Effectiveness of mesenchymal stem cell-conditioned medium in bone regeneration in animal and human models: a systematic review and meta-analysis

Given the limitations of current therapies for the reconstruction of bone defects, regenerative medicine has arisen as a new therapeutic strategy along with mesenchymal stem cells (MSCs), which, because of their osteogenic potential and immunomodulatory properties, have emerged as a promising alternative for the treatment of bone injuries. In vivo studies have demonstrated that MSCs have a positive effect on regeneration due to their secretion of cytokines and growth factors that, when collected in conditioned medium (MSC-CM) and applied to an injured tissue, can modulate and promote the formation of new tissue. To evaluate the effectiveness of application of conditioned medium derived from mesenchymal stem cells in bone regeneration in animal and human models. We conducted a systematic review with a comprehensive search through February of 2018 using several electronic databases (MEDLINE, EMBASE, SCOPUS, CENTRAL (Ovid), and LILACS), and we also used the “snowballing technique”. Articles that met the inclusion criteria were selected through abstract review and subsequent assessment of the full text. We assessed the risk of bias with the SYRCLE and Cochrane tools, and three meta-analyses were performed. We included 21 articles, 19 of which used animal models and 2 of which used human models. In animal models, the application of MSC-CM significantly increased the regeneration of bone defects in comparison with control groups. Human studies reported early mineralization in regenerated bones, and no bone resorption, inflammation, nor local or systemic alterations were observed in any case. The meta-analysis showed an overall favorable effect of the application of MSC-CM. The application of MSC-CM to bone defects has a positive and favorable effect on the repair and regeneration of bone tissue, particularly in animal models. It is necessary to perform additional studies to support the application of MSC-CM in clinical practice.


Background
Reconstruction of bone defects generated by fractures, tumors, infections or congenital diseases is a real challenge in oral and maxillofacial surgery and orthopedics. Although bones have an ability to repair and regenerate, in bone lesions of large size, the process of healing fails, and such injuries do not repair themselves spontaneously (Oryan et al., 2013). Current therapies have focused on the placement of grafts and bone substitutes, which are widely used but also have some limitations and disadvantages in reconstruction of bone defects that exceed the critical size (Oryan et al., 2013;D, 2010;Shrivats et al., 2014). This has stimulated the search for new therapeutic alternatives to produce adequate regeneration and rehabilitation of these defects (Padial et al., 2015;Bertolai et al., 2015). This may give rise to regenerative medicine, which seeks to repair or replace damaged cells and tissues of an organ to restore its normal functioning. Regenerative medicine uses tools from tissue engineering, gene therapy and cellular therapy, the latter of which is mainly represented by the use of mesenchymal stem cells (MSC) (de Santana et al., 2015;Berthiaume et al., 2011;Tatullo et al., 2015). MSCs are a type of adult stem cells, which are multipotent and thus can self-regenerate, proliferate and differentiate into multiple cell lineages (Saeed et al., 2016;Wen et al., 2013;Monaco et al., 2011). There have been multiple reports in the literature revealing the therapeutic effects of the application of MSC for bone regeneration in animal and human models (Cancedda et al., 2007;Ramamoorthi et al., 2015;Wang et al., 2012a). Currently, it has been suggested that their main mechanism of action in tissue regeneration and repair through the release of growth factors, cytokines and extracellular matrix molecules, which have a paracrine effect on host cells, modulating endogenous cell migration, angiogenesis, and cell differentiation, and inducing the repair and regeneration of injured tissues (Liang et al., 2014;Ivanova et al., 2016;Chaparro & Linero, 2016;Linero & Chaparro, 2014). Secreted factors are referred to as a secretome and can be found in the medium where the mesenchymal stem cells are cultivated, known as a conditioned medium (MSC-CM). It has been shown that MSC-CM exerts a beneficial effect on regeneration of bone and tissue, as the secretome participates in stimulation of multiple cellular functions (Ivanova et al., 2016;Clough et al., 2015). Published systematic reviews have evaluated the application of MSC-CM for the treatment of injuries and pathologies in several organs, such as acute renal failure, myocardial infarction, liver failure, lung disease, and nerve injury, where MSC-CM significantly promoted the repair and regeneration of tissue injuries and/or damaged organs (Muhammad et al., 2018;Akyurekli et al., 2015;JA, 2014). However, there have been no systematic reviews specifically evaluating the application of MSC-CM particularly in bone regeneration. Accordingly, the objective of this review was to assess the effectiveness of application of conditioned media derived from mesenchymal stem cells in bone regeneration in animal and human models.

