The “elusive DMOAD”: Aggrecanase inhibition from laboratory to clinic A.-M. Malfait1, M.D. Tortorella2

1Rush University Medical School, ABSTRACT mechanical “wear and tear” disease (3), Department of Medicine, Division From the time of their discovery in 1999, but rather a condition in which well-de- of Rheumatology, Chicago IL, USA; the aggrecanases, and ADAMTS-5 in fined biochemically mediated pathways 2Guangzhou Institutes of Biomedical particular, have been heavily investi- bring about articular cartilage damage Health, Chinese Academy of Sciences, Guangzhou, China. gated as targets for disease-modifying (4). These changing concepts brought osteoarthritis drug (DMOAD) devel- the expectation that, one day, scientists Anne-Marie Malfait Micky D. Tortorella opment. Here, we provide a brief nar- would be able to develop inhibitors that rative review of the discovery efforts Please address correspondence to: prevent, slow down, halt, or even re- Dr Anne-Marie Malfait, to target these , and how this verse cartilage damage. Since then, the Rush University Medical School, led to the current ongoing programmes developing concept that OA is a failure Department of Medicine, that hold promise for the future. We of the joint as an organ, where different Division of Rheumatology, discuss a comparison of inhibition of tissues and the crosstalk between them, 1611 W. Harrison Street, Suite 510, collagen breakdown versus inhibition including articular cartilage, subchon- Chicago, IL 60612, USA. of breakdown. We then sum- dral bone, and the synovium, drives E-mail: [email protected] marise existing programmes that target [email protected] the progression of the disease and as- ADAMTS-5, including small molecule sociated symptoms (5). This has led Received and accepted on September 17, 2019. inhibitors, monoclonal neutralising to the identification of multiple targets antibodies and nanobodies, and gene in different joint tissues that contrib- Clin Exp Rheumatol 2019; 37 (Suppl. 120): S130-S134. editing technologies. We also briefly ute to disease. Drugs that are currently discuss the potential analgesic effects under investigation for their potential © Copyright Clinical and Experimental Rheumatology 2019. this strategy may offer in addition to its DMOAD effects include both anabolic joint-protective effects. (i.e. promoting cartilage repair) and Key words: ADAMTS-5, , anti-catabolic strategies, with some disease-modifying osteoarthritis drug Introduction drugs intended for systemic administra- (DMOAD), antibodies, cartilage Successful disease modification in OA tion while others, such as the anabolic remains an elusive goal, in spite of FGF-18 (Sprifermin, EMDSerono) and significant progress in our understand- the wnt-pathway inhibitor, SM04690 ing of osteoarthritis (OA) pathogen- (Lorecivivint, Samumed), are being esis and sophistication of methods to developed for intra-articular adminis- assess disease state and progression. tration. As discussed in detail in the article by Elsewhere in this Supplement, Oo and Oo and Hunter in this Supplement, a Hunter discuss DMOADs that are in disease-modifying osteoarthritis drug phase 2/3 (1). Exciting novel strategies (DMOAD) is a drug that modifies the are currently in earlier phases (phase underlying OA pathophysiology, there- 1/2), including a senolytic agent for in- by inhibiting structural damage to pre- tra-articular administration, UBX0101 vent or reduce long-term disability and (Unity Biotechnology), which aims Funding: A.