Anabolic Therapies and Strategies to Regenerate the Musculoskeletal System

April 3, 2013

Background and Approach

The overall goals of all NIAMS roundtables are to discuss scientific and clinical needs and opportunities and to listen to the concerns and challenges facing the scientific community. These sessions provide a valuable source of input for the NIAMS planning process.

This roundtable focused on basic and translational research needs related to the development of anabolic therapies (i.e., treatments that build or increase the size and function of a tissue) and strategies to regenerate the musculoskeletal system. In advance of the meeting, participants were encouraged to consult with colleagues regarding the following questions:

  1. What are the most promising anabolic targets (cells, ligands/receptors, intracellular signaling pathways) for diseases and disorders of bone, cartilage, and muscle at this time?
  2. What common pathways affecting multiple tissues of the musculoskeletal system could be explored to develop anabolic therapies? What anabolic therapies for one musculoskeletal tissue may have the potential for synergistic effects on the musculoskeletal system as a whole?
  3. What are the concerns and potential side effects or off-target effects of candidate anabolic therapies? What are potential strategies for limiting these effects?
  4. What are the gaps in our knowledge of the biology of the musculoskeletal system that limit progress in the development of effective anabolic therapies? What research infrastructure, resources, or methodologies would facilitate the rapid development of anabolic agents for the musculoskeletal system?
  5. What innovative research strategies are likely to advance the development of anabolic therapies?

Aspects of these topics were discussed in depth and are summarized here. The NIAMS greatly appreciates the community’s input on these questions.


Musculoskeletal diseases and disorders, such as osteoporosis, osteoarthritis (OA), and muscle atrophy, arise in part from inadequate anabolic tissue regeneration or repair. The participants agreed that strategies for activating anabolic pathways in bone, cartilage, and muscle hold considerable potential to yield new therapeutic approaches for diseases relating to individual tissues and for the musculoskeletal system as a whole.

They reviewed what is known about anabolic processes of bone, cartilage, and muscle, and progress toward developing anabolic therapies for related disorders. Research questions associated with several pathways and the current understanding of their effects on the different musculoskeletal system components were discussed in detail. The Wnt,1 PTH/PTHrP,2 the TGFβ superfamily,3 IGF/IGFBP,4 leptin, adiponectin, and serotonin pathways were among those mentioned. Additional discussion addressed the influence of the central nervous system, physical activity, energy metabolism, and inflammation on anabolic processes.

The participants noted that, while the bone, cartilage, and muscle fields are at different stages of research and therapeutic development, all would benefit from additional studies. The pharmaceutical industry has devoted considerable resources toward anabolic therapies for bone. Teriparatide, a PTH fragment, is the only FDA-approved anabolic therapy for the musculoskeletal system — it is intended to prevent osteoporosis associated fractures. Despite a comparable degree of effort in the joint field, there are no approved disease-modifying therapies to regenerate cartilage affected by OA. Effective pharmacologic strategies to stop muscle atrophy also are lacking.

Researchers have identified several potential treatment targets with anabolic effects. Examples in bone include the Wnt signaling antagonists sclerostin and DKK-1;5 agents that regulate or mimic PTH action (e.g., PTHrP, and calcilytic drugs aimed at the calcium sensing receptor); and components of the BMP/TGFβ6 pathways. Recent studies suggest that several pathways, such as those activated by PTH, calcitonin, and FGF-18,7 may regenerate cartilage, even in joints with advanced OA. A variety of agents are being explored as ways to maintain or increase muscle mass. Inhibition of the TGFβ super family member myostatin/GDF-88 (a natural inhibitor of muscle growth) or its receptor (activin receptor IIA/B) increases muscle mass, prevents muscle wasting, and has potential as a treatment for various muscle conditions. These pathways are at different stages of development, some at preclinical studies, some at phase II/III trials, and some with unexpected side effects that cast doubt on their further therapeutic development.

The mechanistic details of many pathways targeted by current or potential anabolic therapies are not well understood. For example, it remains unclear why intermittent, but not continuous, PTH administration is anabolic, or why the effectiveness of intermittent PTH wanes over time. Participants also cited reports indicating the transient effect of some molecules that target the Wnt pathway. They noted that anabolic agents influence pathways that may be active and important only for select periods during growth and development. The specificity of these networks and processes during development may depend on localized ligands that interact with receptors that are widely expressed in multiple tissues. Thus, systemic delivery of anabolic agents raises concerns about off-target and possibly adverse effects. So far, this burden has limited the practical application of molecules such as IGF-1 and myostatin.

