Zhang 2018

Pharm Res (2019) 36:9 https://doi.org/10.1007/s11095-018-2538-7

RESEARCH PAPER

Physicochemical Characterization and Pharmacokinetics of Agomelatine-Loaded PLGA Microspheres for Intramuscular Injection Hongjuan Zhang 1 & Chenguang Pu 1 & Qiao Wang 1 & Xinyi Tan 1 & Jingxin Gou 1 & Haibing He 1 & Yu Zhang 1 & Tian Yin 1 & Yanjiao Wang 1 & Xing Tang 1 Received: 27 March 2018 / Accepted: 29 October 2018 # Springer Science+Business Media, LLC, part of Springer Nature 2018

ABSTRACT Purpose The aim of this study was to design agomelatine loaded long acting injectable microspheres, with an eventual goal of reducing the frequency of administration and improving patient compliance in treatment of depression. Methods AGM-loaded microspheres were prepared by an O/W emulsion solvent evaporation method. The physicochemical properties and in vitro performance of the microspheres were characterized. The pharmacokinetics of different formulations with various particle sizes and drug loadings were evaluated. Results AGM-loaded microspheres with drug loading of 23.7% and particle size of 60.2 μm were obtained. The in vitro release profiles showed a small initial burst release (7.36%) followed by a fast release, a period of lag time and a second accelerated release. Pore formation and pore closure were observed in vitro, indicating that the release of drug from microspheres is dominated by water-filled pores. Pharmacokinetic studies showed that AGM microspheres could release up to 30 days in vivo at a steady plasma concentration. As well, particle size and drug loading could significantly influence the in vivo release of AGM microspheres. Conclusions AGM-loaded microspheres are a promising carrier for the treatment of major depressant disorder.

KEY WORDS agomelatine . PLGA microspheres . pore formation and pore closure . release in vitro . release in vivo

* Yanjiao Wang [email protected]

1

School of Pharmacy, Shenyang Pharmaceutical University, 103 Wenhua Road, Shenyang 110016, China

ABBREVIATIONS AGM AUC Cmax DALYs DAS DSC EMA FDA MDD PLGA PXRD SD SDS SEM T1/2 Tmax UPLC-MS/MS YLDs

Agomelatine Area under the curve Maximum concentration Disability adjusted life years Drug and statistics software Differential scanning calorimetry European Medicines Agency Food and drug administration Major depressive disorder Poly (D,L-lactide- co-glycolide) Powder X-ray diffraction Standard deviation Sodium dodecyl sulfate Scanning electron microscopy Plasma half-life Peak time Ultra-performance liquid chromatography-tandem mass spectrometry Years lived with disability

INTRODUCTION Depression is a common heterogeneous disorder characterized clinically by low spirit, mental retardation, and exercise suppression. It has been estimated that depression is expected to be the second highest cause of morbidity by 2020 (1). As well, patients with major depressive disorder (MDD) in the risk of death from suicide, accident, coronary heart disease, stroke and respiratory disease is increasing (2–4). The estimated global calculated years lived with disability (YLDs) and disability adjusted life years (DALYs) of MDD are 8.2% (5.9-10.8%) and 2.5% (1.93.2%) respectively (5). Furthermore, not only do patients with MDD suffer from increased serious physical illness, but they are also affected by decreased social function.

9

Page 2 of 11

MDD has a high rate of relapse and high levels of dysfunction, and so has serious consequences for both the depressed individuals and their families (6,7). At present, the clinical treatment of depression is mainly drug based treatments, and the efficiency of the antidepressant drugs is only fully achieved when taken regularly and for a long period. It is recommended that the treatment cycle for depression is at least one year to minimize the risk of recurrence (8,9). However, most studies have shown that more than 40% of patients discontinue the treatment within 1 month, and approximately 56% of patients discontinue the treatment within 4 months (9). It is known that tolerability and psychological problems weaken the capacity of antidepressants, and that a steady and sustained plasma concentration is critical for the treatment of depression (10,11). Hence, in order to achieve good clinical efficiency, it is of importance to develop a long-acting injection of antidepressants. Agomelatine (AGM, Fig. 1) is a novel small molecule antidepressant with melatonergic MT1/MT2 receptor agonist and 5-HT2C receptor antagonist activity. It was approved by the EMA in 2009, and was thought to be able to improve the sleep of people with MDD by synchronizing the circadian rhythms. It also showed improved tolerability and a low propensity for drug withdrawl, sexual dysfunction and weight gain compared to other antidepressants (3,6). According to its chemical structure, agomelatine is a water-insoluble small molecule drug. Its solubility increase with temperature and the effective concentration is very low due to its new mechanism of action (12,13). At present, agomelatine is only commercially available in oral tablets. However, various shortcomings severely restrict clinical use of agomelatine tablets. First, due to its short half-life (1–2 h) and severe first-pass effect in the liver, orally administered agomelatine often shows a low absolute bioavailability (5% of oral dose), severe hepatic metabolism and further elevated aminotransferases

Fig. 1 Chemical structure of agomelatine.

