A porcine collagen-derived matrix as a carrier for recombinant human bone morphogenetic protein-2 enhances spinal fusion in rats

The Spine Journal 9 (2009) 22–30

2008 Outstanding Paper Award Runner-up

A porcine collagen-derived matrix as a carrier for recombinant human bone morphogenetic protein-2 enhances spinal fusion in rats Masashi Miyazaki, MDa, Yuichiro Morishita, MD, PhDb, Wubing He, MDb, Ming Hu, MDb, Chananit Sintuu, MSc, Henry J. Hymanson, BSb, Jonathan Falakassa, BSb, Hiroshi Tsumura, MD, PhDa, Jeffrey C. Wang, MDb,* b

a Department of Orthopaedic Surgery, Oita University, Oita, Japan Department of Orthopaedic Surgery, University of California at Los Angeles, Los Angeles, CA 90095, USA c Department of Bioengineering, University of California at Los Angeles, Los Angeles, CA 90095, USA

Received 6 February 2008; accepted 5 August 2008

Abstract

BACKGROUND CONTEXT: Recombinant bone morphogenetic proteins (rhBMPs) have been used successfully in clinical trials. However, large doses of rhBMPs were required to induce adequate bone repair. Collagen sponges (CSs) have failed to allow a more sustained release of rhBMPs. Ongoing research aims to design carriers that allow a more controlled and sustained release of the protein. E-Matrix is a injectable scaffold matrix that may enhance rhBMP activity and stimulate bone regeneration. PURPOSE: The purpose of this study was to test E-Matrix as a carrier for rhBMPs in a CS and examine its feasibility in clinical applications by using a rat spinal fusion model. PATIENT SAMPLE: A total of 80 Lewis rats aged 8–16 weeks were divided into nine groups. STUDY DESIGN/SETTING: Rat spinal fusion model. OUTCOME MEASURES: Radiographs were obtained at 4, 6, and 8 weeks. The rats were sacrificed and their spines were explanted and assessed by manual palpation, high-resolution microcomputed tomography (micro-CT), and histologic analysis. METHODS: Group I animals were implanted with CS alone (negative control); Group II animals with CS containing 10 mg rhBMP-2 (positive control); Group III animals with CS containing 3 mg rhBMP-2; Group IV animals with CS containing 3 mg rhBMP-2 and E-Matrix; Group V animals with CS containing 1 mg rhBMP-2; Group VI animals with CS containing 1 mg rhBMP-2 and E-Matrix; Group VII animals with CS containing 0.5 mg rhBMP-2; Group VIII animals with CS containing 0.5 mg rhBMP-2 and E-Matrix; and Group IX animals with CS and E-Matrix without rhBMP-2. RESULTS: Radiographic evaluation, micro-CT, and manual palpation revealed spinal fusion in all rats in the BMP-2 and E-Matrix groups (IV, VI, and VIII) and high-dose BMP-2 groups (II and III). Four spines in the 3 mg rhBMP-2 group (V) fused, and one spine in the 0.5 mg rhBMP-2 group (VII) exhibited fusion. No spines were fused in Groups I (CS alone) and IX (E-Matrix alone). The volume of new bone in the area between the tip of the L4 transverse process and the base of the L5 transverse process in Group IV was equivalent to the volumes observed in Group II. CONCLUSION: E-matrix enhances spinal fusion as a carrier for rhBMP-2 in a rat spinal fusion model. The results of this study suggest that E-Matrix as a growth factor carrier may be applicable to spinal fusion and may improve rhBMP-2’s activity at the fusion site. Ó 2009 Elsevier Inc. All rights reserved.

Keywords:

E-matrix; Carrier; Bone morphogenetic protein; Spinal fusion; Rat model

FDA device/drug status: approved for this indication (E-Matrix). The author, JCW, acknowledges a financial relationship (receives royalties from Medtronic, DePuy, SeaSpine, Biomet, Aesculap, Stryker; consultant for Synthes; on the Advisory board of K2M, UCLA Bone-Biologics; received research support from Pioneer), which may indirectly relate to the subject of this article. 1529-9430/09/$ – see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2008.08.009

* Corresponding author. Orthopaedic Spine Service, UCLA Comprehensive Spine Center, 1250 16th Street, Suite 745, Santa Monica, CA 90404, USA. Tel.: (310) 319-3334; fax: (310) 319-5055. E-mail address: [email protected] (J.C. Wang)