Although much of what is known about anabolic pathways comes from developmental biology, participants cautioned against direct extrapolation of this knowledge to adult tissues. They emphasized the value of testing anabolic therapies in preclinical models that mimic human pathological and aging conditions.

Osteoporosis and OA are characterized by excessive tissue degradation (i.e., catabolism). The participants noted that combined anticatabolic and anabolic approaches should be more effective than either alone. However, in bone, where resorption and formation are normally coupled in the bone remodeling process, the combined administration of teriparatide and an anticatabolic bisphosphonate blunts teriparatide’s anabolic effects. This observation suggests that strategies to uncouple the bone formation and resorption processes could impact the development and use of anabolic therapies for bone.

While bone health depends on the balance between bone formation by osteoblasts and bone resorption by osteoclasts, parallels with similar processes in cartilage and muscle have not been well defined. Cartilage homeostasis is generally considered to depend on chondrocyte activities with possible influences from the surrounding tissues including synovium and subchondral bone, while muscle appears to depend on both fully differentiated and immature cells of the muscle lineage. In addition, muscle grows primarily by increasing the size of existing fibers through increased protein synthesis, which can occur over a relatively short time. In contrast, bone growth and cartilage maintenance depend on cell proliferation, differentiation and matrix accumulation, typically occurring on a longer time scale.

Modulation of the body’s intrinsic healing capability could be an important strategy for treating people who have musculoskeletal conditions. Knowledge of endogenous repair processes and how they relate to damage or dysfunction could inform approaches to preserve and restore tissue. The participants also commented that better knowledge of the nature and identity of stem and progenitor cells would be useful. Understanding the cells’ in vivo sources, function, and lineage development, and the mechanisms of controlling their maintenance, proliferation, and differentiation, may accelerate therapy development. The ability to maintain an adequate supply of stem or progenitor cells and appropriately control their growth and differentiation will be important.

The evidence for biochemical and biomechanical signaling interactions among bone, muscle, and cartilage is strong. Joints’ bone/cartilage interfaces may be critical for maintaining healthy articular cartilage, and bone-active agents may be useful, directly or indirectly, for OA treatments. There is increasing interest in endocrine or other biochemical interactions between bone and muscle, in addition to the long-recognized mechanical signals arising from muscle activity. Recent observations further suggest important connections between the regulation of bone mass and the control of energy metabolism, with the central nervous system playing an important role. Given the energy-intensive nature of muscular contraction, it seems likely that muscle also participates in this regulatory network.

The complexity of the musculoskeletal system and the possible anabolic pathways must be considered when developing and testing potential therapies. Some drug targets or pathways exist widely within and outside of musculoskeletal tissues. Some are capable of forming different receptor/ligand combinations, and some may cause multiple different effects. Furthermore, the action of anabolic agents may depend on the local environment and stage of differentiation of the target cell. For example, the Wnt pathways are generally considered anabolic for bone. The relatively limited distribution of the Wnt inhibitor sclerostin confers significant specificity on the action of anti-sclerostin antibody as a bone anabolic agent. However, the less restricted tissue distribution of the Wnt inhibitor DKK-1 potentially limits the utility of anti-DKK-1-based therapies. Activating the Wnt pathway can either positively or negatively affect cartilage, depending on the developmental stage of the chondrocytes. While muscle researchers have studied the Wnt pathway extensively in development, they know relatively little about its role in mature muscle.

Future Directions

Participants commented on the utility of developing methods for delivering drugs to the intended target tissue and strategies for improving a potential treatment’s specificity by controlling the target tissue’s local environment. Such approaches could restrict an intervention’s effects to a localized region and limit the potency and length of time when it is active. Limiting the use of a drug for particular conditions is another way of conferring specificity in anabolic therapy development. When discussing potential undesirable consequences of anabolic therapies, participants reiterated the importance of appropriately controlling growth and differentiation. Cancer is always a concern when developing anabolic therapies that may influence stem cell behavior. The participants noted that while osteosarcoma is a common concern and often observed in preclinical animal studies of anabolic therapies, it occurs relatively rarely in human adults. As additional anabolic drugs become available, health care providers and patients must continue to weigh risks and benefits before beginning a particular treatment.