Methods
This systematic review was designed to answer the following question: What is the effectiveness of application of conditioned medium derived from mesenchymal stem cells in bone regeneration in animal and human models?

Search strategy
We developed a search strategy to identify the studies published before February of 2018 in the electronic databases MEDLINE (OVID), EMBASE, CENTRAL (OVID) EBM Reviews -Cochrane Central Register of Controlled Trials, SCOPUS, Virtual Health Library (IBECS/LI-LACS/CUMED) using the following search terms: "Conditioned medium", "Mesenchymal stem cell", "Paracrine communication", "Secretome", "Tissue engineering", "Regenerative medicine", "Bone regeneration", "Bone repair", "Humans", "Animal model", "Experimental study", "Clinical trial", "Clinical study" and "Case reports". The "snowballing technique" was also used as a search strategy in the lists of references of studies found in electronic databases. (See Appendix 1: Electronic search strategy).

Study selection
The titles and abstracts of studies identified in the systematic search were evaluated independently by two researchers (MB and IL). Disagreements in the selection of articles were resolved by discussion and consensus. After the initial selection, potentially relevant articles were retrieved for a full-text assessment.

Eligibility criteria
We included all experimental in vivo studies that evaluated bone regeneration after the application of MSC-CM in animal and human models reported in articles written in both English and Spanish with a publication date after 2000 and that reported measurable clinical, radiographic and/or histologic outcomes. We excluded studies that applied conditioned medium for regeneration of other tissues than bone, derived from other cell types, in vitro studies and review articles. (See Appendix 2: Exclusion criteria).

Data extraction
After evaluation of full-text studies that met the inclusion criteria, we performed data extraction using a form developed for this review, where we obtained the following information: authors, publication year, objective, number (n) and population characteristics (age, sex, strain), methods and study design, intervention characteristics (cells used, preparation of conditioned medium, administration method, type of bone defect evaluated, duration of intervention, established groups (MSC-CM, comparison and control), outcomes assessed (bone regeneration, secondary outcomes, tests conducted for the measurement of results, most important results, statistical methods) and conclusions of the studies. Studies were grouped according to the following experimentation models: animal models and human models. (See Appendix 3: Data extraction form).

Quality assessment
We assessed the risk of bias of animal studies with the SYstematic Review Centre for Laboratory animal Experimentation (SYRCLE) tool (Hooijmans et al., 2014), and of human studies with the Cochrane risk of bias tool (Higgins & Green, 2011). These risks of bias were classified into high, low or unclear. We used Revman 5.3 software to perform the graphic summary of risk of bias in the studies (Review Manager (RevMan) [Computer program], 2014). When studies were not experimental, we conducted a quality assessment with the "CARE checklist" tool for case reports (Gagnier et al., 2013).

Intervention effect measure and synthesis of results
We performed a narrative description and an analysis of characteristics, findings and primary and secondary outcomes from the studies. Bone regeneration was reported in the original measures used in the studies. The results were treated with standardized mean difference (SSMD) due to the diversity in the measure instruments, measure units used, comparative interventions and the time when the effect was evaluated. These comparative measures were reported with their respective confidence intervals (CI) at 95%. Calculation were made by Revman 5.3 (Review Manager (RevMan) [Computer program], 2014). We explored statistical heterogeneity using I2 and Chi2 tests for bone regeneration outcome, we performed four forest plots, three of which were generated with a global diamond (meta-analysis). The studies were grouped according to the measure of the effect used and the time at which the evaluation of bone regeneration was performed. We assessed the percentage of bone regeneration at 2 and 4 weeks and the volume of bone regeneration at 8 weeks.

Assessment of publication biases
Due to the limited number of studies included within the quantitative evidence of an overall effect of intervention on bone regeneration (meta-analysis), it was not possible to explore reporting bias using "funnel plots".

Assessment of the methodological quality of the systematic review
All phases of this systematic review were performed and reviewed according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) (Moher et al., 2009). (See Appendix 4: PRISMA checklist).