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

23

Introduction Studies on animal models and clinical trials have demonstrated the osteoinductive effects of recombinant bone morphogenetic proteins (rhBMPs) in various surgical procedures, such as fracture repair, healing of critical-size bone defects, and spinal fusion [1–9]. However, the results of clinical trials show that high doses of BMPs are required to induce adequate spinal fusion because the molecules are soluble and can diffuse away from the fusion site easily and become inactivated in vivo [10]. BMPs are highly expensive and their usefulness is therefore limited; moreover, they may cause local adverse effects such as unwanted ectopic bone formation and inflammation, particularly when used in the cervical spine [11,12]. A number of strategies are being developed to provide a safer, less expensive, and more efficacious spinal fusion using rhBMP. Some ongoing strategies aim at designing carriers that allow a more controlled and sustained release of the protein so that the growth factor concentration is maintained locally within the therapeutic range. A collagen sponge (CS) has been used as a carrier matrix for rhBMPs over the last several years because of its excellent biocompatibility, ability to degrade into physiological end-products, and suitable interaction with cells and molecules. Both basic and clinical studies have demonstrated its safety [13–15]. Although collagen sponges allow a longer local retention of rhBMPs at the implantation site than a buffer alone [2], there are perhaps more optimal carriers that may induce the maximum therapeutic effect. E-Matrix is an injectable scaffolding matrix for cellular attachment derived from porcine skin collagen. It has been developed to accelerate wound healing in diabetic foot ulcers [16]. Prior studies on E-Matrix have shown that it enhances the production of growth factors, such as transforming growth factor-b-3 and vascular endothelial-derived growth factor receptors [17]. It is assumed that the interaction of host cells with E-Matrix leads to altered cellular responses, accelerating tissue regeneration. Prior clinical trials have shown E-Matrix to be safe [16]. We hypothesize that the incorporation of EMatrix and rhBMP-loaded absorbable collagen sponge is feasible, that it allows a more sustained release of rhBMPs than the collagen sponge alone, and that the combination will stimulate more efficacious bone regeneration. The purpose of this study was to develop E-Matrix with a collagen sponge as a carrier matrix for rhBMPs and examine its feasibility in clinical application by using a rat spinal fusion model. Materials and methods E-Matrix is a biocompatible scaffold for cellular attachment; it is composed of gelatin alpha chains derived from porcine skin collagen stabilized by copolymerization with a high-molecular weight polysaccharide (500 kDa dextran). It is designed such that the open polar alpha chains have the maximum possible number of hydrogen bonding sites for

Context Recombinant human bone morphogenetic protein (rhBMP) for spinal fusion has demonstrated efficacy in human trials when used in the anterior lumbar spine. The currently available forms use a collagen sponge carrier. The optimum use of rhBMP (concentration, carrier, application timing, etc) in different fusion techniques remains unclear. Negative side effects of rhBMP, such as graft resorption, sterile fluid collections, and airway swelling in the cervical spine, are suspected to be dose and application dependant. Contribution The current study evaluated the efficacy of an additive carrier, called E-Matrix, to rhBMP-2 in a rat spinal fusion model. The investigators hypothesized that previously suggested efficacy of E-Matrix as a growth factor enhancer might augment the effects for rhBMP-2. The authors compared nine different combinations of combinations of rhBMP-2 with E-matrix. They detected improved radiographic fusion scores and the fusion rates with some combination preparations at lower than usual concentrations of rhBMP. Implications The direct clinical implications for human use of this animal study are necessarily limited. Nor is this study comparable to most animal spineBMP research which has been performed using a rabbit model. Also, the long-term safety of the dosages and combinations used have not been assessed in animal or human applications. However, assuming further research can adequately address these limitations, E-matrix or a similar additive carrier may augment fusion at a lower, and perhaps safer, concentration of rhBMP-2. dTSJ Editors polar amino acids. This open polar structure was designed to mimic the open structure of early embryonic dermis [16] (Fig. 1). Pioneer Surgical Orthobiologics (Greenville, NC) provided the E-Matrix for this study. Rat spinal fusion model Preparation of carrier matrices RhBMP-2 was applied to a CS carrier (Helistat; Integra Life Sciences, Plainsboro, NJ) with or without E-Matrix

24

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

posterior midline incision was made in the skin. Next, two separate paramedian incisions were made 3 mm from the midline in the lumbar fascia, and the transverse processes were exposed through these incisions. The transverse processes of L4 and L5 were decorticated using a low-speed burr. Subsequently, CS with or without recombinant BMP-2 and E-Matrix was implanted on each side. The fascial and skin incisions were closed with a 3–0 absorbable suture. The rodents were housed in separate cages and fed food and water ad libitum, and their conditions were monitored on a daily basis.

Fig. 1. Schematic diagram of the E-Matrix.

