Spinal arthrodesis is useful for treating a variety of acquired and degenerative conditions of the lumbar spine. The classic technique for achieving spinal fusion involves placing autogenous bone graft between the decorticated spinal surfaces. In the posterolateral spine, autogenous bone fusion has been shown to mature through a series of steps including inflammation, fibrocartilage formation, endochondral ossification, and final remodeling. 1 Unfortunately, 5% to 40% of attempted lumbar fusions fail to fuse. 2 Fusion failure results in the formation of a spinal pseudarthrosis, and may continue to be the source of spinal symptoms in up to 57% of patients. 3
In addition to pseudarthrosis, autograft donor site morbidity also can compromise the clinical results after spinal fusion surgery. The incidence of donor site complications reportedly is 6% to 25%, and can include chronic pain, hypersensitivity, fracture, hernia, cosmetic deformity, and surrounding anesthesia. 4–7
Allograft bone has been used for spinal fusion to avoid the problems associated with autogenous bone harvest. Unfortunately, in most applications, allograft bone has produced inferior fusion rates and higher rates of graft resorption, as compared with autogenous bone, because it has little or no demonstrable osteoinductive potential. Other concerns with the use of allograft include the risks of host immunologic rejection and infectious agent transmission. Therefore, allograft bone, although useful in certain applications, has not supplanted autograft as the ideal source of bone graft for spinal fusion. 5,8,9
Osteogenic proteins were first reported by Urist 10 in 1965. These related molecules, transcribed from the TGFβ superfamily of genes, are able to stimulate bone formation in extraosseous tissues by inducing differentiation of pluripotent precursor cells along an osteogenic line (Figure 1). The discovery of osteogenic proteins has created significant enthusiasm in recent years because they pose a potential therapeutic modality to induce fracture healing and promote bone fusion. Many investigators and corporations have invested significant resources into the characterization and purification of these osteogenic molecules in the hope that they would extend, enhance, or replace autograft bone. One osteogenic molecule recombinant human osteogenic protein-1 (rhOP-1), also known as bone morphogenetic protein-7 (BMP-7), is the focus of this article.
Osteogenic protein 1 has been isolated by molecular cloning techniques and introduced into a Chinese hamster ovarian cell line, which is able to express rhOP-1. The protein product, a homodimer, is purified by column chromatography to greater than 97% purity. The mature, processed protein has a molecular mass of 30 to 49 kDa as the subunits are differentially glycosylated. 11
The commercially available rhOP-1 product is marketed by Stryker Biotech (Hopkinton, MA). In the commercial form, 3.5 mg of lyophilized rhOP-1 is combined with a carrier comprising 1 g of Type 1 bovine bone collagen for a final rhOP-1 concentration of 0.875 mg/mL. The dry powdered mixture is reconstituted by the addition of saline to form a paste just before implantation (Figure 2). The formulation of the OP-1 implant (3.5 mg of rhOP-1 to 1 g of bovine bone collagen) was designed to allow local delivery of OP-1 to orthopedic sites. The carrier serves to deliver and contain the osteogenic molecules at the site of fusion and acts as a resorbable matrix for bone growth. The addition of 230 mg of carboxymethylcellulose to the OP-1 implant forms a putty that currently is being used in clinical trials of spinal fusion. A biodistribution study performed in rabbits demonstrated that after implantation in an osseous defect, the OP-1 remains well contained in the site and locally metabolized. Approximately 20% to 27% of the OP-1 remains at the surgical site 5 days after implantation. By 28 days after implantation, only 0.02% of the implanted dose remains (unpublished data, C. Toth and A. Pierce).
The dosage selection of OP-1 was based on several considerations. First, little or no species specificity or species-specific sensitivity is expected for the OP-1 protein because there is extensive structural conservation of OP-1 across species. For example, there is 98% sequence identity of the amino acid sequence in the mature region of OP-1 between mouse and human. 12 This lack of species specificity and sensitivity is reinforced by studies involving healing of a critical-size segmental long bone defect. Various species including dog, nonhuman primate, and human have demonstrated the ability of such a defect to heal reliably using the OP-1 implant (3.5 mg of OP-1 to 1 g of collagen). 13–16 This article reviews the animal studies and early clinical trials in humans related to the use of this product to induce spinal fusion.
