Construction and evaluation of a phospholipid-based phase transition in situ gel system for brexpiprazole

The objective of this study was to develop phospholipid-based injectable phase transition in situ gels (PTIGs) for the sustained release of Brexpiprazole (Brex). Phospholipid (Lipoid S100, S100) and stearic acid (SA) were used as the gel matrix which was dissolved in biocompatible solvent medium-chain triglyceride (MCT), N-methyl pyrrolidone (NMP), and ethanol to obtain PTIGs solution. The Brex PTIG showed a solution condition of low viscosity in vitro and was gelatinized in situ in vivo after subcutaneous injection. Both in vitro release assay and in vivo pharmacokinetics study in SD rats displayed that Brex in PTIGs could achieve a sustained release, compared with brexpiprazole solution (Brex-Sol) or brexpiprazole suspension (Brex-Sus). The Brex-PTIGs had good degradability and biocompatibility in vivo with rare inflammation at the injection site. Among the three Brex-PTIG formulations, Brex-PTIG-3 with the SA in the formulation had the greatest gelation viscosity, the lowest initial release rate, and the most stable release profile with sustained release of up to 60 days. The above results indicated that, as a novel drug delivery system, the Brex-PTIGs offered a new option for the clinical treatment of patients with schizophrenia.


Introduction
As one of the main types of central nervous illness, schizophrenia has a prolonged course and a high relapse rate, making it necessary to take long-term medication [1,2].The crucial point is because patients with schizophrenia need to take medication on time every day, there is a problem of poor medication compliance resulting in poor treatment outcomes [3,4].Risperidone as the first-line drug for the clinical treatment of schizophrenia, its microspheres Risperdal Consta® and suspension PERSERI® have been marketed [5,6].Brexpiprazole (Brex) is the first multi-target antipsychotic that exerts agonism at dopamine receptors, partial agonism of the 5-HT 1A receptor, and antagonism of the 5-HT 2A receptor [7], is a new option for the treatment of schizophrenia [8].Brex is currently marketed as oral tablets (marketed in 2015) [9] and orally disintegrating tablets (marketed in 2018) in strengths of 0.5 to 2 mg/day [10].Until now, there was no sustained-release formulation of Brex for marketing.Because of its indications, it was feasible to prepare a sustained-release formulation of Brex.Therefore, a long-acting injection of Brex might be a good choice for the treatment of schizophrenia.
Ran Tao and Li Liu are contributed equally to this paper.

3
The drug depot is a long-acting, sustained-release drug delivery system that can last for days or even months with a single administration, it has drawn tremendous attention for its advantages [11,12].Injectable in situ gels are drug depots that can undergo non-covalent cross-linking phase transitions at the injection site and become local drug delivery depots following administration due to changes in external environmental conditions (e.g., light, temperature, pH, hydrophobicity, ionic strength, etc.) [13,14].In situ gel drug delivery systems offer more benefits, such as multiple routes of administration (subcutaneous, intramuscular, nasal, or ocular, etc.), simple manufacturing techniques, and low-cost manufacturing, than other drug depots such as suspensions, microspheres, or implants [15].Currently, available gel injections included OncoGel®, Eligard®, Atridox-Elyzol®, SomatulineDepot®, etc.All these gel injections provided a delayed release and prolonged the duration of action [16][17][18].This strategy may also apply to enhancing treatment outcomes in schizophrenia.
The drug depot under development includes a thermosensitive gel based on poloxamer, a photosensitive gel based on acrylamide (AM), a PH-sensitive gel based on chitosan, and a solvent exchange gel based on PLGA [19][20][21].Currently marketed injectable in situ gels all utilize local delivery systems based on solvent exchange induction.Most of the long-acting gel injectable products used poly (lactic-coglycolic acid) (PLGA) gel technology (Atrigel®) owning advantages such as ease of minimally invasive administration and reduced dosing frequency [22,23].However, some adverse effects hindered the practical use of PLGA systems, such as inflammation and potential toxicity [24,25].In contrast, phospholipid-based phase transition in situ gels showed excellent clinical potential to address the deficiencies of PLGA gel [25][26][27].As is well known, phospholipids have excellent biocompatibility and safety [28].Xiang et al. developed a phospholipid-based phase transition in situ gelation system to treat hyperlipidemia [29].This system could release pitavastatin slowly with good biocompatibility and a mild inflammatory response.Moreover, solid lipids have been widely applied in pharmaceutics for sustained-release drug delivery [30,31].Zhang et al. developed an in situ hybrid implant composed of stearic acid (SA) and PLGA.The unique advantage of incorporating SA was the improved drug release profiles and reduced burst release caused by morphology altering and phase inversion of the implant [32].Medium-chain triglycerides (MCTs) lacked the hydrophilic domains, but they could act as dopants to replace some molecules in the liquid crystal structure of phospholipids and form a better drug depot [15,33].
This study aimed to develop phospholipid-based phase transition in situ gel systems of brexpiprazole (Brex-PTIGs) with good biocompatibility and improved medication adherence in patients with schizophrenia.Brex-PTIGs consist of phospholipid (S100), medium-chain triglyceride (MCT) or stearic acid (SA), and biocompatible solvents N-methyl pyrrolidone (NMP) and ethanol (Table 1).NMP is one of the FDA-approved solvents with a high safety profile and could dissolve brexpiprazole extremely well.The excipients in the Brex-PTIGs are biodegradable and have been approved by the FDA for approved marketing [34].The Brex-PTIGs were characterized by viscosity, in vitro release of brexpiprazole, in vivo biocompatibility, and biodegradability.In addition, the gelation conditions, rheological properties, and stability were also evaluated.The sustained-release preparations of brexpiprazole in this study have the potential for industrial production and clinical selection.