-M. Malfait (R01AR064251, offer potential symptomatic relief (1). to eliminate senescent chondrocytes R01AR060364, R61AR073576) is Currently, there are no FDA or EMA from articular cartilage. Senescent supported by the US National Institutes approved DMOADs (2), but aggressive cells (SnCs), which have lost prolifera- of Health/National Institute of Arthritis ongoing efforts by many organisations tive potential, accumulate in all tissues and Musculoskeletal and Skin Diseases and teams in academia and industry are with age and promote the ageing pro- (NIH/NIAMS). offering the hope that DMOADs are on cess through the secretion of the “Se- The funding sources had no role in the study. the horizon. As depicted in Fig. 1 [and nescence Associated Secretory Pheno- discussed in (1)], the current DMOAD type” (SASP), a host of inflammatory Competing interests: A.M. Malfait has received consulting fees from EMDSerono pipeline is densely populated with ac- cytokines, chemokines, and and Vizuri, as well as a grant from tive clinical trials. that profoundly alter the tissue micro- Galapagos N.V.; It was not until the 1980s that the con- environment (6, 7). Senescent chon- M. Tortorella is an employee of GIBH. cept took hold that OA is not simply a drocytes are found in cartilage isolated

S-130 Clinical and Experimental Rheumatology 2019 Aggrecanase inhibition in OA / A.M. Malfait & M.D. Tortorella

Fig. 1. Pipeline of potential DMOADs in Phase I. from patients undergoing joint replace- Breakdown of articular cartilage is a carefully. While it is accepted that ment (8, 9). In a surgical mouse model hallmark of OA. Two key targetable preventing loss of collagen from the of OA, SnCs were found to accumu- pathways contribute to the enzymatic cartilage ECM will preserve its elastic late in cartilage and synovium (10). degradation of the cartilage extracellu- properties and integrity, systemic inhi- UBX0101 is a potent senolytic small lar matrix (ECM). Aggrecan and type II bition of collagenase activity may pose molecule inhibitor of the MDM2/p53 collagen make up the two major macro- specific problems. Type II collagen is a interaction, and disruption of molecular components of articular car- hydrophobic molecule and has no self- this interaction triggers the elimination tilage and are essential for maintaining elimination mechanism from the ECM. of senescent cells. In a surgical mouse cartilage function and integrity, with As cells produce more collagen as part model of OA, intra-articular treatment aggrecan providing cartilage with its of its normal anabolic maintenance, with UBX0101 attenuated joint dam- compressibility and collagen providing collagen will accumulate in the matrix age and, importantly, the drug was able its elasticity [reviewed in (11)]. Deg- until it is proteolytically degraded and to reduce development of naturally oc- radation of these macromolecules is removed by collagenase activity. If this curring disease in ageing mice (10). A mediated by proteolytic cleavage, thus activity is blocked, collagen has the phase 1 study to evaluate safety, tolera- representing druggable mechanisms potential to build up in the ECM caus- bility, and pharmacokinetics of a single of intervention requiring the design ing fibroplasia. Indeed, musculoskel- intra-articular injection of UBX0101 in of inhibitors with the proper etal syndrome (MSS, which is defined patients diagnosed with painful OA of pharmacokinetic properties. Aggrecan as painless loss of range of motion in the knee was recently completed. Re- breakdown is mediated by the aggre- large joints, particularly in the shoul- sults have yet to be posted on https:// canases, predominantly ADAMTS-4 ders, joint stiffness and joint swelling, clinicaltrials.gov. and ADAMTS-5, while collagen un- soft tissue pain, and fibrosis of palmar winding and degradation is mediated by tendons) has been observed in clinical Targeting the cartilage matrix the , predominantly MMP- studies with broad-spectrum MMP in- The current pre-clinical and clinical 1, 8, 13 and 14, although MMP-8 has