**Percutaneous Gene-Delivery Mediated Intervertebral Body Fusion-Young Investigator Grant
Matthew E. Cunningham, MD, PhD; Oheneba Boachie-Adjei, MD
Status of Research Progress:
The research outlined in the grant proposal involves delivering genetically modified mesenchymal stromal cells to the disk spaces of rats, with the intent of driving bone formation and fusion of the spinal levels. Our prior work demonstrated that delivery of a heterodimeric form of BMP2/BMP7 to the disk spaces resulted in vigorous bone formation and fusion events in approximately 50% of levels attempted. The current research project was designed to assess (1) the effect of co-delivery of vascular endothelial growth factor (VEGF) in the system (co-delivery of VEGF and BMPs has previously been shown to improve BMP-mediated bone formation by 1.5-2.5 fold), and (2) co-delivery of an agent capable of dissolving the disk tissue. Originally, the factor being considered for chemically ablating the disk was gene-delivery of IL1-beta. Upon further consideration, this factor was decided against, due to desire to better limit the potential damaging effects of IL1-beta on newly formed osteoid, and to deliver an agent that would not be expressed over an extended period of time. Chondroitinase ABC is well described in the literature as a safe agent, and has been used in several models including rat and rabbit to digest disk tissue; we decided it was a better factor to be used in our study (available from Sigma).
Our model also requires the generation of recombinant adenovirus stocks of the genes to be delivered. The generation of AdVEGF-165 and AdBMP-7 stocks (10^13 particle units of each) was quite easily performed as has been done previously in the lab, but generation of the AdBMP-2 virus stock was quite vexing. The ultimate solution to the AdBMP-2 problem was obtaining a new starter sample for the virus expansion from a collaborator. This solution was only obtained after a week of surgical implantations were completed (~50 animals) and several mass cultures of virus yielded no viable recombinant virus (leading the project into being delayed and over-budget).
All of the animal surgeries have been completed, samples (spines) were harvested, and specimen processing is being completed. There are n=15 or 16 for each of the 9 groups (no surgery, Naïve cell delivery, VEGF, chABC, BMP-2/7, VEGF/BMP, VEGF/chABC, chABC/BMP, and triple treatment).
Findings to Date:
The animals underwent analysis of spine mobility, as we described in the grant proposal, preoperatively, and at 4, 8 and 12 weeks postoperatively, and normalized data is depicted in the figure below. As was seen in the previous use of this method, there is an approximately 20% reduction in mobility from non-operated (“Mock”) spines to even the most mobile of the operated spines. The least mobility was observed in the spines from the BMP and VEGF/BMP treated groups, but interestingly addition of chABC to either of these conditions (chABC/BMP or triple treatment) showed mobility more like the Naïve cell implanted and non-fusion control (VEGF treatment).
Analysis of fusion by palpation has been completed, and shows fusion of 7% of levels in the chABC/BMP group, 37% in the BMP group (p=0.01), and 47% (p<0.001). The two most responsive groups (the latter ones in the previous sentence) did not show statistical difference from the prior fusion results (p>0.05). None of the other 6 groups had any specimens with fusion events. Bone production is being assessed currently, using histology, lateral spine x-rays and micro-CT. Examination of lateral radiographs of the animals reveals bone production appears predominantly outside of the disk space, but smaller amounts of bone within the disk space may not be detected with X-ray only and may require the micro-CT for description. Lateral radiographs of the chABC treated animals with BMP-cotreatment suggest that the chABC interferes with the bone production that occurs external to the disk space. Further quantificatons of bone production are pending.
Neoangiogenesis in the disk tissue has been characterized using a V-cadherin immunostain of sagittally sectioned specimens. The findings are preliminary, but is appears that delivery of chABC induces vascularity in the disk tissue. Confirmation of this finding, and further evaluations are pending.
Biomechanical endpoints will be assessed in the upcoming weeks utilizing a previously established 4-point bending assay.
In-vivo Spine Biomechanics: Application of an Innovative Combined MR and Dual Fluoroscopic Imaging Technique
Kirkham B. Wood, MD; Guoan Li, Ph
In the initial application we proposed to recruit 10 healthy human subjects (5 males and 5 females age 50-70 years), using our combined MR and dual fluoroscopic imaging (DFIS) technique. MR scans were used to construct anatomical vertebral models and DFIS was used to record in-vivo spinal motion. Using the image matching technique, the actual spinal motion was then virtually reproduced in our software system.