Animals
Healthy male SD rats, 170±22 g in body mass, were provided by the Experimental Animal Centre of Chongqing Medical University.All rats were provided with standard housing conditions and maintained at 20 ℃ and 70% RH with a normal 12 h light/dark cycle starting 1 week before the experiment.The care of laboratory animals and the animal experimental operation had conforming to the Chongqing Management Approach to Laboratory Animals (Chongqing government order NO.195).

Preparation of the Brex-PTIG system
Brex was dissolved in NMP, and different amounts of S100, SA, MCT, and ethanol were added sequentially to obtain a gel solution by stirring at 37 °C for 30 min.The composition of the formulations was presented in Table 1.To obtain samples in the gel state, different volumes of phosphate buffer (PBS, 0.01 M, pH 7.4) were added to the Brex-PTIGs solution and stirred for 10 min at room temperature [35].

Determination of Brex
The Brex in Brex-PTIGs or releasing medium were analyzed by high-performance liquid chromatography (HPLC) (Agilent 1260infinity, USA).The samples were diluted with methanol and separated on an Agilent Poroshell EC-C18 column (50 mm x 4.6mm, 2.7 µm).The mobile phase was potassium dihydrogen phosphate (pH 2.0)-methanol (35:65, v/v) at a flow rate of 1.0 ml/min.The detection wavelength was 213 nm.The concentration of Brex was calculated by the established standard curve (0.1-8.1 µg/ml).The Brex in plasma was analyzed by LC-MS (Agilent 1290+6460).100 μl of plasma sample was transferred into a 2 ml centrifuge tube, 300 μl of precipitant acetonitrile was added, and centrifuged at 13000 rpm for 5 min.50 µl of the supernatant was taken and 25 µl of internal standard solution (Aripiprazole, 240 ng/ml) and 150 µl acetonitrile were added.
After vortexed for 1 min and then centrifuged at 13000 rpm for 3 min, the supernatant was collected for determination.A liquid chromatographic column (4.6 × 150 mm, 3 μm) (Luna PFP) was employed for the analytical separation.The mobile phases A and B were 0.5 vol% (v/v) formic acid aqueous and formic acid -water -acetonitrile (0.5:15:85), respectively.The flow rate of the mobile phase was 0.5 ml/ min, the injection volume was 20 μl, and the temperature of the chromatographic column was 40 ℃.For the mass spectroscopic measurements, an electrospray ionization source (ESI), positive ion mode detection, full ion scanning mode (100-500 m/z), and a capillary voltage of 7 V were used.The concentration of Brex was calculated by the established standard curve (10-180 ng/ml).