Pharm Res

(2019) 36:9

(2,7). Clinically, the use of agomelatine tablets requires liver function testing, which severely limits the use of the drug (9). Also, daily medication is particularly inconvenient for people with MDD, as they often refuse to take medicine because of psychological problems. Therefore, it is ideal to develop a depot-like system which can retain the therapeutic concentration of agomelatine in plasma for a long time by releasing the drug in a controlled manner for patients with MDD to ensure stable recovery and to prevent relapse. Thus, promising polymeric microspheres can be applied to many hydrophobic drugs. With the development of drug delivery systems, depotlike sustained-release drug delivery systems including injectable drug suspensions, oil-based injectable solutions, polymer-based in-situ formings and polymeric microspheres that can release drugs longer than one week have attracted much attention in the pharmaceutical field (14). These systems offer numerous advantages over traditional drug delivery systems, including predictable drug release period, improved drug stability, reduced dosing frequency, increased patient compliance, decreased side effects and an overall cost reduction of medical care (14). Among various sustained-released drug delivery systems, polymeric microspheres, particularly PLGA–based microspheres, are the most widely used and commercially successful due to their ability to encapsulate a variety of drugs, high bioavailability, biocompatibility and sustained release (15). PLGA, certified by FDA to be used for drug delivery systems, not only has good biodegradability and biocompatibility, but can also tailor the release rate by changing the molecular weight or the ratio of the two monomers. Today, several commercial products are based on PLGA microspheres, including Risperdal Consta® (risperidone, Janssen, Inc.), Vivitrol® (naltrexone, Alkermes, Inc.) and Leupron Depot® (leuprolide acetate, Abbot, Inc.) and more (14,16). Many different types of processes are commonly employed for the preparation of polymer microspheres, including emulsion-solvent extraction/evaporation, spray drying, interfacial polymerization and coacervation (16,17). Until now, the preparation process and many properties of PLGA microspheres including encapsulation efficiency, release and degradation rates have been studied in detail (18–21). However, many obstacles hinder the development of insoluble drug microsphere products. On one hand, traditional emulsion solvent evaporation techniques and spray drying techniques may result in pores in the microspheres during the solvent removal process. Too many drugs can then diffuse through the holes, leading to an unwanted burst release in the initial stage of release. On the other hand, drug loading limitation is a significant problem for insoluble drugs (22). Particularly for hydrophobic drugs with relatively big water solubility, it is difficult to achieve a good encapsulation. In addition,

Pharm Res

(2019) 36:9

potential toxicity caused by residual solvent inside the microspheres can be a concern (22). In this work, in order to improve the bioavailability, reduce the side effects caused by the first pass effect of the oral tablets and achieve sustained release of AGM for the treatment of MDD, AGM-loaded PLGA microspheres were first developed using the traditional emulsion solvent evaporation method. Preparation, characterization, in vitro and in vivo release properties of AGM-loaded PLGA microspheres were subsequently evaluated in detail.

MATERIALS AND METHODS Materials and Animals Poly (D, L-lactide-co-glycolide) (PLGA) (LA/GA: 75:25, Mw: 70 kDa, carboxylic acid end group) was purchased from Jinandaigang Biomaterials (Jinan, China). Poly vinyl alcohol (KURARAY POVAL 22–88, 87.0-89.0% hydrolyzed) was purchased from Kuraray Co., Ltd. (Osaka, Japan). Agomelatine (AGM) was provided by Benyuan Pharmacy Co., Ltd. (Benxi, China). Dichloromethane (DCM), acetonitrile and all the other reagents were of chromatographic grade. Male Sprague-Dawley rats, with the average weight of 180–220 g, were obtained from the Experiment Animal Center of Shenyang Pharmaceutical University (Shenyang, China). All procedures were performed according to the guidelines issued by the Ethical Committee of Shenyang Pharmaceutical University (SYPU-IACUC-C2017–0512001). All efforts were made to minimize animal suffering and to limit the number of animals used with the approval of the Animal Ethics Committee of Shenyang Pharmaceutical University. Preparation of AGM-Loaded PLGA Microspheres An O/W emulsion solvent evaporation technique was used to prepare AGM-loaded PLGA microspheres (23,24). Briefly, 300 mg PLGA and 129 mg AGM were dissolved in 1.5 mL DCM to obtain an organic phase. The organic phase was cooled to 4°C, slowly added to 30 mL pre-cooled 1% (w/v) PVA aqueous solution and then stirred (Ultra-Turrax TP18/10, IKA, Germany) at a set speed for 2 min to form the O/W emulsion. The obtained emulsion was added to 120 mL distilled water to dilute the emulsion, and stirred at 300 rpm for 4 h to evaporate the solvent. The resulting microspheres were collected by centrifugation at 3000 rpm for 5 min, washed three times with distilled water to remove residual organic solvent and PVA, lyophilized (VirTis AdVantage Plus Bench Top Freeze Dryer, SP industries, Inc., USA) and finally stored at 4°C.