hibitors and is a pharmacological effect pipeline comprises different strategies not been reproducibly found in articular likely due to the non-selectivity of col- to block the activity of the collagenases cartilage like other members (unpub- lagenase inhibitors (12, 13). Therefore, and aggrecanases. In the current nar- lished results). it is important that drug discovery rative review, we provide a brief over- scientists design drugs that target the view of the discovery efforts to target Targeting collagen degradation collagenase(s) responsible for catabo- these enzymes, and how this led to the There are pros and cons to inhibiting lism observed in OA, but not the col- current ongoing programmes that hold either collagen versus aggrecan deg- lagenase activity responsible for matrix promise for the future. radation, which need to be considered homeostasis in cartilage and other tis-

Clinical and Experimental Rheumatology 2019 S-131 Aggrecanase inhibition in OA / A.M. Malfait & M.D. Tortorella sues. It is currently thought that MMP- 24). It was also found that Adamts5 null Neutralising antibodies targeting 1 and MMP-14 are homeostatic col- mice are protected from joint damage ADAMTS-5 lagenases in cartilage and some other in the antigen-induced arthritis model GSK developed a humanised AD- tissues, whereas MMP-13 appears to (25), as well as in a surgical model of AMTS-5-selective monoclonal anti- be responsible for the catabolism ob- OA, induced by destabilisation of the body (GSK2394002), which was effi- served in OA (14-16). Designing se- medial meniscus (DMM) (26). Aggre- cacious in the DMM model both in a lective MMP-13 inhibitors has proven canase-resistant aggrecan mutant mice prophylactic and a therapeutic protocol. difficult, however. The first and second are also protected from developing Systemic treatment attenuated joint generation inhibitors that were devel- experimental OA after DMM surgery damage as well as mechanical allo- oped chelated the zinc atom in the cat- (27). Hence, from the time of their dynia, an indicator of pain (32, 33). Un- alytic pocket. Although zinc chelation discovery, the aggrecanases, and AD- fortunately, the programme was halted provides high affinity, it also promotes AMTS-5 in particular, have been heav- because toxicity studies showed that promiscuity and lack of selectivity ily investigated as targets for DMOAD GSK2394002 resulted in modulation (17). More recently, several research development [discussed in (28)]. Cur- of cardiovascular functions that posed groups have discovered what is known rently, three categories of therapeutics considerable challenges for clinical de- as S1’ inhibitors that dock snuggly into targeting these enzymes have been velopment (34). No further information the S1’ loop of the collagenase without explored: small molecule inhibitors, on the mechanisms of this observed ef- binding to the zinc atom, resulting in monoclonal neutralising antibodies, fect is available. highly selective drugs (18). Time will and gene editing technologies. Another anti-ADAMTS-5 antibody was tell if these can be brought forward into developed by Rottapharm, CRB0017, clinic trials. Small molecule inhibitors which targets the ancillary domain of Agg523 is an orally available small ADAMTS-5. It was reported to slow Targeting aggrecan degradation molecule selective inhibitor of AD- down disease progression in a mouse Unlike collagen, aggrecan has a self- AMT-4 and ADAMTS-5 developed by model of spontaneous OA (STR/Ort) clearance mechanism that is independ- Pfizer (Wyeth) that successfully- com upon intra-articular administration ent of proteolytic degradation. For pleted phase I safety trials in the USA. (35). No further information is avail- example, when chondrocytes produce This was in 