Search results
The searches yielded a total of 6500 articles after removing duplicates. After the first screening, 6473 studies didn't met the eligibility criteria, a total of 27 full-text articles were reviewed, from which 6 articles were excluded (Shang-Chun et al., 2016;Otsuru et al., 2018;Li et al., 2018;Byeon et al., 2010;Sakaguchi et al., 2017;Pethő et al., 2018). (See section 2.3: Eligibility criteria and Appendix 2: Exclusion criteria).
We selected 21 articles that met the inclusion criteria, 19 of which described animal models and 2 of which described human models (a case report and a phase I clinical trial) (Fig. 1).

Description of included studies
The first published study in humans is a case report (Katagiri, 2016) (Katagiri et al., 2016) that evaluated the safety and use of MSC-CM for alveolar bone regeneration in eight partially edentulous patients aged 45 to 67 years, which required bone augmentation, including maxillary sinus floor elevation (SFE), guided bone regeneration (GBR) and socket preservation (SP) for subsequent placement of dental implants. The second study was a phase I clinical trial (Katagiri, 2017ª) (Katagiri et al., 2017a) that evaluated the safety of use of the secretome of bone marrow-derived mesenchymal stem cells (MSC-CM) for surgical procedures of maxillary sinus floor elevation and bone grafting in 6 systemically healthy, partially edentulous patients.
We found 19 experimental studies using animal models where the conditioned media derived from mesenchymal stem cells were applied to regeneration of bone tissue. The characteristics of these studies are detailed in Table 1.

Application of MSC-CM
In the human models: Katagiri, 2016(Katagiri et al., 2016 performed procedures of maxillary sinus floor elevation and guided bone regeneration with an implant of beta-tricalcium phosphate (B-TCP) soaked in MSC-CM and socket preservation with an implant of atelocollagen sponge soaked in MSC-CM. Katagiri, 2017ª (Katagiri et al., 2017a also performed maxillary sinus floor elevation procedures with an implant of B-TCP soaked in MSC-CM, and in the control group, B-TCP without MSC-CM was implanted. In the animal models: In ten studies, MSC-CM were applied to circular bone defects, eight of which applied the MSC-CM in calvarial bone defects of 5 mm in diameter (Chang et al., 2015;Katagiri et al., 2013;Katagiri et al., 2017b;Katagiri et al., 2017c;Osugi et al., 2012;Qin et al., 2016;Sanchooli et al., 2017;Wang et al., 2015), one in calvarial bone defects of 8 mm in diameter (Wang et al., 2012b), and another in bilateral bone defects of 10 mm in diameter in the mandibular angles of rabbits (Linero & Chaparro, 2014). In two studies, MSC-CM were applied in periodontal bone defects (Inukai et al., 2013;Kawai et al., 2015). In one study MSC-CM was applied in maxillary sinus floor elevation procedures . In two studies, MSC-CM was applied in models of distraction osteogenesis of tibia (Ando et al., 2014;Xu et al., 2016). In two more studies MSC-CM was evaluated in fractures, one in a femur (Furuta et al., 2016) and the other in the middle third of the fibula in rats with diabetes (Wang et al., 2012b). One study evaluated osseointegration of an implant soaked in MSC-CM in a socket created in a femur (Tsuchiya et al., 2013), and another study evaluated the therapeutic  (Table 1).

Risk of bias of studies
In 19 studies (95%) that underwent the assessment, a high risk of bias was observed for most parameters.
Only one study scored a low risk of bias in 8 of the 9 parameters evaluated (Sanchooli et al., 2017) (Fig. 2).
The bone regeneration was reported in terms of the percentage area of newly formed bone over the total area of bone defect (9 studies) (Linero & Chaparro, 2014;Ando et al., 2014;Chang et al., 2015;Katagiri et al., 2013;Katagiri et al., 2015;Katagiri et al., 2017b;Katagiri et al., 2017c;Osugi et al., 2012;Tsuchiya et al., 2015), the volume of new regenerated bone relative to the total volume of bone defect (3 studies) (Qin et al., 2016;Sanchooli et al., 2017;Wang et al., 2012b), the ratio of bone volume over the volume of tissue (one study) (Xu et al., 2016), the fractions of area and volume of newly formed bone tissue (2 studies) (Inukai et al., 2013;Wang et al., 2015), the volume of the sequestra in the model of bisphosphonate-related osteonecrosis of the jaw (one study) , the osseointegration of the implant measured as the rate of bone contact (one study) (Tsuchiya et al., 2013) and the presence of bridging callus on two cortices in fractures (one study) (Furuta et al., 2016). Only one study reported bone regeneration in a qualitative manner by using the findings of histological analyses  ( Table 2).

