A Brief Overview of the STP 36th Annual Symposium: Musculoskeletal System

Abstract

The title of the 2017 Society of Toxicologic Pathology symposium was “Musculoskeletal System.” A brief overview of the General Scientific Symposium is presented herein and describes the topics presented by each speaker. Symposium speakers addressed subjects pertinent to musculoskeletal system toxicologic pathology and drug development ranging from molecular biology of bone homeostasis to regulatory agency requirements and considerations for registration of bone therapeutics. This overview serves to summarize sessions and is intended as a guide to the individual submissions by speakers and symposium contributors.

The musculoskeletal system is an important and complex system to evaluate due to its composite physiology and sampling and analysis challenges. The Society of Toxicologic Pathology’s (STP) 36th annual symposium sought to dissect and demystify the hurdles involved in musculoskeletal evaluation and developing therapeutics for bone disorders and rare bone diseases. The scientific sessions were designed to provide attendees with a working knowledge of the tools available to analyze the musculoskeletal system, molecular biology of bone homeostasis, effects of perturbation by agents targeting bone and other tissues, safety and regulatory considerations when designing programs for bone drug submissions, and lessons learned from a variety of clinical populations and nonclinical studies intended to support investigational new drug (IND) applications. These topics were addressed by expert speakers with in-depth knowledge of the clinical populations affected and/or the nonclinical packages developed to register drugs intended to treat those patients. Speakers frequently shared data directly from INDs and other nonclinical safety packages as case studies to demonstrate critical points and perspectives.

The General Scientific Symposium on the Musculoskeletal System

The STP’s 36th annual symposium was opened with a dynamic and engaging presentation by Dr. Cathy Carlson. Using a variety of degenerative, toxic, and metabolic degenerative bone disease cases, Dr. Carlson applied a case presentation approach to guide the audience through her decades of osteoarthritis (OA) research. She reviewed worst-case scenarios in tissue sampling, fixation, and trimming and warned of the perils of sampling a disease process in the chronic stages or incorrectly interpreting a change based on plane of section; several published examples were shared to prove the latter point. Special histopathochemical techniques, computed tomography (micro-CT), and the pathophysiology of a variety of musculoskeletal conditions including gangliosidoses, Legg–Calve–Perthes, osteoporosis, and other degenerative joint diseases (DJDs) were showcased from species including humans, nonhuman primates, goats, pigs, dogs, cats, and rodents. Various translational details (e.g., goats do not develop osteochondrosis dessicans, rodents lose cartilage vessels by 1 day old) were of particular interest to audience members working with animal models of DJD. In addition to being an excellent introduction to the broad world of musculoskeletal pathophysiology, Dr. Carlson’s lecture was a fitting segue into session 1.

Monday morning’s session 1, aptly entitled “Toolbox for the Evaluation of the Musculoskeletal System,” was an appropriate entry into the subject of bone and muscle analysis for pathologists. Histopathological, biomarker, biomechanical, and imaging modalities were discussed, as were considerations for applicable animal models and the appropriate terminology and regulatory considerations for such.

Dr. Aurore Varela opened session 1 by reminding the audience that histopathology is not a particularly sensitive method to detect changes in bone. Through several examples in drug development (e.g., peroxisome proliferator (PPAR)γ, Wingless-related integration site (WNT) inhibitors, ibandronate), Dr. Varela discussed sensitive approaches to evaluating bone changes, namely, biomarkers (urinary N-telopeptides, procollagen type 1 N-propeptide, and deoxypyridinoline), imaging, and functional assays to assess bone integrity. Reviewed imaging methods included dual-energy X-ray absorptiometry, micro-CT, and Xtreme CT with a resolution of 30 μm. She explained how deformation and elasticity can be measured by a 3-point bending test or by vertebral compression. Dr. Varela also noted that the timing of bone quality studies critically relies on bone turnover rates, which greatly vary between humans, nonhuman primates, and rodents. She further provided guidelines on the appropriate duration of specific studies and informed the audience that many of the noninvasive techniques discussed were well suited for longitudinal analysis.

