Bone Morphogenetic Protein rhBMP2 Experimental Review and Clinical Update

Brian R. Subach, Regis W. Haid, Jr., Gerald E. Rodts, Jr., Carmen A. Petraglia

The NeuroSpine Institute, Department of Neurological Surgery, Emory University, Atlanta, Ga., USA

Over 30 years ago, Urist et al. [67, 68] identified a group of protein extracts, derived from the ground substance of mature bovine bone, capable of inducing both cartilage and bone formation when implanted into the soft tissues of study animals. Aptly named, bone morphogenetic proteins (BMPs) by Urist, these glycoproteins comprise a subset of the transforming growth factor-^ family of related growth and differentiation factors. Of the more than 20 BMPs isolated to date, 6 appear to be structurally related to each other and capable of initiating the process of endochondral bone formation. The presence of such factors within the matrix of mature bone indicates a likely role in the regeneration and remodeling of bony structures after injury or repetitive stresses [4, 9, 18, 24, 43, 44].

The Basics of BMP Biology

Each of the six known osteoinductive BMPs shares significant similarities on a molecular level. Synthesized within the cell in precursor form, each molecule has a hydrophobic leader or secretory sequence with the mature portion of the protein at the carboxy terminus marked by a highly conserved, seven-cysteine repeat. Each mature BMP begins as 2 monomers of 120 amino acids each, which undergo disulfide linkage dimerization to form either a homologous or a heterologous protein chain. In the case of BMP-2 and BMP-7, the specific structure was first identified by isolating the bovine protein from bone extracts.

Oligonucleotide probes were used to obtain the human complementary DNA (cDNA) sequence. The cDNA clones were then spliced into a viral expression vector and transfected into a carrier cell in a process called recombination. In the case of BMP-2, the cells used were Chinese hamster ovary cells. Such cells produce the pure recombinant differentiation factor rhBMP-2 in large quantities in a process similar to fermentation. This process avoids potential complications related to the transmission of infectious materials from human donor bone tissue and eliminates the possibility of xenograft interactions with human recipients of BMP derived from bovine sources.

An Osteoinductive Role for BMP

In order to function as a suitable graft for bridging bone defects or fusing fracture lines and unstable motion segments, the prospective material would ideally possess three characteristics. The material would provide a source of primitive osteoprogenitor cells, which, under the appropriate influence, would form osteoblasts and osteocytes (osteopromotive). Such precursor cells are unfortunately relatively scarce. Bone marrow, for example, contains a ratio of only one osteprogenitor cell to approximately 50,000 nucleated cells in a young adult. This ratio may dip to 1:200,000 cells in an elderly individual afflicted by degenerative spinal disease [21]. Despite techniques to concentrate marrow extracts, successful efforts have only resulted in a maximum of 5-fold improvement of the unfavorable cellular ratio. Second, the graft material would produce local growth factors to stimulate bone growth and vascularity in the area (osteoinductive). There are numerous reports in the literature detailing the complex interaction of various autocrine and paracrine growth factors released from fibroblasts, platelets, and even the local hematoma that forms at the site of injury [1, 7, 16, 37, 46]. Finally, the third property of the graft material ideally would be its ability to act as a scaffold for bony ingrowth. Such ability is known as osteoconduction.

There are a number of possible reasons, which may account for the osteo-inductive role of BMP. BMP acts as a chemotactic agent, a growth factor, and a differentiation factor. BMP acts as a chemotactic factor and initiates the recruitment of progenitor and stem cells toward the area of bone injury. In vivo studies of the local effects of BMP indicate an initial migration of mesenchymal stem cells to the area of implantation far in excess of that supplied by bone marrow grafting [12]. BMP acts as a growth factor and stimulates both angiogene-sis and the proliferation of stem cells from surrounding mesenchymal tissues. BMP also acts as a differentiation factor by promoting the maturation of stem cells into chondrocytes, osteoblasts, and osteocytes.