GEL in an Eppendorf tube as follows. The collagen sponge was cut with a scalpel into a 5  10-mm strip and placed in a sterile Eppendorf tube. For the groups in which the carrier did not contain E-Matrix, rhBMP-2 was dissolved in phospate-buffered saline. For those in which the carrier contained E-Matrix, rhBMP was dissolved in the mixed buffer composed of E-Matrix GEL and phospate-buffered saline (9:1). It was then loaded onto the CS immediately before implantation. Study groups A total of 80 Lewis rats aged 8–16 weeks were divided into nine groups. Group I (n 5 5) animals were implanted with CS alone (negative control); Group II (n 5 5) animals with CS containing 10 mg rhBMP-2 (positive control); Group III (n 5 10) animals with CS containing 3 g rhBMP-2; Group IV (n 5 10) animals with CS containing 3 mg rhBMP-2 and E-Matrix GEL; Group V (n 5 10) animals with CS containing 1 mg rhBMP-2; Group VI (n 5 10) animals with CS containing 1 mg rhBMP-2 and E-Matrix GEL; Group VII (n 5 10) animals with CS containing 0.5 mg rhBMP-2; Group VIII (n 5 10) animals with CS containing 0.5 mg rhBMP-2 and E-Matrix GEL; and Group IX (n 5 10) animals with CS and E-Matrix GEL without rhBMP-2. All the animals were sacrificed at eight weeks after implantation. Surgical technique for constructing L4–L5 posterolateral spine fusion model Approval was obtained from the UCLA Chancellor’s Animal Research Committee before animal experimentation. The Lewis rats were anesthetized with isoflurane inhalation and monitored by an assistant during the surgery. The posterolateral lumbar fusion in rats has been well established as an acceptable model for measuring bone growth [18]. A

Radiographic analysis The fusion between L4 and L5 was quantified using plain radiographs that were obtained at 4, 6, and 8 weeks. The fusion between the L4 and L5 transverse processes in each rat was recorded as the percentage of the total area between L4 and L5 that was filled with new bone. Three blinded independent observers scored the bone formation in each rat according to a 6-point scale: 0, no bone formation; 1, bone filling less than 25% of the area; 2, bone filling 25% to 50% of the area; 3, bone filling 50% to 75% of the area; 4, bone filling 75% to 99% of the area; and 5, clear evidence of fusion with bone filling all gaps between L4 and L5. Manual assessment of fusion Manual palpation is the most sensitive and specific method for assessing spinal fusion in this model. After eight weeks of implantation, the explanted lumbar spines were manually tested for intersegmental motion by three blinded independent observers. Any motion detected on either sides between the facets or between the transverse processes was considered as a failure of fusion, and unilateral fusion was considered as no fusion. The absence of motion (right and left) and bilateral fusion was considered as successful fusion. The spines were scored as either fused or not fused. Microcomputerized tomography analysis Next, the spines were scanned using high-resolution microcomputerized tomography (micro-CT) that used the 9–20-mm resolution technology of mCT40 (Scanco Medical, Basserdorf, Switzerland). The micro-CT data were collected at 55 kVp and 72 mA and reconstructed using a cone-beam algorithm supplied with the Scanco micro-CT scanner. Visualization and data reconstruction were performed using mCT Ray T3.3 and mCT Evaluation Program V5.0 (Scanco Medical), respectively. Using these software packages, the area from the tip of the L4 transverse process to the base of the L5 transverse process on the micro-CT images were measured in the groups with 100% fusion to compare the volume of new bone formation. Histologic analysis After the rats were sacrificed, the spines were dissected and the specimens were fixed in 40% ethanol, decalcified using standard 10% decalcifying solution HCl (Cal-Ex) (Fisher Scientific, Fairlawn, NJ), washed with running tap water, and then transferred to 75% ethanol. Serial sagittal

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

sections near the transverse processes were cut carefully at the level of the transverse process. The specimens were embedded in wax and sectioned. The sections were stained with hematoxylin and eosin. Three independent observers blindly scored histologic bone formation. Histologic fusion was defined as bony trabeculae bridging from one transverse process to the next. Fusion masses were assessed and the extent of new bone formation was scored using the following scoring criteria: 0, empty cleft; 1, sight bump within the fibrocartilage tissue (filling less than 25% of the gap area); 2, some gaps within the fibrocartilage tissue (filling 25–50% of the gap area); 3, small gaps within the fibrocartilage and bone tissue (filling 75–99% of the gap area); 4, bridged with bone tissue, however, the fusion masses were composed of thin trabecular bone; and 5, completely bridged with abundant mature bone tissue. Statistical methods The computer program Statistical Package for the Social Sciences (SPSS) (V13, SPSS; Chicago, IL) was used. ANOVA was performed for statistical analysis. Interobserver reliability was assessed by reporting both the observed agreement and by computing the k-statistic. The k-statistic corrects the observed agreement for possible chance agreement among observers. The agreement was rated as follows: poor, k 5 0–0.2; fair, k 5 0.21–0.4; moderate, k 5 0.41– 0.60; substantial; k 5 0.61–0.8; and excellent, k O0.81. A value of 1 indicated absolute agreement, whereas a value of 0 indicated agreement no better than chance.