Page 3 of 11

9

Characterization of AGM-Loaded PLGA Microspheres Particle Size and Morphology Particle size and distribution of AGM-loaded PLGA microspheres were investigated by a laser diffraction particle size analyzer (BT-9300S, Bettersize Co., Ltd., Dandong, China). The microspheres were re-dispersed in distilled water prior to analysis. The mean size and size distribution were expressed by D50 (volume weighted mean diameter) and Span (Span value = (D90-D10)/D50) respectively. The surface morphology of the microspheres was examined using scanning electron microcopy (SEM) (SU8010, Hitachi., Ltd., Japan). Freeze-dried microspheres were spread on double-sided conductive adhesive tape, which was previously attached to a copper stub, and then the sample was coated with a thin layer of gold under an argon atmosphere and observed by SEM. SEM photos of the in vitro release study were also obtained. X-Ray Diffraction Analysis To investigate the physical state of AGM within the microspheres, AGM, blank microspheres, physical mixture of AGM and blank microspheres and AGM-loaded microspheres were studied by powder X-ray diffraction (PXRD) (PANanalyical, Almelo, The Netherlands) using Nifiltered Cu Kα radiation. The experiment was conducted at a voltage of 40 KV and 30 mA. Samples were analyzed in a 2θ range of 5–60°. Differential Scanning Calorimetry AGM, blank microspheres, physical mixture of AGM and blank microspheres and microsphere products were studied by differential scanning calorimetry (DSC) technique using a METTLER TOLEDO DSC1 (METTLER TOLEDO, Switzerland). 5 mg samples were placed in an aluminum pan and heated at a rate of 10°C/min, using dry atmosphere of nitrogen as carrier gas, across a temperature range of 30 to 150°C. HPLC-UV Analysis and Drug Loading Quantitative analysis of agomelatine in various samples was performed using a Hitachi high pressure liquid chromatography (HPLC) system consisting of a Hitachi Chromaster 5110 HPLC pump, a Hitachi Chromaster 5210 autosampler, a Hitachi Chromaster 5310 column oven and a Hitachi Chromaster 5410 UV detector. For analysis, a C18 reversed-phase column (Haito Pack ODS, 250 × 4.6 mm, 5 μm) was used, with the mobile phase of acetonitrile:water 50:50 (v/v) and the flow rate was of 1 mL/min. UV detection was performed at 230 nm.

9

Page 4 of 11

Pharm Res

A direct extraction method was used to determine the loading content and encapsulation efficiency of AGM in microspheres accurately. In brief, approximately 10 mg microspheres were dissolved in acetonitrile and sonicated for 1 min. Then the samples were filtered through a 0.22 μm membrane and determined by HPLC. Drug loading and encapsulation efficiency were calculated using the following equations: Drug loadingð%Þ ¼

weight of AGM in microspheres weight of microspheres  100

ð1Þ

Encapsulation efficiencyð%Þ ¼

Actual AGM loading  100 Theoretical AGM loading

ð2Þ

In Vitro Release and Degradation Evaluation The optimal formulation performance in vitro was investigated as follows. Briefly, 10 mg AGM-loaded microspheres were suspended in 10 mL release medium (0.5% SDS, 0.02% NaN3) (25,26). All the samples were incubated in a shaking water bath (Zhicheng Inc., China) at 37 ± 1°C and 100 rpm. At predetermined sampling points, samples were centrifuged at 3000 rpm for 10 min (USTC chuangxing Co. Ltd. Zonkia Branch, China). 9 mL supernatant were separated for the HPLC assay, and fresh medium with equal volume was added in the meantime. All release tests were conducted in triplicate. The amount of released drug during the first 24 h was designated as Bburst release^. In order to evaluate its release mechanism and degradation, SEM was used to observe the microspheres at different time points during the release test. In Vivo Pharmacokinetic Study The in vivo performance of AGM-loaded microspheres was investigated, to further examine the influence of particle size and drug loading on the in vivo release of AGMloaded microspheres. The pharmacokinetics of three AGM-loaded microsphere formulations and AGM solution were investigated. The dosing of rats in the study was calculated based on the oral dose of humans. After 7 days acclimatization, rats were randomly divided into four groups (n = 6). All animals were fasted for 12 h with access to water prior to dosing. Group 1 received AGM solution (AGM carboxymethylcellulose

(2019) 36:9

suspension) as a single dose of 2.6 mg/kg by mouth to investigate the oral bioavailability and the minimum effective blood concentration of AGM in rats. Group 2, Group 3 and Group 4 received different AGM microspheres formulations (formulation MS1, MS2 and MS3 respectively) by intramuscular injection at the right hind-leg muscle as a single dose of 10 mg/kg. At predetermined intervals, approximately 0.3 mL blood samples were collected via the orbital vein with heparinized tubes. The sampling time points for AGM solution was 10, 20, 30, 45, 60, 90, 120, 240, 360, 480, 720, 1440 and 2160 min after administration by mouth and 0.5, 1, 2, 4, 8, 12 h and 1, 2, 3, 5,7, 9, 12, 15, 18, 21, 24, 27, 30 and 35 days for the three AGM microsphere groups. The heparinized blood samples were separated immediately by centrifugation at 4000 rpm for 10 min and stored at −80°C until subsequent extraction and analysis. Plasma samples were processed as follows. First, 100 μL of plasma samples, 20 μL of indapamide (500 ng/mL, internal standard) and 20 μL of acetonitrile were mixed by vortex for 3 min. Next, 2 mL of ethyl acetate were added to the mixture and vortexed for 10 min to extract the mixture. After centrifugation at 10,000 rpm for 10 min, 1.5 mL of the organic layer were removed and dried with nitrogen. Finally, after removal of the solvent, the residue was reconstituted with 100 μL of mobile phase, and the plasma concentration of AGM was d etermined u sing ultra p erformance liqu id chromatography-tandem mass spectrometry (UPLC-MS/ MS, Water Corp., Milford, MA). The obtained data were analyzed using a noncompartmental method with drug and statistics (DAS) software (version 2.0, Mathematical Pharmacology Professional Committee of China, Shanghai, China).