Some cells respond to the growth factor aspect of BMP by altering their rates of proliferation. Yamaguchi et al. [71] demonstrated this in vitro by quantifying cellular proliferation of the rat C26 calvarial osteoprogenitor cells after treatment with BMP-2. This BMP-2 effect, however, appears to maintain specificity for certain cell types. For example, although BMP-7 has been shown to be mitogenic for a human osteosarcoma cell line (TE-85), treatment with BMP-2 showed no measurable effect on proliferation. In contrast, treatment of an osteoblast cell line (MC3T3-E1) with BMP-4 results in an inhibition of growth and a globally diminished proliferative index. By the effects on both mature and immature cell types, it seems reasonable that BMPs must be involved in the regulation of bone growth and maintenance of bone structure.

BMPs may also initiate the differentiation of stem cells into a specific phenotype. For example, the rat calvarial stem cell line (C26) is considered multipotential, in that such cells may be precursors for adipocytes, muscle cells, or osteoblasts. When BMP-2 is added to the culture medium, such cells become mature osteoblasts with increased surface expression of receptors for parathyroid hormone, alkaline phosphatase, and calcitonin [71]. This effect may also be observed in bone marrow cells. For example, the mouse line of marrow cells (W-20-17) may differentiate into either adipocytes or osteoblasts, depending upon the specific hormonal influence. The BMP-2 treatment of such cells results in both the differentiation of the cells into osteoblasts and the surface expression of receptors normally seen on mature cells.

Sources of BMP

At present there are three ways to obtain bone growth and differentiation factors: extraction of the factors from animal or human bone matrix, production of a single factor by cellular hosts using recombinant technology, and direct delivery of the DNA encoding for the factor to cells at the site of desired bone formation.

The first of these was initially employed by Urist et al. [67, 68]. From massive quantities of bovine bone, the group was able to extract a mixture of proteins found to stimulate bone growth in vivo. Under clinical evaluation in Europe as NeOsteoâ„¢ (Sulzer Spinetech, Wheat Ridge, Colo., USA), this mixture of BMPs and other associated proteins is derived through a well-engineered isolation process. The precise combination of factors comprising this substance has not yet been fully characterized, but appears to be reproducible through the manufacturer's process. This substance has shown experimental promise in bridging both segment skeletal defects in dogs and in bringing about spinal fusion in animal models of posterolateral arthrodesis. [12, 14, 17, 26, 27]. Like other growth and differentiation factors, a carrier substance is necessary to maintain adequate concentrations at the site of fusion. Substances such as natural coral (hydroxyapatite), collagen, and calcium sulfate have each been investigated [26, 27].

The second method of obtaining bone growth and differentiation factors has previously been discussed (see Molecular Biology). The process of obtaining recombinant human BMPs such as rhBMP-2 (Medtronic Sofamor Danek, Memphis, Tenn., USA and Genetics Institute, Cambridge, Mass., USA) and rhBMP-7 (Stryker Biotech, Hopkinton, Mass., USA) has been described. Such proteins differ from mixtures of extracted substances, mainly in terms of purity of product. Original studies of these substances focused upon animal models of segmental bone defects in the appendicular skeleton of rats, sheep, and dogs [23, 24, 35]. Cole et al. [23] compared rhBMP in a carrier to autologous grafting in a skeletal defect model with considerable success. Gerhart et al. [35] showed the utility of rhBMP in healing segmental femoral defects in sheep. Shortly thereafter, recombinant BMP was applied to animal models of spinal fusion and later, humans.

The third strategy for engineering bone formation involves gene therapy, or the delivery of the appropriate gene, or cDNA encoding for BMP to the local cells, rather than the actual factor. There are two obvious benefits to the strategy as compared to recombinant technology. First, the cost of genetic manipulation is significantly less than that required to both produce and market the purified rhBMP. Second, the potential for prolonged local production of the factor is greater with gene therapy when compared to the relatively short-lived effect of the rhBMP/carrier complex. Attempts to introduce BMP-2 cDNA into animal models are preliminary, but have met with limited success [1, 17, 29, 70].