Characterization of the Brex-PTIG system
The viscosities of Brex-PTIGs before and after gelation were determined at room temperature using a rotational viscometer (Shanghai Fangrui Instruments Co., Ltd.).The morphological images after gelation were also taken by scanning electron microscopy (Thermo).
The rheological behavior of Brex-PTIGs was investigated using a rheometer (Anton Paar).The energy loss modulus G'' and storage modulus G' were enumerated from vibration measurements at the constant strain in the angular frequency range of 0.01-10 Hz [15,36].

In vitro release
The release of Brex from Brex-PTIGs was monitored according to the literature [37].The release medium was PBS containing 30% (v/v) ethanol (pH 6.5) [36,38,39].Brex powder was dissolved in the release medium to a concentration of 30 mg/ml (Brex-Sol) which served as the control.
100 μl of Brex-PTIGs or Brex-Sol was added to dialysis cassettes (Thermo) with a cut-off molecular weight of 10,000, and each dialysis cassette was immersed in 300 ml of fresh release medium under magnetic stirring at a constant speed of 200 rpm at 37 °C.At each time point, 50 ml of release medium was collected and the same amount of fresh release medium was replenished.The collected release medium was filtered through a 0.22 μm filter and then analyzed by HPLC according to the method mentioned in "Determination of Brex".The cumulative release of Brex was calculated and the cumulative release-time profile was plotted.

In vivo pharmacokinetic studies in rats
Brexpiprazole solution (Brex-Sol) and Brexpiprazole suspension (Brex-Sus) were prepared as controls.Brex powder was dissolved in NMP to form Brex-Sol at a concentration of 30 mg/ml.Brex-Sus was prepared by dissolving the drug in a mixture of pure water containing 25% (v/v) NMP at a concentration of 30 mg/ml.The rats were divided into 6 groups randomly and administrated subcutaneously with saline, or Brex-PTIG-1, Brex-PTIG-2, Brex-PTIG-3, Brex-Sol, and Brex-Sus at a Brex dosage of 100 mg/kg, respectively.Blood was taken from the retroorbital plexus of the rat at each time point and collected into heparinized tubes.The plasma was obtained after centrifuging (5500 rpm, 5 min) at 4 °C.The concentration of Brex in plasma was analyzed by LC-MS according to the method mentioned in section 2.4.The pharmacokinetic parameters of different formulations were fitted using DAS 2.0 software.

Construction of in vitro-in vivo correlation (IVIVC)
The IVIVC model was constructed using the in vitro release (%) and the in vivo absorption (%) in rats.The corresponding in vivo absorption (%) was plotted on the y-axis with the in vitro release (%) as the x-axis, and the best model was selected by linear and polynomial regression.The percentage of drug absorption in rats was calculated according to the Wagner-Nelson method.The Wagner-Nelson relation was given by: where C t represented the concentration of Brex in rat plasma at a specific time t, K e was the elimination rate constant, AUC 0→t was the area under the plasma concentration curve from time 0 to t, and AUC 0→∞ was the area under the plasma concentration curve from time 0 to infinity.
The percentage of in vitro release was calculated as follows: where C t denoted the concentration of Brex in the release medium at a specific time t; C 0 was the total concentration of Brex in the formulation.

Biocompatibility and biodegradability studies
Rats were shaved and randomly divided into 6 groups, injected subcutaneously with saline, Brex-PTIG-1, Brex-PTIG-2, Brex-PTIG-3, Brex-Sol, or Brex-Sus (100 mg/kg for Brex), respectively.Blood was taken from the retroorbital plexus of the rat at each time point and collected into tubes.The serum was obtained after centrifuging (5500 rpm, 5 min) at 4 °C.The assay was performed according to the corresponding enzyme-linked immunosorbent assay kit instructions, and the expressions of cytokines TNF-α and IL-6 were detected.
The appearance of the rat's skin in each group was photographed at 1, 7, 14, and 21 d after injection.A 2×2 cm area of skin around the injection site was cut after executing the rats.The gels maintained at the injection site were separated and weighed.The body weight of rats after administration was also monitored by weighing.