Effect of MSC-CM on bone regeneration
In the studies performed in human models (Katagiri et al., 2016;Katagiri et al., 2017a), the radiographic images and CT showed an early mineralization of regenerated bone without bone resorption or evident inflammation of maxillary sinus membrane. Histological analysis showed increased formation of new bone tissue as well as increased vascularity of regenerated area with little infiltration of inflammatory cells in comparison with the control cases, where the formation of new bone was significantly lower and there was a greater inflammatory infiltrate.
In general, studies in animal models reported that application of MSC-CM to bone defects significantly increased regeneration of bone tissue in comparison with other intervention or control groups (Linero & Chaparro, 2014;Chang et al., 2015;Furuta et al., 2016;Inukai et al., 2013;Katagiri et al., 2013;Katagiri et al., 2017b;Katagiri et al., 2017c;Kawai et al., 2015;Osugi et al., 2012;Qin et al., 2016;Sanchooli et al., 2017;Tsuchiya et al., 2015;Wang et al., 2012b;Wang et al., 2015). Studies that carried out maxillary sinus floor elevation reported early mineralization in the grafted area upon application of conditioned media . The studies that evaluated bone regeneration during distraction osteogenesis reported that the secretome of MSCs accelerated the formation of new bone callus and bone healing, shortening the period required for treatment (Ando et al., 2014;Xu et al., 2016). In models of bone fracture, conditioned medium helped to improve new bone formation, angiogenesis and consolidation of the fracture (Furuta et al., 2016;Wang et al., 2012b). The studies that evaluated bone regeneration in periodontal defects, reported that application of conditioned medium promoted differentiation of stem cells to osteoblastic lineage, endogenous cellular migration and bone regeneration, showing a large amount of new bone and cement formation and minimal infiltration of inflammatory cells (Inukai et al., 2013;Kawai et al., 2015). In the study that evaluated therapeutic effects of MSC-CM in rats with bisphosphonate-related osteonecrosis of the jaw (BRONJ), 63% of the open sockets healed with full coverage by soft tissue . In the study that evaluated implant osseointegration, it was reported that the removal torque was significantly higher in the group where the MSC-CM were applied, than in control groups (Tsuchiya et al., 2013) (Table 2).
Of the 21 articles included, 5 did not report the volume of CM used and of the remaining 17, only 4 report the protein concentration (Linero & Chaparro, 2014;Tsuchiya et al., 2013;Wang et al., 2015;Xu et al., 2016). The non-reporting of MSC-CM dose used, the variability in the amount of CM applied, but above all, not identifying the concentration of proteins contained in the applied conditioned medium, prevent a comparative analysis of the studies and therefore to find a relation between the dose of MSC-CM and therapeutic effectiveness.
In all the studies that performed histological analysis, it was observed that in bone defects treated with MSC-CM there was a greater formation of new, primarily mineralized, regenerated bone with little or no infiltration of inflammatory cells, whereas the control groups showed reduced formation of bone tissue with less mineralization, greater amounts of connective tissue and increased infiltration of inflammatory cells (Linero & Chaparro, 2014;Ando et al., 2014;Inukai et al., 2013;Katagiri et al., 2013;Katagiri et al., 2015;Katagiri et al., 2017b;Katagiri et al., 2017c;Kawai et al., 2015;Osugi et al., 2012;Qin et al., 2016;Tsuchiya et al., 2015;Wang et al., 2012b;Xu et al., 2016).
Most studies reported that cytokines and growth factors contained in conditioned medium act synergistically to stimulate the migration and proliferation of osteoprogenitor cells, promote osteogenesis and bone regeneration and improve the early vascularization (Linero & Chaparro, 2014;Chang et al., 2015;Kawai et al., 2015). MSC-CM contains a mixture of multiple growth factors at relatively low concentrations that promote bone regeneration without causing a severe inflammatory response (Katagiri et al., 2016;Katagiri et al., 2017a;Ando et al., 2014;Chang et al., 2015;Furuta et al., 2016).