As discussed above, many anabolic targets and pathways are shared among different musculoskeletal tissues and play prominent roles in many other cells. Our understanding of the links between bone metabolism, energy metabolism, and the central nervous system is also improving. Recent advances in musculoskeletal research have led to a recognition that, in order to fully explore the potential and consequences of anabolic therapies, bone, cartilage, and muscle must be viewed as components of a network that includes not only other tissues of the musculoskeletal system, but also other organs and systems. Some anabolic therapies may benefit more than one tissue or the musculoskeletal system as a whole; some may produce multi-system synergistic effects. For example, strategies for increasing muscle mass may also result in increased bone mass and strength, and vice versa. On the other hand, attempts at altering an anabolic pathway in one tissue, especially through systemic delivery of therapies, may have multiple consequences. These could include off-target, and possibly adverse, effects.

Taking a system-wide view would represent a paradigm shift, in that the bone, cartilage and muscle communities have traditionally focused on their single tissue of interest. It could produce valuable information if studies exploring potential new treatment targets would incorporate outcome measures that reflect the status of bones, joints, and muscles together, and consider the musculoskeletal system in its larger biological context.

The separation between different traditional scientific disciplines, as well as clinical specialties, may hinder collaborations that could advance the development of anabolic therapies and regenerative strategies for the musculoskeletal system. Today’s knowledge about the different musculoskeletal tissues is rooted in the various fields’ unique cultures—histories that influence the way researchers think, develop hypotheses, and design experiments. Participants noted that physiology research has generated many muscle biology advances, while bone research has strong roots in endocrinology. Although expertise in specific tissues is important, interactions between experts from different fields are critical if studies of potential treatments are to consider musculoskeletal tissues together as an integrated system. Professional societies could facilitate communication among groups of scientists to build the necessary bridges and cross-fertilize musculoskeletal research fields.

Participants emphasized that a more comprehensive assessment of the pathways that could yield targets for anabolic therapies will require tools and skilled researchers in systems biology, bioinformatics, and the management of large datasets. Genomic approaches such as genome-wide association studies were mentioned for their potential to reveal novel loci that influence musculoskeletal phenotypes. Understanding the biological basis for these associations will rely on continued development of resources and methodologies, such as computational tools to model the net effects of large numbers of interacting components and combine information from different biological levels (e.g., genetic variation, epigenetics, gene expression, phenotype), and expanded resources for mouse genetics. The participants also noted that there are rich databases on teriparatide effects from its use in osteoporotic patients. In view of the increasing evidence of the potential beneficial effects of teriparatide on cartilage, it may be appropriate to revisit existing clinical data and carry out secondary data analysis for teriparatide’s effects on OA.

Several participants noted the value of common resources, such as core facilities for animal phenotyping and bioinformatics. Systems, genetic, and computational analysis could also help to explain complexities revealed in clinical studies and clinical trials. The efficacy of potential anabolic therapies may vary with disease subtype or stage, a patient’s genetic background, and environmental factors. Analytical methods that stratify diseases and patient populations with respect to such variables can help to target therapies to those most likely to benefit, and to identify those most vulnerable to adverse events.

In addition to supporting the generation of sophisticated research tools and the ability to analyze large, complex data sets, participants cautioned that established techniques such as those common to the traditional biochemistry and pharmacology fields continue to have value. Research questions should drive investigators’ choice of tools, rather than the reverse. In addition, the participants voiced the view that exploratory studies that generate new hypotheses continue to be important.

Osteoporosis and OA clinical trials are hindered by a combination of slow and unpredictable rates of disease progression and relatively insensitive detection methods. Therefore, the development of surrogate markers for clinical trials of anabolic therapies is desirable. Markers can involve imaging technologies, biochemical analyses, or functional measures. Surrogate markers for osteoporosis therapies should reflect changes in intrinsic bone strength and fracture risk. For OA, the unclear relationship between pain and disease progression complicates the development of surrogates. Surrogate markers for OA treatment, such as those being developed through the FNIH Biomarkers Consortium, will most likely parallel structural changes in the affected joint. There have been recent advances in imaging biomarkers and biochemical measures for muscle, but clinical trials continue to rely on measures of complex function such as timed walking tests. Further studies of biomarkers and validation of early surrogate markers can accelerate the clinical testing of potential therapeutics for musculoskeletal conditions.

