1 Tuning dual-drug release from composite scaffolds for bone2 regeneration3 J.L. Paris a,bQ1 , J. Román a, M. Manzano a,b, M.V. Cabañas a,*, M. Vallet-Regí a,b,*4 aDepartamento de Química Inorgánica y Bioinorgánica, Facultad de Farmacia, UCM, Instituto de Investigación Sanitaria Hospital 12 de Octubre i+12,5 28040 Madrid, Spain6 bCentro de Investigación Biomédica en Red de Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Spain

A R T I C L E I N F O

Article history:Received 30 January 2015Received in revised form 20 March 2015Accepted 21 March 2015Available online xxx

Keywords:Dual-drug delivery systemDesigned porous scaffoldIbuprofenZoledronic acidCo-delivery releaseTissue engineering

A B S T R A C T

This work presents the tuning of drug-loaded scaffolds for bone regeneration as dual-drug deliverysystems. Two therapeutic substances, zoledronic acid (anti-osteoporotic drug) and ibuprofen (anti-inflammatory drug) were successfully incorporated in a controlled manner into three dimensionaldesigned porous scaffolds of apatite/agarose composite. A high-performance liquid chromatographymethod was optimized to separate and simultaneously quantify the two drugs released from the dual-drug codelivery system. The multifunctional porous scaffolds fabricated show a very rapid delivery ofanti-inflammatory (interesting to reduce inflammation after implantation), whereas the anti-osteoporotic drug showed sustained release behaviour (important to promote bone regeneration).Since ibuprofen release was faster than desired, this drug was encapsulated in chitosan spheres whichwere then incorporated into the scaffolds, obtaining a release profile suitable for clinical application. Theresults obtained open the possibility to simultaneously incorporate two or more drugs to an osseousimplant in a controlled way improving it for bone healing application.

ã 2015 Published by Elsevier B.V.

7 1. Introduction

8 In recent decades, social development has led to an aging9 population as well as a considerable increase in the number of

10 traffic accidents. Both factors, as a consequence of the increased life11 expectancy and the way of life in developed countries, have12 contributed significantly to an exponential increase in different13 bone diseases. Consequently, treatment of bone defects continues14 generating a considerable effort in the scientific community15 (Arcos et al., 2014; Drosse et al., 2008). In this sense, tissue16 engineering (TE) (Fisher and Mauck, 2013; Langer and Vacanti,17 1993) has played a key role as it has introduced new strategies18 based on the development of composite systems that integrate19 cells, growth factors, and scaffolds, (Drosse et al., 2008; Eisenbarth,20 2007; Vallet-Regí and Ruiz-Hernández, 2011) avoiding major21 drawbacks associated to autografts and allografts (O’Brien, 2011).22 TE develops substrates that restore, maintain or improve the

23function of damaged tissues (Langer, 2000). In case of damaged24bone, TE scaffolds with added drug delivery function, have25emerged as an attractive approach in recent years (Catauro26et al., 2015a; Mouriño et al., 2013). Q327In bone TE, biocompatible, biodegradable and interconnected28highly porous scaffolds in a three dimensional (3D) geometry are29desirable. In this sense, composite scaffolds which combine30biopolymers and bioactive ceramics are being developed for this31application (Catauro et al., 2015b; Garg and Goyal, 2014; Mallick32and Cox, 2013; Mouriño et al., 2013). Great attention has been33focused Q4on hydrogels as biopolymers because of its resemblance to34the extracellular fluids (Fisher et al., 2014; Malda et al., 2013). They35have been studied for different medical applications, such as drug36delivery systems and bionanotechnology (Ankareddi and Brazel,372007; Gaharwar et al., 2014; Peppas et al., 2006). In particular, the38natural polysaccharide agarose, which is a thermo sensitive39hydrogel, has been successfully used in tissue engineering and40other biological applications (Cheng et al., 2007; Gruber et al.,412006; Luo and Shoichet, 2004; Marras-Marquez et al., 2014; Rotter42et al., 1998). Inclusion of hydroxyapatite (the mineral phase43present in bone) on hydrogels improves both the mechanical44stability and bioactivity (Juhasz et al., 2010) and also induces an45osteoinductive and osteoconductive behaviour. In this sense,46carbonate-substituted hydroxyapatite shows better bioactivity

* CorrespondingQ2 authors at: Departamento de Química Inorgánica y Bioinorgá-nica, Facultad de Farmacia, UCM, Instituto de Investigación Sanitaria Hospital 12 deOctubre i+12, 28040 Madrid, Spain. Tel.: +34 913941843; fax: +34 913941786.

E-mail addresses: vcabanas@ucm.es (M.V. Cabañas), vallet@ucm.es(M. Vallet-Regí).

http://dx.doi.org/10.1016/j.ijpharm.2015.03.0480378-5173/ã 2015 Published by Elsevier B.V.

International Journal of Pharmaceutics xxx (2015) xxx–xxx

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsev ier .com/locate / i jpharm

47 in vitro and in vivo than stoichiometric hydroxyapatite48 (Porter et al., 2005). Moreover, nanoscale particulate apatites49 are being increasingly considered in composite scaffolds to closely50 mimic the nanosized features of natural bone.51 Following the implantation of a scaffold, the material is still not52 vascularized. Therefore, systemic administration of drugs would53 not be effective at this stage. Drug release from the scaffolds could54 provide an adequate therapeutic concentration of the drug in the55 target site. Moreover, the simultaneous release of two or more56 therapeutic substances with different pharmacological activity57 may improve the outcome in different pathologies (Aderibigbe58 et al., 2015; Lee et al., 2008).59 Most of the fabrication processes to obtain scaffolds with 3D60 designed porosity employ organic solvents, high temperature or61 acidic conditions that preclude the incorporation of bioactive62 molecules and drugs (Fierz et al., 2008; Martínez-Vázquez et al.,63 2013). In previous work we have described an easy and inexpensive64 method to obtain agarose/hydroxyapatite scaffolds with hierar-65 chical porosity at room temperature for bone repair application66 (Peña et al., 2010; Román et al., 2011). Biocompatibility studies67 demonstrate that these scaffolds allow the culture of osteoblasts68 inside and outside the material (Alcaide et al., 2010; Cabañas et al.,69 2014). Recently, these scaffolds were investigated as protein70 release matrices with different release kinetics depending on the71 strategy used to incorporate the protein into the scaffold (Cabañas72 et al., 2014).73 This work aims to improve the functionality of these74 biodegradable scaffolds, turning them into dual-drug delivery75 systems (DDDS) for bone regeneration. In this sense, we have76 designed a DDDS for bone regeneration with two drugs widely77 used in traumatology: zoledronic acid (a bisphosphonate anti-78 resorptive drug) and ibuprofen (a nonsteroidal anti-inflammatory79 drug). Bisphosphonates are used in standard clinical practice for80 the treatment of diseases associated with increased bone81 resorption, such as osteoporosis, and might enhance bone82 regeneration when included in scaffolds (Cattalini et al., 2012;83 Coleman et al., 2011; Giger et al., 2013; Nancollas et al., 2006;84 Rosenqvist et al., 2014). On the other hand, ibuprofen is commonly85 prescribed to relieve pain due to inflammation that arises from86 different osseous pathologies or surgical treatments (Mouriño and87 Boccaccini, 2010; Rainsford, 2013), and it is also commonly88 prescribed simultaneously with bisphosphonates in order to89 minimize one of its main side effects: the “flu-like syndrome”90 (Coleman et al., 2011). This DDDS should have a long-term release91 of zoledronic acid (to enhance bone regeneration) and a relatively92 fast delivery of ibuprofen (to reduce inflammation).