Animal Studies of rhOP-1
Cook et al 17 performed posterolateral fusions in a dog model using either autograft bone or OP-1 (2.8 mg of rhOP-1 to 800 mg of bovine collagen). In this study, nine dogs were divided into groups of three and subjected to posterolateral fusion using four implant substances including OP-1 with a collagen carrier, bovine bone collagen carrier alone, autologous iliac crest bone, or no implant material. The implants were randomized to vertebral locations in such a way that each animal received all four types of implants. The three groups of animals were killed at 6, 12, and 26 weeks, and the spines were subjected to CT imaging, MRI imaging, nondestructive mechanical testing, and histology.
All the OP-1 treated levels demonstrated stable fusion by 6 weeks and complete fusion by 12 weeks. The autologous bone–treated levels demonstrated a slower progression to fusion by 26 weeks. No fusion was observed at the levels treated with either the collagen carrier alone or no implant. Histologic observations were consistent with the radiologic results, with OP-1 inducing bone formation faster than autograft, which resulted in earlier, more mature fusion masses. Biomechanical testing demonstrated that the OP-1 fusion sites were significantly stiffer than those treated with the carrier alone or no implant sites by 6 weeks. By 26 weeks, both the autograft-treated and OP-1–treated sites were statically stiffer than those of the control subjects (P < 0.05).
Magin and Delling 18 compared OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen) with an osteoconductive hydroxyapatite bone graft substitute or autograft bone in a sheep interbody fusion model. In their study, 30 sheep underwent interbody fusion via a posterolateral approach and supplemental transpedicular instrumentation. The sheep were evaluated at intervals with plain radiographs, bone scintigraphy, CT imaging, and MRI. At 6 months after the surgical procedure, the animals were killed, and the spines were subjected to mechanical testing and histology. Fluorochrome labeling was performed to study bone turnover. To evaluate the presence of fusion, a scoring system was used to grade both plain radiographs and CT images. The amount of bone formation was statistically higher in the OP-1–treated animals than in either the autograft- or hydroxyapatite-treated animals by 4 months (P < 0.05). Mechanical testing and histology confirmed the maturity and stiffness of the fusion in the OP-1–treated animals, which were significant, as compared with those qualities in the hydroxyapatite group (P < 0.05) (Figure 3). Bone scintigraphy demonstrated significantly less activity in the OP-1 group at the 6-month time point than in the other groups, signifying a more mature fusion mass in these animals.
Chirossel et al 19 studied interbody fusion in a sheep model using either a polyetheretherketone (PEEK) cage or a titanium cage. The cages were filled with OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC), autograft bone, or no filling material. Fusion was performed in 22 animals at either L3–L4 or L4–L5. The animals were evaluated using radiographs, histomorphometry, fluorochome labeling, CT imaging, and histology. Two animals were killed early because of non–device-related complications and removed from the analysis. The remaining animals were killed at 24 weeks. Solid bony fusion was observed in three of four titanium cage/autograft animals, three of five titanium cage/rhOP-1 animals, two of four PEEK/autograft animals, and four of five PEEK/rhOP-1 animals (Figure 4). Although the numbers were too small for statistical significance, histology showed mature trabecular bone within the fusion sites in the OP-1–treated animals and immature woven bone in the autograft-treated fusion sites (Figure 5). An almost complete absence of the 4-week fluorochrome label was observed in the PEEK/rhOP-1–treated group, signifying nearly a complete turnover of the bone formed at this early time point.
Cunningham et al 11 compared OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen) with autograft bone in a skip-level sheep thoracic spine model using threaded fusion cages. In this study, 12 sheep underwent thoracoscopic surgery at T5–T6, T7–T8, and T9–T10. The levels were randomly treated by either a control method (nonsurgical, reamed disc space, or BAK device without graft) or an experimental method (autograft dowel, BAK and autograft, or BAK and rhOP-1 putty). The sheep were killed 16 weeks after surgery. The spines were subjected to CT imaging, biomechanical testing, microradiography, histomorphometry, and histology. Microradiographs showed fusion in one of six reamed disc space sites, two of six BAK without graft sites, four of eight autograft dowel sites, five of eight BAK and autograft sites, and six of eight BAK and rhOP-1 sites (Figure 6). Both the autograft and OP-1–filled cage were noted to be significantly stiffer than the empty BAK device (P < 0.05), but were not significantly different from one another.
