Simvastatin Embedded into Poly(Lactic-Co-Glycolic Acid)-Based Scaffolds in Promoting Preclinical Bone Regeneration: A Systematic Review

: Simvastatin embedded into poly(lactic-co-glycolic acid) (PLGA)-based scaffolds can stimulate bone regeneration in preclinical models. However, the ideal pharmacological dose has not been evaluated. This systematic review reports on the simvastatin doses used in preclinical studies and evaluates the regeneration of critical-sized bone defects. References were selected in a two-phase process. Electronic databases (Embase, LILACS, LIVIVO, PubMed, SCOPUS, and Web of Science) and grey literature databases (Google Scholar, Open Grey, and ProQuest) were searched until September 2022. The risk of bias was considered to be low based on the SYRCLE tool. We identiﬁed four studies in rat, two in parietal and two in calvaria bone, one in mouse parietal bone, and one in rabbit femur bone. Simvastatin, ranging from 8 to 100 µ g, signiﬁcantly increased bone formation in ﬁve studies, as compared to the scaffold alone based on µ -computed tomography, histomorphometric, and radiography analysis. The median increase in bone formation caused by simvastatin was 2.1-fold compared to the PLGA-based scaffold alone. There was, however, no signiﬁcant correlation between the relative bone gain and the doses of simvastatin ( p = 0.37). The data suggest that relatively lower doses of simvastatin can consistently promote preclinical bone regeneration. However, the interpretation of these data must consider the heterogenicity of the PLGA-scaffolds, the defect anatomy, the observation period, and the evaluation method.


Introduction
Since the alveolar process depends on tooth function, this bone will undergo atrophy following the tooth extraction [1]. Consequently, the loss of tissue dimension can lead to clinical aspects that difficult, or even prevent, prosthetic rehabilitation due to the esthetic impairment and/or the limitation of installing the dental implant in the correct position [2]. Therefore, bone regeneration procedures have been performed in several clinical situations of tooth loss. However, vertical bone reconstructions are clinically unpredictable and hard to achieve. Autologous bone grafts have limitations, including the morbidity of the donor site, limited bone availability in some harvesting areas, high rates of bone remodeling, and unpredictable degradation rate over time. Consequently, approaches, including drug delivery systems with different graft materials and growth factors/active substances, have been proposed. However, despite the efforts made, biomaterials and surgical techniques

Protocol and Registration
This systematic review followed the checklist Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [19]. The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the identification number CRD42021206667.

Inclusion Criteria
The PICOS acronym (population, intervention, comparison, outcome, and type of study) was used to create the focused question of this systematic review; Population (P): Animals who received SIM embedded into PLGA-based scaffolds in critical bone defects. Intervention (I): Local delivery of SIM embedded into PLGA-based scaffolds in critical in vivo bone defects. Comparison (C): PLGA-based scaffolds without SIM. Outcome (O): SIM dose required for bone formation. Studies (S) were considered eligible when they met the following inclusion criteria: Evaluate the bone formation through the local delivery of SIM embedded into PLGA-based scaffolds in critical in vivo bone defects, as compared to PLGA-based scaffolds without SIM. No publication time restrictions were applied.

Exclusion Criteria
The following exclusion criteria were considered: (1) In vitro studies; (2) studies evaluating human patients (clinical trials); (3) studies evaluating non-critical in vivo bone defects; (4) studies evaluating SIM systemic administration; (5) studies evaluating scaffolds with a different composition than PLGA; (6) studies with insufficient data regarding bone formation or cytotoxic effect; (7) studies with less than 28 days of follow-up; (8) studies not published in the Roman Latin alphabet; (9) review articles, case reports, protocols, short communications, personal opinions, letters, posters, conference abstracts, or book chapters; (10) full text not available; and (11) duplicate data (e.g., dissertations/thesis in which correspondent published articles were available).

Information Sources
A detailed research strategy was developed for each following electronic database: Embase, Latin American and Caribbean Health Sciences (LILACS), Leibniz Information Centre for Life Sciences (LIVIVO), PubMed, SCOPUS, and Web of Science. As additional literature, a search strategy for Google Scholar web search (first 100 references), Open Grey, and ProQuest (Dissertations and Thesis) was elaborated. Other than that, reference lists of potentially relevant articles were hand-searched to identify any studies that could have been missed in the previous steps. All databases search was conducted in September 2022. Detailed search strategies are available in Figure A1. Reference lists of included studies were manually searched, as recommended by Greenhalgh and Peacock [20]. A software (EndNote X7, Thomson Reuters, Canada) was used to manage the references.