932. Materials and methods

942.1. Materials

95Nanocrystalline hydroxycarbonateapatite (nHCA) was prepared96by a precipitation method from Ca(NO3)2!4H2O, (NH4)2HPO4 and97(NH4)2CO3 aqueous solutions at pH 9.2 (NH4OH solution) and 37 "C98(Padilla et al., 2008).99Agarose polymer (for routine use) was purchased from100Sigma–Aldrich, Steinheim, Germany; as well as Chitosan (from101crab shells, min. 85% deacetylated) and Sodium Tripolyphosphate,102TPP (Tech., 85%). Zoledronic acid (ZOL) monohydrate and ibuprofen103(IBU) antinflamatory were kindly provided by Novartis Pharma-104ceuticals AG (commercial name Zometa) and by Normon105Laboratories S.A. (Madrid, Spain), respectively.

1062.2. Dual-drug delivery systems preparation and characterization

107Three dimensional interconnected porous agarose/nHCA scaf-108folds containing drugs were fabricated using a shaping process109patented by the authors: GELPOR3D (Peña et al., 2010; Vallet-Regí110et al., 2010). The scaffolds were loaded with two drugs in two111different steps (during the scaffold fabrication and after its112consolidation). Briefly, the fabrication method consists of the113following steps (Fig. 1): the agarose powder was suspended in114deionized water (3.5% w/v) and heated to 90 "C with continuous115stirring; once a translucent sol was achieved, temperature was116gradually decreased to 45 "C and then the ceramic powder (nHCA)117was added. In this step, one of the drugs (ZOL or IBU alternatively)118was added as powder to the suspension (drug 1), which was119incorporated through the whole volume of the material. The slurry120so obtained was poured into a designed mold (a cube constituted121by rigid filaments of stainless steel 1 mm in diameter arranged122parallel to each other). After five minutes at room temperature,123complete consolidation of the bodies was reached. After with-124drawal of the designed mold, the scaffolds, freshly prepared, were125easily cut and shaped (Fig. 1). Then, the pieces were freeze-dried126(Heto Drywinner, freezing temperature and sublimation pressure127of #86 "C and 0.05 mbar, respectively).128Once the scaffolds containing drug 1 were dried, the same129amount of another drug (drug 2, IBU or ZOL alternatively) was130incorporated into the scaffolds by injection of an aqueous solution,131exploiting its hydrogel-like behaviour (Fig. 1). Afterwards, the132systems were freeze-dried again.133The composition employed to prepare the DDDS (50% ceramic134and 50% hydrogel) was chosen according to previous results

Fig. 1. Fabrication method of dual-drug delivery systems.

2 J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

135 (Román et al., 2011): compositions with low contents of agarose136 result in an insufficient ceramic particle-polysaccharide binder137 interaction that may cause the particles migration once implanted.138 On the other side, slurries containing high ceramic loads may139 result too viscous to be poured in the mould and fill the interstices140 between the rigid filaments.141 Thus, two types of systems containing 45% nHCA, 45% agarose142 and 10% drugs (5% IBU + 5% ZOL) (data are expressed in weight %)143 were fabricated: AH-zol-ibu sample (where zoledronic acid was144 incorporated during the scaffold preparation, and the ibuprofen145 was added after scaffold consolidation) and AH-ibu-zol sample (in146 which ibuprofen was added before consolidation and the147 zoledronic acid was incorporated after) (Fig. 1).148 The fabricated DDDS were characterised by X-ray diffraction149 (XRD) with a Philips X-Pert MPD diffractometer, scanning electron150 microscopy (SEM) in a JEOL 6400 and Hg intrusion porosimetry151 using a Micromeritics AutoPore III 9410 porosimeter. Fourier152 Transform Infrared (FTIR) spectra were obtained in a Nicolet Nexus153 spectrometer equipped with a Smart Golden Gate Attenuated Total154 Reflectance (ATR) accessory.

155 2.3. Chitosan spheres preparation and characterization

156 Ibuprofen-loaded chitosan spheres were prepared by the157 following procedure(Agnihotri et al., 2004): ibuprofen (150 mg)158 was dispersed in 10 mL of 1.5% chitosan solution in 0.5% acetic acid159 in deionized water. This suspension was added drop wise to a 10%160 Sodium Tripolyphosphate (TPP) solution in deionized water under161 gentle stirring. Chitosan spheres formed almost immediately and162 were cured for 20 min. Then they were filtered and dried in an oven163 at 37 !C overnight (sample CT-ibu).164 The spheres containing IBU were then introduced in the165 composite scaffolds following the same procedure previously166 described to introduce drug 1 (sample AH-CT-ibu).167 The systems were characterised by scanning electron micros-168 copy (SEM) in a JEOL 6400, Fourier Transform Infrared (FTIR)169 spectra were obtained in a Nicolet Nexus spectrometer equipped170 with a Smart Golden Gate ATR accessory.