The following were the two specific aims that we proposed:
Specific Aim 1
To quantify the 6DOF kinematics of the vertebral bodies of L2, L3 and L4 in space during standing position, flexion and extension, right and left bending and medial and lateral rotation of the trunk of the body.
Specific Aim 2
To determine the 3D deformation of the intervertebral disc, the articular contact motion of the apophyseal joints, and finally the elongation of the posterior longitudinal ligament during standing position, flexion and extension, right and left bending and medial and lateral rotation of the trunk of the body.
Upon completion of the project, we have recruited 11 subjects. However, after the MR scan, two subjects were found to have the presence of early disc degeneration in the absence of clinical symptoms. Additionally, one subject was found to have early scoliosis without symptoms. These three subjects were excluded from further investigation. Thus our effective subject group is 8 healthy subjects (50-70 years old, 2 males and 6 females). We quantified the 6DOF kinematics of L2, L3, L4 and additionally L5 during standing, flexion-extension, left-right bending and left-right twisting as stated in Aim 1. We then determined the 3D deformation of the intervertebral disc (IVD) and the apophyseal (facet) joint motion during various postures as stated in the first part of Aim 2.
In addition, we studied the interspinous process (ISP) kinematics as a new pilot study which was not stated in the original specific aims. ISP devices have been introduced for clinical use in the lumbar spine and have showed promising results. However, the fundamental kinematics of spinous processes during daily weightbearing postures has not been reported in the literature. Our method can efficiently provide quantitative data on the motion of the spinous process, which can thereby provide baseline information that can be used for further improvement in treatments of spinal diseases that involve the use of ISP implants.
We also studied the vertebral motion characters in sagittal and transverse planes. Knowledge of motion patterns of lumbar vertebrae is important for treatment of intervertebral disc diseases using dynamic fusion or total disc replacement surgeries with the aim to reserve spinal motion. We analyzed on the motion centers of the vertebrae in living human subjects, which has not been reported previously in literature.
We didn’t conduct the articular contact of the apophyseal joints and the elongation of the posterior longitudinal ligament studies that were stated in Aim 2. One major difficulty has been the limitations funding. We greatly appreciate the one year support from the North American Spine Society research grant. However, the study was proposed to be completed with two years funding support. In order to complete our stated goals, we would like to request continued funding support from NASS if at all possible. Given our progress to date, we are confident that this study can be completed.
Findings and Conclusions
In the following text we will outline the ongoing lumbar spine biomechanical studies that we have conduct using our novel imaging system. Upon completion, we obtained data for validation of the technique, vertebral kinematics, intervertebral disc (IVD) deformation, facet joint kinematics, ISP kinematics and vertebral motion characters.
Specific Aim 1:
1. Wang S, Passias P, Li G, Li G, Wood K (2008) Measurement of vertebral kinematics using noninvasive image matching method-validation and application. Spine 33:E355-361.
Conclusion of the manuscript: The combined MR and DFIS technique was first validated for its accuracy in an in-vitro ovine study. The translation positions of the ovine spine could be determined with a mean accuracy less than 0.40 mm. The technique was then applied on a living subject perform various weightbearing positions. The repeatability of reproducing in-vivo human spine 6DOF kinematics was less than 0.3 mm in translation and less than 0.7° in orientation. The technique showed sufficient accuracy and repeatability for human lumbar biomechanics studies.
2. Li G, Wang S, Passias P, Xia Q, Li G, Wood K (2009) Segmental in vivo vertebral motion during functional human lumbar spine activities. Eur Spine J 18:1013-1021.
In-vivo Lumbar Vertebrae Kinematics. Poster, ORS 55th, Las Vegas, NV.
Conclusion of the manuscript: The combined MR and DFIS technique was applied to measure lumbar segmental kinematics of L2-L5 (Fig 1). The data demonstrated that the upper vertebrae had a larger range of flexion during active flexion-extension motion while the lower levels had a larger range of bending during left-right bend, and no statistical difference in left-right twist. This could be related to the different anatomic orientation of the facet joints at different levels as the L2-3 facet orients more vertically than L4-5 which facilitates flexion. This difference may also be attributed to the lordosis of the lumbar spine. Besides the primary rotations, coupled translations were found to be greater in the left-right and anterior-posterior than proximal-distal directions. Coupled bending and twisting motions were found to have a larger range of motion than coupled flexion. This data provides baseline information on the normal ROM of the lumbar vertebrae on which there is no consensus at present.