Grauer et al 20 compared rhOP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC) with autograft or collagen–CMC carrier alone in a rabbit posterolateral fusion model. Posterolateral fusion was performed in 31 rabbits at L5–L6 using one of the three substances: OP-1, autograft, or carrier alone. The animals were killed 5 weeks after the procedure. The spines were evaluated by manual palpation, biomechanical testing, and histology. Seven rabbits (23%) were killed early because of complications and excluded from the analysis. By manual palpation, five of eight (63%) rabbits in the autograft group, none of eight (0%) rabbits in the carrier group, and all eight rabbits (100%) in the OP-1 group achieved a solid fusion. Statistical significance was achieved in both the autograft and OP-1 groups, as compared with the carrier-alone group (P < 0.05). Biomechanically, flexion stiffness was greatest in the OP-1 group and least in the carrier-alone group (P < 0.05). Histologically, predominant fibrocartilage was observed at the autograft fusion sites, whereas the OP-1 sites demonstrated maturing trabecular bone surrounded by a cortical shell.
Cunningham et al 21 studied the fusion time course in a skip-level posterolateral dog model using autograft alone, autograft and OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC), or OP-1 alone. In this study, 36 animals were divided into four groups (4-, 8-, 12-, and 24-week sacrifice) and subjected to fusion at L3–L4 and L5–L6, yielding 18 sites treated with each implant at each time point. The dog spines were evaluated using radiography and biomechanical testing to determine the fusion status. At 4 weeks, radiographic fusion was present in 0 of 18 (0%) autograft sites, 7 of 18 (38%) autograft and rhOP-1 sites, and 4 of 18 (22%) rhOP-1 sites. At 8 weeks, 4 of 18 (22%) autograft sites, 16 of 18 (88%) autograft and rhOP-1 sites, and 12 of 18 (66%) rhOP-1 sites demonstrated fusion. At 12 weeks, 5 of 18 (27%) autograft sites, 15 of 18 (83%) autograft and rhOP-1 sites, and 13 of 18 (72%) rhOP-1 sites demonstrated fusion. At each time point, the difference between the autograft alone and the OP-1–containing sites was statistically significant (P < 0.01) (Figures 7–9). Mechanical testing failed to demonstrate a significant difference between the groups at 4 weeks, but did demonstrate significantly increased stiffness in the OP-1 sites, as compared with the autograft-alone sites, at the 8- and 12-week time points.
Paramore et al 22 studied the toxicity of OP-1 by intentionally placing it into the subarachnoid space during a lumbar laminectomy and fusion in a dog model. In this study, 30 dogs underwent posterior exposure of the spine followed by a L2 laminectomy and durotomy. Four control animals underwent dural closure and autologous posterolateral fusion using bone from the laminectomy site. The study animals underwent implantation of OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen) with and without 230 mg of CMC within the dural sac followed by dural closure. Posterolateral fusion was performed by placing the remaining OP-1 mixture and the autograft from the laminectomy site between the decorticated transverse processes. The animals were killed after 16 weeks. Assessment of the spines was carried out by palpation for fusion as well as CT scanning and histology of the spine and spinal cord. Two animals in the OP-1 group were killed early because of paraplegia caused by epidural hematomas. According to palpation criteria, 80% of the OP-1 animals and 25% of the autograft animals achieved a solid fusion (P < 0.05). The animals subjected to implantation of OP-1 in the subarachnoid space demonstrated bone formation adjacent to the spinal cord that caused mild spinal cord compression. Spinal cord histology demonstrated inflammation adjacent to the newly formed bone, but no evidence of spinal cord inflammation or neuron cell death.
Patel et al 23 studied the ability of OP-1 to overcome the inhibitory effects of nicotine in a rabbit posterolateral fusion model. In this study, 18 rabbits underwent single-level posterolateral fusions at L5–L6 using either autograft bone or OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC). Nicotine was administered to all the animals via a subcutaneous mini-osmotic pump. The animals were killed 5 weeks after surgery, and the fusion sites were assessed by biomechanical testing, palpation, radiography, and histology. Three rabbits had complications and were excluded from the assessment. As determined by manual palpation, only two of eight (25%) autograft sites were solidly fused, whereas all of the OP-1 fusion sites (7/7) were solidly fused (P < 0.01). Biomechanical testing confirmed the results of manual palpation. Histologically, the autograft-treated animals demonstrated very immature fusion masses, whereas the OP-1–treated animals demonstrated predominantly mature bone at the fusion sites. The authors concluded that OP-1 was able to overcome the inhibitory effects of nicotine on a spinal fusion in the rabbit model.