2008, and the programme able at this time. aggrecan as part of their anabolic main- was later discontinued. Currently, An ADAMTS-5-inhibiting bifunctional tenance, the aggrecan molecules that do Galapagos N.V. has a programme Nanobody, M6495, is being developed not bind to hyaluronan via link protein centering on an orally available selec- by EMDSerono. Nanobodies are a nov- are normally lost from the ECM through tive small molecule antagonist of AD- el class of proprietary therapeutic pro- diffusion driven by charge repulsion. AMTS-5, GLPG1972. It was reported teins based on single-domain antibody This charge-based repulsion mecha- to dose-dependently inhibit aggrecan fragments. M6495 is a bifunctional Na- nism is an effective clearance system turnover in human cartilage explants nobody (Ablynx) of 28.1 kDa that binds for aggrecan, which relies only on the (IC50<1μM) and it has a protective ADAMTS-5, but not ADAMTS-1, -4, highly negatively charged properties of effect when administered prophylacti- or -15, and inhibits its enzymatic activi- the molecules and is not dependent on cally in 2 animal models, an 8-week ty. In this molecule, the target arm bind- proteolytic cleavage. Therefore, target- murine DMM model and a 3-week rat ing ADAMTS-5 is conjugated to a half- ing the aggrecan degradation pathway meniscectomy model (29, 30). In a re- life extension arm for serum albumin. It through inhibition of the aggrecanases cently completed phase 1 study, doses has been reported that M6495 dose-de- may be a safer alternative compared to of 300 mg, 600 mg, and 1050 mg/day pendently inhibited aggrecan turnover inhibition of collagen degradation. In for two weeks were reported to be gen- in human cartilage explants (36). In an theory, systemic inhibition of the ag- erally safe and well tolerated in healthy 8-week murine DMM model, it slowed grecanases should only prevent loss male subjects, and a favourable PK/PD progression of joint damage when ad- of healthy cartilage aggrecan during profile was observed (31). Interesting- ministered prophylactically (37). Phase disease, without the accumulation of ly, at all 3 doses, the drug demonstrated 1 clinical trials with this nanobody, newly synthesised aggrecan in cartilage the ability to reduce aggrecan ARGS administered subcutaneously, were re- and/or other tissues. neoepitope levels in plasma, indica- cently completed and results have not In 1999, ADAMTS-4 and ADAMTS-5 tive of suppressing ongoing cartilage yet been posted. were cloned and characterised (19-21). degradation. Currently, the compound Subsequent in vitro studies on bovine is being tested in phase 2. Collective- Gene editing technologies and human cartilage explants showed ly, all information that has been made At the Guangzhou Institutes of Bio- that these enzymes are the major pro- publicly available regarding small medicine & Health, a drug candidate teases responsible for aggrecan deg- molecule aggrecanase inhibitors sug- was developed, B001-5, which is a radation, and blocking their activity gests that systemic inhibition of these mixture of two chemically modified inhibited both aggrecan and collagen enzymes is well tolerated and safe, at small interfering RNAs targeting AD- degradation in response to IL-1 (22- least at the levels dosed. AMTS-5 and ADAM-17 that have a