Results Rat spinal fusion model No abnormal rat behavior was noted in the 81 operated rats, and no rats showed any neurologic deficits before or after the surgical procedure, or at sacrifice. One rat in this

25

study died of infection and another rat was added to the group. Radiographic analysis Radiographs of the rat spines were obtained at 4, 6, and 8 weeks. At four weeks, the spines of the rats in Groups II, III, IV, VI, and VIII showed evidence of bone formation between the L4 and L5 transverse processes, and bony bridging was detected. It was difficult to detect bony bridging in the spines in Groups V and VII; Groups I and IX showed no evidence of bone formation. By eight weeks, all the spines in Groups II, III, IV, VI, and VIII appeared fused on plain radiographs. In contrast, even at eight weeks, some spines in Groups V and VII showed minimal evidence of new bone formation and no fusion, and the spines in Groups I and IX showed no fusion (Fig. 2). The radiographic scores are shown in Table 1. With regard to the 3 mg rhBMP-2 groups, there were no differences between the groups with E-Matrix (Group III) and those without E-Matrix (Group IV), and these radiographic scores are similar to those of the positive control group (Group II). On the other hand, with regard to the 1 mg rhBMP-2 and 0.5 mg rhBMP-2 groups, the groups with E-Matrix (Groups VI and VIII) had significantly (p!.05) higher radiographic scores than those without E-Matrix (Groups V and VII) at every time point. The scores of Group IX are low and similar to those of the negative control group (Group I). Consistent agreement (k 5 0.848) was noted among the three independent observers. Manual palpation Table 2 shows the proportions of rats in each group that were judged to have achieved fusion by three independent evaluators. Consistent agreement (k 5 0.893) was noted among these observers. The spines of all the rats in Groups II (n 5 5) and Groups III, IV, VI, and VIII (n 5 10, each) were assessed as fused (fusion rate, 100%). Four spines

Fig. 2. Radiographs of the Lewis rat spines obtained at eight weeks. The materials implanted in the rats are indicated in parentheses. (A) Group III (CS containing 3 mg rhBMP-2), (B) Group IV (CS containing 3 mg rhBMP-2 and E-Matrix), (C) Group V (CS containing 1 mg rhBMP-2), (D) Group VI (CS containing 1 mg rhBMP-2 and E-Matrix), (E) Group VII (a CS containing 0.5 mg rhBMP-2), and (F) Group VIII (CS containing 0.5 mg rhBMP-2 and E-Matrix). By eight weeks, all the spines in Groups III, IV, VI, and VIII (A, B, D, F) appeared fused on the plain radiographs. In contrast, even at eight weeks, some spines in Groups V and VII (C, E) showed minimal evidence of new bone formation and no fusion, and the spines of the rats in Groups I and IX showed no fusion.

26

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

Table 1 Radiographic scores at 4, 6, and 8 weeks Group I II III IV V VI VIII VIII IX

Treatment

No. studied radiographically

Score at 4 wk

Score at 6 wk

Score at 8 wk

CS alone 10 mg rhBMP-2 þ CS 3 mg rhBMP-2 þ CS 3 mg rhBMP-2 þ E-Matrix GEL þ CS 1 mg rhBMP-2 þ CS 1 mg rhBMP-2 þ E-Matrix GEL þ CS 0.5 mg rhBMP-2 þ CS 0.5 mg rhBMP-2 þ E-Matrix GEL þ CS E-Matrix GEL without rhBMP þ CS

5 5 10 10 10 10 10 10 10

0.13 3.47 3.07 3.27 1.93 3.17* 1.33 2.57* 0.33

0.20 3.73 3.43 3.67 2.43 3.37* 1.63 2.93* 0.60

0.27 4.13 3.93 4.10 2.93 3.87* 1.97 3.73* 0.77

*p!.001vs.without E-Matrix.

in the Group V rats (n 5 10) were assessed as fused (fusion rate, 40%), and one spine in Group VII rats (n 5 10) exhibited fusion (fusion rate, 10%). Group I (n 5 5) and Group IX rats (n 5 10) showed no spine fusion (fusion rate, 0%). With regard to the 1 mg rhBMP-2 and 0.5 mg rhBMP-2 groups, the groups with E-Matrix (Groups VI and VIII) had significantly higher fusion rates than those without EMatrix (Groups V and VII; p!.05 and p!.001). Micro-CT analysis At eight weeks, we sacrificed the rats and performed a three-dimensional micro-CT analysis. The spines of all the rats in Groups II, III, IV, VI, and VIII exhibited considerable new bone formation. The new bone mass was solidly fused, and no gaps were detected between the transverse processes. Multiple cut sections were reconstructed to evaluate the presence of a bony bridge between the transverse processes. Trabeculae bridging the transverse processes were consistently observed on the cut-plane images of all spine samples of Groups II, III, IV, VI, and VIII. Four spines in Group V and one spine in Group VII showed evidence of new bone formation and fusion; however, the fusion masses were composed of thin trabecular bone; six spines in Group V and nine spines in Group VII showed no fusion. The spines of the Group I and IX rats did not exhibit any bony bridging between the transverse processes, and a cleft between the L4 and L5 transverse processes was observed in all the spines with nonunion as well as in those with a fair amount of new bone mass over the targeted areas (Figs. 3 and 4). Computer analysis of the