Storage Stability A preliminary stability study of the optimized formulation was carried out. Freeze-dried microspheres were kept at 4°C and room temperature, respectively. Samples were taken on 0 d, 7 d and 28 d to observe its fluidity, and their in vitro release were determined.

Statistical Analysis A paired student’s t test was used for statistical analysis, with p < 0.05 considered the minimum level of significance. Pharmacokinetic parameters were obtained using a non-compartmental method with drug and statistics (DAS) software.

Pharm Res

(2019) 36:9

Page 5 of 11

RESULTS AND DISCUSSION Optimizing the Formulation of Microspheres The traditional emulsification solvent evaporation method was employed to prepare AGM-loaded microspheres. The parameters of the preparation process and formulation were optimized by assessing major factors including drug loading, encapsulation efficiency, particle size distribution and burst release (in terms of the cumulative drug release after 1 day). In the preliminary study, it was found that a high encapsulation efficiency could be achieved by increasing the concentration of PLGA, but the burst release of AGM-loaded PLGA microspheres was mainly related to the particle size. In order to obtain microspheres with low burst release, the formulation was optimized by preparing microspheres of different particle sizes through varying the stirring speed. As shown in Table I, the particle size of the microspheres could be adjusted by changing the stirring speed of emulsification, with a higher stirring speed resulting in smaller microspheres. It was also found that as the particle size increased, burst release from the microspheres decreased significantly (27). Finally, considering that a very large particle size could restrict the syringeability of the formulation, formulation E was chosen as the optimal formulation. A simple direct extract method was used to determine the loading content and encapsulation efficiency of AGM in the microspheres accurately. The determined drug loading of the optimized formulation was 23.7% and the average encapsulation efficiency was 78.5%. It is well known that the solubility of the drug in water is one of the major factors determining the encapsulation efficiency of drugs in an emulsion solvent evaporation process (23,27). Drugs with high water solubility always tend to leak into the outer water phase, causing a low encapsulation efficiency (27). However, AGM is an insoluble drug with a slightly higher solubility, and the solubility of AGM in water increases with temperature, which is not conducive to encapsulation of the drug. In order to improve the encapsulation efficiency, the organic phase and the water phase was precooled to a low temperature, which could Table I Relationship of Particle Size and Burst Release of AGM-Loaded PLGA Microspheres Formulation

A

B

C

D

E

F

Stirring speed (rpm) Mean particle size (μm) Drug loading (%) Encapsulation efficiency (%) Burst release (%)

12,000 23.0 21.8 72.5 25.6

10,000 31.0 22.9 75.9 22.6

8000 36.5 22.5 74.5 16.4

6000 50.4 23.4 77.8 11.4

4000 60.2 23.7 78.5 7.3

2000 100.9 22.5 77.4 7.1

9

prevent the drug from migrating to the water phase by decreasing the solubility of AGM in water and increasing the viscosity of the organic phase. In addition, it was reported that increasing the viscosity of the organic phase by increasing the polymer concentration or molecular weight could increase the drug encapsulation (28). Thus, PLGA with a relatively high molecular weight and a high concentration of 200 mg/mL was used to achieve high drug encapsulation and long duration of drug release. Characterization of AGM Microspheres The optimized AGM microspheres (Formulation E) were characterized in terms of particle size, size distribution, morphology, drug loading and encapsulation efficiency. As shown in Fig. 2, AGM microspheres have a mean diameter of 60.2 μm and a narrow particle size distribution (Span = 0.2). It has been reported that larger particle size of microspheres could lead to a higher the encapsulation efficiency. (28). As well, the particle size and its distribution can significantly affect the in vitro release in the microspheres system. Therefore, microspheres with a relatively big particle size are beneficial for obtaining a high drug encapsulation and slow drug release. The SEM images (Fig. 3a and b) revealed smooth surfaces structure with no pores and drug crystals. They confirmed that the particle size of AGM microspheres was evenly distributed, which is consistent with that obtained from the laser diffraction particle size analyzer. The physical state of AGM within the microspheres was investigated using the PXRD technique. As the physical properties of high molecular weight PLGA are not suitable for PXRD testing, blank microspheres instead of PLGA was used in this study. Figure 4 showed the PXRD patterns of AGM, blank microspheres, microspheres with 20% AGM loading, the physical mixture of AGM and blank microspheres (AGM: blank microspheres = 20: 80). Several distinct peaks in the PXRD of AGM were consistent with reports indicating that AGM is present in a crystalline form II (29). However, they showed that only a faint crystal diffraction peak appeared in microspheres when the drug loading was 20%, and thus it can be concluded that the drug in microspheres with 20% AGM loading mainly existed in the amorphous state, and a small part of AGM existed in a crystalline state. Further investigation of the physical state of AGM within the microspheres was performed using DSC. Figure 5 showed the DSC pattern of AGM, blank microspheres, physical mixture of AGM and blank microspheres and AGM-loaded microspheres (20% drug loading). The AGM powder and physical mixtures showed obvious endothermic peaks at approximately 109°C. By contrast, there were two small endothermic peaks at near 78°C and 97°Cin the microspheres with drug

9

Page 6 of 11

Pharm Res

(2019) 36:9

Fig. 2 Particle size distribution of AGM microspheres (75/25 PLGA 70 kDa; stirring speed: 4000 rpm/ min; drug loading = 20.31%; encapsulation efficiency = 70.23%).