S-132 Clinical and Experimental Rheumatology 2019 Aggrecanase inhibition in OA / A.M. Malfait & M.D. Tortorella long retention time in joints follow- peptide (but not a scrambled peptide) oarsi.org/sites/default/files/docs/2016/oarsi_ ing intra-articular injection. The oli- elicited knee hyperalgesia in wild-type white_paper_oa_serious_disease_121416_1. pdf. (2017). gonucleotides were also conjugated to but not Tlr2 null mice. Remarkably, 3. HUSKISSON EC, DIEPPE PA, TUCKER AK, small molecule cholesterol analogs for “Chloe mice”, a transgenic line in which CANNELL LB: Another look at osteoarthritis. enhanced chondrocyte and synovial fi- the MMP cleavage site (N341↓342F) in Ann Rheum Dis 1979; 38: 423-28. broblast penetration. The IND applica- the aggrecan interglobular domain is 4. DINGLE JT: Articular damage in arthritis and its control. Ann Intern Med 1978; 88: 821-26. tion with the Chinese FDA for approval mutated, thus preventing production of 5. LOESER RF, GOLDRING SR, SCANZELLO CR, to conduct phase I safety trials will be the 32-amino acid fragment (27), were GOLDRING MB: Osteoarthritis: a disease of completed in 2020 (38). protected from knee hyperalgesia after the joint as an organ. Arthritis Rheum 2012; DMM surgery despite exhibiting more 64: 1697-707. 6. COPPÉ JP, PATIL CK, RODIER F et al.: Sene- Will ADAMTS-5 inhibition severe cartilage damage (42). Thus, this scence-associated secretory phenotypes provide symptomatic benefit? single TLR-2 ligand, which is a prod- reveal cell-nonautonomous functions of on- The disconnect between the extent of uct of aggrecanase-mediated aggrecan cogenic RAS and the p53 tumor suppressor. PLoS Biol 2008; 6: 2853-68. joint damage and severity of pain has cleavage, may play a central role in 7. CAMPISI J, ROBERT L: Cell senescence: role often been cited as a big hurdle for suc- driving knee pain but not joint damage in aging and age-related diseases. Interdiscip cessful development of DMOADs [see in murine OA. It will be of interest to Top Gerontol 2014; 39: 45-61. also (1)]. After all, will patients care test if circulating levels of this 32-mer 8. MARTIN JA, BUCKWALTER JA: Human chon- drocyte senescence and osteoarthritis. Bior- that their cartilage is no longer being fragment can be developed as a marker heology 2002; 39: 145-52. degraded if they still feel pain? Recent of OA pain. 9. PHILIPOT D, GUÉRIT D, PLATANO D et al.: reviews have discussed how clinical re- p16INK4a and its regulator miR-24 link search is increasingly revealing specific Summary senescence and chondrocyte terminal differ- entiation-associated matrix remodeling in structural changes that are correlated to In conclusion, despite the substantial osteoarthritis. Arthritis Res Ther 2014; 16: pain and sensitisation in patients with unmet medical need in OA, the devel- R58. OA (for a good review on this subject opment of DMOADs has proven elu- 10. JEON OH; KIM C, LABERGE RM et al.: please see ref. 39). sive. Despite decades of research and Local clearance of senescent cells attenuates the development of post-traumatic osteoar- In the case of ADAMTS-5 blockade, development efforts by many research thritis and creates a pro-regenerative envi- it should be noted that preclinical evi- groups, no drug has made it into the ronment. Nat Med 2017; 23: 775-81. dence suggests that blocking the activ- hands of patients. Reasons for this in- 11. CHUBINSKAYA S, MALFAIT, AM, WIMMER ity of ADAMTS-5 will be accompanied clude the complexity of OA disease M: Form and Function of Articular Cartilage. In: Orthopedic Basic Science. EINHORN T, by an analgesic effect. In Adamts5 null and its multi-gene nature, less support MD, O’KEEFE R, MD, CHU C, MD, JACOBS J, mice, attenuated joint damage after for musculoskeletal disease research MD (Eds.) 2013, American Academy of Or- DMM surgery was associated with pro- by funders and big pharma compared thopedic Surgeons. 