Study Selection
A two-phase selection process using online software was performed (Rayyan, Qatar Computing Research Institute, Qatar). In phase 1, two reviewers (E.B.M. and L.O.M) independently conducted title and abstract reading to identify potentially eligible studies. The same reviewers performed the full-text reading of eligible articles in phase 2. In both selection phases, disagreements were solved in a consensus discussion. A third reviewer (R.B.C.) was involved in making the final decision if a consensus was not reached. If data were missing or unclear, an attempt to contact the corresponding authors was made to resolve or clarify the issue.

Data Collection Process
Data collection was performed by one author (E.B.M) and a second author (L.O.M) cross-checked the information. Disagreements were resolved by discussion. If needed, a third author (R.B.C.) was involved in making the final decision. The following data were recorded: Study characteristics (author, year, and country of publication), population characteristics (total animals/defects, control group, test group, animal species, bone defect area, and bone defect dimension), scaffold properties (SIM dose, drug delivery system, and PLA/PGA ratio), and outcome measures (analyses methods, experimental time, main findings, and p-value). In the case of uncertainty, the authors were contacted.

Quality and Risk of Bias Assessment
The risk of bias (RoB) of the included articles was assessed independently by two reviewers (E.B.M. and L.O.M.) using the Systematic Review Centre for Laboratory Animal Experiments (SYRCLE) tool [21]. This tool is based on the Cochrane Collaboration RoB Tool. It has been adapted to evaluate the bias aspects in animal experiments aiming to assess the methodological quality of the studies. The possible answers to each of the RoB questions were "Yes", "No" or "Unclear'. Briefly, the following points and questions were considered: Selection bias (sequence generation, baseline characteristics, and allocation concealment), performance bias (random housing and blinding of study personnel), detection bias (random and blinding of outcome assessors), attrition bias (incomplete outcome data), reporting bias, and other sources of biases ( Figure 1).

Summary Measures
A qualitative analysis of the results based on the quantification of the bone formation in critical defects in pre-clinical models due to the SIM embedded into PLGA-based scaffolds was performed. The articles that described the quantification of the bone formation using histological, µ-computed tomographic (µCT), and/or radiographic analyses were considered.

Summary Measures
A qualitative analysis of the results based on the quantification of the bone formation in critical defects in pre-clinical models due to the SIM embedded into PLGA-based scaffolds was performed. The articles that described the quantification of the bone formation using histological, µ-computed tomographic (µCT), and/or radiographic analyses were considered.

Synthesis of Results
A qualitative analysis of the results based on the SIM dose embedded into PLGAbased scaffolds, required to promote bone formation in critical defects in pre-clinical models (reported or calculated), was performed. The statistical pooling of data using metaanalysis was planned if studies were considered sufficiently homogeneous regarding methodology and data availability.

Study Selection
In phase 1, 738 references were retrieved electronically from the following databases: EMBASE (169), LILACS (03), LIVIVO (75), PubMed (144), SCOPUS (96), Web of Science (67), ProQuest (43), Google Scholar (140), and Open Grey (01). Additional references were not identified manually. After removing duplicates, 266 references remained. Subsequently, title and abstract evaluation were performed, and 18 articles were included in phase 2 for full-text reading. Finally, after full-text analyses, six studies matched the inclusion criteria and were included for further analyses, while 12 articles were excluded Were the groups similar at baseline or were they adjusted for confounders in the analysis?
Was the allocation adequately concealed?
Were the animals randomly housed during the experiment?
Were the caregivers and/or investigators blinded from knowledge which intervention each animal received during the experiment?
Were animals selected at random for outcome assessment?
Was the outcome assessor blinded?
Were incomplete outcome data adequately addressed?

SELECTIVE OUTCOME REPORTING
Are reports of the study free of selective outcome reporting?

