (p= .15). In Study II, the DSC analyses showed that PNIPAAm-AAm containing 10% acrylamide had an appropriate phase transition temperature of 42◦C. SEM images showed that the surface coating consisted of a 200 nm thick uniform polymer layer. The QCM-D analysis coupled with in situ NIR irradiation demonstrated a dramatic shift in frequency that was attributed to mass being released from the surface upon irradiation. This mass release correlated well with the drug release profile as determined using UV/VIS spectroscopy with phenol as a model drug. For Study III, clarithromycin release from PLGA microspheres revealed sustained release for around 4 weeks with 50% release during the first week. Histologically, new bone formation was evident at 2 and 4 weeks of healing in all groups and bone formation increased as a function of healing time. At 12 weeks, Group 4 (β-TCP with PLGA microspheres loaded with clarithromycin) showed significantly higher amount of newly formed bone compared to Group 1 (sham). The µCT showed that Group 4 expressed significantly higher bone formation compared to Group 1 at all time points. In Study IV, the SEM images showed TiO2 films with porous structures covering the entire surface with pore sizes determined to be 6 nm for P123 and 7.2 nm for P123-PPG. The QCM-D analysis revealed an absorption of 3300 ng/cm2 of Sr-ranelate on the 7.2 nm TiO2 films, which was about 3 times more than the observed amount on the 6 nm TiO2 films (1200 ng/cm2). The histomorphometric analyses revealed higher percentages of bone implant contact (BIC) and bone area (BA) for implants with Sr-ranelate compared to implants in the control group after 2 and 6 weeks of healing. However, these differences were found not to be significant (BIC with a p-value of 0.43 after 2 weeks and 0.172 after 6 weeks), (BA with a p-value of 0.503 after 2 weeks, and 0.088 after 6 weeks). The mean BIC and BA values within the same group showed significant increase among all groups after comparing 2 and 6 weeks. Conclusions: Most studies assessing local drug/chemical compound release systems in implants evaluated the influence of the use of BPs, CaP, and BMPs on bone healing. The use of local chemical compound delivery systems around implants could significantly improve implant osseointegration in animal models (Study I). In addition, on demand-release of the antibiotic agent vancomycin from the coating induced by NIR light resulted in a clear inhibition zone around a coated substrate in a bacteria culture test, thereby providing proof of concept of the developed drug delivery system (Study II). The in vivo findings showed that β-TCP with clarithromycin-loaded microspheres can enhance bone formation in bone defects (Study III). Meanwhile, the in vivo findings on Sr-ranelate study (Study IV) could not confirm the positive effects of Sr-ranelate on implant incorporation in bone made by other authors.