Proceedings from the Workshop on Science-Based Assessment: Accelerating Product Development of Combination Medical Devices (2004)

Barbara D. Boyan

Georgia Institute of Technology

Marshall Urist was a pioneer in the 1970s in understanding the principle of osteoinduction and in determining whether it could be used clinically. His work focused on determining whether the phenomenon involved a protein and, if it did, whether one or more proteins were involved. Dr. Urist coined the term “bone morphogenetic protein” because he recognized that the agent or agents that he was investigating were responsible for initiating the cascade of developmental events leading to bone morphogenesis. In contrast, growth factors were believed at that time to regulate cell proliferation.

Today, we know that bone morphogenetic proteins (BMPs) are members of the transforming growth factor beta (TGFβ) superfamily and that at least 15 BMPs exist. We also know that BMPs function in bone formation by recruiting mesenchymal progenitor cells and initiating endochondral ossification in heterotopic sites, that they stimulate osteogenesis in orthotopic sites by acting on multiple cell types, and that they not only affect processes related to bone formation but also are actively involved in the formation of other musculoskeletal tissues as well as nonskeletal tissues such as the kidney and the cardiovascular system. Research today on BMPs focuses on mesenchymal stem cells (MSCs), progenitor cells, and committed cells; ALK receptors and SMAD signaling; how BMPs are synthesized by multiple cell types; autocrine and paracrine actions; and the inhibitors that are cosecreted. During the past 2 years, the first commercial products using BMPs have entered the marketplace.

Challenges that remain in the basic science of BMPs include determining whether response varies among cell types and at different rates of maturation within the same lineage; how target cell specificity can be controlled clinically; what the specific role of each BMP is; and how BMP activity is regulated in vivo.

Amy J. LaForte

Stryker Biotech

Bone morphogenetic proteins (BMPs) are a class of proteins that induce bone formation. New technology in orthopedic repair combines BMPs with a carrier matrix to create a combination product that, when implanted in a bony defect, initiates bone formation (osteoinduction) and provides local containment and cell adhesion (osteoconduction). However, orthopedic surgeons currently lack standardized information about the potency to initiate bone formation of either new manufactured products or allograft/ autograft bone tissues. To put this in perspective, please note that labeling standards exist that require sunscreens to have an SPF rating, generators to specify a power output, and food to be labeled with a caloric assessment. Development of new products intended to generate bone would be facilitated if a standardized measurement of osteoinductivity were developed. In addition, investment is needed in the development and validation of new methods of imaging and quantifying new bone formation in humans in order to better assess the clinical utility of products and tissues. New methods of evaluating clinical endpoints combined with advances in imaging could shorten clinical development cycles and increase the accuracy of safety and efficacy evaluations.

William McKay

Medtronic Sofamor Danek

On July 2, 2002, the U.S. Food and Drug Administration (FDA) approved INFUSE™ bone graft, which contains a recombinant human bone morphogenetic protein (rhBMP-2), for use as a combination biologic medical device in conjunction with a titanium interbody fusion device (LT CAGE™). INFUSE™ bone graft is the first bone morphogenetic protein (BMP) that FDA determined to be safe and effective as a replacement for autogenous bone graft. FDA approval came 16 years after the discovery of rhBMP-2 and after hundreds of millions of dollars had been spent in research and development. The phase that took the most time in the product development process was

regulatory clearance, including the investigational device exemption (IDE) and the premarket approval (PMA) application, which took 6 years from the time that the final concentration and carrier had been identified. Research and development leading up to this final product identification required standard preclinical safety and effectiveness studies that could not have been avoided or accelerated. During this process, however, it was discovered that nonhuman primate bone biology is the most similar to that of human bone biology and the most predictive of effectiveness in humans. Studies using nonhuman primate bone biology were therefore recommended as the standard for testing.

INFUSE™ bone graft consists of rhBMP-2 at a concentration of 1.5 mg/ml delivered on a collagen sponge. Theoretically, rhBMP-2 could be used in any bone grafting procedure, but because FDA currently requires approval for each specific spinal infusion technique or bone grafting procedure, limitations on company resources prohibit the development of products for all procedures. Companies must selectively choose the more commonly used bone grafting procedures for clinical study and FDA approval.

Initial FDA approval of a bone graft replacement requires a well-designed, prospective randomized study demonstrating the safety and effectiveness of the product. Since the contribution of the bone graft replacement in a combination biologic medical device is the initiation of bone formation, the clinical protocol should involve the utilization of thin-slice computed tomography (CT) scans to assess the degree of bone formation and fusion. CT scans have become the gold standard in assessing bone formation and quality.

Less stringent FDA clearances for use of the identical bone graft replacement product in expanded clinical bone grafting indications would significantly accelerate the process development. Adoption of a more methodical, streamlined approach would benefit the patient, companies, and the FDA. The extent of safety and effectiveness data required for expanded indications should be based on how different they are from the original cleared indication. For example, clearance of INFUSE™ bone graft with the same concentration and carrier in any interbody fusion cage should require only abbreviated PMA supplement justification, since animal studies have shown that BMP is effective in all types of interbody fusion devices. Further clearance of INFUSE™ bone graft with the same concentration and carrier used in the posterolateral fusion technique should require a PMA supplement with only limited supporting clinical data indicating safety and effectiveness. If the BMP concentration and carrier are changed, however, a prospective randomized clinical trial should probably be required to support a new PMA application.

In summary, all new bone graft replacement products should undergo rigorous prospective randomized clinical investigations involving CT assess-

ment of bone formation. Subsequent approvals of the same growth factor (BMP) concentration and carrier for expanded indications should be approved via PMA supplements with only animal data or limited clinical data (see Table 1).