From the Laboratory

Early work in the isolation of proteins with osteoinductive activity suggested that BMP-2 and BMP-7 were primarily responsible for the effects observed in vivo [3, 22, 23, 53, 59]. As a result, rhBMP-2, produced in a Chinese hamster ovary cell line, was the first of these molecules studied in detail. Implantation of the recombinant factor in a rat model resulted in ectopic bone formation with a dose-effect relationship temporally identical to that of bone-derived extracts; however, the amount of pure rhBMP-2 required to induce formation of a given amount of bone was approximately 10-fold less than that required of the bone extract [5, 12, 13, 51]. Subsequent studies in nonhuman primates show no difference in the dose required to effect consistent posterolateral spinal fusions.

The ability to form bone at ectopic sites, however, had little application to current spinal fusion techniques. Realizing the limitations inherent in

Fig. 1. Photomicrograph of intertransverse spinal fusion using carrier matrix alone (a) as a control and rhBMP-2/carrier (b). Histologic section demonstrates transverse processes at inferior-lateral corners with bridging collagen scar tissue and minimal bone formation (a) and abundant new membranous bone formation (b).

Fig. 1. Photomicrograph of intertransverse spinal fusion using carrier matrix alone (a) as a control and rhBMP-2/carrier (b). Histologic section demonstrates transverse processes at inferior-lateral corners with bridging collagen scar tissue and minimal bone formation (a) and abundant new membranous bone formation (b).

autogenous and allogeneic bone grafting, investigators began applying BMP technology to animal models of spinal fusion procedures [32, 36, 38, 62]. Multiple studies, involving various concentrations of BMP in a variety of carrier substrates, have shown remarkable results. Early work by Boden et al. [12-15] and Holliger et al. [45] compared rhBMP-2 to autologous bone graft in a rabbit posterolateral lumbar fusion model. Remarkably, all BMP-treated animals (100%) attained solid, bony fusions across the operated level, which were biomechani-cally stiffer and stronger than the autograft-only fusions observed in 42% of the control group. Similar studies in a canine model also confirmed the efficacy of rhBMP-2 in producing mature fusion masses [28]. The canine study by David et al. [28] demonstrated a dose dependence to the BMP effect, with greater concentrations producing greater effects; however, this contradicts a study by Sandhu et al. [58-60] in a similar canine model, which shows BMP to be more effective than autologous bone graft, but in a dose-independent manner [63]. Most investigators developed the opinion that bone induction was relatively simple in lower species, but was only indirectly applicable to human models. As a result, research focused on developing spinal fusion models in primates. As a developmentally higher species, primates provide a more realistic test environment for evaluating the effectiveness of BMP [40, 41]. Boden et al. [10] applied this belief to a nonhuman primate model of intertransverse spinal fusion and demonstrated effective fusion rates using rhBMP-2 on a ceramic carrier delivered by a minimally invasive approach (fig. 1). Sandhu et al. [60] went a step further to demonstrate clinically, mechanically, and radiographically equivalent spinal arthrodesis using rhBMP-2 without decortication of the prospective fusion bed. Boden et al. [7, 16] and Martin et al. [51], by adding rhBMP-2 to autograft, were able to demonstrate subsequent induction of BMP-6, osteocalcin, and collagen within the graft itself.

Control 0.75 mg/ml 1.50 mg/ml

Fig. 2. Threaded titanium interbody cage implanted into the lumbar spine of a primate. Control group is cage alone. Experimental groups are rhBMP-2-impregnated sponges at concentrations of 0.75 and 1.50mg/ml. Actual spines at 6 months from surgery. Control demonstrates a pseudarthrosis with fibrous scar tissue within the cage. BMP groups both show mature bone bridging the interspace within the cage.

Control 0.75 mg/ml 1.50 mg/ml

Fig. 2. Threaded titanium interbody cage implanted into the lumbar spine of a primate. Control group is cage alone. Experimental groups are rhBMP-2-impregnated sponges at concentrations of 0.75 and 1.50mg/ml. Actual spines at 6 months from surgery. Control demonstrates a pseudarthrosis with fibrous scar tissue within the cage. BMP groups both show mature bone bridging the interspace within the cage.