Dr. John Vahle’s presentation on the proper use and contextualization of diagnostic criteria, and the introduction of the International Harmonization of Nomenclature and Diagnostic Criteria for Lesions in Rats and Mice (INHAND) publications and best practices, fittingly preceded the sessions utilizing this terminology. While recognizing the utility, sensitivity, and specificity of techniques described and demonstrated in Drs. Carlson’s and Varela’s lectures, Dr. Vahle acknowledged that it is not practical to utilize all these techniques on standard toxicology studies. The “Nonproliferative and Proliferative Lesions of the Rat and Mouse Skeletal Tissues (Bones, Joints, and Teeth)” guidance was designed to avoid “diagnostic” terms that may confuse clinicians and regulatory reviewers and include a lexicon that could be semiquantitative in nature (e.g., “increased osteoblastic surface”). Although alignment in terminology is typically a desirable attribute across toxicology studies in support of a program or across programs for a given target, Dr. Vahle demonstrated how variability in choice of terms (e.g., increased osteoid, increased osteoblastic surface, increased bone, osteoblast hyperplasia) could gain alignment. Dr. Vahle concluded his talk by challenging the audience to define adversity with 2 examples of treatment-related increased osteoid.

Dr. Matthew R. Allen ensued in leading the audience through a discussion of longitudinal bone growth, intracortical remodeling, collagen–mineral interactions, and aging, utilizing both compelling and humorous narratives. Emphasizing that rats and mice don’t cease longitudinal bone growth, he used bisphosphonate (e.g., alendronate and zolendronate) examples to illustrate how osteoclast inactivation leading to unresorbed bone has been misinterpreted as new bone. Dr. Allen described the physiology of human intracortical remodeling, and how this differs from the rodent, which remodels bone at the endocortical surface and on the trabeculae. He also described pitfalls of using rodents to study collagen–mineral interactions, including the inability of rodents to adequately model accumulation of cross-linked collagen or microdamage, decreases in mineral heterogeneity, or the effects of material-level bone properties. He did note that microcracks in rodent bone may be induced by vigorous loading activity. Dr. Allen also reviewed the comparative collagen orientation between species and discussed how this contributes to bone hydration, which is essential to bone strength. During aging, a bone modeling unit (BMU) imbalance materializes, leading to bone mass and quality changes. Dr. Allen emphasized the difference between studying and recognizing “healthy aging,” and “unhealthy aging” in bone, and how the latter must take into account comorbidities that affect the BMU including cardiometabolic diseases.

Session 2 on Monday afternoon was entitled “Bone Therapeutics: Safety Considerations.” The speakers were exceptionally experienced in the regulatory aspects of nonclinical filing packages for bone therapeutics. Topics ranged from study planning to fulfill pharmacology and toxicology facets of a program to the challenges associated with developing therapies for rare bone disease.

The session opened with a lecture defining bone safety assessments as an analysis of bone quality. This review of bone mass regulation included descriptions of porosity/age correlations and the coalescence of bone pores leading to permanent trabeculae loss. Bisphosphonates, sclerostin, and sodium fluoride were used as examples of bone density modulators, and several nonclinical strategies were presented as examples on how the requirements for IND applications to support appropriate clinical trials for bone agents could be satisfied. Some bone agents have qualified for a 2-species exemption based on pharmacology, but that for rare bone diseases, transgenic mice might be considered as pharmacology models.

Next, Dr. Jacques Brown shared his perspective as a clinician treating patients with antiresorptive therapies for bone diseases. Dr. Brown advised the audience on the diagnosis and classification of osteonecrosis of the jaw, and how this rare condition must be differentiated from medication-related osteonecrosis of the jaw (MRONJ). He explained how risk factors for MRONJ are not limited to treatment with bisphosphonates or denosumab, but also include antiangiogenesis drugs such as VEGF inhibitors, as well as tooth extractions and decreased oral hygiene. The latter part of Dr. Brown’s talk focused on subtrochanteric and diaphyseal atypical femoral fractures, which may occur with bisphosphonate use. He further explained how these fractures are not considered typical for postmenopausal osteoporosis (PMO), as PMO fractures are typically transverse and occur in the femoral neck or vertebrae.