OTHER BIAS
OTHER SOURCES OF BIAS Was the study apparently free of other problems that could result in high risk of bias? BLINDING Figure 1. Graphical risk of bias summary assessed by systematic review center for laboratory animal experiments.

Synthesis of Results
A qualitative analysis of the results based on the SIM dose embedded into PLGAbased scaffolds, required to promote bone formation in critical defects in pre-clinical models (reported or calculated), was performed. The statistical pooling of data using meta-analysis was planned if studies were considered sufficiently homogeneous regarding methodology and data availability.

Study Selection
In phase 1, 738 references were retrieved electronically from the following databases: EMBASE (169), LILACS (03), LIVIVO (75), PubMed (144), SCOPUS (96), Web of Science (67), ProQuest (43), Google Scholar (140), and Open Grey (01). Additional references were not identified manually. After removing duplicates, 266 references remained. Subsequently, title and abstract evaluation were performed, and 18 articles were included in phase 2 for full-text reading. Finally, after full-text analyses, six studies matched the inclusion criteria and were included for further analyses, while 12 articles were excluded (Table A1). Figure 2 shows a flowchart describing the complete process of identification, inclusion, and exclusion of studies.
Appl. Sci. 2022, 12, x FOR PEER REVIEW 5 of 14 (Table A1). Figure 2 shows a flowchart describing the complete process of identification, inclusion, and exclusion of studies.

Study Characteristics
The characteristics of the selected studies are shown in Table 1 and Figure 3. The included studies were published in the English language from 2013 up to 2017. The studies were conducted in China (3) and Brazil (3). Different experimental animal models were tested, including rats (4), rabbits (1), and mice (1). In total, 333 animals were analyzed. As expected, due to the selection criteria, all the studies evaluated critical-size defects to assess the osteogenic capacity of the implanted drug delivery systems. The bone defects were made in the parietal [22][23][24], calvaria [8,18], and femur [25] bones. PLGA-based scaffolds, PLGA-based scaffolds embedding SIM, and no treatment were used in the bone defects. The bone formation was evaluated by histological [8,23], µCT [18,24,25], and radiographic [22] analyses.

Study Characteristics
The characteristics of the selected studies are shown in Table 2 and Figure 3. The included studies were published in the English language from 2013 up to 2017. The studies were conducted in China (3) and Brazil (3). Different experimental animal models were tested, including rats (4), rabbits (1), and mice (1). In total, 333 animals were analyzed. As expected, due to the selection criteria, all the studies evaluated critical-size defects to assess the osteogenic capacity of the implanted drug delivery systems. The bone defects were made in the parietal [22][23][24], calvaria [8,18], and femur [25] bones. PLGA-based scaffolds, PLGA-based scaffolds embedding SIM, and no treatment were used in the bone defects. The bone formation was evaluated by histological [8,23], µCT [18,24,25], and radiographic [22] analyses.

Risk of Bias (RoB) in Individual Studies
The RoB was assessed using the SYRCLE tool [21]. In summary, the RoB was considered low for most items evaluated in the studies (Table 3). However, all the included studies failed to report if the allocation sequence was adequately generated and applied, as well as if the caregivers/investigators and outcome assessors were blinded to knowledge of the received intervention of each animal during the experiment [8,18,[22][23][24][25]. Additionally, the question related to the animals selected at random for outcome assessment was unclear for all the studies [8,18,[22][23][24][25].