A safety and efficacy study has been completed in humans comparing autograft alone to autograft augmented with OP-1 for posterolateral spinal arthrodesis. 24 Sixteen patients with degenerative lumbar spondylolisthesis and spinal stenosis were randomized to uninstrumented posterolateral fusion with autograft and OP-1 (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC) or autograft alone. The clinical outcome was assessed using the Oswestry score, whereas fusion status was graded radiographically by two blinded radiologists using dynamic and static radiographs. At the 6-month follow-up assessment, 9 of 12 (75%) patients in the autograft and OP-1 group were graded as fused, whereas 2 of 4 (50%) patients in the autograft-alone group achieved fusion (P = 0.547). Clinical success, defined by at least a 20% improvement in the Oswestry score, was achieved in 83% of the autograft and OP-1 group, but in only 50% of the autograft-alone group (P = 0.245). No adverse side effects were observed in the OP-1–treated patients.
A similar study involving patients with degenerative spondylolisthesis is ongoing in Australia. After decompression, the patients undergo an uninstrumented posterolateral arthrodesis with an autograft on the one side and OP-1 putty (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC) on the contralateral side. Preliminary results for this study have been reported by Speck 25 at the 6-month time point. Using CT scan to measure bone formation between the transverse processes, the fusion masses of five patients have been evaluated. In all cases, bone formation was noted to be equal or greater on the OP-1 side as compared with the autograft side (Figures 10–12).
An additional pilot trial of OP-1 for posterolateral spinal fusion is ongoing. Patients with lumbar spinal stenosis and degenerative spondylolisthesis (L3–L5 segments) are treated with decompression, followed by uninstrumented posterolateral fusion using either iliac crest autograft alone or OP-1 putty (3.5 mg of rhOP-1 to 1 g of bovine bone collagen to 230 mg of CMC) alone. Outcome is assessed clinically by the Oswestry scale, and fusion status is graded radiographically by blinded radiologists. To be considered a clinical success, patients must achieve at least a 20% improvement in their Oswestry score and demonstrate a solid arthrodesis with bony bridging and no motion on flexion–extension radiographs. The 6-month results for 36 enrolled patients have been reported. 26 The overall clinical success rate was noted to be 32% higher in the OP-1–treated patients than in the autograft-treated patients. Although the difference between autograft and OP-1 in this study is not yet statistically significant, the trend suggests that the OP-1 group may achieve a statistically significant benefit with further patient enrollment. At this writing, no OP-1–related adverse events have been observed.
Therapeutic growth factors appear to have significant potential for enhancing spinal fusion. Animal studies and early human clinical trials have shown the efficacy of OP-1 as an alternative or enhancer of autologous bone for spinal fusion. Importantly, no serious adverse effects of OP-1 have been observed in these trials. Further work is necessary to determine the optimal dosing and carrier for further improvement on the preliminary results that have been reported so far. It is not yet clear whether OP-1 will replace or augment autograft bone for the wide spectrum of spinal pathology encountered clinically. Each clinical scenario (anterior or posterior with or without instrumentation) will have to be analyzed individually to determine the efficacy and cost effectiveness of OP-1 before there can be widespread acceptance of bone morphogenetic proteins in spinal applications. As with all new developments, each answered question creates new areas requiring investigation.
- OP-1 (BMP-7) is a strongly osteoinductive bone morphogenetic protein manufactured by recombinant DNA technology.
- OP-1 has shown efficacy in promoting spinal fusion in animal models in both the anterior and posterolateral environments.
- OP-1 has shown efficacy as a bone graft enhancer and bone graft substitute in an animal model of posterolateral spinal fusion.
- OP-1 is able to overcome the negative effects of nicotine on spinal fusion in an animal model.
- Early human trials support the safety and efficacy of OP-1 putty for posterolateral spinal arthrodesis.
1. Heiple KG, Goldberg VM, Powell AE. Biology of cancellous bone grafts. Orthop Clin North Am 1987; 18: 179–85.
2. Cotler JM, Star AM. Complications of spinal fusion
. In: Cotler JM, Cotler HB, eds. Spine fusions: Science and Technique. New York: Springer-Verlag, 1990.
3. Watkins, MB, Bragg C. Lumbosacral fusion
: Results with early ambulation. Surg Gynecol Obstet 1956; 102: 605.
4. Cockin J. Autologous bone grafting: Complications at the donor site. J Bone Joint Surg Br 1971; 52: 153.
5. Damien CJ, Parsons JR. Bone graft and bone graft substitutes: A review of current technology and applications. J Appl Biomater 1991; 2: 187–208.
6. Summers BN, Eisenstein SM. Donor site pain from the ilium: A complication of lumbar spine fusion
. J Bone Joint Surg Br 1989; 71: 677–80.