Specific Aim 2:
1. Wang S, Xia Q, Passias P, Wood K, Li G (2009) Measurement of geometric deformation of lumbar intervertebral discs under in-vivo weightbearing condition. J Biomech 42:705-711.
In-vivo Lumbar Intervertebral Disc Geometric Deformation during Weightbearing Posture (2009). Poster, ORS 55th, Las Vegas, NV.
The details of IVD deformation such as the magnitude and direction of tension and shear have yet to be clearly defined in-vivo mainly due to technical limitations. We investigated lumbar IVD geometric deformation from adjacent level translation and orientation of the endplates using the combined MR and DFIS. Tensile and shear deformation was quantified by comparing the weightbearing (standing) position to non-weightbearing (supine) position at the L2-L5 vertebral levels (Fig 2).
Disc deformation during weightbearing differed based on the segmental level. The tension and compression portion altered within the vertebral levels. The magnitude of the deformation decreased from the cephalad L2-3 level to the caudad L4-5 level. This may be related to the physiologic lordosis and inherent weightbearing patterns of our subjects (for instance; every subject in our study was right-handed). The shear deformation was limited to within 30% for each level, which can be attributed to constraint from the strong ligamentous and muscular attachments of the lumbar spine together with the disc material. We also found that L2-3 and L4-5 experienced shear from opposing directions and L3-4 had a small deformation. Overall the spine maintains stability by balancing shear in different directions and the anatomic inflection point is roughly located at the L3-4 disc.
2. Kozanek M, Wang S, Xia Q, Passias P, Xia Q, Li G, Bono C, Wood K, Li G Range of Motion and Orientation of the Lumbar Facet Joints In Vivo. Submitted to Spine
In-vivo Kinematics of Lumbar Facet Joints (2009). Poster, ORS 55th, Las Vegas, NV.
Alterations in motion of the facet joints have been thought to be associated with various types of lumbar spine pathology including disc degeneration, facet degeneration and neural impingement. However, determination of normal in-vivo motion of the lumbar facet joints in-vivo remains elusive. We applied the combined MR and DFIS to non-invasively investigate the kinematics of lumbar facet joints in vivo during flexion-extension, left-right bending and left-right tension. The standing position was compared to the reference MR (supine) position and the ranges of motion from the end-points of flexion, bending and twisting of trunk were also determined. We noted that for such a small joint the ranges of motion were of considerable magnitudes (up to 4 mm in translations and up to 8° in rotations). Further, during flexion-extension of the trunk, the facet joints were demonstrated to rotate primarily along the medio-lateral axis and shift in the proximal-distal direction. During the other studied activities of sidewise bend and twist, the facet joints did not move in one predominant direction of rotation or translation. Instead, the resulting motion was a kinematic coupling of rotations and translations in different directions. This may be related to the different orientation of the facet joints at cranial and caudal levels. Additionally, we observed that the translations were not perfectly symmetrical between the left and right side of the same level. This can be explained by the inherent asymmetry in lumbar facet orientation.
New Experimental Information
1. Xia Q, Wang S, Passias P, Kozanek M, Li G, Grottkau BE, Wood KB, Li G (2009) In vivo range of motion of the lumbar spinous processes. Eur Spine J
In-vivo Range of Motion of the Lumbar Spinous Processes (2009). Poster, ORS 55th, Las Vegas, NV.
In a new pilot study, we investigated the spinous process kinematics of eight asymptomatic subjects using the combined MR and dual fluoroscopic imaging system.
We found that the changes in the interspinous process (ISP) distance observed in the normal subjects were within 10 mm magnitudes. Significance in ISP distances was detected for the supine, standing, flexion and extension positions at L2-3 and L3-4, but not at L4-5. We observed that the distance changes for L2-3 and L3-4 were larger than that of L4-5. The data indicates that interspinous distance change is posture dependent.
The quantitative data on the motion of the spinous processes determined by our study can serve as baseline information for further improvement of treatments of spine diseases that involve the ISP devices.
2. Xia Q, Wang S, Passias P, Li G, Wood K (2009) Lumbar Spine Vertebral Motion Characters in Sagittal and Transverse Planes, Submitted to Spine (attachment 10)
Lumbar spine motion of L2, L3 and L4 was reproduced using a combined dual fluoroscopic imaging system (DFIS) and MR imaging technique during flexion-extension and left-right twisting of the body. Based on geometrical features of the vertebrae, ranges of motion (ROM) of 3 representative locations were measured from anterior to posterior: the vertebral body center, the spinal cord canal center and the spinous process tip. Rotational centers of the vertebral segments were then calculated.