micro-CT images revealed the volume of new bone in the area between the tip of the L4 transverse process and the base of the L5 transverse process. The volume of new bone in the Group IV rats is equivalent to that in the Group II rats (Fig. 5). Histologic analysis Histologic analysis of the spines in Groups III and IV demonstrated abundant bone bridging between the transverse processes. The CS carrier had disappeared; mature bones were observed, and they displayed osteoid tissue, connected to form trabeculae, and were surrounded by well-developed bone marrow cavities (Fig. 6). The nonfusion samples in Group V showed that the new mature bones formed considerably later and the distribution of cartilaginous and fibrous tissues was more prominent in the nonunion bridge area (Fig. 7, left). In contrast, analysis of the Group VI rats demonstrated new bone formation between the transverse process and showed mature osteoid tissue, trabeculae constructs, and bone marrow cavities (Fig. 7, right). The nonfusion samples of the Group VII rats showed patterns identical to those observed in Group V in the bridge area without mature bone formation (Fig. 8, left). In contrast, the spines in Group VIII showed fusion, with mature bone formation; however, the fusion masses were composed of thin trabecular bone (Fig. 8, right). Histologic fusion scores are shown in Table 3. With regard to the 3 mg rhBMP-2 groups, there were no differences between the groups with E-Matrix (Group III) and those without EMatrix (Group IV). These histologic scores are similar to

Table 2 Assessment of spinal fusion via manual palpation Group I II III IV V VI VIII VIII IX

Treatment

No. assessed manually for fusion

No. assessed as fused

Fusion rate (%)

CS alone 10 mg rhBMP-2 þ CS 3 mg rhBMP-2 þ CS 3 mg rhBMP-2 þ E-Matrix GEL þ CS 1 mg rhBMP-2 þ CS 1 mg rhBMP-2 þ E-Matrix GEL þ CS 0.5 mg rhBMP-2 þ CS 0.5 mg rhBMP-2 þ E-Matrix GEL þ CS E-Matrix GEL without rhBMP þ CS

5 5 10 10 10 10 10 10 10

0 5 10 10 4 10 1 10 0

0 100 100 100 40 100* 10 100y 0

*p!.01; yp!.001vs.without E-Matrix.

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

27

Fig. 3. Three-dimensional reconstruction of microcomputerized tomographic images of rat spines in the anteroposterior view. The materials implanted in the rats are indicated in parentheses. (A) Group III (CS containing 3 mg rhBMP-2), (B) Group IV (CS containing 3 mg rhBMP-2 and E-Matrix), (C) nonfusion sample in Group V (CS containing 1 mg rhBMP-2), (D) Group VI (CS containing 1 mg rhBMP-2 and E-Matrix), (E) nonfusion sample in Group VII (CS containing 0.5 mg rhBMP-2), (F) Group VIII (CS containing 0.5 mg rhBMP-2 and E-Matrix). The spines of all the rats in Groups III, IV, VI, and VIII (A, B, D, F) exhibited considerable new bone formation. The new bone mass was solidly fused into the L4 and L5 transverse processes, and the gaps between the L4 and L5 transverse processes could not be detected. Six spines in Group V (C) and nine spines in Group VII (D) showed no fusion.

those of the positive control group (Group II). However, with regard to the 1 mg rhBMP-2 and 0.5 mg rhBMP-2 groups, the groups with E-Matrix (Groups VI and VIII) had significantly (p!.05) higher histologic scores than those without E-Matrix (Groups V and VII). The scores of Group IX are low and similar to those of the negative control group (Group I). Consistent agreement (k 5 0.958) was noted among the three independent observers. Discussion It has been demonstrated that rhBMP-2 is more efficacious in combination with a carrier matrix [19]. An ideal carrier should be biocompatible, biodegradable, adequately porous to allow cell and blood infiltration, and amenable to

sterilization. It should be adhesive to adjacent bone and have appropriate mechanical stability and affinity for BMPs. The function of this carrier is to allow retention and sustained release of the protein for a sufficient period of time [20,21]. Although none of the existing carrier matrices exhibit all of these properties sufficiently, various carriers have been tested in basic and clinical studies [22–26]. The efficacy of E-Matrix in wound healing has been studied. This porcine collagen-derived matrix is designed to mimic tertiary embryonic connective tissue and stimulate fetal wound repair mechanisms. It has been shown to have beneficial effects on tissue growth [17] and stimulate healing of diabetic foot wounds in humans [16]. Prior studies demonstrated that tissues treated with E-Matrix were associated with significantly increased angiogenesis. E-Matrix

Fig. 4. Three-dimensional reconstruction of microcomputerized tomographic images of rat spines (cut-plane images). The materials implanted in the rats are indicated in parentheses. (A) Group III (CS containing 3 mg rhBMP-2), (B) Group IV (CS containing 3 mg rhBMP-2 and E-Matrix), (C)nonfusion sample in Group V (CS containing 1 mg rhBMP-2), (D) Group VI (CS containing 1 mg rhBMP-2 and E-Matrix), (E) nonfusion sample in Group VII (CS containing 0.5 mg rhBMP-2), and (F) Group VIII (CS containing 0.5 mg rhBMP-2 and E-Matrix). Trabeculae bridging the transverse processes were consistently observed on the cut-plane images of all spine samples in Groups III, IV, VI, and VIII (A, B, D, F). Six spines in Group V (C) and nine spines in Group VII (E) showed a cleft between the L4 and L5 transverse processes with no fusion.