Diseases such as osteoporosis and OA are common and associated with variants in multiple genes. However, research into rare monogenic musculoskeletal diseases has contributed and is expected to continue to inform the development of anabolic therapies. Study of the rare conditions sclerosteosis and van Buchem disease, for example, led to the development of the promising sclerostin-based anabolic strategies for bone that are in preclinical and clinical testing. Anabolic therapies developed as treatments for common diseases may also help people who suffer from rare genetic diseases. Osteogenesis imperfecta, many forms of which are characterized by low bone mass with high bone turnover, was cited as one such example.

Engagement by the private sector is likely to be essential for developing and testing any anabolic therapy before it can be accepted for widespread clinical use. Roundtable participants included scientists with private sector experience, who pointed out that, while the pharmaceutical industry and academic researchers share the goal of translating fundamental discovery into clinical application, the communities’ approaches and priorities differ. There are numerous steps along the drug development pathway during which the academic investigators could interact with the private sector. Effective collaborations require an understanding and appreciation of the cultural and philosophical differences that drive decisions of the academic world and private sector.

Looking forward to the development of future anabolic therapies for the musculoskeletal system, participants reiterated a few themes that occurred throughout the discussion. There is a need for an integrative, global view in developing anabolic therapies. Anabolic therapies for one musculoskeletal tissue should take into consideration their effects on other musculoskeletal tissues, and the musculoskeletal system as a whole, while also including organs and systems outside of the musculoskeletal system. This will benefit from a better understanding of the crosstalk between the musculoskeletal tissues, as well as other tissue systems. The heterogeneity of tissues, pathways, developmental stages and disease status further adds to the complexity of mechanistic studies and the development and testing of interventions. More in-depth knowledge about these areas is needed for researchers to develop better strategies to control specificity, to limit off target and adverse effects, to control the proliferation and differentiation of progenitor and stem cells, and to combine the benefits of anabolic and anticatabolic therapies. Participants further emphasized that effective collaboration between experts from different fields and team science will accelerate the development of novel anabolic therapies.


BARTON, Elisabeth R., Ph.D., University of Pennsylvania
BELLIDO, Teresita, Ph.D., Indiana University/Purdue University Indianapolis
CHUBINSKAYA, Susan, Ph.D., Rush University Medical Center
CLEMENS, Thomas L., Ph.D., Johns Hopkins University (Co-chair)
ESSER, Karyn, Ph.D., Department of Physiology
HART, Tracy, Osteogenesis Imperfecta Foundation
KARSENTY, Gerard, M.D., Ph.D., Columbia University Medical Center
KRONENBERG, Henry, M.D., Massachusetts General Hospital
MATTHEWS, Gloria L., D.V.M., Ph.D., Genzyme
MBALAVIELE, Gabriel, Ph.D., Washington University School of Medicine
O’KEEFE, Regis, M.D., Ph.D., University of Rochester
PACIFICI, Maurizio, Ph.D., Children's Hospital of Philadelphia
ROSEN, Clifford, M.D., Maine Medical Center Research Institute
ROSEN, Vicki, Ph.D., Harvard School of Dental Medicine
WILLIAMS, Bart, Ph.D., Van Andel Research Institute
WILLIAMS, John, Ph.D., National Institute on Aging/National Institutes of Health


CARTER, Robert, M.D.
CHEN, Faye H., Ph.D. (Co-chair)
DRUGAN, Jonelle K., Ph.D., M.P.H.
KATZ, Stephen I., M.D., Ph.D. (Co-chair)
LESTER, Gayle, Ph.D.
LINDE, Anita M., M.P.P.
McGOWAN, Joan A., Ph.D.
MOEN, Laura K., Ph.D.
SHARROCK, William J., Ph.D.

1 Wnt = wingless-type mouse mammary tumor virus (MMTV) integration site family
2 PTH = parathyroid hormone; PTHrP = parathyroid hormone-related protein
3 TGFβ = transforming growth factor beta
4 IGF = insulin-like growth factor; IGFBP = insulin-like growth factor-binding protein
5 DKK = dickkopf-related protein
6 BMP = bone morphogenetic protein
7 FGF = fibroblast growth factor
8 GDF = growth/differentiation factor