Posterolateral fusion models attempted to replace autograft with a BMP/ carrier complex, but still required internal fixation. Attention was then focused upon interbody spinal fusion techniques, which could possibly obviate the need for both autograft and fixation [19, 20, 34, 42, 47, 48, 55, 56, 64, 65]. Zdeblick et al. [73] performed three level anterior cervical fusions in goats using a BAK cage filled with either local autograft or a collagen impregnated rhBMP-2 sponge [49]. Eleven of 21 animals (52%) in the autograft group had histologic evidence of pseudarthrosis, while only 1 (5%) of the BMP group failed to form a solid bony fusion. The biomechanical stiffness of the BMP construct was equal to that of an autograft/cervical-plated level. Boden et al. [8] performed the same procedure in the lumbar spine of primates. Using rhBMP-2 on a collagen carrier in both titanium-threaded interbody cages and threaded bone dowels, both were delivered laparoscopically with a documented improvement of fusion rates over empty cages and autograft-filled cages [8, 57] (fig. 2, 3).

Preliminary results of human clinical studies have been encouraging. In a recent report of a randomized, prospective controlled clinical pilot study, Boden et al. [8] demonstrated solid bony fusions by both clinical and radiographic

Bone Morphogenetic Protein

Fig. 3. Bone dowel interbody device implanted into the lumbar spine of a primate. Control group is dowel with autograft. Experimental group is a bone dowel with an rhBMP-2-impregnated sponge inside. Actual spines at 6 months from surgery. Control group demonstrates a solid fusion (a) and reabsorption of the bone dowel/autograft construct (c). BMP animals (b, d) both show mature bone bridging the interspace with resorption of the dowel.

Fig. 3. Bone dowel interbody device implanted into the lumbar spine of a primate. Control group is dowel with autograft. Experimental group is a bone dowel with an rhBMP-2-impregnated sponge inside. Actual spines at 6 months from surgery. Control group demonstrates a solid fusion (a) and reabsorption of the bone dowel/autograft construct (c). BMP animals (b, d) both show mature bone bridging the interspace with resorption of the dowel.

criteria in 11 patients undergoing anterior lumbar interbody fusion procedures with a tapered, threaded titanium cage filled with an rhBMP-2-impregnated collagen sponge (fig. 4). Pain scores, as documented by the Oswestry Disability and the Short Form-36 questionnaires, improved concomitantly as fusion progressed [8, 30, 39, 56, 66, 69].

Based upon the findings in both the primate models and human trials, the United States Food and Drug Administration (FDA) approved rhBMP-2 for limited use in patients with spinal disorders. Specifically, the FDA approved BMP for use as an adjunct to spinal fusion in patients with degenerative disc disease undergoing anterior lumbar interbody fusion using titanium cages (LT cages).

Doctor Patient Impregnation

Fig. 4. Computed tomographic reconstructions of a human patient receiving a threaded, titanium cage (Lordotec cage, Medtronic Sofamor Danek, Memphis, Tenn., USA) with rhBMP-2 on a collagen sponge carrier. a Sagittal view of the left cage at 6, 12 and 24 months. b Right cage at the same time intervals. c Coronal views at the same time intervals. All images show fusion with increasing bone density over time.

Fig. 4. Computed tomographic reconstructions of a human patient receiving a threaded, titanium cage (Lordotec cage, Medtronic Sofamor Danek, Memphis, Tenn., USA) with rhBMP-2 on a collagen sponge carrier. a Sagittal view of the left cage at 6, 12 and 24 months. b Right cage at the same time intervals. c Coronal views at the same time intervals. All images show fusion with increasing bone density over time.

The cages attempt to both restore and maintain intervertebral height and protect the BMP from exposure to diluting substances such as blood and irrigation fluid. Although approved, BMP has yet to reach the market. Concerns raised by experts include the improper use of the BMP such that vascular and neural elements may come in contact with the protein causing injury, lower fusion rates due to improper or unapproved implantation techniques, and stimulation of infectious or neoplastic processes due to use in patients with such diseases.

Discussion

Research over the past decade has shown the utility of using the growth and differentiation factor, BMP-2, to promote bone formation at the site of bone loss or injury. The in vivo role of BMP-2 and its complex interaction with other growth and differentiation factors remains to be clarified. The use of BMP-2 as a means of replacing harvested autograft and obviating the need for internal fixation, each with its attendant morbidity, appears likely as a result of outcomes from both animal and human studies. Although dose-effect relationships and carrier substrates may provide continued investigational challenges, the use of recombinant technology and gene therapy in the field of bone fusion have been firmly established. In the past 30 years since Marshall Urist first coined the term bone morphogenetic protein, one doubts that he could have envisioned the monumental strides and clinical progress, which researchers in the field have achieved to this point.

