Commentary & Perspective

Commentary & Perspective on
"Osteogenic Activity of the Fourteen Types of Human Bone Morphogenetic Proteins (BMPs)"
by Hongwei Cheng, MD, PhD, et al.
and on
"A Single Percutaneous Injection of Recombinant Human Bone Morphogenetic Protein-2 Accelerates Fracture Repair"
by Thomas A. Einhorn, MD, et al.

Commentary & Perspective by
Gary E. Friedlaender, MD*, Wayne O. Southwick Professor and Chair, Department of Orthopaedics and Rehabilitation, Yale University School of Medicine, New Haven, CT

Is the Standard Still Gold?

This is an exciting time for those interested in bone regeneration and its application to skeletal homeostasis, fracture repair, and bone-graft incorporation. Much is understood about osteoblasts, osteoclasts and macrophages, the cell populations responsible for osseous repair, and the cytokines that control the cascade of events that are the bone remodeling cycle. But, before this evolving knowledge can be optimally translated to clinical application, many questions must be answered. For example, which cytokine or cytokine combinations and in what doses; when and where are they to be used; and how are they best delivered? Two excellent examples of these investigations are published here, one by Cheng and colleagues entitled "Osteogenic Activity of the Fourteen Types of Human Bone Morphogenetic Proteins (BMPs)," and the other by Einhorn and coworkers entitled "A Single Percutaneous Injection of Recombinant Human Bone Morphogenetic Protein-2 Accelerates Fracture Repair."

Our challenge as clinicians and scientists is to understand the important contributions provided by such work, recognize the limitations of conclusions based upon the experimental design, and bring appropriate information into our patient care decision-making process. Just for starters, we must understand the basic nature of bone regeneration at the cellular and even at the molecular levels. We must also be aware of the potential differences among well-designed in vitro studies that study in isolation what are really complex in vivo interactions. If that is not enough, we are constantly faced with the risky task of extrapolating any of these observations into the clinical arena while controlling for the many biases inevitable with new knowledge.

Both of the articles discussed here advance our knowledge and raise unanswered questions.

There are at least fifteen members of the TGF-beta superfamily, some of which have been demonstrated to cause the recruitment, differentiation, and upregulation of cells involved in osteogenesis. The study by Cheng and colleagues aimed to clarify the relative osteogenic activity of fourteen BMPs. To do so, adenoviruses expressing single BMP molecules were prepared and then used to infect standardized cell populations in culture. These cell lines individually reflected a pluripotential character (mouse C3H10T1/2), were precommitted to an osteoblast lineage (mouse C2C12), or were cells with an established osteoblastic behavior (human osteosarcoma line TE-85). Measured increased activity by these cells indicated that the specific BMP caused either osteoblastic differentiation or upregulation of osteoblastic activity. By testing each BMP against the same cell lines, relative activity could be determined. Highest responses in these cell lines were seen with BMPs 2, 4, 6, 7, and 9, with BMPs 2, 6, and 9 quantitatively most active under the circumstances assessed.

This study provides valuable insight into the relative activity of isolated BMPs on well-characterized cell populations that represent varying positions along the osteoblast differentiation pathway. Furthermore, this model can be used to assess combinations of molecules introduced at different periods of time.

There are, importantly, limitations with this approach if one wishes to establish the relative efficacy of selected BMPs in human disorders. For example, the approach does not take dose into account and depends on surrogate cell populations. Perhaps most importantly, osteoinduction and osteogenesis in vivo involve the cooperation of several cell types rather than a single, isolated cell type, and these events have been demonstrated to require various BMP (and other) molecules. Indeed, it has been shown that introducing one BMP may cause cells to produce other BMPs. Thus, there is an artificial nature to this experimental model that may not predict clinical behavior. Nonetheless, it is an important step that helps to clarify aspects of the physiologic mechanisms at play. This type of study can be very helpful in selecting future directions worthy of closer attention.

The study by Einhorn and coworkers has the advantage of making quantitative observations in an intact experimental model. As such, it begins to clarify end results with the potential to vary the dose, the timing of treatment, and the carriers used to implant a specific BMP. Other measures of bone remodeling can be assessed by using histomorphometric techniques, and the expression of various BMPs and their relative timing of appearance have also begun to be explored in similar models.

This model demonstrates, as predicted by the investigators, that the application of a single BMP delivered percutaneously accelerates fracture repair in the rat. The study was not intended to compare the efficacy of various single BMPs or combinations of BMPs, and it uses a single time, dose, and carrier that clearly works in the rat but may not be optimal either in rats or in humans. These issues can be explored in the future but were not goals of the present study, nor should they be inferred from this work as the authors carefully caution.

What is particularly exciting about these and similar studies is the notion that we can now, as clinicians, begin to favorably manipulate the biology of fracture repair. We will learn which fractures require intervention (perhaps even the majority, if speed of repair can be predictably enhanced) along with the doses, timing, and carriers most likely to work best. Investigators will clarify which patient diseases represent contraindications to the use of osteoinductive molecules, and which disorders, habits (such as smoking), drugs (including some NSAIDs and chemotherapeutics), and physical modalities (including irradiation) with known negative effects on the skeleton can be compensated for by the use of adjunctive osteoinductive biomolecules.

Clinical trials are enormously expensive such that large, multicenter, outcomes-oriented studies comparing different BMPs (controlling for the best dose, timing, and carriers) are unlikely. However, as improved results can be documented through rigorous outcomes research, the issue of funding new technology in the clinical arena will become an important priority.

Exciting times, excellent research, and new knowledge all portend reassessment of our past gold standards, including autogenous bone graft. Is it time to change the standard? Perhaps not yet, but soon!

*The author did not receive grants or outside funding in support of the research or preparation of this manuscript. The author did receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity (Stryker Biotech, Biometric Pharmaceuticals). No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the author is affiliated or associated.