Cell therapy in bone healing disorders.

In addition to osteosynthetic stabilizing techniques and autologous bone transplantations, so-called orthobiologics play an increasing role in the treatment of bone healing disorders. Besides the use of various growth factors, more and more new data suggest that cell-based therapies promote local bone regeneration. For ethical and biological reasons, clinical application of progenitor cells on the musculoskeletal system is limited to autologous, postpartum stem cells. Intraoperative one-step treatment with autologous progenitor cells, in particular, delivered promising results in preliminary clinical studies. This article provides an overview of the rationale for, and characteristics of the clinical application of cell-based therapy to treat osseous defects based on a review of existing literature and our own experience with more than 100 patients. Most clinical trials report successful bone regeneration after the application of mixed cell populations from bone marrow. The autologous application of human bone marrow cells which are not expanded ex vivo has medico-legal advantages. However, there is a lack of prospective randomized studies including controls for cell therapy for bone defects. Autologous bone marrow cell therapy seems to be a promising treatment option which may reduce the amount of bone grafting in future.

Introduction

Treating bone healing disorders represents a huge challenge for orthopedic and trauma surgeons and frequently produces unsatisfactory results. Critical size bone defects, in particular, which appear after tumor surgery or trauma do not heal spontaneously and require special therapy. There are also diseases which, despite surgical intervention and the application of all conventional therapies to promote bone regeneration, are accompanied by insufficient bone healing. These include aneurysmal bone cysts, enchondroma and congenital pseudarthrosis. In the broadest sense, bone defects also include avascular osteonecrosis which is defined by the death of osteoblasts. In addition to successful bone healing through the use of growth factors, increasingly positive results of osseous regeneration through stem cells have been published in recent years.[1] This article describes the current state of cell-based therapy for osseous regeneration.

Established treatments in bone healing disorders

Autologous bone transplantation

Autologous bone transplantation is the therapy of choice for treating bone healing disorders. Despite its high efficiency in regenerating bone tissue, autologous (cancellous) bone transplantation does have numerous disadvantages. These include a longer surgery time, damage by surgical exposure (e.g. subcutaneous nerves), persisting pain and swelling at the donor site, and impaired esthetics due to scar formation or osseous malformation. Furthermore, the potential for osseous regeneration of autologous bone grafts in elderly people is low compared to an increased donor site morbidity in this population.[2] The growing interest among experts can also be seen in the increasing number of publications dealing with donor site morbidity(Figure 1).

Figure 1

The increasing frequency of publications on “donor site morbidity” and “bone” listed by Medline reflects the growing interest and examination of bone harvesting-related damage.

Callus distraction

New bone formation in long bones is achieved using callus distraction, including the so-called segment transport and external fixation devices. There must be osseous interruption which is fracture-related or created by an osteotomy. Disadvantages include the fact that the process can continue for months, the risk of infections transmitted via the pin tracks of the fixation, and the lack of application possibilities to the pelvis, spine, thorax, skull or to the hand and foot skeletons.

Ultrasound and shock waves

Using extracorporeal shock wave therapy to regenerate bone is mostly restricted to treating atrophic pseudarthrosis. Critical size bone defects cannot be healed by this non-operative therapy.

Biological fundamentals and rationale of cell-based therapy of bone defects and bone healing disorders

The rationale for a cell-based therapy to induce bone tissue regeneration is based on the high osteogenic potency of undifferentiated or almost undifferentiated osteoblastic progenitor cells of various origins. This has been documented in a now vast number of pre-clinical studies.[3] For ethical and biological reasons, stem cell therapy on the musculoskeletal system is limited to autologous transplantation of postpartum progenitor cells. Omnipotent (the potential to regenerate a complete, viable organism) or totipotent (potential to regenerate different types of tissue) embryonic stem cells, on the other hand, are used only in experimental studies.