Meta-analysis of MSC-CM effect on bone regeneration
We grouped 7 studies that shared similar characteristics to evaluate through a meta-analysis of the effect of MSC-CM on bone regeneration in terms of the amount of newly formed tissue and time of tissue regeneration. (Fig. 3) details the effect of the application of MSC-CM compared with PBS control at 2 weeks Katagiri et al., 2017b;Katagiri et al., 2017c). We observed an overall favorable effect of MSC-CM with SMD: 3.16 (95% CI 2.42, 3.49), which indicates a significant difference between the MSC-CM and PBS groups. Regarding the comparison of MSC-CM with an empty defect, the favorable MSC-CM effect was maintained (SMD 4.09, 95% CI 1.82 to 6.36), indicating a significant difference between the MSC-CM and empty defect groups Katagiri et al., 2017b;Katagiri et al., 2017c) (Fig. 4).
At 4 weeks of bone regeneration, (Fig. 5) details the effect of the application of MSC-CM compared with PBS. In Osugi 2012(Osugi et al., 2012, there was a favorable MSC-CM effect with a statistically significant difference (SMD: 8.69, 95% CI 2.55, 14.82);  also presented a favorable MSC-CM effect; however, the difference between the two interventions was not significant (SMD: 1.30, 95% CI -0.35, 2.95). Due to high heterogeneity (I2: 81%), it was not possible to obtain the overall effect of the intervention.
( Fig. 6) details the results of analysis of the overall effect of the intervention after the application of MSC-CM compared with the implantation of scaffolds (gel of type 1 collagen, Hydrogel) as measured by the volume of regenerated bone at 8 weeks (Qin et al., 2016;Sanchooli et al., 2017) when the favorable MSC-CM effect was maintained with statistically significant differences (SMD: 1.78, 95% CI 0.77, 2.78).

Inflammatory response
In the human studies (Katagiri et al., 2016;Katagiri et al., 2017a), the clinical observations and blood tests showed no abnormal findings except for lesser signs of inflammation after surgery. No local or systemic alterations were observed, and no patient showed abnormal swelling or delayed healing.
Animal studies reported that there was no inflammatory response to the application of MSC-CM, and histological analyses showed reduced infiltration of inflammatory cells in MSC-CM groups in comparison with other groups (Inukai et al., 2013;Katagiri et al., 2013;Katagiri et al., 2017c;Kawai et al., 2015;Osugi et al., 2012;Xu et al., 2016).

Angiogenesis
In Katagiri, 2017(Katagiri et al., 2017b, the results indicated that the presence of vascular endothelial growth factor (VEGF) in MSC-CM promoted the migration of endothelial cells and endogenous stem cells, which allowed the formation of more blood vessels and bone tissue in the defect. They also observed that neutralization of VEGF in MSC-CM abolished angiogenesis, which caused only a minor migration of endogenous stem cells into the defect and reduced new bone formation. Similarly,  demonstrated that MSC-CM strongly promoted angiogenesis by increasing the expression levels of angiogenic markers, such as VEGF-A, ANG-1 and ANG-2, in MSCs cultured with MSC-CM. In addition, the results in Osugi 2012 (Osugi et al., 2012) indicated that MSC-CMs have the potential to mobilize endogenous MSCs to promote angiogenesis and bone regeneration.

Antiresorptive activity
In Ogata 2015 , the application of MSC-CM in rats with induced BRONJ generated an effect of maintaining the osteoclast function. The results indicated that 63% of rats with BRONJ in the MSC-CM group healed with a full coverage of connective tissue, while in the control group, exposed necrotic bone and inflamed soft tissue were observed. The anti-apoptotic, anti-inflammatory and angiogeneic effects of MSC-CM dramatically regulated the turnover of local bone, generating positive results in the treatment of BRONJ.