28

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

Fig. 5. The volumes of new bone in the area between the tip of the L4 transverse process and the base of the L5 transverse process. The volume of new bone in the Group IV rats is equivalent to that in the Group II rats.

mechanism of action may result from its ability to make available to cells the primary amino acid sequence of the gelatin monomer which has cell attachment sequences normally obscured within the collagen triple helix. This primary amino acid sequence also contains sequences with growth factor-like and growth factor binding activity. In E-Matrix, these sequences are available to cells and may facilitate the observed effects. It is postulated that E-Matrix may stimulate bone regeneration, and the open polar structure may efficiently hold proteins such as the BMPs. We hypothesize that E-Matrix, in combination with CS, allows a more sustained release of rhBMPs and accelerates bone formation at the fusion site in vivo. In this study, we tested the efficacy of E-Matrix in enhancing the osteogenic activity of rhBMP-2 in a well-accepted rodent spine fusion model. According to our previous data, 10 mg of rhBMPs was sufficient to form a fusion mass and for use as positive control. A recent study reported that the minimum amount of BMP-2 required to achieve 100% fusion rate in Lewis rats was 3.2 mg [27].

The groups were segregated according to doses (3, 1, and 0.5 mg of rhBMP-2) with or without E-Matrix. The radiographic scores in Groups VI and VIII were higher than those in Groups V and VII at every time point. Additionally, the fusion rates in Groups VI and VIII were 100% and considerably higher than those in Groups V (40%) and VII (10%). In contrast, the radiographic scores in Groups III and IV were similar, and 3 mg rhBMP achieved 100% fusion irrespective of the presence or absence of E-Matrix. This implies that it is difficult to achieve 100% fusion with doses less than 3 mg of rhBMP with CS alone, and matrices could be added to increase the fusion rate. In this regard, E-Matrix was demonstrated to be efficacious and it significantly enhances spinal fusion from the early phase. Reconstructed micro-CT images clearly demonstrated the thickness and quality of the bony structures objectively, which is consistent with the histologic results. The cutplane micro-CT images of all samples in Groups II, III, IV, VI, and VIII showed a fusion mass composed of trabecular bone, and these samples histologically demonstrated maturity of bone tissue through endochondral formation with increased trabecular and bone marrow formation. In contrast, the cut-plane micro-CT images of nonfusion samples in Groups V and VII clearly showed a cleft between the L4 and L5 transverse processes; these samples histologically demonstrated distribution of cartilaginous and fibrous tissues in the nonunion attempted fusion area. The microCT allows researchers to evaluate precise images and data. Calculations using computer software revealed the volume of the new bone in the target area. The volume of new bone in the posteriolateral section in Group IV was similar to that in Group II and represents the efficacy of E-Matrix; E-Matrix may enhance spinal fusion mass by more than three times. Thus, currently, micro-CT has arguably become the gold standard of bone evaluation in small rodent fusion studies. Although rhBMP therapy is promising for spinal fusion, there is a concern that it could lead to heterotopic bone formation. Possible methods for countering this side-effect

Fig. 6. Cross-section of the L4–L5 transverse processes of a rat spine. The materials implanted in the rats are indicated in parentheses. (Left) Group III (CS containing 3 mg rhBMP-2), (Right) Group IV (CS containing 3 mg rhBMP-2 and E-Matrix); the collagen sponge carrier had disappeared, and mature bones were observed, and they displayed osteoid tissue, connected to form trabeculae, and were surrounded by well-developed bone marrow cavities (magnification 40).

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

29

Fig. 7. Cross-section of the L4–L5 transverse processes of a rat spine The materials implanted in the rats are indicated in parentheses. (Left) Nonfusion sample in Group V (CS containing 1 mg rhBMP-2); the new mature bones were formed considerably later and the distribution of cartilaginous and fibrous tissues was more prominent in the nonunion bridge area (magnification 40). (Right) Group VI (CS containing 1 mg rhBMP-2 and E-Matrix); mature osteoid tissue, trabeculae constructs, and bone marrow cavities appeared (magnification 40).