loading of 20%. This confirms that a small portion of AGM exists in a crystalline state, but the majority of drug was dispersed in the polymer matrix in an amorphous state, which is consistent with the PXRD data. The endothermic peaks near 78°C and 97°C were below the melting point of the drug, which inferred that a small part of AGM and PLGA form cocrystals, as AGM contains a secondary amide group which can be involved in supramolecular heterosynthons with the carboxyl end of PLGA (30). It is well known that hydrophobic drugs tend to crystallize inside the microspheres, and phase separation occurs within microspheres at high drug loadings, whereas the drug was dispersed in the polymer in an amorphous state at lower drug loading (17,31). In our study, both PXRD and DSC results confirmed that there was a very small amount of AGM dispersed in a crystal state in the microspheres, and thus it could be concluded that the maximal drug embedding capacity in the PLGA was less than 20%. In Vitro Release Behavior of AGM Microspheres Drug release from PLGA-based drug delivery systems can follow a combination of several mechanisms. In general, diffusion and degradation/erosion are two main release mechanisms of microspheres, It is commonly believed that the release rate of microspheres is diffusion-controlled Fig. 3 SEM images of AGM microspheres (75/25 PLGA 70 kDa; stirring speed: 4000 rpm/ min; D50 = 61.33; drug loading = 20.07%; encapsulation efficiency = 71.20%).

initially, and degradation/erosion controlled during the final stages of the release period (32) . Figure 6 showed the in vitro release profile of AGM-loaded microspheres. Only about 7.36% AGM was released from PLGA microspheres in the first day in vitro, indicating a small burst release of AGM-loaded PLGA microspheres. From day 2 to day 14, drug was released quickly from the microspheres at a nearly constant rate of about 1.45% per day. However, from day 14 to day 70, a lag time occurred and the drug was released at a rate of approximately 0.45% per day. In the next 35 days, the second accelerated release occurred and drug was released quickly at a nearly constant rate of 1.06% per day in this stage. The above results indicated that the pattern of drug release in vitro was different from the typical triphasic release profile, which can be divided into three stages: an initial burst release, lag time depending on the Mw and end-capping of the polymer with a slow or absent diffusion-controlled release, and finally a second erosion-accelerated release (17). However, in this study it showed that a rapid burst release phase was followed by a period of near zero-order quick release, a lag phase and a final second accelerated release. Except for the first day’s burst release, the drug release profiles of different release durations of AGM-loaded microspheres could all be fitted to a zero order model, and acceptable regression coefficients were achieved for all release phases (Table II). These results indicated that the drug release from

Pharm Res

(2019) 36:9

Page 7 of 11

9

Fig. 4 The PXRD curve of AGM (a), blank microspheres (b), AGMloaded microspheres (c), physical mixture of AGM and blank microspheres (d).

PLGA microspheres was predominately governed by diffusion of drug through connected channels in the microspheres, which are formed by the presence of drug. To further investigate the release mechanism of AGMloaded microspheres in vitro, SEM images of AGM-loaded microspheres in different release stages were obtained (Fig. 7). It can be observed that the microspheres kept their shape and integrity over 10 weeks, until they disintegrated completely, suggesting that the slow-degradation region was at the surface, that is, that the internal degradation of the microspheres was faster than at the surface. At the same time, the surface of the microspheres began to show a small quantity of big hydrophilic pores after 1 day, which was formed due to Fig. 5 The DSC curves of AGM (a), blank microspheres (b), AGMloaded microspheres (c), physical mixture of AGM and blank microspheres (d).

the diffusion of the drug. In other words, the initial burst release was attributed to the diffusion of non-encapsulated drug particles on the surface or drug molecules close to the surface. As the release continued, the surface of the microspheres became uneven and the hydrophilic pores became larger and more numerous. In this phase, a fast release occurred at a nearly constant rate of 1.45% per day and this is consistent with SEM results. However, after 2 weeks, the SEM images clearly showed the closure of the big pores at the surface, and the pores seemed to be still closed after 10 weeks, which caused the lag time of drug from day 14 to day 70. Next, many detectable small pores began to form after 10 weeks which can be seen in Fig. 7. This suggested that

9

Page 8 of 11

Pharm Res

(2019) 36:9

Fig. 6 In vitro release profiles of AGM-loaded microspheres (the optimal formulation: drug loading = 20%, D50 = 60 μm, containing 2.0 mg AGM) in release medium (0.5% SDS solution, containing 0.02% NaN3). Each point represents the mean ± SD; n = 3.

the second accelerated release was attributed to a large number of newly formed small pores. The above in vitro results indicated that the release of drug from the microspheres was mainly through waterfilled pores. The initial burst release was attributed to the diffusion of the drug in the surface of the microspheres. However, the microsphere release in the later phase was mostly the result of diffusion through waterfilled pores. Figure 8 showed the in vitro release process of AGM-loaded microspheres. In general, the overall release mechanism of microspheres can be explained as follows. First, when the microspheres are immersed in an aqueous release medium, the drug from the surface spreads to the medium and hydration occurs, which is faster than erosion (33,34). PLGA absorbs a large amount of water and temporarily increases the pressure inside the microspheres, which leads to the formation of big pores in the low density area on the surface of the microspheres. As the amount of the water increases inside the microspheres, and the auto-catalytic phenomenon within the microspheres means the internal degradation of the microspheres is faster than at the surface, the pores become bigger and more numerous (32,34). Drug is mainly released from the big pores in this phase. Then, as the release progresses, the volume of the water inside increases, as well as the mobility and the ability for Table II Evaluation of Drug Release Kinetics of AGM-Loaded Microspheres Phase