171 2.4. In vitro drugs release study

172 In vitro delivery assays were performed by soaking scaffolds of173 1 cm side in 10 mL saline solution (NaCl 0.9%) at 37 !C. The pH of the174 release medium was adjusted to 7.4 (physiological pH) with Tris-175 Buffer. To avoid limitation of the delivering rate by external176 diffusion constrains, continuous stirring was maintained during177 the delivery assays. The volume of release medium employed was178 chosen to ensure sink conditions: The concentration of the drugs in179 the medium was never higher than 20% of their solubility in that180 medium. Aliquots of 150 mL were extracted at different times.181 The concentration of released drugs from the loaded scaffolds182 was monitored by reverse-phase high-performance liquid chro-183 matography (RP-HPLC) in a Waters Alliance automatic analysis184 system with a Model #2695 separations module coupled to a185 Model #2996 photo-diode array detector. Chromatographic186 separations were carried out with a 250 " 4.6 mm prepacked187 analytical column Mediterranea Sea18 (Teknokroma Inc.) contain-188 ing 5 mm C-18 functionalized silica beads.189 In order to separate and quantify the drugs introduced into the190 scaffolds, specific analytical conditions were developed and the191 following method was employed. The gradient mobile phase was a192 mixture of aqueous phosphate buffer 10 mM at pH 3 and methanol193 (Vallet-Regí et al.,1998; Vila et al., 2013). The best resolution of two194 drugs was achieved with the following mobile composition: the195 initial mixture composition was 90:10 (v/v) for the first 5 min, and196 then was changed to 15:85 (v/v) over a period of 10 min, keeping

197the flow with the mobile phase for other 25 min. The flow rate of198the mobile phase was 0.7 mL min#1 and the injection volume was19910 mL. The column temperature was maintained at 37 !C.200The detection of the released drug was performed by UV at201different wavelengths: 210 nm for ZOL and 219 nm for IBU.202Chromatograms were recorded using Millenium Software. Under203these conditions, the retention time of the ZOL was 4 $ 0.5 min,204whereas the IBU retention time was 25 $ 0.6 min (Fig. 2). Several205solutions at concentrations ranging from 0.003 to 0.5 mg mL#1

206(ZOL) or 0.003 to 0.125 mg mL#1 (IBU) in aqueous NaCl 0.9%207buffered at pH 7.4 were used as standards. In addition, the208chromatographic peaks were stable appearing at the same209retention time during the analysis and no degradation peaks were210present in the chromatograms. Results were processed with the211software OriginPro8 (OriginLab Northampton, MA)

2123. Results and discussion

213Multifunctional 3D scaffolds based on nano-hydroxycarbona-214teapatite/agarose composites have been fabricated at room215temperature through a shaping method Q5named GELPOR3D216(Peña et al., 2010; Román et al., 2011; Vallet-Regí et al., 2010).217These scaffolds were loaded with two drugs, (ZOL and IBU), in two218different steps. In these systems, the drug included during scaffold219fabrication, drug 1, is mentioned first and the drug included after220scaffold consolidation, drug 2, is stated last, giving two different221materials: AH-zol-ibu and AH-ibu-zol.222The method used to obtain these DDDS allows a complete223control over the amount of drugs present in the scaffolds. This224constitutes a significant advantage versus other designed porosity225systems that depend on immersing the scaffolds in highly226concentrated solutions of the drugs to load them into the material227(a very inefficient process).228The addition of the drugs (10 wt.%) during the scaffold229fabrication does not affect the body consolidation process neither230the textural features of the scaffolds as demonstrated by SEM and231Hg porosimetry studies. These DDDS show a macroporous232structure constituted by a designed 3D network of interconnected233pores of around 900 mm, and a connected porosity due to freeze-234drying with diameters between 100 and 200 mm (Fig. 3). Scaffolds235fabricated with similar composition and drying method to those236used in this work (Román et al., 2011) showed a porosity around23795%, with 30% for the pores with the highest diameter, around 65%

Fig. 2. Typical HPLC chromatogram of ZOL and IBU standard mixture at 219 nm.

J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx 3

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

238 for the 100–200 mm pores and less than 3% for pores around239 50–100 nm (due to the structure resulting from the coating of240 nHCA particles by the agarose).241 The presence of loaded drugs into the scaffolds was confirmed242 by FTIR spectra. Fig. 4 shows FTIR spectrum corresponding to AH-243 ibu-zol scaffold. (The FTIR spectrum corresponding to AH-zol-ibu244 was very similar and it is not included in the figure for the sake of245 clarity). The spectra corresponding to IBU, ZOL and the scaffold246 without any loaded drug (AH) or containing just zoledronic acid247 (AH-zol) are also included for comparative purposes.248 FTIR spectrum of AH-ibu-zol shows the presence of character-249 istic bands associated to functional groups corresponding to nHCA250 and agarose (see FTIR spectrum of sample AH) and additional251 bands due to the drugs were also detected. The most intense band252 of ibuprofen, at 1720 cm!1, that corresponds to carboxylic acid253 group, -COOH, (Szegedi et al., 2012) is not observed. However, a254 new band at 1545 cm!1, corresponding to carboxylate ion COO!,255 can be appreciated in the co-loaded scaffold. This indicates the256 deprotonation of the acid group of IBU during the synthesis257 process, due to the basic environment provided by the nHCA, as it258 has been observed in other basic materials (Krupa et al., 2010). This259 deprotonation was not observed in scaffolds prepared without260 nHCA (Supporting Information). On the other hand, the bands261 corresponding to the aromatic ring chains of zoledronic acid262 (Pascaud et al., 2012), ca. 1548 and 1578 cm!1, can be observed in263 the fabricated scaffolds with ZOL, indicating that aromatic rings264 remain unmodified. Moreover, the vibration bands of ZOL at265 1300 and 1321 cm!1 attributed to P—O bond (Waghe et al., 2006)266 are shifted to 1286 and 1307 cm!1, respectively, in the spectra267 corresponding to scaffolds containing ZOL (AH-zol and AH-ibu-zol268 scaffold). This displacement can be attributed to the interaction269 between the ZOL and the nHCA, as the bisphosphonates possess a270 pyrophosphate-like chemical structure that confers a strong271 affinity of calcium by a chelate formation (Bujoli et al., 2006;272 Juillard et al., 2010; Vargas-Becerril et al., 2013).273 Systems containing ZOL (forming a chelate with Ca from the274 ceramic component) and/or IBU (with its carboxylic group275 deprotonated) were immersed in aqueous solution to study the276 co-release of the drugs. To quantify zoledronic acid and ibuprofen,277 a RP-HPLC method was developed in order to determine the278 concentration of both drugs (since they are being released279 simultaneously, and their UV–vis spectra overlap). Firstly, the280 specific analytical conditions were studied to separate both drugs