should be established. The present study does not directly discuss this topic with or without the use of E-Matrix. However, one possible strategy to control the heterotopic bone formation is using fibrin glue. It was reported that fibrin glue was able to limit rhBMP-2 diffusion and control rhBMP-2 stimulated bone growth [27,28]. The combination of E-Matrix and fibrin glue may make rhBMP therapy safer and more efficacious. Another important adverse effect of rhBMP-2 is soft-tissue inflammation; some reports have cautioned against the use of high-dose rhBMP-2 for cervical anterior spinal fusion [11,12]. In this study, no soft-tissue inflammation was noted over time. Controlling the diffusion and maintaining the ideal dose of rhBMP-2 by using ideal carrier matrices may prevent this effect, however, further studies are required to define this effect. This study showed successful results of enhancing spinal fusion by adding E-Matrix to CS. However, this study was conducted using a rat model, and therefore the results of this study do not necessarily apply directly toward application of this technology to humans. There are many differences between rats and humans including the fact that

rats walk on four legs and humans walk on two resulting in different spine biomechanics. Additionally, the physiological microenvironments present in each species are different. Previous studies have shown a species-dependent effect of rhBMP-2 on spine fusion with higher species requiring larger doses [29,30]. Accordingly, different doses of rhBMP-2 and E-Matrix will likely be needed to achieve the desired effects in each species and it is expected that a high bone-forming ability is required for adequate posterolateral fusion in humans [8,9,15]. At this point, the relevance and the applicability of E-Matrix to human spine fusion are uncertain. Although E-Matrix has previously been tested in humans for its ability to treat diabetic foot ulcers and no significant adverse responses were associated with its use [16], it is still currently unknown whether E-Matrix is a safe carrier matrix for rhBMP-2 in humans. To clarify the efficacy and the safety of E-Matrix, further studies on primates and humans are required. In conclusion, this study demonstrated that E-Matrix, an injectable scaffolding matrix for cellular attachment derived from porcine skin collagen, enhanced rat posterolateral spinal fusion as a carrier for rhBMP-2. Its use is

Fig. 8. Cross-section of the L4–L5 transverse processes of a rat spine The materials implanted are indicated in parentheses. (Left) Nonfusion sample in Group VII (CS containing 0.5 mg rhBMP-2); the distribution of cartilaginous and fibrous tissues was shown in the nonunion bridge area (magnification 40). (Right) Group VIII (CS containing 0.5 mg rhBMP-2 and E-Matrix); mature bone tissue was observed; however, the fusion masses were composed of thin trabecular bone (magnification 40).

30

M. Miyazaki et al. / The Spine Journal 9 (2009) 22–30

Table 3 Histologic fusion scores at 8 weeks Group I II III IV V VI VIII VIII IX

Treatment

Score at 8 wk

CS alone 10 mg rhBMP-2 þ CS 3 mg rhBMP-2 þ CS 3 mg rhBMP-2 þ E-Matrix GEL þ CS 1 mg rhBMP-2 þ CS 1 mg rhBMP-2 þ E-Matrix GEL þ CS 0.5 mg rhBMP-2 þ CS 0.5 mg rhBMP-2 þ E-Matrix GEL þ CS E-Matrix GEL þ CS without rhBMP

0.27 5.00 4.87 5.00 3.47 4.63* 2.33 4.20 0.57

*p!.05 vs.without E-Matrix.

expected to decrease the required doses of rhBMP-2 in clinical use, thus reducing medical expenses. However, its efficacy and safety for large animals and humans are not clear and need to be examined through further animal studies before clinical application. Despite these limitations, E-Matrix with CS will be an attractive carrier for rhBMP2 for various orthopedic surgeries. Acknowledgments Funding was provided by Pioneer Surgical Orthobiologics, Greenville, NC. The authors also thank Pioneer Surgical Orthobiologics for providing the E-Matrix. References [1] Welch RD, Jones AL, Bucholz RW, et al. Effect of recombinant human bone morphogenetic protein-2 on fracture healing in a goat tibial fracture model. J Bone Miner Res 1998;13:1483–90. [2] Bouxsein ML, Turek TJ, Blake CA, et al. Recombinant human bone morphogenetic protein-2 accelerates healing in a rabbit ulnar osteotomy model. J Bone Joint Surg Am 2001;83-A:1219–30. [3] Jones AL, Bucholz RW, Bosse MJ, et al. Recombinant human BMP-2 and allograft compared with autogenous bone graft for reconstruction of diaphyseal tibial fractures with cortical defects. A randomized, controlled trial. J Bone Joint Surg Am 2006;88:1431–41. [4] Hollinger JO, Schmitt JM, Buck DC, et al. Recombinant human bone morphogenetic protein-2 and collagen for bone regeneration. J Biomed Mater Res 1998;43:356–64. [5] Boden SD, Martin GJ Jr, Horton WC, et al. Laparoscopic anterior spinal arthrodesis with rhBMP-2 in a titanium interbody threaded cage. J Spinal Disord 1998;11:95–101. [6] Boden SD, Martin GJ Jr, Morone MA, et al. Posterolateral lumbar intertransverse process spine arthrodesis with recombinant human bone morphogenetic protein 2/hydroxyapatite-tricalcium phosphate after laminectomy in the nonhuman primate. Spine 1999;24: 1179–85. [7] Burkus JK, Transfeldt EE, Kitchel SH, et al. Clinical and radiographic outcomes of anterior lumbar interbody fusion using recombinant human bone morphogenetic protein-2. Spine 2002;27:2396– 408. [8] Dimar JR, Glassman SD, Burkus KJ, et al. Clinical outcomes and fusion success at 2 years of single-level instrumented posterolateral fusions with recombinant human bone morphogenetic protein-2/compression resistant matrix versus iliac crest bone graft. Spine 2006;31: 2534–9. [9] Garrison KR, Donell S, Ryder J, et al. Clinical effectiveness and costeffectiveness of bone morphogenetic proteins in the non-healing of fractures and spinal fusion: a systematic review. Health Technol Assess 2007;11:1–150.