Discussion
Bone regeneration is a physiological process that requires the migration and proliferation of specific cells in a biological substrate of soluble factors and proteins, which coordinate the formation of new tissue, thus restoring bone structure and function. Local unfavorable conditions, such as inadequate blood supply, damage to the surrounding soft tissues, mechanical instability, extensive loss of bone tissue and local infection, cause a delay in the repair process and persistence of bone defects (Rosset et al., 2014). Although the exact mechanisms that regulate the process of bone regeneration at the molecular level are not yet fully understood (Dimitriou et al., 2011), several methods have been proposed for bone reconstruction, ranging from autografts, allografts, xenografts and bone substitutes (Pilipchuk et al., 2015). These treatment strategies have several drawbacks. An autologous bone graft is widely used for its osteoinductive, osteoconductive and osteogenic properties and immunogenic compatibility, but this implies the need for a double surgical procedure, which causes morbidity at the donor site, thus making it difficult to use (Goulet et al., 1997); in addition, the absence of cell populations in allografts and xenografts results in poor osteogenic and osteoinductive properties (Padial et al., 2015). To overcome these limitations, regenerative medicine aims to replace or regenerate tissues or organs to restore and stabilize their normal functions (Mason & Dunnill, 2008) using different tools, such as tissue engineering, gene therapy, cell therapy and therapy based on growth factors. An interest in cell therapy and specifically in mesenchymal stem cells and their clinical application has grown exponentially in the past 25 years (Le Blanc & Davies, 2018). MSCs are relatively easy to harvest and expand ex vivo, are able to modulate the immune system, and are able to repair injured tissues in particular; therefore, MSCs have become an attractive source for many applications in regenerative medicine (Le Blanc & Davies, 2018;Klyushnenkova et al., 2005;Caplan & Correa, 2011). Several studies showed the beneficial effects of stem cell therapy in diseases such as osteoarthritis (Yang et al., 2015), acute myocardial infarction , wound healing (Yoshikawa et al., 2008), kidney damage (Ma et al., 2013), peripheral nerve injury (Wang et al., 2009), bone defects, and others (Linero & Chaparro, 2014). Thanks to the large amount of scientific research on in vitro and in vivo models and 799 clinical trials reported by the US National Institutes of Health (NIH) (clinical trials.gov) (consultation carried out in June 2018), we know that MSC therapy is a safe and effective method for treatment of certain diseases and/ or conditions. Originally, it was hypothesized that due to their proliferative and multipotent capacities, transplanted stem cells differentiated into the cells of interest and replaced the injured tissue (Spees & Lee, 2016;Ankrum & Karp, 2010); however, the results of animal studies and clinical trials have demonstrated that a curative effect can be attributed to their ability to secrete growth factors, cytokines, chemokines, and extracellular matrix molecules at the receptor site, which modulate endogenous cell migration and stimulate angiogenesis and differentiation of the patient's stem cells, thus inducing the formation of new tissues (Muhammad et al., 2018;Chen et al., 2008). The secreted factors are referred to as the secretome and can be found in the environment where mesenchymal stem cells grow; that is, mesenchymal stem cell-conditioned medium (MSC-CM). MSC-CM contains the regenerative agents capable of promoting and modulating the formation of new tissues (Muhammad et al., 2018;JA, 2014). The application of MSC-CM has been shown to be effective in diseases such as focal cerebral ischemia (Inoue et al., 2013), Alzheimer's disease (Mita et al., 2015), acute renal failure (Matsushita et al., 2017), rheumatoid arthritis (Ishikawa et al., 2016), diabetes (Izumoto-Akita et al., 2015), and other diseases (Muhammad et al., 2018;Akyurekli et al., 2015;Shimojima et al., 2016;Yamaguchi et al., 2015;Wakayama et al., 2015;Fukuoka & Suga, 2015) and in conditions that affect the bone tissue, such as nonunion fractures and bone defects (Linero & Chaparro, 2014;Shang-Chun et al., 2016;Otsuru et al., 2018).
Molecular mechanisms and key factors involved in the therapeutic effects of MSC secretome are still unknown (Bari et al., 2019). Some studies have compared the biological effects of secretome with those of stem cells and in general terms, most of them have shown that the secretoma has greater or equal efficacy to that of cells (L et al., 2019;Tran & Damaseer, 2015). Porzionato and collaborators, demonstrated in a model of bronchopulmonary dysplasia, that the extracellular vesicles contained in the MSC secretome obtained better results in terms  Lee, et al.; found no significant differences in the potential to induce immune tolerance in the animals to which MSC vs MC-MSC were applied in an allogeneic mouse skin transplant model , likewise, our research group in a previous study, in a rabbit model where bicortical mandibular bone defects were performed, we found that the amount of neoformed bone tissue, bone density, the arrangement of collagen fibers, maturation and calcification of the inorganic matrix, were very similar on the side treated with MSC vs the side treated with the MC-MSC, demonstrating morphologically, radiologically and histologically, that there are no significant differences between the transplantation of MSC and the application MSC-CM in bone regeneration (Linero & Chaparro, 2014). Three other articles included in this review compared the application of MSC vs MSC-CM for bone regeneration. Which reported that although increased bone regeneration was observed in all groups where MSC-CM were applied, the difference with the MSC groups was not significant (Ando et al., 2014;Osugi et al., 2012;Sanchooli et al., 2017). The therapeutic differences between the application of cells and conditioned medium, perhaps arises from the possibility of using a cell-free product, which offers advantages over cell therapy. Although it has been reported that the application of MSC is safe, using only the proteins they secrete and not the cells, avoids the risk of emboli formation after intravenous administration and decreases the risk of pathological and tumorogenic transformation due to uncontrolled cellular differentiation (Bari et al., 2019). In addition, the application of cells is subject to problems such as poor cellular survival in the host after transplantation, poor ability to differentiate from transplanted cells, sequestration at non-target sites and failure of cells to graft in the long term (L et al., 2019). Secretome preserves the composition of the parental cells while maintaining the same privileged immunity of the MSC, allowing its allogeneic application without immune activation. Conditioned medium can be manipulated, stored and characterized more easily than cells, sterilization is possible without loss of efficacy, and they are ready for immediate use.
In this review, we systematically collected all the available data in the literature and critically evaluated whether the conditioned medium derived from MSC significantly promoted bone regeneration in animals and humans, making an objective and clear assessment of the scientific evidence published, resulting in a systematic review developed specifically to evaluate the effect of MSC-CM on regeneration of bone tissue.