Quantitatively relevant amounts of mesenchymal, multipotent progenitor cells are found not only in human bone marrow, but particularly also in the periosteum and in adipose tissue.[4-6] On the other hand, stem cells with osteoblastic potency, occur in lesser quantities in numerous other tissues, such as muscle,[7] umbilical cord blood,[8] placenta,[9] skin,[10] cartilage[11] and synovium.[12] Osteoblastic differentiation of mesenchymal stem cells passes through numerous intermediary stages, whereby it is less the original tissue of the MSC than the local environment with correspondingly different stimuli that influences the kinetics, gene expression and protein synthesis of the cells. The mechanisms of intracellular signal transduction are complex and most clinically oriented orthopedic surgeons can barely grasp the overall picture (Figure 2). Depending on the degree of differentiation of the osteoblastic precursor, different typical proteins and antigens are expressed in different amounts (Figure 2). As differentiation increases, the cellular proliferation rate falls.

Figure 2

Summary of important intracellular pathways of signal transduction during osteoblastic differentiation. Cytomechanic stimuli, BMPs and inflammatory stimuli, in particular, encourage osteoblastic differentiation. The expression of some of the listed KO-factors, such as Lef1/Tcf7,[13] decreases towards the end of osteogenic differentiation. On the other hand, other expression factors (e.g. Lef1▽N), increase in terminal osteoblastic differentiation.[14] The differentiation paths of adipoblasts and osteoblasts from a common progenitor cell separate relatively late, whereby adipose tissue in addition to human bone marrow is suitable as the original tissue used in cell therapies for bone regeneration.[15] Due to the small or even lack of expression of MHC-II, mesenchymal progenitor cells have a low immunogenetic potential.[16] Moreover, in contrast to other cell types, they have an immunosuppressive effect on neighboring cells. ALK: activin receptor-like kinase; ALP: alkaline phosphatase; APC: activated protein C; BMP: bone morphogenic protein; cbfa: core binding factor; Cdk: cyclin-dependent kinases; CHOP: CCAAT enhancer-binding protein (C/EBP) homologous protein; CTGF: connective tissue growth factor; cAMP: cyclic adenosine monophosphate; COX: cyclo-oxygenase; ERK: extracellular signal-related kinase; LRP: LDL receptor-related protein; MAP: mitogen-activated protein kinase; MHC: major histocompatibility complex; OAZ: Olf-1/EBF-associated zinc finger; PG: prostaglandin(s); PPAR: peroxisome proliferator activated receptor; SF: short form; SnoN: Ski-related novel oncogene; STAT: transducer and activator of transcription; Tob: transducer of erbB2; VEGF: vascular derived growth factor; Wnt: wingless gene; sFRP: secreted frizzled related protein, Lef: lymphoid enhancer binding factor.

Other characteristics which make autologous mesenchymal progenitor cells an attractive candidate for the treatment of bone defects are:

simple availability and an uncomplicated harvesting technique by aspiration without the disadvantage of significant harvesting morbidity;

the standardized and well-established isolation technique using density gradient centrifugation or flow cytometry technology (e.g. fluorescence activated cell sorter, FACS);

a consensual definition of the term “mesenchymal stem cells” (MSC) that has now been worked out, with well-defined biological properties[17-19] (Table 1);

Table 1

Consensual definition of the term “mesenchymal stem cell”. A large number of synonyms exist, however, in scientific literature, e.g. precursors of non-hematopoietic tissue, colony forming units-fibroblasts, marrow stromal cells, bone marrow stroma/stem cells.

Properties of mesenchymal stem cells
Expression of mesenchymal markers	CD49a, CD73, CD90, CD105, CD146, Stro-1,
Expression of matrix receptors	CD44, CD29, CD71
Absence of hematopoietic markers	CD45, CD34, CD14, CD11b, HLA-DR
Biological properties	Spindle-shaped morphology Good adherence to plasticEvidence of CFU-FSelf-renewal potentialMesenchymal multipotency (osteoblast, chondroblast, adipoblast, myoblast, fibroblast and their precursors)High proliferation rate, particularly in the presence ofdefined growth factors, e.g. FGF-2

the simple cultivation technique for in vitro expansion and determination of the proliferation rate through colony forming units (CFU);

that osteoblastic differentiation can be well controlled in in vitro cultivation with stimuli such as dexamethasone, ascorbic acid and β-glycerol phosphate (DAG), and the decades of experience with this stimulation method;

the availability of defined cellular expression markers by which osteoblastic differentiation can be reliably documented (e.g. osteocalcin, osteopontin, osteoprotegerin, Cbf1/Runx2, collagen type I, alkaline phosphatase, osterix, bone sialo protein, signs of biomineralization, RANKL);[20]

the immunosuppressive and immunomodulatory effects of MSC which lead to a limitation of local inflammatory reactions at the transplantation site;[21]

the good adherence to surfaces, which favors the use of scaffolds and advocates local concentration of in vivo introduced cells at the transplantation site;[22]

the decades of experience in bone marrow transplantations in hemato-oncology, which is supported by the low transplantation risk especially in autologous transplantations.