Dr. Rogely Boyce closed the first half of session 2 with a toxicology perspective on antiresorptive bone drugs. She began with a set of case examples reminding attendees that typical sections aren’t adequate to identify subtle treatment–related effects. Sharing histopathology of growth plate changes from both bisphosphonate and denosumab toxicology studies, Dr. Boyce demonstrated how fluorochrome pulse methodologies can be critical in understanding bone effects. Antiresorptive bone agent labels include nonclinical information regarding maternal and fetal effects, and Dr. Boyce shared data from the toxicology package supporting denosumab registration, which showed fetal effects in an early pre- and postnatal development study.

The second half of session 2 was opened by Dr. Luc Chouinard, who shared his extensive experience as a study pathologist evaluating chronic toxicology studies of bone agents. Dr. Chouinard reviewed the function of the bone basic multicellular unit basic multicellular unit (BMU) and the balanced and coupled effects in bone remodeling. The physiology of parathyroid hormone (PTH) actions and osteoblast differentiation was also reviewed. Dr. Chouinard discussed osteosarcoma formation observed in studies with both PTH1-34 and parathyroid hormone-related peptide (PTHrP) molecules. With the identification of the SOST gene as a compelling target for various bone diseases came the concern of uncontrolled bone formation as a result of inhibition of the SOST gene product sclerostin. Dr. Chouinard directed investigators to carefully consider all potential mechanisms, incorporate any existing data, and fully characterize any proliferative signals. The designs for the IND-enabling toxicology studies to support the antisclerostin antibody romosozumab were discussed, as to exclude animals that developed antidrug antibodies, and complement gross and histologic examinations with radiographic, peripheral quantitative computed tomographic imaging.

Dr. Frank Rauch educated the audience in rare bone diseases, focusing on osteogenesis imperfecta (OI) and X-linked hypophosphatemia (XLH). He reviewed the classification of OI based on qualitative or quantitative defects in specific collagen isoforms and explained how treatment regimens for children with OI can be very efficacious but have specific durations of treatment during the bone growth phase to prevent recurrence of bone resorption. He also noted how XLH-affected individuals most commonly have hypophosphatemic rickets, which present with hypomineralized periosteocytic lesions. Dr. Rauch discussed the current treatments of XLH-related lesions, which target downstream effects of the genetic mutation affecting osteocyte function, but do not address the genetic alteration itself.

Dr. Boyce closed session 2 with a toxicology perspective on the rare bone diseases discussed at length by Drs. Brown and Rauch. Dr. Boyce stressed that although many disease models were discussed at the symposium, they are not sufficient to satisfy the pharmacologic and toxicologic concerns of a program for regulatory consideration. The Food and Drug Administration (FDA) draft guidance on drug development issues for rare diseases is a useful guide to understand the framework needed to successfully prosecute a target for IND submission.

Tuesday morning of the 36th annual STP symposium began with session 3 entitled “Unintended Pharmacologic Effects on Bone.” This session was designed to approach bone findings resulting from unintended direct and indirect pharmacology.

Session 3 began with Dr. Alexander G. Robling highlighting that there are few true anabolic bone agents (e.g., teriparatide and abaloparatide) that pharmacologically activate bone formation, but there are myriad avenues for inhibiting osteoclasts. Dr. Robling then elegantly dissected the sclerostin story, beginning with research into osteoporosis pseudoganglioma syndrome, which maps to a mutation in low-density lipoprotein receptor 5 (LRP5). Additionally, he explained that high bone mass patients were discovered to have single amino acid substitutions in LRP5. Further investigation revealed WNT inhibition through sclerostin binding to a specific LRP5 motif specific for osteoblasts and osteocytes. Dr. Robling discussed numerous transgenic approaches to interrogate LRP5-related factors including family members LRP4 and 6. He concluded his talk by discussing off-target bone effects in oncology and requested that the audience not neglect bone mass evaluations clinically and nonclinically in programs targeting the WNT pathway.