Histological analysis
The skin removed was fixed in 4% (v/v) paraformaldehyde for 48 hours.Paraffin sections were processed with hematoxylin-eosin (HE) staining and observed.

Data and statistical analysis
All experiments were repeated at least three times and data are presented as mean ± standard deviation (SD).Differences between experimental groups were assessed using a two-way analysis of variance by the analysis software GraphPad InStat 3.3 following a normal distribution test. (1)

Preparation of the Brex-PTIG system
Brex is a novel multi-target antipsychotic drug for the treatment of schizophrenia, and its preparations on market are oral tablets [9].To enhance medication adherence in patients with schizophrenia, there is an urgent need to develop psychotropic drug delivery systems that can deliver medication that lasts for months rather than days.Phospholipid-based organogels are of interest due to their non-toxicity, good biocompatibility, and injectability [40].In this study, we designed in situ gel systems for Brex based on the phenomenon that high concentrations of phospholipids form in situ drug depots by precipitating, prolonging the duration of drug action.The Brex-PTIGs consist of S100, MCT or SA, NMP, and ethanol (Table 1).Brex-PTIGs were obtained only by mixing and stirring of various components, which makes the preparation process simple, cost-effective, and suitable for mass production.As shown in Fig. 1, the obtained Brex-PTIGs solution were all clear yellow homogeneous solutions.After mixing with PBS, the Brex-PTIGs solution underwent a fast phase transition and convert to a semi-solid state (Fig. 1A2, B2, and C2).
Although extensive research was focusing on various insitu implant delivery systems, only solvent diffusion-based in-situ polymer precipitation systems are commercially available, such as doxycycline (AtridoxElyzol®) for periodontal delivery and leuprolide (Eligard®) for the treatment of prostate cancer [37,38].In this study, we used S100 as the main gel matrix, which is a biodegradable and biocompatible excipient with low toxicity that has been approved by the FDA for marketing [31].Based on the solvent diffusion polymer precipitation system, the in-situ gel in this study utilized the diffusion of ethanol and NMP after encountering the aqueous phase, which resulted in the gel's transition from a liquid to a semi-solid.
Brexpiprazole was a weakly basic compound with a pKa value of 7.8, which is practically insoluble in water, but freely soluble in NMP (an FDA-approved solvent for injection) [36].According to the formulation of the marketed in situ phase transition gel product buprenorphine longacting injection (SUBLOCADE®), the content of NMP in it reached up to 50%, and the maximum amount of a single dose was 833 mg [41].Whereas the highest content of NMP in our Brex-PTIG in a single dose was 463 mg.Therefore, the NMP content of 45% (v/v) in Brex PTIG was considered to be safe.
Ethanol is a good solvent in injections, but when injections contain high concentrations of ethanol, they often cause severe pain and even local necrosis at the injection site [15,37].To improve tolerance and reduce adverse effects in patients with schizophrenia, we strictly control the ethanol dosage to less than 10% (v/v).Moreover, studies had shown that the higher concentration of ethanol in sustained-release gels, the more severer the initial burst release of drugs [40].Meanwhile, MCT and SA were also used to regulate viscosity, improve the release profile of the Brex and alleviate the irritation response caused by ethanol [36,42].