281by optimizing the composition of the mobile phase (Fig. 2):282methanol and aqueous phosphate buffer in various ratios. After283separation they were quantified by UV at different wavelengths:284210 nm for ZOL and 219 nm for IBU. An important concern285regarding drug release from scaffolds is to ensure the drug286stability during its incorporation and release. Drug stability in this287study was assured by the unmodified UV/vis spectra of both drugs288throughout the analysis time. Again, this proves the versatility of289the scaffold preparation procedure, based on the gelation of290agarose, carried out at mild conditions, allowing the incorporation291of active molecules during the process without modifying them.292Similar systems rely on the setting reaction of cements, an293exothermic process that can actually reduce the activity of the294incorporated drugs (Vorndran et al., 2010).295Fig. 5 shows the release profiles of IBU and ZOL as a function of296the incorporation method of the drug at pH 7.4. Data depicted in297this figure show that ibuprofen has a short-term release while298zoledronic acid has a long term and sustained release behaviour.299Moreover, independent release behaviours of the drugs were300observed, since the presence of a drug does not influence the301release of the other (single drug systems have also been studied as302reference, data not shown).

Fig. 3. AH-zol-ibu scaffold: photograph (a); SEM micrographs at differentmagnifications (b,c); schematic representation of the dual matrix (d).

Fig. 4. FTIR spectra corresponding to (from top to bottom): AH scaffold; AH-ibu-zolscaffold; AH-zol scaffold; zoledronic acid and ibuprofen.

4 J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

303 In a physiological pH saline solution, IBU is practically released304 in 3 h, for both types of incorporation, whereas for this time, just305 around a 3% of ZOL is released when it is added during the scaffold306 fabrication (AH-zol-ibu) and 9% when it was incorporated after307 fabrication (AH-ibu-zol). Data collected in this figure show that, at308 72 h, only 20% (AH-ibu-zol) or 10% (AH-zol-ibu) of zoledronic acid309 has been released.310 Contrarily to what one would expect, both loading modes311 (during and after scaffold manufacture) resulted in rather similar312 in vitro release profile of ibuprofen: a very fast release of the drug,313 related to the rapid swelling of the scaffolds in aqueous solution314 observed in previous studies. (Cabañas et al., 2014) On the other315 hand, the retention of antiosteoporotic drug is related to the316 complexation between the phosphonate groups of zoledronic acid317 and the Ca ions of nHCA, previously mentioned. This bonding318 nHCA-ZOL has been found to yield protection against the observed319 cytotoxic effects of ZOL in vitro. (Murphy et al., 2014) The lower320 release observed for AH-zol-ibu indicates that the complexation321 reaction is facilitated when ZOL is introduced during the322 fabrication process.323 Even though these results match the initial objective of this324 work (a very slow release of zoledronic acid and a very fast release325 of ibuprofen), ibuprofen release is too fast for the clinical326 application we are aiming at. The drug should be released during327 a short period of time (a few days) in order to yield a relevant anti-328 inflammatory effect. However, ibuprofen release should not last329 much longer, since prolonged anti-inflammatory treatment might330 difficult the regeneration process and, therefore, deteriorate the331 clinical outcome (Thomas and Puleo, 2011).332 In order to extend ibuprofen release, and because of the333 versatility of the scaffold fabrication process, the drug was334 encapsulated in chitosan-TPP spheres (CT) (648.016 ! 22.03 mm335 in diameter) before being introduced in the scaffold. Chitosan is a336 biodegradable natural polysaccharide widely used in drug delivery337 systems (Agnihotri et al., 2004). Drug-loaded chitosan spheres,338 containing a 26 wt.% of IBU (CT-ibu) were introduced in the slurry339 before its addition to the 3D-mold. These CT-ibu spheres isolated,340 or included in the composite scaffolds (AH-CT-ibu) together with a341 schematic representation of the system are shown in Fig. 6. FTIR342 spectrum of CT-ibu spheres shows that IBU is present with its343 carboxylic group protonated (Supporting information).344 Ibuprofen release studies from spheres isolated (CT-ibu) and345 scaffolds with spheres within them (AH-CT-ibu) were carried out

346in the same conditions as those used previously. The results are347shown in Fig. 7. We can see a slower release from the sphere-348loaded scaffolds than from the isolated chitosan spheres (and, in349both cases, slower than non-encapsulated ibuprofen-loaded350scaffolds). In the case of the scaffold with encapsulated ibuprofen,351with minimised burst effect, its release lasts more than 2 days,352what would be suitable for bone healing.353The release profiles of Figs. 5 and 7 were fitted to different354kinetic models, in order to try to determine the mechanism driving355drug release (Costa and Sousa Lobo, 2001). Table 1 summarizes the356kinetic parameters resulting from the better fittings of zoledronic357acid and ibuprofen release.358For the release of zoledronic acid, the best fit was Q6obtained by359the power law equation known as Korsmeyer–Peppas equation360(Peppas, 2000) (Eq. (1)):

Mt

M" ktn (1)

361362where Mt/M is the accumulative amounts of drug release at time t;363k is a kinetic constant and n is an exponent which characterised the364mechanism of drug release and it ranges between n = 0.5

Fig. 5. Release profiles of ZOL (5) and IBU (*) from the DDDS: AH-zol-ibu (—) andAH-ibu-zol (- - - -) at pH 7.4.

Fig. 6. Photograph and SEM micrograph of CT-ibu spheres (a, b); SEM micrograph ofAH-CT-ibu scaffold with arrows pointing CT-ibu spheres (c); schematic represen-tation of the new dual matrix (d).

0,0 10,0 20,0 30,0 40,0 50,0

0

20

40

60

80

100

% D

rug

rele

ased

Time (h)

AH- ibu

CT-ibu

AH-CT-ibu

Fig. 7. Release profiles of IBU from non-encapsulated IBU-loaded scaffolds (AH-ibu), IBU-loaded chitosan spheres (CT-ibu) and composite scaffolds containingdrug-loaded spheres (AH-CT-ibu) at pH 7.4.