Conclusions

The widespread use of spinal fusion procedures in the management of spinal disorders has led investigators to explore the use of growth and differentiation factors in such procedures. Either as an adjunct to allograft bone or as a replacement for harvested autograft, BMPs appear to improve fusion rates after spinal arthrodesis in both animal and human models, while reducing the donor site morbidity previously associated with such procedures [2, 6, 11, 14, 23, 25,

50, 54, 72]. The use of recombinant genetic technology in the production of BMP has improved the efficiency, cost-effectiveness, and safety of producing and using such materials. rhBMP-2, as one of the first factors identified in the process of endochondral bone formation, has been extensively researched over the past decade. The efficacy and dose profile of this differentiation factor in the context of various carrier substrates has been investigated [15, 26, 27, 33,

51, 58, 61]. Based upon the encouraging results of preliminary studies, the future role of rhBMP-2 may lie in the elimination of autologous bone grafting, the reduction of the need for instrumented fixation, and augmentation of accepted fusion rates.

Acknowledgments

The authors wish to thank Medtronic Sofamor Danek for their continued support and assistance in the preparation of the figures.

References

1 Aronow MA, Gerstenfeld LC, Owen TA, Tassinari MS, Stein GS, Lian JB: Factors that promote progressive development of the osteoblast phenotype in cultured rat calvarial cells. J Cell Physiol

1990;143:213-221.

Arlington ED, Smith WJ, Chambers HG, Bucknell AL, Davino NA: Complications of iliac crest bone graft harvesting. Clin Orthop 1996;329:300-309.

Asahina I, Sampath TK, Nishimura I, Hauschka PV: Human osteogenic protein-1 induces both chondroblastic and osteoblastic differentiation of osteoprogenitor cells derived from newborn rat calvaria. J Cell Biol 1993;123:921-933.

Aspenberg P, Jeppsson C, Wang JS, Bostrom M: Transforming growth factor beta and bone mor-phogenetic protein 2 for bone ingrowth: A comparison using bone chambers in rats. Bone 1996; 19:499-503.

Aspenberg P, Turek T: BMP-2 for intramuscular bone induction: Effect in squirrel monkeys is dependent on implantation site. Acta Orthop Scand 1996;67:3-6.

Banwart JC, Asher MA, Hassanein RS: Iliac crest bone graft harvest donor site morbidity: A statistical evaluation. Spine 1995;20:1055-1060.

Boden SD, Hair G, Titus L, Racine M, McCuaig K, Wozney JM, Nanes MS: Glucocorticoid-induced differentiation of fetal rat calvarial osteoblasts is mediated by BMP-6. Endocrinology 1997;138:2920-2928.

Boden SD, Martin GJ, Horton WC, Truss TL, Sandhu HS: Laparoscopic anterior spinal arthrod-esis with rhBMP-2 in a titanium interbody threaded cage. J Spinal Disord 1998;11:95-101. Boden SD, McCuaig K, Hair G, Racine M, Titus L, Wozney JM, Nanes MS: Differential effects and glucocorticoid potentiation of bone morphogenetic protein action during rat osteoblast differentiation in vitro. Endocrinology 1996;137:3401-3407.

Boden SD, Moskovitz PA, Morone MA, Toribitake Y: Video-assisted lateral intertransverse process arthrodesis: Validation of a new minimally invasive lumbar spinal fusion technique in the rabbit and nonhuman primate (rhesus) models. Spine 1996;21:2689-2697.

Boden SD, Schimandle JH: Biologic enhancement of spinal fusion. Spine 1995;20(suppl):

Back Pain Revealed

Back Pain Revealed

Tired Having Back Pains All The Time, But You Choose To Ignore It? Every year millions of people see their lives and favorite activities limited by back pain. They forego activities they once loved because of it and in some cases may not even be able to perform their job as well as they once could due to back pain.

Get My Free Ebook


Post a comment