Clinical application of cell therapies in bone healing disorders

In contrast to the extensive in vitro and animal experiment data, there are only a few studies that show clinical results for cell therapy treatments to regenerate bone.

There are two clinical application forms of cell therapies to regenerate bone. Besides the biological differences, various health law-related consequences also emerge for the manufacturer and the orthopedic surgeon in attendance.

Cell therapies without expansion in culture:

with the isolation of defined primary cells;

without the isolation of defined primary cells.

Cell therapies with ex vivo expansion:

with typical cell differentiation;

without typical cell differentiation.

Cell therapies without expansion in culture

What is generally meant here is cell therapies that are harvested or produced during an operation. The tissue used for this does not leave the operation theatre or operation area and is, therefore, under the direct supervision and responsibility of the operator in attendance. Bone marrow aspiration concentrate (BMAC) is a typical example of this form of application. At the beginning of the operation, a defined volume of bone marrow is harvested by Jamshidi-vacuum aspiration of the ventral or dorsal iliac crest and suspended in an anti-coagulating heparin and ACDA solution in a transfusion bag. Mononuclear cells are then isolated from the harvested bone marrow aspirate in a density gradient centrifuge in the closed system that we have been using since 2005.

Possible quality controls of the cell therapy (BMAC) are to compare the number of cells in the BMAC with that in the initial aspirated bone marrow, and determine the CFU-F and ALP activity during in vitro cultivation.[23-25] Despite these quality parameters, the individual potency of in vivo applied cell therapies cannot be reliably predicted. Some publications indicate, however, that compared to the transplantation of a defined type of cell, applying mixed populations of mesenchymal and hematopoietic progenitor cells at different stages of differentiation is more effective for osteogenic regeneration.[26]

In a prospective clinical study and in various experimental treatments, our research group has so far successfully treated over 100 patients with local bone healing disorders using a BMAC biomaterial composite. Fifty percent of the bone defects were grafted with autologous cancellous bone and the remaining 50% with a BMAC biomaterial composite (hydroxylapatite, Orthoss®, Geistlich, Wolhusen, Switzerland vs. collagen sponge, Gelaspon®, Chauvin Ankerpharm, Berlin, Germany). So far, our study has found that the use of BMAC reduces the harvest of autogenous bone by 50% with no slowing down or absence of bone healing being observed.[27, 28] No complications with the application were observed in any of the patients. The low complication risk of this procedure[29] and the osteogenic potency in the parallel application of different biomaterials has also been reported by other research groups.[30, 31]

A high variance in the number of harvested cells was observed in the human bone marrow aspirate.[32, 33] To maximize the yield of bone marrow cells, the following procedure is recommended.

Create a sufficiently high vacuum: this is necessary to create sufficient local force to retrieve the (in contrast to hematopoietic cells) strongly adherent mesenchymal cells from the tissue mass. If the vacuum is too low, the amount of peripheral blood in the aspirate will be higher. As well as the plunger pressure created by the operator, the geometry of the syringe also plays an important role in creating sufficient negative pressure(Figure 3).

Figure 3

In iliac crest-vacuum aspiration, the geometry of an aspiration syringe influences the proportion of MSCs in the aspirate. The pressure required to retrieve the mesenchymal cells is exerted at the tip of the needle and is defined by the formula: pressure (P) = force (F)/area (A), whereby the force used to create a vacuum is created by withdrawing the plunger of the syringe. This force remains relatively constant. Narrow, long syringes are, therefore, advantageous when harvesting MSCs using bone marrow aspiration.

Draw several small portions in small aspiration volumes: the number of mesenchymal progenitors (and the vacuum) per aspiration volume decreases during an aspiration procedure.[34] It is recommended, therefore, that you draw a maximum 5 mL of bone marrow and then create a new vacuum.