The ROMs of L2 with respect to L3 (L23) and L34 increased proportionally from anterior to posterior locations. During flexion-extension motion, the rotational centers of both L23 and L34 segments were located at posterior one-third of the vertebral body. During left-right twisting, the rotational centers of both L23 and L34 segments were located approximately 30 mm anterior to the front edge of the vertebral body. (Fig. 5)
Different portions of the lumbar vertebrae exhibit different motion characters during the body motion. The data may be useful for improving the treatment of spinal diseases such as using the total disc replacement or dynamic fusion for motion preservation.
**Biological Repair of Intervertebral Disc Degeneration
Fackson Mwale, PhD; John Antoniou, MD, PhD; Peter Roughley
We have made major progress on all of the proposed aims of the grant on this topic. From these studies 2 papers are currently in preparation. In addition 2 abstracts will be sent to the 2010 Orthopedic Research Society Conference and the annual NASS conference for podium presentation. Final results will be submitted in the next two months as we await the results of the analysis from the final samples.
Findings or Conclusions to Date
We have demonstrated that Link N, the N-terminal peptide of link protein generated by proteolytic degradation during tissue turnover, can stimulate the synthesis of proteoglycans and collagens in a rabbit model of disc degeneration. Thus, Link N represents a potential growth factor able to stimulate ECM proteins to generate an NP-like phenotype. Furthermore, being a synthetic peptide, link N has considerable financial benefits for clinical use over recombinant growth because it is extremely cheap to produce. In the present study, the use of link N represents a new and original avenue to delay further disc degeneration. Furthermore, the in vivo rabbit model of disc degeneration is known to results in a number of slowly progressive and reproducible changes that show similarities to changes seen in human intervertebral disc degeneration.
The long term goal is to promote nucleus pulposus repair in the degenerated intervertebral disc. We hypothesized that the injection of link N, a natural peptide with growth factor properties, in the presence of stem cells can stimulate the repair of the degenerated intervertebral disc in Thomson grades 2 and 3 discs.
The specific objectives are:
- To establish a baseline for the disc degeneration model.
- To determine if link N alone can stimulate repair of the degenerated disc.
- To determine if mesenchymal stem cells can stimulate repair of the degenerated disc.
Cytokine Profile in Intervertebral Disc Tissues from Patients with Discogenic Axial Back Pain Confirmed by Discography
Yejia Zhang, MD, PhD; D. Greg Anderson, MD
Abstract
Background Context: Back pain affects millions of people worldwide and is thought to be most commonly related to symptomatic degeneration of the intervertebral disc. Unfortunately, imaging studies showing degeneration have a poor correlation with patient symptoms, leaving the clinician without a good tool to determine the location and source of the pain. It is known that degenerating discs produce a wide variety of cytokines which may play a role in the pathophysiology of discogenic pain. Our study sought to quantify the cytokines produced by the intervertebral discs of human patients with severe discogenic pain as compared to those with non-painful discs.
Study Design: Cytokine, chemokine and growth factor expression levels were measured and compared for intervertebral discs from patients with severe discogenic pain and controls.
Patient Sample: Disc tissue was collected from patients undergoing ALIF for severe discogenic low back pain and patients undergoing ALIF for adult scoliosis (controls).
Outcome Measures: Cytokine, chemokine and growth factor expression levels.
Methods: We collected surgically-removed intervertebral disc tissues from patients undergoing ALIF for severe back pain with well-documented preoperative provocative discography and patients undergoing for adult scoliosis (controls). The discs were matched for degree of degeneration according to the Pfirrmann grading scheme and then subjected to protein extraction. The extracted proteins were analyzed using a human cytokine array which detects the presence and levels of 42 cytokines, chemokines and growth factors. The quantitative data was then compared between the discogram positive discs and controls.
Results: Multiple specific cytokines were elevated in the painful discs, compared with control discs of the same grade of degeneration. Cytokines found at significantly higher levels in the painful discs were: Growth-Regulated Oncogene (GRO), Interleukin (IL)-8, Monocyte Chemoattractant Protein (MCP)-1, and Regulated upon Activation, Normal T-cell Expressed and Secreted (RANTES).
Conclusion: Because certain cytokines are elevated in painful discs, these factors deserve further study to determine if these peptides play a causative role in the pain process within the intervertebral disc. Causative cytokines may serve targets for pharmacologic or biologic manipulation in the future as a treatment mechanism for discogenic pain.
*Abstracts/permission forms not received at the time of publication
**Current and ongoing research