J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx 5

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

365 (for Fickian diffusion), and n = 1 (for Case II transport), respectively,366 being the inter-mediate values indicative of anomalous transport.367 In our case the value of the exponent n obtained was lower (ca.368 0.2). This low value may be attributed to a non-typical Fickian369 diffusion mechanism with some physico-chemical interference370 (Chen and Zhu 2012; Tamimi et al., 2008). These data agree with371 the nHCA and ZOL interaction described above, the formation of a372 chelate and detected by FTIR spectra. Comparing the kinetic fitting373 for ZOL release, a higher value of k when ZOL is introduced after the374 scaffold consolidation can be observed, what is in agreement with375 the higher amount of the drug released from this scaffold.376 On the other hand, the release of non-encapsulated ibuprofen377 from the scaffolds showed in Fig. 5 follows a first order exponential378 decay model (according to the Noyes–Whitney equation)379 (Peppas, 2000) (Eq. (2)):

Mt

M! A"1 # e#Kt$ (2)

380381 with A = (Mt/M)max being the maximum number of biomolecules382 released. The release rate constant, K, gives information about the383 solvent accessibility and the diffusion coefficient through the384 scaffold channels. In this case, no appreciable difference is385 observed in the kinetic fitting of IBU release introduced before386 or after scaffold consolidation (Table 1). To determine the387 mechanism underlying ibuprofen release, a scaffold AH-ibu was388 immersed in saline solution at pH 6.5, and the release profile of the389 drug was analyzed (Supporting information). IBU release at pH390 6.5 was slower than at pH 7.4, as can be observed in the release391 profile and in the kinetic fitting to Eq. (2) (A = 94 % 4%, K = 0.0044392 h#1%0.0005, R2 = 0.987). Given that solubility of IBU is lower at393 acidic pH, and that the dissolution rate is a function of the394 solubility of a molecule (Costa and Sousa Lobo, 2001), the change in395 the release profile of IBU between the two cases (at pH 7.4 and pH396 6.5) might indicate that the mechanism driving non-encapsulated397 ibuprofen release from this scaffolds is merely the dissolution of398 the drug.399 The slower ibuprofen release from chitosan spheres (inside a400 scaffold or isolated) can be described using, again, the Korsmeyer–401 Peppas equation (Table 1). In both cases, the n value is close to 0.5;402 this implies that the mechanism behind drug release is diffusion403 through the walls of the chitosan spheres. Even though in both404 cases, the same kinetic model can be used, the kinetic constant (k)405 is higher in the isolated spheres, what agrees with the faster406 release observed in that case, since in the spheres embedded inside407 a scaffold, ibuprofen has to get out of the scaffold once it has408 diffused out of the chitosan spheres.409 According to the drug release data, the proposed strategy with410 the dual pharmacological effect could be a versatile option for the411 treatment of bone injuries that would require bone replacement.412 The apatite/agarose scaffold itself could allow bone regeneration,413 since previous in vitro studies (Cabañas et al., 2014) showed that

414the material stimulated osteoblast proliferation and the cells415colonized the interconnected macroporous scaffolds. At the same416time, the drugs included in it would provide an environment417where bone regeneration would be enhanced (by the slow release418of ZOL) and the inflammation in the area would be reduced in the419short-term (by the fast release of IBU).

4204. Conclusions

421Multifunctional 3D designed porous scaffolds for bone regen-422eration were fabricated by a simple conformation method carried423out in mild conditions that allow introducing many different424biomolecules avoiding their degradation during scaffold manufac-425ture. Two biological active substances, zoledronic acid (an426antiresorptive drug) and ibuprofen (an antiinflammatory drug),427have been incorporated into the scaffolds and were co-delivered to428the medium following different tuneable release profiles. The429composition of the scaffolds allows the bisphosphonate to be430retained by the ceramic, nHCA, and the inclusion of ibuprofen-431loaded chitosan spheres enables the adjustment of its release432behaviour.433The development of a liquid chromatographic method allowed434the separation and simultaneous quantification of both drugs435which are released following different mechanisms, which can be436described employing several kinetic models.437The results obtained show the potential of these agarose/438nanohydroxycarbonateapatite scaffolds to retain and slowly439release zoledronic acid, as well as to quickly deliver ibruprofen440to the surrounding tissue. Although these DDDS have been studied441for simultaneous delivery of drugs with antiinflamatory and442antiosteoporotic properties, this approach allows the development443of multifunctional scaffolds to deliver multiple pharmaceutical or444biological agents for bone tissue regeneration.

445Acknowledgments

446Financial support from Q7Ministerio de Economía y447Competitividad, Spain (Project MAT2012-35556 and Project448CSO2010-11384-E, Ageing Network of Excellence) is gratefully449acknowledged. The XRD and SEM measurements were performed450at C.A.I Difracción de Rayos X and Microscopia Electrónica,451Universidad Complutense, respectively. We thank Novartis Phar-452maceuticals AG and Normon Laboratories S.A. (Madrid, Spain) for453providing the zoledronic acid and ibuprofen, respectively.

454Appendix A. Supplementary data

455Supplementary data associated with this article456can be found, in the online version, at http://dx.doi.org/10.1016/457j.ijpharm.2015.03.048

Table 1Kinetic parameters Q9of zoledronic acid (ZOL) and ibuprofen (IBU) released from different materials.

Drug Material Model Kinetic parameters Fit parameters

ZOL AH-ibu-zol Korsmeyer–Peppas k = 2.5 (h#n) % 0.2 R2 = 0.993Mt/M = ktn n = 0.24 % 0.01

AH-zol-ibu Korsmeyer–Peppas k = 1.03 (h#n) % 0.04 R2 = 0.997Mt/M = ktn n = 0.206 % 0.005

Non-encapsulated IBU AH-ibu-zol First order release A = 96.1 % 1 R2 = 0.991Mt/M = A(1 # e#Kt) K = 0.056 h#1% 0.004

AH-zol-ibu First order release A = 97.23 % 2 R2 = 0.992Mt/M = A(1 # e#Kt) K = 0.051 h#1% 0.004

Encapsulated-IBU CT-ibu Korsmeyer–Peppas k = 9.69 (h#n) % 2.85 R2 = 0.987Mt/M = ktn n = 0.43 % 0.06

AH-CT-ibu Korsmeyer–Peppas k = 1.28 (h#n) % 0.215 R2 = 0.992Mt/M = ktn n = 0.525 % 0.023