The number of mesenchymal progenitor cells falls with repeated aspiration procedures at the same spot. A maximum 3 aspirations at the same spot before positioning the needle at another spot is, therefore, recommended.

More cells are retrieved with parallel insertions of the aspiration needle than with diverging insertions in one area(Figure 4).

Figure 4

The areas of the iliac crest reached by the tip of the aspiration needle if inserted at diverging angles at the same point overlap, so that areas that have been perforated and aspirated once already are subjected to the procedure several times. This leads to a drop in the amount of MSCs per volume of bone marrow aspirate. If the positions of the inserted needle are parallel, then new MSC harvesting areas will always be accessed.

Also ensure sufficient anti-coagulation of the bone marrow aspirate during the harvesting procedure. Heparin and ACDA solutions are used for this. The aspiration needles and syringes should be flushed with the solution before use. Density gradient centrifugation is particularly suitable for isolating mononuclear cells for bone regeneration therapy.[23]

Cell therapy treatments with temporary ex vivo expansion

In orthopedics and traumatology, autologous cell therapies have been used regularly on the musculoskeletal system after ex vivo cultivation, at least since the clinical introduction of autologous chondrocyte transplantation (ACI). Unlike cartilage regeneration, for which ACI was used in more than 12,000 patients between 1987 and 2005,[35, 36] there are no reliable data on osseous regeneration after temporary in vitro cultivation. In the treatment of necrosis of the femoral head, for instance, whereas numerous one-step transplantations are documented, only three case studies with a maximal observation period of three months can be found. Here, a mixed cell population from bone marrow cells (so-called tissue repair cells, TRCs), was expanded over 12 days under GMP conditions and then transplanted autologously together with a scaffold made of tricalcium phosphate (TCP) within the framework of core decompression.[37]

The particular drawbacks of temporary cultivation of MSCs lie not only in the considerable logistical effort to ensure the quality of the cell therapy treatment but especially in the biological characteristics of this cell population. As soon as MSCs are isolated from their tissue mass and transferred to a culture dish, differentiation proceeds in accordance with the culture conditions.[38-40] The yet inconclusive biological effects when fetal bovine serum is used in the culture, as well as telomere shortening, and thus cell aging with ex vivo cultivation also have to be considered. Furthermore, analysis of 170 neoplasia-associated DNA promoters was able to show that despite the relatively high genetic stability of MSCs from human bone marrow or adipose tissue, damage in the genome could occur at later stages.[41] The question as to whether these genotoxic effects of prolonged in vitro cultivation are also clinically manifested after re-transplantation remains unanswered, however. The potential effects of changes in the chromatin structure due to epigenetic factors at the beginning of osteoblastic differentiation also remain largely unknown.[42]

Cell therapy in local bone defects, bone healing disorders and osteonecrosis

Other research groups have also reported positive clinical results after using human bone marrow cells. Giannini et al. showed that in patients with osteochondral defects in the talus, functional improvements were achieved through autologous bone marrow cell transplantation by arthroscopic surgery.[43] As early as 1991, Conolly et al.[44] reported equivalent healing rates for autologous bone marrow grafting to treat post-traumatic pseudarthrosis of the tibia. Other authors also support the high osseous regeneration potency of the percutaneous implantation of autologous bone marrow concentrate to treat pseudarthrosis[32, 33] and discuss supplementary osteoblastic stimulation using platelet rich plasma (PRP).[45]

Some authors, on the other hand, have reason to believe that bone regeneration through cell therapy also depends very much on the transplantation site and the local blood supply. According to a study by Kitho et al., in which ex vivo -cultivated MSCs were used together with PRP in 51 lengthening osteotomies, cell therapy accelerated bone healing in the femur compared to in the tibia.[46] Overall, however, the cell therapy showed no advantages over the untreated control group and, moreover, no relation between the bone healing rate and the number of transplanted cells or the PRP concentration was found. Hernigou et al., however, reported that grafting over 50,000 osteoblastic progenitor cells particularly encouraged healing in atrophic tibial pseudarthrosis.[33]

Minimally invasive cell therapy of solitary or aneuryamal bone cysts via percutaneous implantation of autologous bone marrow is favored by a number of authors owing to the healing rate of over 80% [47, 48](Figure 5). On the other hand, a randomized clinical study showed that in the treatment of simple bone cysts, autologous bone marrow injections were inferior to local steroid injections.[49]

Figure 5

Healing course after autologous cell therapy with bone marrow aspiration concentrate (“BMAC”) augmented with HA granules in a 4-year old male patient with a large aneurysmal bone cyst of the proximal femur. Ten months after surgery, relevant new bone formation starting within the transplant is observed in computed tomography scan (star, *). The clinical and radiological 3.5 year follow up after treatment showed no recurrence and an asymptomatic patient. Based on the reduced amount of autologous bone available for grafting, pediatric patients in particular might benefit from the minimally invasive cell therapy.