Dr. Kendall Frazier’s presentation logically followed with an extensive discussion on the physeal effects observed with a variety of therapeutics including oncology drugs. He explained how physeal lesions are frequently the result of cartilage or vessel perturbations and may have dissimilar pathophysiology to other bone lesions. Drug-induced lesions at the physis are typically dose dependent and time dependent and have been observed with VEGF (Vascular endothelial growth factor), ALK5 (Transforming growth factor beta type I receptor kinase), TGF-β (Transforming growth factor beta), and FGFr (Fibroblast growth factor receptor) inhibitors among others. Dr. Frazier used an example of ALK5-associated increased width of the zone of hypertrophy to illustrate the importance of getting matched sections from treated and control animals. Beautiful examples of Von-Kossa’s and Movat’s histochemistry on fixed and frozen bone sections illustrated the ability to differentiate subtle changes in physeal elements. Dr. Frazier pointed out that although partial physeal closures will not recover, other physeal changes may, and commented that many drug-induced physeal changes are considered adverse.

Dr. Melissa Schutten presented an in-depth review of the bone effects of vismodegib, a small molecule inhibitor of smoothened. Smoothened is a protein in the hedgehog (HH) signaling pathway, which is critical in development. Inhibition of smoothened by vismodegib resulted in treatment-related findings in tissues where HH is expressed in adults, such as the hair, taste buds, ovary, teeth, and growth plates. There did not appear to be a dose–response relationship, and toxicities were observed at subpharmacologic doses. Dr. Schutten shared the mitigating strategy to prevent premature growth plate closure in patients taking vismodegib, which included a label update and a “Dear Health Care Provider” letter.

Dr. Katy Gropp provided an excellent foil to Dr. Frazier’s physeal lecture by presenting numerous practical examples of evaluation of cancellous bone. Cancellous bone may be analyzed in standard histology sections of proximal tibia, the distal femur, and sternum, although in chronic mouse studies lumbar vertebrae may be useful due to their trabeculation. Dr. Gropp provided PTH, sodium-glucose cotransporter 2 (SGLT2), and somatostatin treatment examples to illustrate osteoclast defects. Dr. Gropp further coached the audience on the criteria for bone necrosis including associated marrow and adipocyte necrosis and showed examples of the separation of the osteoclasts and osteoblasts from the marrow space by the canopy.

Dr. Lise Loberg finished session 3 with several informative case studies of monoclonal antibody oncology molecules that caused bone and marrow constituent changes suggestive of a cytokine pathophysiology. The cases were from rodent toxicology studies in which endosteal new bone formation and hematology changes were observed, with concurrent marrow histology findings of fibrosis, hypocellularity, increased granulocytes, and stromal matrix accumulation. Dr. Loberg shared that the commonalities in these studies were similar to those observed with cytokine and/or growth factor–induced proliferative bone lesions and provided a cross-species review of growth factors known to cause increased new bone formation and marrow fibrosis. The bone findings were not expected in the clinic, as the hematology changes were generally dose limiting.

In Wednesday morning’s session 4, the emerging reality of bone as an endocrine organ was covered by lectures, which each took a deep dive into 4 aspects: energy metabolism, interactions between bone and the nervous system, pleiotropic actions of fibroblast growth factor 23 (FGF23), and interactions between bone and the immune system.

Dr. Kathryn Motyl opened session 4 by reviewing bone energy metabolism: energy sources (protein, carbohydrates, and fats), metabolic processes (oxidative phosphorylation and glycolysis) of bone cells, and how bone cell energy balance affects bone mass and bone turnover. Effects on bone marrow adipocytes and mesenchymal stem cells by PTH, PPARs and caloric restriction were reviewed as well as some interactions between hormones and bone cells (such as osteoclasts on insulin sensitivity and the effects of risperidone on bone energy expenditure).

Dr. Baldock continued session 4 by exploring the interrelationship between bone and the nervous system by concentrating on the impact of sympathetic nerve fibers around trabecular bone, and the role of leptin and its downstream mediator neuropeptide Y (NPY), on cortical bone. Osteocalcin and NPY share some structural and sequence homology, and systemic osteocalcin has been shown to signal in the arcuate nucleus, leading to lower bone formation and bone volume with more osteoclast surfaces. A novel bone-derived insulin sensitizer, osteoglycin, demonstrates that bone is a regulating organ as well as a regulated organ.