12. BROWN PD: Ongoing trials with matrix met- tection from mechanical allodynia, an to other therapeutic areas, fewer re- alloproteinase inhibitors. Expert Opin Inves- indicator of pain and sensitisation of the searchers working in the OA field as tig Drugs 2000; 9: 2167-77. sensory nervous system (40). In agree- compared to other more “trendy” areas 13. KRZESKI P, BUCKLAND-WRIGHT C, BALINT ment with this finding, it was reported of research, as well as other factors (for G et al.: Development of musculoskeletal tox- icity without clear benefit after administration that inhibiting ADAMTS-5 with a neu- example, the erroneous perception that of PG-116800, a matrix in- tralising antibody blocked mechanical OA is not a severe disease, as discussed hibitor, to patients with knee osteoarthritis: a allodynia after DMM surgery, both in a in this Supplement) (43). However, randomized, 12-month, double-blind, place- prophylactic and in a therapeutic treat- with candidate drugs slowly making bo-controlled study. Arthritis Res Ther 2007; 9: R109. ment protocol (32, 33). Likewise, the their way into clinical trials, there is 14. HOLMBECK K, BIANCO P, CATERINA J et al.: M6495 Nanobody improved gait per- hope that a “true” DMOAD will finally MT1-MMP-deficient mice develop dwarfism, formance in a surgical rat model in a make its way to patients in the foresee- osteopenia, arthritis, and connective tissue dose-dependent manner (37). able future. As we have discussed here, disease due to inadequate collagen turnover. Cell 1999; 99: 81-92. It is worth highlighting that AD- inhibition of ADAMTS-5 remains an 15. NEUHOLD LA, KILLAR L, ZHAO W et al.: AMTS-4/5-mediated degradation of ag- attractive strategy that may offer pain Postnatal expression in hyaline cartilage of grecan may have a direct effect on no- relief in addition to structural protec- constitutively active human collagenase-3 (MMP-13) induces osteoarthritis in mice. ciceptors and thus directly cause pain. tion, and the activity of which can be J Clin Invest 2001; 107: 35-44. Indeed, we recently reported that a 32 monitored quite easily. 16. STICKENS D, BEHONICK DJ, ORTEGA N et al.: amino-acid fragment that is generated Altered endochondral bone development in by the orchestrated enzymatic cleavage References 13-deficient mice. 1. OO WM, HUNTER DJ: Disease modification Development 2004; 131: 5883-95. of the aggrecan core protein by AD- in osteoarthritis: Are we there yet? Clin Exp 17. KALVA S, SARANYAH K, SUGANYA PR, NI- 342 AMTS-5 and MMPs (“32-mer”, F– Rheumatol 2019; 37 (Suppl. 120): S135-40. SHA M, SALEENA LM: Potent inhibitors pre- E373) (41) can activate cultured nocic- 2. Osteoarthritis: A Serious Disease, in White cise to S1’ loop of MMP-13, a crucial target eptors, specifically through Toll-like Paper Submitted to the U.S. Food and Drug for osteoarthritis. J Mol Graph Model 2013; Administration 2016, Pre Competitive Con- 44: 297-310. receptor (TLR)-2 (42). Furthermore, sortium for Osteoarthritis Osteoarthritis Re- 18. BARAGI VM, BECHER G, BENDELE AM et intra-articular injection of this 32-mer search Society International: https://www. al.: A new class of potent matrix metallopro-

Clinical and Experimental Rheumatology 2019 S-133 Aggrecanase inhibition in OA / A.M. Malfait & M.D. Tortorella

teinase 13 inhibitors for potential treatment 27. LITTLE CB, MEEKER CT, GOLUB SB et al.: 35. CHIUSAROLI R, VISENTINI M, GALIMBERTI of osteoarthritis: Evidence of histologic and Blocking aggrecanase cleavage in the aggre- C et al.: Targeting of ADAMTS5’s ancillary clinical efficacy without musculoskeletal tox- can interglobular domain abrogates cartilage domain with the recombinant mAb CRB0017 icity in rat models. Arthritis Rheum 2009; 60: erosion and promotes cartilage repair. J Clin ameliorates disease progression in a sponta- 2008-18. Invest 2007; 117: 1627-36. neous murine model of osteoarthritis. Osteo- 19. ARNER EC, PRATTA MA, TRZASKOS JM, 28. FOSANG AJ: ADAMTS-5 takes centre stage in arthritis Cartilage 2013; 21: 1807-10. DECICCO CP, TORTORELLA MD: Generation new developments for aggrecanase inhibitors. 