Initial results are also available on cell therapy treatments to promote osseous fusion at the spine. After an observation period of 34 months, fusion rates of over 90% were found for tricalcium phosphate (TCP)-bone marrow composite transplantations.[50] For an HA-collagen-I composite incubated with autologous bone marrow, the posterolateral lumbal fusion rate was found to be the same as for autologous bone transplantation, but fusion rates were not the same for intracorporeal fusions.[51]

Cell therapies have been used successfully to treat avascular osteonecrosis (AVN) for many years by Hernigou et al.[52, 53] and, in the meantime, also by other research teams in experimental treatments and clinical studies.[54-56] Patients with sickle cell anemia-related AVN, in particular, benefit from a local injection of mononuclear bone marrow cells. Patients with steroid-related AVN have a worse prognosis when treated with MSCs to regenerate bone.[57]

The number of implanted cells and their proliferation potency, as measured by the CFU-F, are positive predictors for successful bone marrow concentrate therapy in the treatment of osteonecrosis.[52] It is unclear whether the reduced number of MSCs in the proximal femur observed in patients with AVN is an independent risk factor in the development of an AVN, or is resulting from AVN.[57] Other authors, however, report comparatively high numbers of osteoblasts in the major trochanter region with necrosis of the femoral head.[58]

It has been shown that for ARCO stages I and II local cell therapy with autologous bone marrow in combination with core decompression diminished the risk of a medium-term progression of necrosis of the femoral head.[59]

A multi-center study on patients with peripheral artery disease (PAD) amply documented that mixed cell populations from human bone marrow not only have an osteoblastic but also an angiogenetic effect.[60]

Because of the lack of control groups, however, cell therapy is mentioned in the current S1-recommended treatments of atraumatic necrosis of the femoral head in adults under “Operation methods without good documentation support”.[61] Besides direct transplantation of the cell suspension into the AVN area within the framework of a core decompression, a cancellous bone graft can also be combined with the autologous cell therapy.[62] For this, a Krohn hollow mill is used to extract a cylinder of cancellous bone and the AVN area subjected to curettage through the resulting cavity. After removing the macroscopic avascular tissue for histopathological diagnosis, the cylinder of cancellous bone is incubated with the cell therapy and then re-implanted in the osseous defect. Medium- and long-term results of this cell therapy treatment are pending. Table 2 is a summary of the results of clinical applications of cell therapies to regenerate bone.

Table 2

A summary of published studies including more than 10 patients after cell therapy in bone defects or bone healing disorders. The medline review showed that cell therapy by bone marrow is not a new technique but has been applied for almost 20 years. According to the scaffold, causative disease, body region and other patient-related factors, most studies demonstrate promising results for bone regeneration by bone marrow cells.