The results of this systematic review indicate that research on this topic has been conducted mainly in animals. Critical evaluation of interventions in this type of models is a real challenge, since the reporting of methodological aspects and results is generally poor, the random allocation of animals into experimental and control groups is not a standard practice, the sample size is relatively small, and several details of the experimental designs were not included in the publications. For this reason, it is important to assess the similarity in the base characteristics between the control group and the experimental group as a necessary parameter (Hooijmans et al., 2014). In the assessment of risk of bias in the articles included in this review, we observed that 95% of these studies presented a high risk of bias in most parameters, mainly due to not reporting randomization in the selection of the animals, concealment of sequence and blinding of the evaluators. This observation applies to many animal studies published at the global level, since most of them have a high risk of bias for the abovementioned aspects (Hirst et al., 2014). Currently, the "SYRCLE's RoB" tool, available from the year 2014, which was developed to establish consistency in the assessment of risk of bias of systematic reviews carried out with animal studies, facilitates the critical evaluation of the evidence and improves the abilities of these studies to transfer to human models (Hooijmans et al., 2014). It is worth noting that one of the studies included, which was published in 2017 (Sanchooli et al., 2017), presented a low risk of bias in 8 of the 9 evaluated parameters. This allows us to conclude that authors are currently reporting all these aspects, improving the quality of the studies and facilitating the transfer of basic research to clinical practice.
The results obtained and the outcomes highlighted in the application of mesenchymal stem cell-conditioned medium for bone regeneration in this review, allow us to indicate that in general, a positive and favorable effect on bone tissue regeneration with this intervention in human and animal models was observed. In animal models, the metaanalysis established an overall favorable effect of intervention with MSC-CM, indicating statistically significant differences in the percentage of bone regeneration between the MSC-CM groups and groups with other treatments. This demonstrates that the mechanism through which the MSC-CMs exert their biological effect, is primarily mediated by the action of growth factors, cytokines, and other constituent molecules, which stimulate and induce the migration of endogenous mesenchymal stem cells, endothelial cells and osteoprogenitor cells, promote their differentiation and expression of osteogenic and angiogenic markers, and stimulate angiogenesis, osteogenesis, repair and regeneration of bone tissue (Inukai et al., 2013;Katagiri et al., 2013;Katagiri et al., 2015;Katagiri et al., 2017b;Katagiri et al., 2017c;Kawai et al., 2015;Ogata et al., 2015;Osugi et al., 2012;Qin et al., 2016;Sanchooli et al., 2017). These effects obtained by the conditioned media are consistent with what has been previously reported in other studies where MSC-CM were applied for the regeneration of different tissues. Chen in 2008and 2014(Chen et al., 2008Chen et al., 2014) mentioned that application of MSC-CM stimulated wound-healing due to the presence of high levels of cytokines that induced angiogenesis, migration and cell proliferation, thereby accelerating injury repair. Shen and Bangh in 2015 and 2014, respectively, indicated that factors present in MSC-CM have chemotactic properties, which are involved in the blood vessel formation and remodeling, angiogenesis stimulation and tissue repair (Shen et al., 2015;Bhang et al., 2014), Zhang et al., showed favorable results and effectiveness of MSC-CMs in repair and regeneration of cartilage , Nakamura in 2015 showed that the secretome of mesenchymal stem cells accelerated regeneration of skeletal muscle (Nakamura et al., 2015), and Monsel and colleagues identified secretome effectiveness for the treatment of lung inflammatory diseases through activation of anti-inflammatory and antiapoptotic pathways (Monsel et al., 2016).
In most of the scientific papers evaluated in this review, the doses of MSC-CM used are expressed in volumetric units ranging from 10 μl to 6 ml, identifying a fairly wide range of dosage, which is likely to respond to variables such as the size of the bone defect and the scaffold used for the application of MSC-CM. But in our concept, to find a relationship between dose and therapeutic effectiveness, it is necessary to identify the concentration of total proteins contained in the applied conditioned medium, not just the volume used. Of the 21 articles included, 4 report the protein concentration (Linero & Chaparro, 2014;Tsuchiya et al., 2013;Wang et al., 2015;Xu et al., 2016); but only one makes a comparative analysis of bone regeneration of jaw defects where conditioned medium were applied with a protein concentration of 100 mg/ml vs twice protein concentration (200 mg/ml); identifying that there are no statistically significant differences in morphometric, radiographic and histological assessments (Linero & Chaparro, 2014). This finding suggests that the biological system has a saturation point where even if there are more proteins in the wound bed the therapeutic effect is not potentialized. However, we consider that more preclinical research is necessary to clarify the relationship between the dose, in terms of protein concentration, and the therapeutic effect.
The results found in studies performed in human models suggest a positive effect of MSC-CM application on bone regeneration (Katagiri et al., 2016;Katagiri et al., 2017a), blood vessel formation, osteogenesis, and bone tissue repair and regeneration without causing an inadequate inflammatory response or adverse effects. However, the evidence reported is not sufficient, and therefore, it is necessary to implement the development of phase I and II clinical trials to verify these effects in humans and allow for the implementation of MSC-CM for bone regeneration procedures in clinical practice.
With all this evidence, we can suggest that MSC-CM application will become a therapeutic alternative with a great potential for the treatment of bone defects. Implementing this new strategy will allow taking advantage of the clinical benefits of cell therapy, using a product free of cells that can be administered as a medicine, more easily adaptable to the therapeutic needs in individuals, allowing the translation of scientific research into clinical development, generating promising prospects for the thousands of patients who would benefit from this type of technological development.
Despite an exhaustive search of the literature, one of the main limitations of this review is the presence of bias related to the low number of published studies. In addition, 95% of the animal studies included were categorized with the high risk of bias due to the absence of randomization, concealment of sequence and blinding in the assessment of the results; therefore, it is likely to generate an overestimate of the treatment effect. It is important that the authors of preclinical studies begin to use the SYRCLE tool to improve the quality of their studies and reduce the biases that frequently occur in this type of research.