Results of Individual Studies
Assaf et al. (2013) [23] evaluated 32 male Wistar rats (250-300 g) divided in two groups (n = 16 each). In each rat, two critical-size defects of 5.3 mm in diameter were created in the dorsal part of the parietal bone. The defect on the right side was the experimental group, while the left side was the control (no treatment). In the first group, the right-side defect was filled with a PLGA scaffold, and the second group received a PLGA-based scaffold embedding SIM (20 µg/scaffold). According to the histological analysis, the PLGA-SIM group promoted the highest length of the bone formation, filling the defects on days 28 and 56 (0% and 96%, respectively, p < 0.05). The control group showed 38% and 52% of bone formation on days 28 and 56, respectively, while the PLGA group filled 71% of the defects on days 28 and 56. group promoted the highest area of bone formation (1.5 × 10 4 mm 2 , p < 0.05). Control, PLGA, PLGA-MSC, and PLGA-SIM-MSC groups showed 4 × 10 3 mm 2 , 7 × 10 3 mm 2 , 5 × 10 3 mm 2 , and 2 × 10 3 mm 2 of bone formation, respectively (p > 0.05). Zhang et al. (2015) [25] created critical defects (6 mm × 10 mm) on the lateral femoral condyle of 30 New Zealand rabbits weighing about 1000 g that were divided into three groups as follows: Sham-operation; PLGA-calcium phosphate composite (CPC); SIM-PLGA-CP (100 µg/scaffold). The bone formation was determined using µCT analysis on days 42 and 84. The SIM-PLGA-CPC group promoted the highest bone formation (25.78 ± 6.89% and 68.0 ± 11.62% on days 42 and 84, respectively, p < 0.05). The shamoperation group showed 3.40 ± 2.25% and 6.10 ± 4.48% of bone defects repaired on days 42 and 84, respectively. The PLGA-CPC group demonstrated a bone coverage of 12.89 ± 5.75% and 29.24 ± 9.25% on days 42 and 84, respectively.

Synthesys of Results
The data were normalized and a correlation analysis was performed (Figures 4 and 5; r= −0.48, p = 0.336. These data suggest that there is no obvious impact of the SIM dose on the overall stimulation of bone regeneration. The impact of the species (mouse/rat versus rabbit) and defect location (femur versus calvaria/parietal) was also not significant (r= −0.46, p = 0.429. Thus, it is hard to predict the ideal dose of SIM using PLGA-based scaffolds. It does not require raising the SIM doses above 10 µg, at least in rodent models. Nevertheless, this analysis must be interpreted with caution due to the heterogeneity of studies concerning the biomaterial composition, preclinical model, and SIM dose. PLGA-HA group demonstrated 0.5% and 3.9% on days 28 and 56, respectively. Liu et al. (2014) [24] evaluated a 4-mm diameter critical-sized defect created at the left side of the calvarium of 32 ICR mice (4 weeks old), divided into four groups: PLGA scaffold; PLGA-SIM (35 μg/scaffold); PLGA-stromal cell-derived factor 1 (SDF1); and PLGA-SIM-SDF1. According to the µ CT performed 42 days after the implantation, the PLGA-SIM-SDF1 group promoted the highest volume of bone formation (1.1 mm 3 , p < 0.05). Conversely, the PLGA group did not lead to bone formation. PLGA-SIM and PLGA-SDF1 showed 0.18 mm 3 and 0.41 of bone formation, respectively. Mendes et al. (2017) [8] evaluated an 8-mm bone defect in the calvaria of 35 Wistar rats (three months old) divided into five groups: control (blank default); PLGA-based scaffold; PLGA-SIM (40 μg/scaffold); PLGA-mesenchymal stem cells (MSC); and PLGA-SIM-MSC. After 56 days, according to the histomorphometric analyses, the PLGA-SIM group promoted the highest area of bone formation (1.5 × 10 4 mm 2 , p < 0.05). Control, PLGA, PLGA-MSC, and PLGA-SIM-MSC groups showed 4 × 10 3 mm 2 , 7 × 10 3 mm 2 , 5 × 10 3 mm 2 , and 2 × 10 3 mm 2 of bone formation, respectively (p > 0.05). Zhang et al. (2015) [25] created critical defects (6 mm x 10 mm) on the lateral femoral condyle of 30 New Zealand rabbits weighing about 1000 g that were divided into three groups as follows: Sham-operation; PLGA-calcium phosphate composite (CPC); SIM-PLGA-CP (100 μg/scaffold). The bone formation was determined using µ CT analysis on days 42 and 84. The SIM-PLGA-CPC group promoted the highest bone formation (25.78  6.89% and 68.0  11.62% on days 42 and 84, respectively, p < 0.05). The sham-operation group showed 3.40  2.25% and 6.10  4.48% of bone defects repaired on days 42 and 84, respectively. The PLGA-CPC group demonstrated a bone coverage of 12.89  5.75% and 29.24  9.25% on days 42 and 84, respectively.