Characterization of the Brex-PTIGs
Different volumes of PBS were added to investigate the effect of water on the gelation of the Brex-PTIGs.As shown in Fig. 2A, the viscosity of the Brex-PTIGs increased with the addition of PBS when the PBS content was less than 7% (w/w).The insolubility of phospholipids in an aqueous solution may be the main reason for the increase in viscosity of the Brex-PTIGs [43].Ethanol, NMP, and water are miscible, providing the basis for the rapid precipitation of S100 and phase separation of Brex-PTIGs.It was demonstrated that water would enter and diffuse into the gel after injection, thus triggering an exchange of solvent and water [42].However, when the PBS content was higher than 7% (w/w), the dilution effect of PBS caused the viscosity of the Brex-PTIGs to decrease (Fig. 2A).After gelation, the viscosity of the Brex-PTIGs increased significantly (p<0.001,Fig. 2B).
Viscosity is an important factor in evaluating long-lasting injectable gels.The viscosity of injections needed to be less than 300 cP, while a higher viscosity is required for the gel to form a depot after injection for sustained release of the drug [36,44].Therefore, the gel must undergo a significant viscosity shift before and after injection.As shown in Fig. 2B, Brex-PTIGs were all suitable for injectable use, with viscosities below 300 cP in the solution state, and underwent effective phase transition after injection, with a significant increase of viscosity.Among the three Brex-PTIGs, BPPG-3 showed the greatest change in viscosity, increasing from a viscosity of 27.11 cP in a solution state to 1634.96 cP in a gelled state.The steep increase in viscosity suggested a shift in the state of the gel solution, which could help reduce the initial release of Brex and prolong its release time [45].
The surface morphologies of the gelation gels were photographed by scanning electron microscopy (Fig. 3A, B).It was found that Brex-PTIG-1 presented a loose porous, favored "honeycomb" distribution.Brex-PTIG-2 was smoother and structurally intact than Brex-PTIG-1, but large areas of uneven "pitting" and "bulging" could be observed.From the perspective of formulation phase transition fusion, it was speculated that Brex-PTIG-2 might have insufficient phase transition or incompletely formed parts [46].Brex-PTIG-3 was smoother than Brex-PTIG-2 and had only small "sharp corners".Therefore, it was inferred that in the three formulations, the amount of materials in Brex-PTIG-3 reached a better proportion and could maintain the complete formation of the gel.Brex-PTIG-3 had some "sharp corners", which might be caused by S100 and SA in Brex-PTIG-3 being better combined and bound to each other to form a denser structure [37].This tightly bound structure might impact drug release.The tighter the binding between matrices, the more "difficult" the release of the drug, thus achieving a sustained release performance.The complex modulus of Brex-PTIGs after gelation was also measured (Fig. 4).The magnitude of the coefficient of storage modulus (G') and the coefficient of loss modulus (G'') was associated with the denaturing of the materials [47,48].The viscoelastic behavior of these three formulations was similar, and the G' was much greater than G'', indicating that they were themselves able to undergo morphological changes, that was, a transition from the liquid state to the solid state, which was similar to previous observations (Fig. 1).The G' of the three formulations increased in turn, while the G'' did not change significantly.As shown in Fig. 4, it could be observed that the G' increased from 780 in Brex-PTIG-1 to 1400 in Brex-PTIG-3, while the G'' increased only by about 200 in the three groups of formulations.When the G' increases (in the case of similar changes in the G''), the material tended to transform into a solid state, and its ability to resist deformation increased.For the release rate of the drug from the material, it was reasonable to speculate that when the formulation formed a gel in situ, the smaller the deformation, the drug was "confined" in the gel and the less easily released, the more sustained-release effect can be achieved.

Drug release studies in vitro
Brex is insoluble in water and its release could be facilitated in release media containing ethanol.In addition, ethanol in the release medium increases the concentrationdependent diffusion outside the gel matrix, thus accelerating the degradation of the gel matrix [42,49].We also adjusted the pH of the release medium to 6.5 [39].The above conditions were set to mimic the biodegradation situation of in situ gels in vivo since there might be certain acids and enzymes at the injection site [15,39].
The release profiles of the three Brex-PTIGs were generally similar (Fig. 5).However, among the three Brex-PTIGs, Brex-PTIG-3 had the slowest release rate.The release rate of Brex in Brex-PTIG-3 was significantly lower than that of Brex-PTIG-1 at 1 h (p<0.05).This indicated the SA was necessary for improving the initial sudden release of Brex from Brex-PTIG.