Burst release(%)

2–14 day

14–70 day

70–105 day

Slope R2

7.36

1.45 0.9806

0.448 0.9878

1.056 0.9989

Fig. 7 SEM images of AGM microspheres in the different release stages. Each arrowhead represents a pore.

rearrangement of the PLGA chains. This means that any significant increase in pressure can be compensated by swelling and arrangement of the polymer chains. The big pores in the surface of the microspheres later started to close in the release medium (35–37), but with further hydrolysis of the PLGA and internal self-catalysis, a lot of new pores formed and the drug tended to release from the newly formed pores. As well, a small amount of crystallized or aggregated hydrophobic drug may reduce the release rate from another aspect. In general, the release rate of microspheres is governed by the pore formation and pore closure which occur simultaneously. The release rate of microspheres in different release phases are kept

Pharm Res

(2019) 36:9

Page 9 of 11

9

Fig. 8 The in vitro drug release process of AGM-loaded microspheres.

constant, which is common for drugs encapsulated in PLGA drug delivery systems. Studies showed that small porous microspheres prepared by high-Mw, hydrophobic PLGA with low polymer chain mobility is more likely to release drugs in a constant rate (34). In Vivo Pharmacokinetics of AGM Microspheres The in vitro release studies showed that the AGM-loaded microspheres exhibited a promising drug release behavior, as reflected by a small burst release followed by a slow release. To further evaluate the in vivo release behavior of different AGM microspheres, the pharmacokinetics of oral AGM solution, which could evaluate the oral bioavailability and the minimum effective blood concentration of AGM in rats, and three different formulations (MS1, MS2 and MS3) of AGM microspheres were investigated. The mean plasma concentration-time profiles of AGM in rats after oral administration of AGM solution and intramuscular administration of AGM microspheres are shown in Fig. 9. The main calculated pharmacokinetic parameters were listed in Table III. As can be seen in Fig. 9a and Table III, the AGM solution was absorbed by the mouth quickly and the plasma concentration reached the peak concentration (Cmax) 32.6 ng/mL in approximately 1 h. The plasma concentration decreased below 5 ng/mL after a single oral dose of 2.6 mg/mL AGM solution to the rats after 24 h, suggesting that the minimun effective plasma concentration was about 5 ng/mL. As seen in Fig. 9b, after intramuscular injection of three different AGM microspheres, the plasma concentration of AGM reached a peak concentration (109.5, 37.4 and 20.3 ng/mL for MS1, MS2 and MS3 respectively), and then the concentration declined gradually. The blood concentration dropped below 5 ng/mL in 30 days. Compared to AGM solution administrated intragastrically, the T1/2 value of AGM microspheres was prolonged to 96.5 h, 99.6 h and 178.4 h, respectively, which was approximately 6.3, 6.6 and 11.8-fold longer than that of oral AGM solution. Furthermore, the AUC of three different AGM microspheres was approximately 4.8, 3.3 and 2.5-fold more than the calculated AUC of AGM oral solution of the equivalent dose, respectively. These results further confirmed that AGM microspheres could provide a sustained release of approximately

30 days with almost no burst release in vivo, and the bioavailability of AGM was dramatically increased when it was prepared into sustained release microspheres. Three different AGM microspheres were released in vitro for 3 months, with a burst release of 25%, 7.0% and 7.0% respectively, but when released in vivo for 30 days demonstrated a burst release of 10.0%, 9.0% and 6.0% (calculated by the AUC (0-1 day) versus AUC (0-∞)), respectively. This indicated that in vitro release test could predict the burst release of AGM microspheres

Fig. 9 Pharmacokinetic results of AGM. (a) mean concentration-time profile of AGM after a single oral dose of 2.6 mg/mL AGM solution to the rats, (b) plasma concentration- time profile of AGM after intramuscular injection of AGM-loaded microspheres (10 mg/kg). MS1: drug loading = 20%, D50 = 30 μm, MS2: drug loading = 7%, D50 = 30 μm, MS3: drug loading = 20%, D50 = 60 μm.

9

Page 10 of 11

Table III The NonCompartmental Model Parameters of AGM after i.g. Administration of 2.6 mg/mL AGM Solution and Intramuscular Administration of AGM Microspheres (10 mg/mL) (Mean ± SD; n = 6)

Pharm Res

Formulations

AGM solution

MS1

MS2

MS3

AUC(0-t) (mg/L·h) AUC(0-∞) (mg/L·h) AUC(0-1 day) (mg/L·h) T1/2 (h) Cmax(mg/L)

330.9 ± 52.5 461.3 ± 106.5 – 15.2 ± 7.3 32.6 ± 12.5

6779.0 ± 1106.3 7285.6 ± 1149.6 680.6 ± 111.1 96.5 ± 16.6 109.5 ± 17.6

5025.2 ± 841.6 7564.1 ± 3940.2 451.6 ± 75.6 99.6 ± 33.3 37.4 ± 1.4

3709.2 ± 778.7 5560.0 ± 996.3 224.4 ± 47.1 178.4 ± 27.0 20.3 ± 4.7

Tmax (h) Burst release (%)