[10] Louis-Ugbo J, Kim HS, Boden SD, et al. Retention of 125I-labeled recombinant human bone morphogenetic protein-2 by biphasic calcium phosphate or a composite sponge in a rabbit posterolateral spine arthrodesis model. J Orthop Res 2002;20:1050–9. [11] Shields LB, Raque GH, Glassman SD, et al. Adverse effects associated with high-dose recombinant human bone morphogenetic protein-2 use in anterior cervical spine fusion. Spine 2006;31: 542–7. [12] Vaidya R, Carp J, Sethi A, et al. Complications of anterior cervical discectomy and fusion using recombinant human bone morphogenetic protein-2. Eur Spine J 2007;16:1257–65. [13] Chvapil M. Collagen sponge: theory and practice of medical applications. J Biomed Mater Res 1977;11:721–41. [14] Geiger M, Li RH, Friess W. Collagen sponges for bone regeneration with rhBMP-2. Adv Drug Deliv Rev 2003;55:1613–29. [15] Glassman SD, Carreon L, Djurasovic M, et al. Posterolateral lumbar spine fusion with INFUSE bone graft. Spine 2007;7:44–9. [16] Marston WA, Usala A, Hill RS, et al. Initial report of the use of an injectable porcine collagen-derived matrix to stimulate healing of diabetic foot wounds in humans. Wound Repair Regen 2005;13: 243–7. [17] Klann R, Lloyd W, Lacy S. A novel hydrogel copolymer reduces scar formation and increase TGF b3 gene expression. In: Presented at the 15th Annual Symposium on Advanced Wound Care. Baltimore (MD); April 29 2002. [18] Wang JC, Kanim LE, Yoo S, et al. Effect of regional gene therapy with bone morphogenetic protein-2-producing bone marrow cells on spinal fusion in rats. J Bone Joint Surg Am 2003;85-A: 905–11. [19] Seeherman H, Wozney J, Li R. Bone morphogenetic protein delivery systems. Spine 2002;27:S16–23. [20] Brekke JH, Toth JM. Principles of tissue engineering applied to programmable osteogenesis. J Biomed Mater Res 1998;43:380–98. [21] Kirker-Head CA. Potential applications and delivery strategies for bone morphogenetic proteins. Adv Drug Deliv Rev 2000;43: 65–92. [22] Suh DY, Boden SD, Louis-Ugbo J, et al. Delivery of recombinant human bone morphogenetic protein-2 using a compression-resistant matrix in posterolateral spine fusion in the rabbit and in the non-human primate. Spine 2002;27:353–60. [23] Bessho K, Carnes DL, Cavin R, et al. Experimental studies on bone induction using low-molecular-weight poly (DL-lactide-co-glycolide) as a carrier for recombinant human bone morphogenetic protein-2. J Biomed Mater Res 2002;61:61–5. [24] Ruhe PQ, Hedberg EL, Padron NT, et al. rhBMP-2 release from injectable poly(DL-lactic-co-glycolic acid)/calcium-phosphate cement composites. J Bone Joint Surg Am 2003;85-A(Suppl 3):75–81. [25] Liao SS, Guan K, Cui FZ, et al. Lumbar spinal fusion with a mineralized collagen matrix and rhBMP-2 in a rabbit model. Spine 2003;28:1954–60. [26] Morisue H, Matsumoto M, Chiba K, et al. A novel hydroxyapatite fiber mesh as a carrier for recombinant human bone morphogenetic protein-2 enhances bone union in rat posterolateral fusion model. Spine 2006;31:1194–200. [27] Patel VV, Zhao L, Wong P, et al. An in vitro and in vivo analysis of fibrin glue use to control bone morphogenetic protein diffusion and bone morphogenetic protein-stimulated bone growth. Spine J 2006;6:397–403. [28] Patel VV, Zhao L, Wong P, et al. Controlling bone morphogenetic protein diffusion and bone morphogenetic protein-stimulated bone growth using fibrin glue. Spine 2006;31:1201–6. [29] Boden SD, Schimandle JH, Hutton WC. The use of an osteoinductive growth factor for lumbar spinal fusion. Part II: Study of dose, carrier, and species. Spine 1995;20:2633–44. [30] Barnes B, Boden SD, Louis-Ugbo J, et al. Lower dose of rhBMP-2 achieves spine fusion when combined with an osteoconductive bulking agent in non-human primates. Spine 2005;30: 1127–33.