Pharmacokinetic studies
To investigate the pharmacokinetic properties of Brex-PTIGs in vivo, healthy SD rats were injected subcutaneously with Brex-Sol, Brex-Sus, Brex-PTIG-1, Brex-PTIG-2, or Brex-PTIG-3 with a single dose of Brex at 100 mg/kg.As shown in Fig. 6, the plasma concentration of Brex in the Brex-Sol group showed a significant increase within 1 h after injection, with a C max of 875.52 ng/ml followed by rapid clearance and Brex could not undetectable after 24 h.Similarly, the concentration of Brex in the Brex-Sus group reached a peak quickly, with a C max of 548.68 ng/ml (Table 2), and then cleared quickly (Fig. 6), and Brex was undetected in plasma after 7 d.In contrast, as shown in Table 2 and Fig. 6, the Brex-PTIGs groups had significantly delayed peaking, compared with Brex-sol and Brex-sus (all p<0.001).The peak concentration of Brex was observed around 2 h postinjection for Brex-PTIG-1, Brex-PTIG-2, and Brex-PTIG-3, with the C max of 139.30 ng/ml, 113.83 ng/ml, and 89.45 ng/ ml respectively, which were significantly lower than that of Brex-sol (875.52 ng/ml) and Brex-sus (548.68 ng/ml) (all p<0.001).A major challenge in the clinical implementation of in situ molded implants was the control of the initial burst release of drugs, particularly for in situ precipitation systems.Compared to solution and suspension, the Brex-PTIGs provided a significantly improved initial burst release of Brex and were able to release Brex smoothly for two months.Undoubtedly, the decreased initial burst release was attributed to the formation of Brex-PTIGs gel depots.There was no significance between the three Brex-PTIGs in the area under the concentration-time curves (AUC (0-∞) ), indicating the same degree of absorption of the three Brex-PTIGs.All Brex-PTIGs released Brex stably for more than 60 d, which might facilitate medication adherence in patients with schizophrenia.
Among the three Brex-PTIGs (Table 2), the C max of Brex-PTIG-3 was significantly lower than Brex-PTIG-1 (p<0.01), and the t 1/2z of Brex-PTIG-3 was significantly longer than that of Brex-PTIG-1 (p<0.05).This was correlated with the release assay of Brex-PTIGs in vitro, which indicated that although a sustained-release profile of Brex-PTIGs was observed, the composition of the gel matrix might have influenced the initial burst release.The results of in vitro release and pharmacokinetics assay suggested that MCT in Brex-PTIG-1 might lessen the initial burst release of phospholipid-based phase transition in situ gel, but it was not as effective as that of SA in Brex-PTIG-3, suggesting SA was a more suitable component for Brex-PTIGs.

Correlative study of in vitro release and in vivo absorption of Brex
Based on the analysis of data from three formulations, there was a 2nd order polynomial correlation between in vitro release (%) (X) and in vivo absorption (%) (Y), with correlation coefficient (R 2 ) values greater than 0.9 (Fig. 7).General, R 2 greater than 0.9 was considered to be well correlated.Therefore, the IVIVC model was successfully constructed.When the in vitro release medium was PBS containing 30% (v/v) ethanol (pH 6.5), the release data of the Both in vitro release and in vivo absorption results of Brex displayed that Brex-PTIG-3 had a more desirable release behavior (compared with Brex-PTIG-1 and 2), which echoed the results of viscosity properties and rheological behavior.The higher the viscosity of gel after phase transition, the greater its G', and the stronger the ability to resist deformation.Therefore, Brex could be better stored in the phospholipid to achieve sustain-release.