6 J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

458 References

459 Aderibigbe, B., Sadiku, E., Jayaramudu, J., Sinha Ray, S., 2015. Controlled dual release460 study of curcumin and a 4-aminoquinoline analog from gum acacia containing461 hydrogels. J. Appl. Polym. Sci. 132, n/a–n/a. doi:http://dx.doi.org/10.1002/462 app.41613.463 Agnihotri, S.A., Mallikarjuna, N.N., Aminabhavi, T.M., 2004. Recent advances on464 chitosan-based micro- and nanoparticles in drug delivery. J. Control. Release465 100, 5–28. doi:http://dx.doi.org/10.1016/j.jconrel.2004.08.010.466 Alcaide, M., Serrano, M.-C., Roman, J., Cabañas, M.-V., Peña, J., Sánchez-Zapardiel, E.,467 Vallet-Regí, M., Portolés, M.-T., 2010. Suppression of anoikis by collagen coating468 of interconnected macroporous nanometric carbonated hydroxyapatite/469 agarose scaffolds. J. Biomed. Mater. Res. A 95, 793–800. doi:http://dx.doi.org/470 10.1002/jbm.a.32901.471 Ankareddi, I., Brazel, C.S., 2007. Synthesis and characterization of grafted472 thermosensitive hydrogels for heating activated controlled release. Int. J.473 Pharm. 336, 241–247. doi:http://dx.doi.org/10.1016/j.ijpharm.2006.11.065.474 Arcos, D., Boccaccini, A.R., Bohner, M., Díez-Pérez, A., Epple, M., Gómez-Barrena, E.,475 Herrera, A., Planell, J.A., Rodríguez-Mañas, L., Vallet-Regí, M., 2014. The476 relevance of biomaterials to the prevention and treatment of osteoporosis. Acta477 Biomater. 10, 1793–1805. doi:http://dx.doi.org/10.1016/j.actbio.2014.01.004.478 Bujoli, B., Roussière, H., Montavon, G., Laïb, S., Janvier, P., Alonso, B., Fayon, F., Petit,

M., Massiot, D., Bouler, J.-M., Guicheux, J., Gauthier, O., Lane, S.M., Nonglaton, G.,479 Pipelier, M., Léger, J., Talham, D.R., Tellier, C., 2006. Novel phosphate–480 phosphonate hybrid nanomaterials applied to biology. Prog. Solid State Chem.481 34, 257–266. doi:http://dx.doi.org/10.1016/j.progsolidstchem.2005.11.039.482 Cabañas, M.V., Peña, J., Román, J., Ramírez-Santillán, C., Matesanz, M.C., Feito, M.J.,483 Portolés, M.T., Vallet-Regí, M., 2014. Design of tunable protein-releasing484 nanoapatite/hydrogel scaffolds for hard tissue engineering. Mater. Chem. Phys.485 144, 409–417. doi:http://dx.doi.org/10.1016/j.matchemphys.2014.01.011.486 Catauro, M., Bollino, F., Papale, F., Gallicchio, M., Pacifico, S., 2015a. Influence of the487 polymer amount on bioactivity and biocompatibility of SiO2/PEG hybrid488 materials synthesized by sol–gel technique. Mater. Sci. Eng. C 48, 548–555. doi:489 http://dx.doi.org/10.1016/j.msec.2014.12.035.490 Catauro, M., Bollino, F., Papale, F., Pacifico, S., 2015b. Modulation of indomethacin491 release from ZrO2/PCL hybrid multilayers synthesized via sol–gel dip coating. J.492 Drug Deliv. Sci. Technol. 26, 10–16. doi:http://dx.doi.org/10.1016/j.493 jddst.2014.12.004.494 Cattalini, J.P., Boccaccini, A.R., Lucangioli, S., Mouriño, V., 2012. Bisphosphonate-495 based strategies for bone tissue engineering and orthopedic implants. Tissue496 Eng. Part B Rev. 18, 323–340. doi:http://dx.doi.org/10.1089/ten.TEB.;1;497 2011.0737.498 Chen, F., Zhu, Y., 2012. Chitosan enclosed mesoporous silica nanoparticles as drug499 nano-carriers: sensitive response to the narrow pH range. Microporous500 Mesoporous Mater. 150, 83–89. doi:http://dx.doi.org/10.1016/j.501 micromeso.2011.07.023.502 Cheng, S.-Y., Heilman, S., Wasserman, M., Archer, S., Shuler, M.L., Wu, M., 2007. A503 hydrogel-based microfluidic device for the studies of directed cell migration.504 Lab Chip 7, 763–769. doi:http://dx.doi.org/10.1039/b618463d.505 Coleman, R., Burkinshaw, R., Winter, M., Neville-Webbe, H., Lester, J., Woodward, E.,506 Brown, J., 2011. Zoledronic acid. Expert Opin. Drug Saf. 10, 133–145. doi:http://507 dx.doi.org/10.1517/14740338.2011.540387.508 Costa, P., Sousa Lobo, J.M., 2001. Modeling and comparison of dissolution profiles.509 Eur. J. Pharm. Sci. 13, 123–133. doi:http://dx.doi.org/10.1016/S0928-0987(01)510 95-1.511 Drosse, I., Volkmer, E., Capanna, R., De Biase, P., Mutschler, W., Schieker, M., 2008.512 Tissue engineering for bone defect healing: an update on a multi-component513 approach. Injury 39 (Suppl. 2) doi:http://dx.doi.org/10.1016/S0020-1383(08)514 70011-1 S9–20.515 Eisenbarth, E., 2007. Biomaterials for tissue engineering. Adv. Eng. 9, 1051–1060.516 doi:http://dx.doi.org/10.1002/adem.200700287.517 Fierz, F.C., Beckmann, F., Huser, M., Irsen, S.H., Leukers, B., Witte, F., Degistirici, O.,518 Andronache, A., Thie, M., Müller, B., 2008. The morphology of anisotropic 3D-519 printed hydroxyapatite scaffolds. Biomaterials 29, 3799–3806. doi:http://dx.520 doi.org/10.1016/j.biomaterials.2008.06.012.521 Fisher, M.B., Mauck, R.L., 2013. Tissue engineering and regenerative medicine:522 recent innovations and the transition to translation. Tissue Eng. Part B Rev.19,1–523 13. doi:http://dx.doi.org/10.1089/ten.TEB.2012.0723.524 Fisher, S., a, R.Y., Tam, Shoichet, M.S., 2014. Tissue mimetics: engineered hydrogel525 matrices provide biomimetic environments for cell growth. Tissue Eng. Part A526 20, 895–898. doi:http://dx.doi.org/10.1089/ten.tea.2013.0765.527 Gaharwar, A.K., Peppas, N.A., Khademhosseini, A., 2014. Nanocomposite hydrogels528 for biomedical applications. Biotechnol. Bioeng. 111, 441–453. doi:http://dx.doi.529 org/10.1002/bit.25160.530 Garg, T., Goyal, A.K., 2014. Biomaterial-based scaffolds–current status and future531 directions. Expert Opin. Drug 11, 767–789. doi:http://dx.doi.org/10.1517/532 17425247.2014.891014.533 Giger, E.V., Castagner, B., Leroux, J.-C., 2013. Biomedical applications of534 bisphosphonates. J. Control. Release 167, 175–188. doi:http://dx.doi.org/535 10.1016/j.jconrel.2013.01.032.536 Gruber, H.E., Hoelscher, G.L., Leslie, K., Ingram, J.A., Hanley, E.N., 2006. Three-537 dimensional culture of human disc cells within agarose or a collagen sponge:538 assessment of proteoglycan production. Biomaterials 27, 371–376. doi:http://539 dx.doi.org/10.1016/j.biomaterials.2005.06.032.