Dr. Erben’s lecture on the pleiotropic actions of FGF23 confirmed that FGF23 is secreted by osteocytes and osteoblasts in response to calcitriol, impaired bone mineralization, and immune stimulation, but that the evidence is not as clear now for secretion in response to increased phosphate. Key actions of FGF23 (with cofactor α-klotho on fibroblast growth factor receptor 1c (FGFR1c)) in the kidney are decreased calcitriol and a direct effect on a NaPO₄ cotransporter on proximal tubule epithelium resulting in phosphaturia. Another effect can occur in distal tubules leading to increased calcium resorption. A heart bone axis for FGF23 has been discovered; FGF23 secretion is upregulated by myocardial damage and hypertrophy, but the molecular drivers of this upregulation need to be understood before FGF23 can be developed as a biomarker of myocardial infarction.

Dr. Weitzman closed session 4 with his talk which concentrated on the role of the immune system on bone turnover and bone loss, particularly on the effects of cytokines on osteoclastogenesis. For example, tumor necrosis factor α can amplify receptor activator of nuclear factor κB (RANK) ligand (RANKL) while decreasing osteoprotegerin (OPG) resulting in a high RANKL/OPG ratio, favoring osteoclastogenesis. Any process (such as immune deficiencies or estrogen deficiency) that increases the RANKL/OPG ratio can easily lead to bone loss and increased fracture risk.

Session 5

Wednesday afternoon’s session 5 covered structural approaches to bone and joint repair, including evaluating devices for bone and joint repair, biologic joint repair strategies, an update on animal models of joint damage and repair, and regulatory considerations for medical devices.

Dr. Wancket began session 5 with a comprehensive overview of the process of evaluating devices for bone and joint repair. The similarities and differences between evaluating medical devices and pharmalogic test articles were presented, and it was pointed out that the relative amount of preclinical testing prior to approval can be quite similar. With evaluation of devices, time, not dose, is a critical parameter, and trauma is an unavoidable confounding factor; differentiation between surgical trauma and device trauma is necessary. Material composition as well as shape and size all play a role in what kind of effects a device will have in a living system. In regard to bone device evaluation, one must find out whether host bone will grow on or into the device, is the device biocompatible (safe) in bone, and does the device work as intended. It is important to keep in mind translatability issues including surgical techniques, appropriately sized animal models, biologic status (such as age), and so on.

The next lecture was given by Dr. Kuroki on biologic joint repair strategies. Articular cartilage has poor regenerative capacity and limited healing potential. Currently, there are no disease-modifying agents for treatment of OA. Surgical attempts at repair range from drilling holes through the articular cartilage to access bone marrow cells (forming type I fibrocartilage not type II articular cartilage) to total joint replacement using metal and/or plastic implants. Tissue engineering using viable articular cartilage cells grown in an anatomic mold for 4 to 6 weeks has shown long-term success in animal models. Osteochondral allografts can be optimized to have high cellular viability and good biomechanical properties and has promising results in clinical trials.

Dr. LeBranche gave attendees an update on animal models of joint damage and repair, beginning with the mono-iodoacetate (MIA) injection model of joint pain. The MIA injection particularly targets weight bearing chondrocytes, causing cartilage necrosis. Osteophytes and marginally located chondrocytes are less affected; prostaglandins increase in the synovium and joint ligaments. Subchondral bone is decreased in thickness and can collapse. Surgical models of OA include the medial meniscal tear (MMT) model, the destabilized medial meniscal (DMM) model that both mimic posttraumatic OA. For the MMT model, damage at the point of the meniscal tear is the most severe compared with the remainder of the joint. Most of the degenerative joint changes occur between postsurgical days 7–21, although in the rat, disease progression can take 12 months. The DMM model progresses more slowly and is less severe than the MMT model. Adaptive responses to the destabilized joints (whether MMT or DMM) include increased subchondral bone (some of which may be woven bone), growth plate thickening, increased trabecular bone in the epiphysis and metaphysis, and osteophyte formation. There may be mesenchymal change (osteoblasts and a fibrovascular network) in subchondral bone along with depletion of bone marrow cellular elements. The thickened growth plate (if present) and increased trabecular bone may be in response to increased focal biomechanical loading.

Dr. Manhart-Byrnes closed the Wednesday scientific sessions with a talk on regulatory considerations for medical devices, beginning with an overview of the FDA entities that are responsible for reviewing the different classes and types of devices based on a tiered risk-based classification system. Class I and II devices are lower risk and are either cleared or not cleared by the agency. Class III devices are higher risk and require a premarketing application including clinical trial data. An International Organization for Standardization (ISO) biocompatibility standard for effects of local implantation was issued in 2016, and a draft guidance on general considerations for animal studies of medical devices is currently out for public comment.