Biodegradability and biocompatibility of the Brex-PTIG system in vivo
The results of the studies presented by the Elisa kit assay showed the levels of TNF-α and IL-6 in rat serum.Compared with the saline group, the remaining groups showed no significant inflammatory reaction within 14 days (no statistical difference) (Fig. 8).
The state of the skin around the injection site was observed on days 0, 7, 14, and 21 after the injection.As shown in Fig. 9, on Day 0 an injection bulge could be seen in the Brex-PTIGs and Brex-sus groups, while the skin seemed smooth and flat with no discernable bulge in the Brex-sol group.On Day 7 after the injection, the volume of the injection bulge significantly decreased in all Brex-PTIGs groups.By Day 14, the skin at the injection site was almost flat in the Brex-PTIGs group.
To further evaluate the degradation of Brex-PTIGs in vivo, the depots formed after subcutaneous injection of Brex-PTIGs were dissected from the rat skin and weighed on different days after injection.The weight change of the residual gel was displayed in Fig. 10A.The weight of the residual Brex-PTIG-1 and Brex-PTIG-2 increased during the 0-5 d after injection and then decreased.This was consistent with the previous study.In a comparative study of in situ gels performed by Zhang et.al., the gel formulation with the best performance (better viscosity and best sustained-release  2 Pharmacokinetic parameters after a single administration of three Brex-PTIGs (mean±SD, n=15) *** p<0.001 compared to Brex-Sol, a p<0.001 compared to Brex-Sus, b p<0.05 and c p<0.01 compared to Brex-PTIG-1 characteristics) showed a slow increase and degradation of the gel weight in vivo [15].The residual weight of Brex-PTIGs at Day 60 all decreased to 10% (w/w) of the initial weight.
To investigate the physical status of the rats following subcutaneous administration, the body weights of the rats were monitored (Fig. 11).A slight decrease in body weight was observed during the first 6 hours after dosing.It is reasonable to assume that there was a lack of energy and poor appetite due to serial blood collecting from the retroorbital plexus.During the period from Day 1 to Day 60 of dosing, the rats' body weight increased continuously, with no significant difference from the saline group.It indicated that there was no significant effect on the body growth of rats after administration.
Organic solvents located in the in situ precipitation system may irritate the injection site.The inflammatory reaction is the main side effect after the local injection of in situ gel [46].Thus, we examined the compatibility and inflammatory effect of Brex-PTIGs at the injection site after a single subcutaneous injection.At first, the skin around the injection site was dissected and observed on 7 d after injection.As shown in Fig. 10B, the yellowish gel-like depot of Brex-PTIGs was with smooth and flat edges and was covered by a transparent biofilm.However, in Brex-Sol and Brex-Sus groups, signs of bleeding and inflammations were observed.A partial white fatty accumulation at the injection site was also observed.
Then, the skin tissues at the injection site were fixed, sliced, and stained with hematoxylin and eosin (HE) for histological analysis.As shown in Fig. 12, none of the Brex-PTIGs showed a severe inflammatory response at the injection site over the 21d following administration.However, for Brex-Sol and Brex-Sus groups, on the first day after injection, the structure of adipocytes was severely disrupted with massive cell necrosis, in addition, infiltration of lymphocytes, granulocytes, and other inflammatory cells into the tissues was also observed on Day 7 after injection, with the formation of large amounts of granulation and fibrous connective tissue.On Day 21, the skin conditions of all groups recovered to normal.These results indicated the Brex-PTIGs were more biocompatible than the solution and suspensions in vivo.Compared to the scanning electron microscopy results in Fig. 3 above, angular morphology in the microscopic state did not cause a severe inflammatory response, and in contrast, it was most pronounced in Brex-PTIG-3, which also had the slowest release rate of Brex (compared with other formulations).
The biocompatibility and mild inflammatory response of Brex-PTIGs might be owing to the biocompatible excipients and their reasonable adding amount.The matrix in Brex-PTIGs was all biocompatible, including S100 and SA.The solvents used in the Brex-PTIGs were MCT, ethanol, and NMP, which were seen as safe [46,50].As an FDA-approved solvent for injectable use, the safety of NMP could be effectively assured, and the low percentage of ethanol (less than 10% (v/v)) might help reduce skin irritation, as demonstrated in previous studies [36].The content of solvents in Brex-PTIGs was controlled strictly to ensure the safety of Brex-PTIGs.

Conclusion
In this study, an injectable in situ gel of Brexpiprazole (Brex) based on phospholipid-based phase transition consisting of S100, MCT/SA, NMP, and ethanol was constructed.Brex-PTIGs in solution state were suitable for injection, and when got in contact with water in vivo, they rapidly transformed to a semi-solid gel state.The Brex-PTIGs had a good sustained release of Brex in vivo for 60 d.The preliminary safety profile was also confirmed.In the following study, the polarization characteristics of the gel, the stability parameters, and the mechanism for sustained release still need to be explored.This study for the first time validated the feasibility of injectable in situ gel based on phospholipid-based phase transition for the sustained delivery of Brex, which holds high potential in developing a valid, safe, easy-producing, and low-cost preparation of Brex for the long-term treatment of schizophrenia.