36. WERKMANN D et al.: In vitro characteriza- and characterization of aggrecanase. A solu- Osteoarthritis Cartilage 2015; 23: 1231-32. tion of the ADAMTS-5 specific nanobody® ble, cartilage-derived aggrecan-degrading ac- 29. CLEMENT-LACROIX P et al.: GLPG1972: a M6495. Osteoarthritis Cartilage 2018; 26. tivity. J Biol Chem 1999; 274: 6594-601. potent, selective, orally available ADAMTS-5 37. BRENNEIS C et al.: Structural and sympto- 20. TORTORELLA MD, BURN TC, PRATTA MA et inhibitor for the treatment of OA. Osteoarthri- matic benefit of a half-live extended, sys- al.: Purification and cloning of aggrecanase-1: tis Cartilage 2017; 25: S58-59. temically applied anti-ADAMTS-5 inhibitor a member of the ADAMTS family of . 30. CLEMENT-LACROIX P et al.: ADAMTS-5 in- (M6495). Osteoarthritis Cartilage 2018; 26: Science 1999; 284: 1664-66. hibition with the potent and highly selective S299-300. 21. ABBASZADE I, LIU RQ, YANG F et al.: Cloning inhibitor GLPG1972 results in strong disease- 38. Patent#201410827675.3, ADAMTS-5 and and characterization of ADAMTS11, an ag- modifying OA drug effects in the rat menis- ADAM-17 siRNA. grecanase from the ADAMTS family. J Biol cectomy model. Osteoarthritis Cartilage 39. O’NEILL TW, FELSON DT: Mechanisms of Chem 1999; 274: 23443-50. 2018; 26: S26. osteoarthritis (OA) pain. Curr Osteoporos 22. MALFAIT AM, LIU RQ, IJIRI K, KOMIYA S, 31. van der AAR EM et al.: ADAMTS-5 inhibitor Rep 20\18; 16: 611-16. TORTORELLA MD: Inhibition of ADAM-TS4 GLPG1972, a potential new treatment in os- 40. MALFAIT AM, RITCHIE J, GIL AS et al.: and ADAM-TS5 prevents aggrecan degrada- teoarthritis, shows favorable safety, pharma- ADAMTS-5 deficient mice do not develop tion in osteoarthritic cartilage. J Biol Chem cokinetics and pharmacodynamics in healthy mechanical allodynia associated with osteo- 2002; 277: 22201-8. subjects. Osteoarthritis Cartilage 2018; 26: arthritis following medial meniscal destabi- 23. PRATTA MA, YAO W, DECICCIO C et al.: S310. lization. Osteoarthritis Cartilage 2010; 18: Aggrecan protects cartilage collagen from 32. LARKIN J, LOHR TA, ELEFANTE L et al.: 572-80. proteolytic cleavage. J Biol Chem 2003; 278: Translational development of an ADAMTS-5 41. FOSANG AJ, NEAME PJ, HARDINGHAM TE, 45539-45. antibody for osteoarthritis disease modifi- MURPHY G, HAMILTON JA: Cleavage of 24. SONG RH, TORTORELLA MD, MALFAIT AM et cation. Osteoarthritis Cartilage 2015; 23: cartilage between G1 and G2 al.: Aggrecan degradation in human articular 1254-66. domains by stromelysins. J Biol Chem 1991; cartilage explants is mediated by both AD- 33. MILLER RE, TRAN PB, ISHIHARA S, LARKIN 266: 15579-82. AMTS-4 and ADAMTS-5. Arthritis Rheum J, MALFAIT AM: Therapeutic effects of an an- 42. MILLER RE, ISHIHARA S, TRAN PB et al.: 2007; 56: 575-85. ti-ADAMTS-5 antibody on joint damage and An aggrecan fragment drives osteoarthritis 25. STANTON H, ROGERSON FM, EAST CJ et mechanical allodynia in a murine model of pain through Toll-like receptor 2. JCI Insight al.: ADAMTS5 is the major aggrecanase in osteoarthritis. Osteoarthritis Cartilage 2016; 2018; 3. mouse cartilage in vivo and in vitro. Nature 24: 299-306, 43. PINCUS T, CASTREJON I, BERGMAN MJ, 2005; 434: 648-52. 34. LARKIN J et al.: The highs and lows of trans- YAZICI Y, GIBSON KA: Osteoarthritis is as se- 26. GLASSON SS, ASKEW R, SHEPPARD B et al.: lational drug development: antibody-mediat- vere as rheumatoid arthritis: evidence over 40 Deletion of active ADAMTS5 prevents carti- ed inhibition of ADAMTS-5 for osteoarthritis years according to the same measure in each lage degradation in a murine model of osteo- disease modification. Osteoarthritis Carti- disease. Clin Exp Rheumatol 2019; 37 (Suppl. arthritis. Nature 2005; 434: 644-48. lage 2014; 22: S483. 120): S7-S17.

S-134 Clinical and Experimental Rheumatology 2019