540Juhasz, J.A., Best, S.M., Bonfield, W., 2010. Preparation of novel bioactive nano-541calcium phosphate–hydrogel composites. Sci. Technol. Adv. Mater. 11, 014103.542doi:http://dx.doi.org/10.1088/1468-6996/11/1/014103.543Juillard, A., Falgayrac, G., Cortet, B., Vieillard, M.-H., Azaroual, N., Hornez, J.-C., Penel,

G., 2010. Molecular interactions between zoledronic acid and bone: an in vitro544Raman microspectroscopic study. Bone 47, 895–904. doi:http://dx.doi.org/54510.1016/j.bone.2010.07.018.546Krupa, A., Majda, D., Jachowicz, R., Mozgawa, W., 2010. Solid-state interaction of547ibuprofen and Neusilin US2. Thermochim. Acta 509, 12–17. doi:http://dx.doi.548org/10.1016/j.tca.2010.05.009.549Langer, R., 2000. Biomaterials in drug delivery and tissue engineering: one550laboratory’s experience. Acc. Chem. Res. 33, 94–101. doi:http://dx.doi.org/55110.1021/ar9800993.552Langer, R., Vacanti, J.P., 1993. Tissue engineering. Science 260, 920–926.553Lee, J.S., Bae, J.W., Joung, Y.K., Lee, S.J., Han, D.K., Park, K.D., 2008. Controlled dual554release of basic fibroblast growth factor and indomethacin from heparin-555conjugated polymeric micelle. Int. J. Pharm. 346, 57–63. doi:http://dx.doi.org/55610.1016/j.ijpharm.2007.06.025.557Luo, Y., Shoichet, M.S., 2004. A photolabile hydrogel for guided three-dimensional558cell growth and migration. Nat. Mater 3, 249–253. doi:http://dx.doi.org/55910.1038/nmat1092.560Malda, J., Visser, J., Melchels, F.P., Jüngst, T., Hennink, W.E., Dhert, W.J.A., Groll, J.,561Hutmacher, D.W., 2013. 25th anniversary article: engineering hydrogels for562biofabrication. Adv. Mater. 25, 5011–5028. doi:http://dx.doi.org/10.1002/563adma.201302042.564Mallick, K.K., Cox, S.C., 2013. Biomaterial scaffolds for tissue engineering. Front.565Biosci. 5, 341–360 (Elite Ed.).566Marras-Marquez, T., Peña, J., Veiga-Ochoa, M.D., 2014. Robust and versatile pectin-567based drug delivery systems. Int. J. Pharm. 479, 265–276. doi:http://dx.doi.org/56810.1016/j.ijpharm.2014.12.045.569Martínez-Vázquez, F.J., Miranda, P., Guiberteau, F., Pajares, A., 2013. Reinforcing570bioceramic scaffolds with in situ synthesized e-polycaprolactone coatings. J.571Biomed. Mater. Res. A 101, 3551–3559. doi:http://dx.doi.org/10.1002/jbm.572a.34657.573Mouriño, V., Boccaccini, A.R., 2010. Bone tissue engineering therapeutics: controlled574drug delivery in three-dimensional scaffolds. J. R. Soc. Interface 7, 209–227. doi:575http://dx.doi.org/10.1098/rsif.2009.0379.576Mouriño, V., Cattalini, J.P., Roether, J., a Dubey, P., Roy, I., Boccaccini, A.R., 2013.577Composite polymer-bioceramic scaffolds with drug delivery capability for bone578tissue engineering. Expert Opin. Drug 10, 1353–1365. doi:http://dx.doi.org/57910.1517/17425247.2013.808183.580Murphy, C.M., Schindeler, A., Gleeson, J.P., Yu, N.Y.C., Cantrill, L.C., Mikulec, K.,581Peacock, L., O’Brien, F.J., Little, D.G., 2014. A collagen-hydroxyapatite scaffold582allows for binding and co-delivery of recombinant bone morphogenetic583proteins and bisphosphonates. Acta Biomater. 10, 2250–2258. doi:http://dx.doi.584org/10.1016/j.actbio.2014.01.016.585Nancollas, G.H., Tang, R., Phipps, R.J., Henneman, Z., Gulde, S., Wu, W., Mangood, A.,586Russell, R.G.G., Ebetino, F.H., 2006. Novel insights into actions of587bisphosphonates on bone: differences in interactions with hydroxyapatite.588Bone 38, 617–627. doi:http://dx.doi.org/10.1016/j.bone.2005.05.003.589O’Brien, F.J., 2011. Biomaterials & scaffolds for tissue engineering. Mater. Today 14,59088–95. doi:http://dx.doi.org/10.1016/S1369-7021(11)70058-X.591Padilla, S., Izquierdo-Barba, I., Vallet-Regí, M., 2008. High specific surface area in592nanometric carbonated hydroxyapatite. Chem. Mater. 20, 5942–5944. doi:593http://dx.doi.org/10.1021/cm801626k.594Pascaud, P., Bareille, R., Bourget, C., Amédée, J., Rey, C., Sarda, S., 2012. Interaction595between a bisphosphonate, tiludronate and nanocrystalline apatite: in vitro596viability and proliferation of HOP and HBMSC cells. Biomed. Mater. 7, 054108.597doi:http://dx.doi.org/10.1088/1748-6041/7/5/054108.598Peña, J., Román, J., Victoria Cabañas, M., Vallet-Regí, M., 2010. An alternative599technique to shape scaffolds with hierarchical porosity at physiological600temperature. Acta Biomater. 6, 1288–1296. doi:http://dx.doi.org/10.1016/j.601actbio.2009.10.049.602Peppas, N., 2000. Hydrogels in pharmaceutical formulations. Eur. J. Pharm.603Biopharm. 50, 27–46. doi:http://dx.doi.org/10.1016/S0939-6411(00)90-4.604Peppas, N.A., Hilt, J.Z., Khademhosseini, A., Langer, R., 2006. Hydrogels in biology605and medicine: from molecular principles to bionanotechnology. Adv. Mater. 18,6061345–1360. doi:http://dx.doi.org/10.1002/adma.200501612.607Porter, A., Patel, N., Brooks, R., Best, S., Rushton, N., Bonfield, W., 2005. Effect of608carbonate substitution on the ultrastructural characteristics of hydroxyapatite609implants. J. Mater. Sci. Mater. Med. 16, 899–907. doi:http://dx.doi.org/10.1007/610s10856-005-4424-1.611Rainsford, K.D., 2013. Fifty years of ibuprofen: advancing pain and fever612management. Int. J. Clin. Pract.(Suppl. 1–2) doi:http://dx.doi.org/10.1111/613ijcp.12050.614Román, J., Cabañas, M.V., Peña, J., Vallet-Regí, M., 2011. Control of the pore615architecture in three-dimensional hydroxyapatite-reinforced hydrogel616scaffolds. Sci. Technol. Adv. Mater. 12, 045003. doi:http://dx.doi.org/10.1088/6171468-6996/12/4/045003.618Rosenqvist, K., Airaksinen, S., Vehkamäki, M., Juppo, A.M., 2014. Evaluating optimal619combination of clodronate and bioactive glass for dental application. Int. J.620Pharm. 468, 112–120. doi:http://dx.doi.org/10.1016/j.ijpharm.2014.04.017.621N., Rotter, J., Aigner A., Naumann, H., Planck C., Hammer G., Burmester M., Sittinger,6221998. Cartilage reconstruction in head and neck surgery: comparison Q8of623resorbable polymer scaffolds for tissue engineering of human septal cartilage. J.