Author	Year	Journal	Bone defect	N patients	Results
Connolly et al.[44]	1991	CORR	Pseudarthrosis	20	Application of autologous bone marrow (BM) in tibial pseudarthrosis or “non-union”. Post-operative treatment with plaster cast. Additional intramedullary nailing in 10 cases. The authors report that autologous BM application produced the same results as for autologous bone transplantation.
Lokiec et al.[63]	1996	JBJS-Br	Simple	10 bone cysts	Percutaneous injection of autogenous bone marrow: all the patients became pain-free after two weeks and resumed full activities within six weeks. The cysts were radiologically consolidated and showed remarkable remodeling within four months. Bone healing was achieved 12–48 months after treatment (no complications).
Köse et al.[64]	1999	Bull Hosp J T Dis	Simple	12 bone cysts	Autologous bone marrow injection in bone cysts: complete healing occurred in 2 patients, whereas 3 cysts showed residual defects. In 6 patient, cysts recurred. Authors concluded that factors such as the size, multi-loculation, and completeness of the filling of the cyst with bone marrow grafting might influence the post-operative outcome.
Hernigou et al.52	2002	CORR	AVN (Hip)	116 (189 hips)	Evaluation of the clinical outcome 5–10 years after core decompression in combination with injection of autologous BM concentrate in the treatment AVN of the femoral head. Very good results in pre-collapse stages (ARCO I-II): 9 out of 145 hips were replaced endoprosthetically. In post-collapse stages, 25 out of 44 hips replaced endoprosthetically. Better results with higher CFU-F and cell numbers.
Rougraff et al.[65]	2002	JBJS-Am	Unicameral	23 bone cyst	Percutaneous injection of allogeneic demineralized bone matrix augmented with autogenous bone marrow is an effective treatment for unicameral bone cysts.
Chang et al.[66]	2002	JBJS-Br	Unicameral bone cyst	79	14 patients treated with BM ( 27 injections) vs. 65 patients with steroid application (99 injections). Repeated injections were required in 57% of patients after BM had been used and in 49% after steroid. No complications. No advantage could be shown for the use of autogenous injection of BM compared with injection of steroid in the management of unicameral bone cysts.
Price et al.[67]	2003	Spine	Spinal fusion	77	Retrospective study with 3 different bone grafting techniques: autologous iliac crest bone graft (ICBG) vs. freeze-dried corticocancellous allograft vs. composite graft of autologous bone marrow (BM) and demineralized bone matrix. Segmental instrumentation with dual-rod fixation was used in 77 patients. No BM aspiration-associated morbidity. Fusion rates were comparable for ICBG and BM group.
Docquier et al.[68]	2003	J Pediatr Orthop	Simple bone cyst	17	Percutaneous aspiration and injection of BM. FU: 33.9 months. Slow regression of the cyst and progressive healing: 13 cases (76%). No response: 2 cases (12%), recurrence: 2 cases (12%).
Gangji et al[59]	2004	JBJS-Am	AVN (hip)	13 (18 hips)	Necrosis of the femoral head in ARCO stages I-II. Core decompression (vs. core decompression + BM aspirate (10 patients). Within 24 months, significant reduction in pain, functional improvement and lower AVN progression rate after cell therapy. No transplantation-related complications.
Hernigou et al.[33]	2005	JBJS-Am	Pseudarthrosis /non-unions (atrophic, tibia)	60	Injection of 20 cm3 BM concentrate: 612±34 progenitor cells/cm3 in the aspirate compared to 2579±1121 progenitor cells/cm3 after density gradient centrifugation: healing in 53 cases. Positive correlation between callus regeneration and the number of CFUs.
Kanellopoulos[69]	2005	J Pediatric Orthop	Active unicameral bone cyst	19	BM injection in bone cysts. All patients were asymptomatic at the latest follow up. Two patients required a second intervention to achieve complete cyst healing. Radiographic outcome was improved in all patients according to the Neer classification at the latest FU. There were no significant complications related to the procedure, nor did any fracture occur after initiation of the above regimen.
Neen et al.[51]	2006	Spine	Spinal fusions	50	Therapy using HA-collagen I composite incubated with autologous BM aspirate (incubation time: 20 min) vs. autologous bone transplantation The same posterolateral lumbar fusion rates for both groups, similar functional results for both groups. Autologous bone transplantations raised the fusion rate in “interbody fusions”, but donor-site morbidity in 14% of the cases.
Yan et al.[70]	2006	Chin J Traumatol	AVN (hip)	28 (44 hips)	Percutaneous multiple hole decompression combined with autologous BMCs. The earlier the stage, the better the result. A randomized prospective study needed in the future to compare with routine core decompression.