Session 6

In the final session on Thursday morning, 4 lectures covering skeletal muscle (SKM) evaluation, an update on biomarkers of SKM injury, sarcopenia, and biomarkers for assessment of muscular dystrophy as well as a student presentation on proliferative cartilaginous lesions in a transgenic mouse line were presented.

The first lecture of the last morning was given by Dr. Gropp, who gave an overview of evaluation of SKM, including a very brief overview of SKM biomarkers, followed by a guide to handling and processing of SKM samples (particularly how to avoid contraction and freeze artifacts), and a review of muscle fiber types and topography (including muscle and species variation and choice of SKM for investigations). Light microscopic stains for frozen and formalin-fixed paraffin-embedded SKM were shown next, followed by common light microscopic findings in SKM (pathologic and incidental), an overview of image analysis and electron microscopy for SKM; a brief mention of other ancillary methods beyond the scope of the lecture and a summary that emphasized good tissue handling, awareness of species, muscle, and topological variations of fiber types; and appropriate use of INHAND terminology.

Mr. Goldstein followed with a comprehensive review of the challenges of biomarker qualification for SKM injury. The lecture began with an overview of drug-induced myopathy (myocyte degeneration/necrosis) including causes and current diagnostic methods, including available biomarkers to evaluate drug-induced SKM injury and their limitations. After this introduction, the process of biomarker qualification was thoroughly explained using the Predictive Safety Testing Consortium’s methodology with an SKM biomarker panel as an example. The lecture concluded with a look ahead at a new potential SKM biomarker, myostatin.

Next came the student presentation by Dr. Macleod on proliferative cartilaginous lesions in the tibiotarsal joints of humanized (CD34+ hematopoietic cells) NOD/Shi-scid/IL-2Rγnull (HuNOG) mice. After reviewing the genetic background of this transgenic mouse strain, an unusual finding of unilateral tibiotarsal joint swelling with hind limb paresis in 2 of the 9 mice at 2 weeks of age was described. In the first mouse, there was an expansile mass at the insertion of the calcaneal tendon with chondroid cell proliferation and acanthosis of overlying skin. In the second mouse, not only was there chondroid cell proliferation in the calcaneal tendon near the insertion, but there was also endochondral ossification with marrow space at the same tendinous insertion. The differential diagnoses for these findings were chondroid metaplasia or a low-grade neoplasia. The skin acanthosis was considered secondary to the chondroid proliferation.

Dr. Buford followed with his lecture entitled “Sarcopenia: The Lion in the Path.” There has been extensive debate on the clinical definition and diagnosis of sarcopenia with no firm consensus. SKM function deficit has been proposed as a unifying term. Decreased muscle strength is a central feature of this condition; morphologic characteristics include internal nuclei, ring fibers, ragged red fibers, disarrangement of myofilaments and Z lines with proliferation of smooth endoplasmic reticulum and t-tubule systems, and accumulation of lipofuscin and nemaline rod structures. By MRI, fat infiltration can be detected and increases with age. A large number of factors play into the onset and progression of sarcopenia, particularly inflammation, comorbidities, denervation, mitochondrial dysfunction, and so on.

In the final lecture of the symposium, Dr. Wilson delved into pathway-specific biomarker quantification in muscular dystrophy: comparing molecular and diagnostic pathology approaches. Duchenne’s muscular dystrophy (DMD) and Becker’s muscular dystrophy were reviewed as to relative incidence in the general human population, the type of genetic effect on dystrophin (loss vs. mutation), and some morphologic features (multifocal/polyphasic myofiber necrosis and regeneration with fibrosis. The mechanism of muscle damage from physical or functional loss of dystrophin is complex and involves increased intracellular calcium and increased muscle contractions. Many animal models of DMD exist, most notably in the dog and mouse. The dog models are more similar to human DMD than are the mouse models, but the mouse models can more closely match the human phenotype. A brief review of current treatments, methods of assessing efficacy and surrogate end points, including biomarkers, MRI, gene expression, methods of measuring dystrophin and digital image analysis techniques followed.