Synthesys of Results
The data were normalized and a correlation analysis was performed (Figures 4 and 5; r= −0.48, p = 0.336. These data suggest that there is no obvious impact of the SIM dose on the overall stimulation of bone regeneration. The impact of the species (mouse/rat versus rabbit) and defect location (femur versus calvaria/parietal) was also not significant (r= −0.46, p = 0.429. Thus, it is hard to predict the ideal dose of SIM using PLGA-based scaffolds. It does not require raising the SIM doses above 10 µ g, at least in rodent models. Nevertheless, this analysis must be interpreted with caution due to the heterogeneity of studies concerning the biomaterial composition, preclinical model, and SIM dose.

Discussion
Considering that bone regeneration is still a challenge in oral and maxillofacial surgeries, the search for biomaterials and predictable techniques that enhance bone regeneration continues [26]. Since SIM has been proposed to support bone formation, drug deliv-

Discussion
Considering that bone regeneration is still a challenge in oral and maxillofacial surgeries, the search for biomaterials and predictable techniques that enhance bone regeneration continues [26]. Since SIM has been proposed to support bone formation, drug delivery systems were evaluated [27,28]. This systematic review evaluates the SIM dose embedded into PLGA-based scaffolds necessary to promote bone regeneration in preclinical models. We observed that, from the six included studies, five studies confirmed SIM embedded into PLGA-based scaffolds promotes bone formation. The required dose ranged from 8 to 50 µg SIM/scaffold in rodents, and 100 µg SIM/scaffold in rabbit. In one study, SIM failed to support bone regeneration. This review may contribute to the experimental design of future studies on SIM in bone regeneration.
Adequate drug release from scaffolds is critical for bone formation, and efforts to find an appropriate SIM dosing and delivery system are made [29]. High SIM doses are associated with an exacerbated inflammatory responses and impaired bone formation [24,30,31], also due to cytotoxicity and blocked cholesterol synthesis [32,33]. Conversely, low SIM doses may not reach the pharmacologically relevant concentration. In this context, three articles evaluated SIM release varying from 4% (1 day) [8], 15% (2 days) [18], to >60% (7 days) [25]. Moreover, SIM release of approximately 30% in 30 days [8], 23% in 56 days [18], and 100% in 21 days [25] was reported. Drug release from PLGA can be controlled by varying the molecular weight and the ratio of lactide to glycolide [11]. Different ratios of lactide to glycolic were used to produce the PLGA scaffolds. The proportion most often used was 50:50 [23,25,28], followed by 82:18 [22], 85:15 [18], and 75:25 [24]. One article did not report the lactide to glycolic ratio [8]. Thus, the lactide-to-glycolic rations may affect the SIM release kinetic and, consequently, the bone regeneration capacity in vivo.
Concerning the RoB judgment, a low RoB was attributed to most items evaluated. Low RoB judgments denote that none or minor methodological flaws occurred in the assessed studies. Consequently, none or small deviations from the true effect estimation befallen, providing confidence in interpreting the results [34]. No study reports whether or not the allocation sequence was adequately generated and applied, as well as whether or not the caregivers/investigators and outcome assessors were blinded to the intervention [8,18,22,24,25]. The animals selected at random for outcome assessment also remained unclear [8,18,22,24,25]. Future studies should put more emphasis on reporting methodological details.
Regarding the limitations of this review, only PLGA-based scaffolds were evaluated. Thus, further studies assessing the dose and release of SIM embedded into different scaffolds are required. The included studies used µCT, histological, and radiographical analysis to quantify the bone regeneration. Due to high-resolution 3D information, µCT is the most reliable to evaluate bone regeneration [35,36]. Moreover, µCT provides information concerning the volume, texture, and external and internal structures of the implanted scaffold. Histological analysis, however, is ideal to study the cellular aspects of bone regeneration, hence any potential adverse effects of the scaffold [37]. Further studies assessing the dose of SIM embedded into scaffolds with different chemical compositions than PLGA are suggested. Additionally, the evaluation of SIM doses used clinically is proposed.

Conclusions
The collected data suggest that simvastatin at 10 to 100 µg/scaffold can approximately double the amount of bone regeneration in rodent and rabbit models. Moreover, the species and the location of the bone defect did not affect the simvastatin dose for stimulating bone regeneration. However, these data must be interpreted under the premise of the heterogenicity of PLGA-scaffolds, the defect anatomy, the observation period, and the evaluation method.