Conclusion
The results of this systematic review indicate that the application of MSC-CM in the animal models is an effective therapy to stimulate bone regeneration and reduce healing time, thus favoring the quantity and quality of newly formed tissue without causing inflammatory reactions or adverse effects. The studies reported in the human models also suggest that MSC-CM improve the process of bone regeneration and may prove to be a safe and effective therapy. Thus, phase I and phase II clinical trials are required to support these findings and to support the application of conditioned medium as a potential therapeutic strategy for the treatment of bone defects.

Author
Title Exclusion reason

Protocol and registration
5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number.

No
Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale.

Yes
Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched.

Yes
Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated.

Yes
Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis).

Yes
Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators.

Yes
Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made.

Risk of bias in individual studies
12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level), and how this information is to be used in any data synthesis.

Appendix 3
Please insert Table 3 here

Appendix 4
Please insert Table 4 here.

Study selection 17
Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.

Yes
Study characteristics

18
For each study, present characteristics for which data were extracted (e.g., study size, PICOS, followup period) and provide the citations.

Yes
Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12).

Yes
Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.

21
Present results of each meta-analysis done, including confidence intervals and measures of consistency.

Risk of bias across studies 22
Present results of any assessment of risk of bias across studies (see Item 15).

Yes
Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]).

Summary of evidence 24
Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).

Yes
Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).

Conclusions 26
Provide a general interpretation of the results in the context of other evidence, and implications for future research.

Funding 27
Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review.

Yes
Modified Eagle Medium; DO: Distraction osteogenesis; CT: Computerized tomography; VEGF: Vascular endothelial growth factor