7. Younger EM, Chapman MW. Morbidity at bone graft donor site. J Orthop Trauma 1989; 3: 192–5.
8. Muschler GF, Lane JM, Dawson EG. The biology of spinal fusion
. In: Cotler JM, Cotler HB, eds. Spine Fusion
: Science and Technique. New York: Springer-Verlag, 1990.
9. Horowitz MC, Friedlaender GE. The immune response to bone grafts. In: Friedlanger GE, Goldberg WM, eds. Bone and Cartilage Allografts. Park Ridge, IL: American Academy of Orthopaedic Surgeons, 1991.
10. Urist MR. Bone formation by autoinduction. Science 1965; 150: 893–9.
11. Cunningham BW, Kanayama M, Parker LM, et al. Osteogenic protein versus
autologous interbody arthrodesis
in the sheep thoracic spine. Spine 1999; 24: 509–18.
12. Ozkaynak E, Schnegelsberg PN, Oppermann H. Murine osteogenic protein (OP-1): High levels of mRNA in kidney. Biochem Biophys Res Commun 1991; 179: 116–23.
13. Geesink RG, Hoefnagels NH, Bulstra SK. Osteogenic activity of OP-1 bone morphogenetic protein (BMP-7) in a human fibular defect. J Bone Joint Surg [Br] 1999; 81: 710–18.
14. Cook SD, Baffes GC, Wolfe MW, et al. Recombinant human bone morphogenetic protein-7 induces healing in a canine long-bone segmental defect model. Clin Orthop 1994; 302–12.
15. Cook SD, Baffes GC, Wolfe MW, et al. The effect of recombinant human osteogenic protein-1 on healing of large segmental bone defects. J Bone Joint Surg [Am] 1994; 76: 827–38.
16. Cook SD, Wolfe MW, Salkeld SL, et al. Effect of recombinant human osteogenic protein-1 on healing of segmental defects in nonhuman primates. J Bone Joint Surg [Am] 1995; 77: 734–50.
17. Cook SD, Dalton JE, Tan EH, et al. In vivo
evaluation of recombinant human osteogenic protein (rhOP-1) implants as a bone graft substitute for spinal fusion
. Spine 1994; 19: 1655–63.
18. Magin M, Delling G. Improved lumbar vertebral interbody fusion
using rhOP-1. Spine 2001; 26: 469–78.
19. Chirossel JP, Gay E, Boutrand JP, et al. Stryker spine PEEK and titanium cages: Interbody fusion
with OP-1 (BMP-7) vs
autograft in a sheep model. Presented at Eurospine, Antwerp, Belgium, 2000.
20. Grauer JN, Patel TC, Erulkar JS, et al. Evaluation of OP-1 as a graft substitute for intertransverse process lumbar fusion
. Spine 2000; 26: 127–33.
21. Cunningham BW, Shimanoto N, Sefter JC, et al. Posterolateral Spinal Arthrodesis
Using Osteogenic Protein-1: An in Vivo
Time-Course Study Using a Canine Model. NASS, New Orleans, LA, 2000.
22. Paramore CG, Lauryssen C, Rauzzino J, et al. The safety of OP-1 for lumbar fusion
with decompression: A canine study. Neurosurgery 1999; 44: 1151–6.
23. Patel TC, Erulkar JS, Grauer JS. OP-1 overcomes the inhibitory effects of nicotine on lumbar fusion
. NASS, New Orleans, LA, 2000:2000.
24. Patel TC, Vaccaro AR, Truumees E, et al. A Safety and Efficacy Study of Op-1 (Rhbmp-7) as an Adjunct to Posterolateral Lumbar Fusion
. Presented at the North American Spine Society Meeting, Seattle, Washington, 2001.
25. Speck G. Posterolateral Fusion
Using OP-1: A model Using Degenerative Spondylolisthesis. Presented at the Australian Spine Meeting, Adelaide, Australia, 2000.
26. Patel TC, McCulloch JA, Vaccaro AR, et al. A Pilot Safety and Efficacy Study of OP-1 (rhBMP-7) in Posterolateral Lumbar Fusion
as a Replacement for Iliac Crest Autograft. Presented at the North American Spine Society 16th Annual Meeting. Seattle, Washington, 2001.
Keywords:© 2002 Lippincott Williams & Wilkins, Inc.
arthrodesis; fusion; OP-1, bone morphogenetic protein; review]Spine 2002;27:S59–S65