Dallari et al.[45]	2007	JBJS-Am	Proximal	33 tibia osteotomies	Prospective, randomized study with 2 therapy groups: lyophilized bone chips + PRP (A, 11 patients) vs. lyophilized bone chips + PRP + bone marrow (B, 12patients). Control group: lyophilized bone chips only. CT-controlled biopsies six weeks post-OP showed increased callus formation in A and B compared to the control group. Improved bone healing in A and B within one year.
Deng et al.[71]	2007	Chin J Regen Reconstr Surg	Bone cyst	13	Transplantation of the autologous bone marrow combined with the allograft bone. Complete healing within 3.5–8 months (Ø 5.2 months). No recurrence, no pathological fracture occurred. Complete recovery of function.
Cho et al.[72]	2007	JBJS-Br	Bone cysts	28 (58)	30 patients treated by steroid injection vs. 28 individuals by bone marrow grafting. Overall success rates: 86.7% vs. 92.0%, respectively (P>0.05). Initial success rate: 23.3% in the steroid group vs. 52.0% in the BM group. Mean number of procedures: 2.19 (1 to 5) vs. 1.57 (1 to 3) (P<0.05). Average healing interval: 12.5 months (4–32) P =14.3 months (7–36) (P>0.05). Rate of recurrence after initial procedure: 41.7% vs. 13.3% (P<0.05). Although the overall rates of success of both methods were similar, the steroid group showed higher recurrences after a single procedure and required more injections to achieve healing.
Wright et al.[49]	2008	JBJS-Am	Bone cysts	77	Randomized, prospective study. Two therapy groups: injection of autologous BM (A) vs. injection of methylprednisolone (B). Healing rate within two years: 23% (A) vs. 42% (B). No significant difference in the functional outcome.
Park et al.[47]	2008	Foot Ankle	Bone cysts	20 (23 cysts)	Therapy of unicameral bone cysts of the calcaneus. Two therapy groups: open surgery application of avital allogenic donor bone + autologous BM (A) vs. injection of demineralized bone powder + autologous BM (B). Healing rate within 49.4 months: A: 9 out of 13 cysts vs. B: 5 out of 10 cysts. No infections.
Gan et al.[50]	2008	Biomaterials	Spinal fusions	41	Application of TCP incubated with BM concentrate (duration circa 2 h). Concentration factor (CFUs-ALP: 4.3). Drop in MSCs with increasing age, but no dependency on gender. After 34.5 months, spinal fusion in 95.1% of the cases.
Zamzam et al.[48]	2008	Int Orthop	Solitary bone cysts	28	A minimum one-off percutaneous injection of autologous BM. No complications. Within 34.7±6.87 months, bone healing in 82% of the cases.
Jäger et al.[73]	2009	CSCRT	Bone defects	10	Significant bone regeneration through bone marrow concentrate (BMAC) in combination with autologous cancellous bone.
Hendrich et al.[29]	2009	Orthop Rev	Bone defects, AVN	101	Proof of the low complication risk of autologous BMAC in 101 applications.
Giannini et al.[43]	2009	CORR	Osteochondral lesions (talus)	48	Functional improvements after arthroscopy-assisted application of autologous BM aspirate in osteochondral defects in the talus.
Sir et al.[74]	2009	Vnitr Lek	Fracture-related bone defects, pseudarthrosis	11	Local and one-step injection of MSCs from human BM. Results pending.
Kitoh et al.[46]	2009	J Pediatr Orthop	Tibial vs.femoral lengthening osteotomies	28 (51 osteotomies)	Retrospective study. Application of ex vivo cultivated MSCs together with PRP Control group: 60 patients without MSC/PRP. No stimulation of bone healing by MSC/PRP. Worse results for the tibia.
Hernigou et al.[56]	2009	Indian J Orthop	AVN (hip)	342 (534 hips)	Autologous cell therapy in ARCO stages I–II in combination with a core decompression. After 8–18 years, 94 endoprosthetic hip replacements. Predictor for a therapy success was a high number of progenitor cells.
Wang et al.[75]	2009	Arch Orthop Trauma Surg	AVN (hip)	45 (59 hips)	BMAC injection in AVN of the femoral head (ARCO stage I–III).Clinically successful in 79.7%. Hip replacement within FU in 11.9% of the hips. Radiologically, 14 of the 59 hips exhibited femoral head collapse or narrowing of the joint space. Overall failure rate: 23.7%. The concentration factor of mononuclear cells from BM vs. BMAC was about 3.
Miller et al.[76]	2010	Int Orthop	Non-union or segmental defect	13	Bone marrow cells harvested by a reamer-irrigator-aspirator (RIA) were treated by dexamethason and transplanted into segmental bone defects. Promising results were achieved using this technique; and given the complexity of these cases, the observed success is of great value and warrants controlled study into both standardization of the procedure and concentration of the grafting material.
Yamasaki et al.[77]	2010	JBJS-Br	AVN (hip)	22 (30 hips)	Transplantation of bone-marrow-derived mononuclear cells (BMMNCs) combined with hydroxypapatite (HA) vs. HA only in AVN of the femoral head. Reduction of the osteonecrotic lesion was observed subsequent to hypertrophy of the bone in the transition zone in the BM group. In 3 patients of the BMMNC group, progression to extensive collapse occurred. Control group showed bone hypertrophy, but severe collapse of the femoral head occurred in 6 of 8 hips.