1.0 ± 0.0 –

1.0 ± 0.9 10.0

3.3 ± 4.0 9.0

– 6.0

in vivo. It is significant to develop an in vitro release assay with good in vitro and in vivo correlation (IVIVC), which can be used not only to predict the in vivo behavior of the formulation and optimize the formation during formulation development, but also be used to control the quality of the product in the preparation of the product process, and so further work will focus on establishing a new in vitro release assay with a good in vitro and in vivo relationship. Comparing MS1 with MS2, it can be inferred that drug loading could significantly influence the burst release of the microsphere in vivo. The lower the drug loading could result in the smaller the burst release in vivo. However, the AUC of MS1 was bigger than that of MS2, indicating that AGM was eliminated quickly in the blood and the rate of elimination of AGM is faster than absorption. Comparing MS1 with MS3, it was apparent that particle size had a significant effect on the release of microspheres in vivo. It can be seen that the bigger the particle size, the smaller the burst release in vivo, suggesting that a relatively big particle is beneficial for achieving a sustained release in vivo with a small burst release. As is shown in Table III, T1/2: MS1 < MS2 < MS3, Cmax: MS1 > MS2 > MS3, burst release in vivo: MS1 > MS2 > MS3. The results concluded that MS3 showed the best sustained release characteristics and minimal burst release, which was very important for reducing the side effects caused by a too high concentration in the blood in vivo. Storage Stability The stability of the optimized formulation was measured at 4°C and room temperature for 30 days, respectively. As shown in Table IV, microspheres at room temperature tended

Table IV The Stability Results of the Optimized Formulation

(2019) 36:9

Condition

Fluidity Burst release in vitro

to agglomerate and its burst release became slowly. However, there was no significant change in fluidity and the burst release when the microspheres were stored at 4°C.

CONCLUSION AGM-loaded PLGA microspheres with smooth surface and narrow particle size distribution were prepared by the emulsion solvent evaporation method. The drug loading content and encapsulation efficiency of AGM microspheres were 23.7% and 78.5% respectively. DSC and PXRD analysis demonstrated that the majority of drug was dispersed in the polymer matrix in an amorphous state, and only a small part of AGM existed in a crystalline state. In vitro release studies showed that AGM microspheres released slowly with a small burst release of 7.0% followed by a fast release, a lag time and a second accelerated release in vitro. SEM images showed pore formation and pore closure in different release stages, indicating that the main release mechanism of the microspheres was diffusion through water filled pores in vitro. Over 30 days, a sustained and steady plasma concentration of AGM could be achieved after a single intramuscular injection of AGM microspheres. In vivo release studies also showed that particle size and drug loading had significant effects on the release of AGM-loaded microspheres. These results suggested the potential use of AGM-loaded PLGA microspheres for treatment of depression over a long time period.

0 day

7 days

28 days

4 °C

25 °C

4 °C

25 °C

4 °C

25 °C

good 8.96

good 8.49

good 8.18

Slightly agglomeration 6.99

good 8.88

agglomeration 6.94

Pharm Res

(2019) 36:9

REFERENCES 1.

2. 3.

4.

5.

6.

7.

8. 9. 10.

11.

12.

13.

14. 15. 16. 17.

18.

19.

20.

Kessler RC, Chiu WT, Demler O, Walters EE. Prevalence, Severity, and Comorbidity of twelve-month DSM-IV disorders in the National Comorbidity Survey Replication (NCS-R). Arch Gen Psychiatry. 2005;62:617–27. Dolder CR, Nelson M, Snider M. Agomelatine treatment of major depressive disorder. Ann Pharmacother. 2008;42:1822–31. Sanand L, Arranzb B. Agomelatine: a novel mechanism of antidepressant action involving the melatonergic and the serotonergic system. Eur Psychiatry. 2008;23(6):396–402. Garcia-Cebrianabcde A, Grassifgahai L, et al. Factors influencing depression endpoints research (FINDER): study design and population characteristics. Eur Psychiatry. 2008;23(1):57–65. Ferrari AJ, Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJ, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med. 2013;10:e1001547. Dubovsky SL, Warren C. Agomelatine, a melatonin agonist with antidepressant properties. Expert Opin Investig Drugs. 2009;18: 1533–40. De BC, Guardiolalemaitre B, Mocaër E, Renard P, Muñoz C, Millan MJ. Agomelatine, the first melatonergic antidepressant: discovery, characterization and development. Nat Rev Drug Discov. 2010;9:628–42. Kennedy SH. Agomelatine: efficacy at each phase of antidepressant treatment. Cns Drugs. 2009;23(Suppl 2):41–7. Demyttenaere K. Agomelatine: a narrative review. Eur Neuropsychopharmacol. 2011;21(Suppl 4):S703–S709. Serna MC, Cruz I, Real J, et al. Duration and adherence of antidepressant treatment (2003 to 2007) based on prescription database. Eur Psychiatry. 2010;25(4):206–213. Kessler RC, Coccaro EF, Fava M, Jaeger S, Jin R, Walters E. The prevalence and correlates of DSM-IV intermittent explosive disorder in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2006;63:669–78. Du W, Zhou Y, Gong Y, Zhao C, et al. Investigation of physicochemical properties and in-vitro in-vivo evaluation of agomelatine polymorphs. Asian J Pharm Sci. 2013;8(3):181–190. Yan L, Gao H, Ren F, Ren G. Solubility of Agomelatine crystal form I and form II in pure solvents and (isopropanol + water) mixtures. J Chem Eng Data. 2015;60:151019103620003. Rhee YS, Park CM, DeLuca PP, Mansour HM. Sustained-release injectable drug delivery. Pharm Tech. 2010:34;s6–s13 Freiberg S, Zhu XX. Polymer microspheres for controlled drug release. Int J Pharm. 2004;282:1–18. Varde NK, Pack DW. Microspheres for controlled release drug delivery. Expert Opin Biol Ther. 2004;4:35. Wischke C, Schwendeman SP. Principles of encapsulating hydrophobic drugs in PLA/PLGA microparticles. Int J Pharm. 2008;364: 298–327. Zhang Z, Bi X, Li H, et al. Enhanced targeting efficiency of PLGA microspheres loaded with Lornoxicam for intra-articular administration. Drug deliv. 2011;18(7):536–544. Wang W, Cai Y, Zhang G, Liu Y, Sui H, Park K, et al. Sophoridine-loaded PLGA microspheres for lung targeting: preparation, in vitro, and in vivo evaluation. Drug Delivery. 2016;23: 3674–80. Zhang Y, Wischke C, Mittal S, Mitra A, Schwendeman SP. Design of controlled release PLGA microspheres for hydrophobic fenretinide. Mol Pharm. 2016;13:2622–30.