J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx 7

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048

624 Biomed. Mater. Res. 42, 347–56. 10.1002/(SICI) 1097-4636(19981205)625 42:3<347::AID-JBM2>3.0.CO;2-J.626 Szegedi, A., Popova, M., Goshev, I., Klébert, S., Mihály, J., 2012. Controlled drug627 release on amine functionalized spherical MCM-41. J. Solid State Chem. 194,628 257–263. doi:http://dx.doi.org/10.1016/j.jssc.2012.05.030.629 Tamimi, F., Torres, J., Bettini, R., Ruggera, F., Rueda, C., López-Ponce, M., Lopez-630 Cabarcos, E., 2008. Doxycycline sustained release from brushite cements for the631 treatment of periodontal diseases. J. Biomed. Mater. Res. A 85, 707–714. doi:632 http://dx.doi.org/10.1002/jbm.a.31610.633 Thomas, M.V., Puleo, D.A., 2011. Infection, inflammation, and bone regeneration: a634 paradoxical relationship. J. Dent. Res. 90, 1052–1061. doi:http://dx.doi.org/635 10.1177/0022034510393967.636 Vallet-Regí M., Peña J., Román J., Cabañas M.V., 2010. Method for the low-637 temperature preparation of bioceramic parts with patterned and638 interconnected three-dimensional porosity. Spain Patent 2010, WO 2010/639 037881 A1.640 Vallet-Regí, M., Granado, S., Arcos, D., Gordo, M., Cabañas, M., Ragel, M.V., Salinas, C.

V., Doadrio, A.J., San Román, J., 1998. Preparation, characterization, and in vitro641 release of ibuprofen from AI2O3/PLA/PMMA composites. J. Biomed. Mater. Res.642 39, 423–428.

643Vallet-Regí, M., Ruiz-Hernández, E., 2011. Bioceramics: from bone regeneration to644cancer nanomedicine. Adv. Mater. 23, 5177–5218. doi:http://dx.doi.org/10.1002/645adma.201101586.646Vargas-Becerril, N., Patiño-Carachure, C., Rodriguez-Lorenzo, L.M., Téllez-Jurado, L.,

2013. Synthesis of hybrid compounds apatite–alendronate by reactive milling647and effects on the structure and morphology of the apatite phase. Ceram. Int. 39,6483921–3929. doi:http://dx.doi.org/10.1016/j.ceramint.2012.10.239.649Vila, M., Cicuéndez, M., Sánchez-Marcos, J., Fal-Miyar, V., Manzano, M., Prieto, C.,650Vallet-Regi, M., 2013. Electrical stimuli to increase cell proliferation on carbon651nanotubes/mesoporous silica composites for drug delivery. J. Biomed. Mater.652Res. A 101, 213–221. doi:http://dx.doi.org/10.1002/jbm.a.34325.653Vorndran, E., Klammert, U., Ewald, A., Barralet, J.E., Gbureck, U., 2010. Simultaneous654immobilization of bioactives during 3D powder printing of bioceramic drug-655release matrices. Adv. Funct. Mater. 20, 1585–1591. doi:http://dx.doi.org/65610.1002/adfm.200901759.657Waghe, A., Kanan, S.M., Abu-Yousef, I., Jensen, B., Tripp, C.P., 2006. Infrared study of658UV-irradiated tungsten trioxide powders containing adsorbed dimethyl methyl659phosphonate and trimethyl phosphate. Res. Chem. Intermed. 32, 613–623. doi:660http://dx.doi.org/10.1163/156856706778400280.

8 J.L. Paris et al. / International Journal of Pharmaceutics xxx (2015) xxx–xxx

G Model

IJP 14757 1–8

Please cite this article in press as: Paris, J.L., et al., Tuning dual-drug release from composite scaffolds for bone regeneration. Int J Pharmaceut(2015), http://dx.doi.org/10.1016/j.ijpharm.2015.03.048