Cell therapy in skeletal diseases

In addition to the local application of MSCs, data are also available for MSCs in the successful treatment of skeletal diseases which are accompanied with deficiencies in the bone structure.[78, 79]

In 1999, after numerous in vitro experiments and animal experimental studies, Horwitz et al., for the first time, treated 3 children with osteogenesis imperfecta (OI) with allogenic transplantations of mesenchymal bone marrow cells.[80] The cells were introduced intravenously after ablative pre-treatment of the patient (chemotherapy and immunosuppression). Post-operative bone biopsies after 216 days and bone density measurements showed a significant quantitative and qualitative improvement in the bone structure. In another publication from the year 2001, the same research group[81] reported their findings on 5 OI patients who had been treated with cell therapy and on 2 other OI patients without cell therapy treatment (OI type III). After an investigation period of six months, children who had received cell-based therapy showed an accelerated growth rate.

Osteopetrosis is another skeletal disease involving insufficient osteoclast activity that is currently being treated with autologous bone marrow transplantation. Driessen et al.[82] found that the probability of 5-year disease free survival was 73% after cell-based therapy. Treatment before the age of three years improved the chances of success of cell therapy in osteopetrosis.[83] However, due to the severe side-effects and possible complications (e.g. graft rejection, hypercalcemic crises, pulmonary hypertension, delayed hematopoiesis, veno-occlusive disease), allogeneic cell therapy treatment of patients with osteopetrosis is limited to severe manifestation of disease.[84] Three case reports also report the successful treatment of an 8-month old infant with infantile hypophosphatasia who underwent transplantation of T-cell depleted bone marrow from the sister. The positive effects of the cell therapy ceased, however, after six months. Twenty-one months after the first transplantation, a second transplantation of ex vivo expanded bone marrow cells took place resulting in an increase in bone mass. At the age of six, the patient in question still showed signs of stunted growth but displayed normal intelligence.[85] Another approach in cell therapy treatment of infantile hypophosphatasia consists of intra-peritoneal, subcutaneous or intraosseous bone transplantation from a related donor parallel to the intravenous bone marrow injection. The postulate here is that migration of the donor MSCs in the recipient organism will positively influence bone healing and the rejection reactions.[86, 87] It is unclear whether cell-based therapy will also gain acceptance in other skeletal diseases, such as osteoporosis.

Outlook for the future

Due to the accelerated aging of osteoblastic progenitor cells after in vitro cultivation, the limited resources, the diminished osteoblastic potency with increasing age and the improved standardizations, immortalized human MSCs are currently undergoing pre-clinical investigations for their suitability for cell therapy.[88, 89] One way of avoiding aging of MSCs is to transfer the cDNA of telomerase reverse transcriptase (hTERT). With this enzyme, the telomeres that have been shortened during the course of replication are returned to their original length. Some authors were able to demonstrate a high osteoblastic potency in vitro and in animal experiments with this process.[88] Nevertheless, in view of the current legislation, it is uncertain whether these new therapy procedures can also be tested clinically. In addition to cell-based therapies, there are other innovative “orthobiologics” with bone regeneration as the goal. These include anchor proteins that stimulate osteoblastic adherence, e.g. RGD sequences, fibronectin or peptide 15. α-granules of thrombocytes, in particular, contain large amounts of growth factors with osteoblastic and cell proliferation potency, such as transforming growth factor-β (TGF), vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF).[90] Whether it makes sense and is necessary, within the context of the clinical application of cell therapies, to increase the osseous potency of osteoblastic progenitor cells using additional growth factors must first be investigated in controlled clinical studies given that the hitherto existing data are contradictory and not sufficiently reliable.[46,91,92]