Page 11 of 11 21.

9

Rodríguez VJ, Bravo-Osuna I, Herrero-Vanrell R, Molina Martínez IT, Guzmán NM. Optimising the controlled release of dexamethasone from a new generation of PLGA-based microspheres intended for intravitreal administration. Eur J Pharm Sci. 2016;92:287–97. 22. Guo Y, Yang Y, He L, Rong S, Pu C, Xie B, et al. Injectable sustained-release depots of PLGA microspheres for insoluble drugs prepared by hot-melt extrusion. Pharm Res. 2017;34:2211–22. 23. Ramazani F, Chen W, van Nostrum CF, Storm G, Kiessling F, Lammers T, et al. Strategies for encapsulation of small hydrophilic and amphiphilic drugs in PLGA microspheres: state-of-the-art and challenges. Int J Pharm. 2016;499:358–67. 24. Alaina FJ, Galperin A, Ratner BD. Drug encapsulated polymeric microspheres for intracranial tumor therapy: a review of the literature ☆ ☆☆. Adv Drug Deliv Rev. 2015;91:23–37. 25. Hirota K, Doty AC, Ackermann R, et al. Characterizing release mechanisms of leuprolide acetateloaded PLGA microspheres for IVIVC development I: In vitro evaluation. J Control Release. 2016;244:302–313. 26. Pu C, Wang Q, Zhang H, et al. In Vitro-In Vivo Relationship of Amorphous Insoluble API (Progesterone) in PLGA Microspheres. Pharm Res. 2017;34(12):2787–2797. 27. Guo W, Quan P, Fang L, et al. Sustained release donepezil loaded PLGA microspheres for injection: Preparation, in vitro, and in vivo, study. Asian J Pharm Sci. 2015;10(5):405–414. 28. Ogawa Y, Yamamoto M, Okada H, Yashiki T, Shimamoto T. A new technique to efficiently entrap leuprolide acetate into microcapsules of polylactic acid or copoly(lactic/glycolic) acid. Chem Pharm Bull. 1988;36:1095–103. 29. Souvie J C, Gonzalez-Blanco I, Thominot G, et al. Process for the synthesis and crystalline form agomelatine: U.S. Patent 7,544, 839[P]. 2009-6-9. 30. Prohens R, Barbas R, Portell A, Fontbardia M, Alcobé X, Puigjaner C. Polymorphism of Cocrystals: the promiscuous behavior of Agomelatine. Cryst Growth Des. 2016;16:1063–70. 31. Mao S, Shi Y, Li L, Xu J, Schaper A, Kissel T. Effects of process and formulation parameters on characteristics and internal morphology of poly(d,l-lactide-co-glycolide) microspheres formed by the solvent evaporation method. Eur J Pharm Biopharm. 2008;68:214–23. 32. Pu C, Wang Q, Zhang H, Gou J, Guo Y, Tan X, et al. In vitro-in vivo relationship of amorphous insoluble API (Progesterone) in PLGA microspheres. Pharm Res. 2017:1–11. 33. Gu B, Sun X, Papadimitrakopoulos F, et al. Seeing is believing, PLGA microsphere degradation revealed in PLGA microsphere/ PVA hydrogel composites. J Control Release. 2016;228:170–178. 34. Fredenberg S, Wahlgren M, Reslow M, Axelsson A. The mechanisms of drug release in poly(lactic-co-glycolic acid)-based drug delivery systems—a review. Int J Pharm. 2011;415:34–52. 35. Fredenberg S, Jönsson M, Laakso T, Wahlgren M, Reslow M, Axelsson A. Development of mass transport resistance in poly(lactide-co-glycolide) films and particles – a mechanistic study. Int J Pharm. 2011;409:194–202. 36. Kang J, Schwendeman SP. Pore closing and opening in biodegradable polymers and their effect on the controlled release of proteins. Mol Pharm. 2007;4:104. 37. Fredenberg S, Wahlgren M, Reslow M, Axelsson A. Pore formation and pore closure in poly(D,L-lactide-co-glycolide) films. J Control Release. 2011;150:142–9.