Purpose: Aggrecanase activity, most notably ADAMTS-5, has been widely implicated as a pathogenic factor in cartilage degradation. Therefore, highly selective and potent monoclonal antibodies (mAbs) to both ADAMTS-5 and ADAMTS-4 were developed and experimental systems were utilized to assess target engagement and modulation of disease-related and safety endpoints with the intent of selecting a candidate and supporting clinical development in osteoarthritis (OA). Methods/Results: In a surgical mouse model of OA, both ADAMTS-5 and ADAMTS-4-specific mAbs bound within cartilage following systemic administration, demonstrating access to the anticipated site of action, whereas structural disease modification and associated alleviation of pain-related behavior were only observed with ADAMTS-5 mAb treatment. Likewise, ex vivo treatment of human OA cartilage demonstrated a preferential role for ADAMTS-5 inhibition over ADAMTS-4, as measured by aggrecan-derived 374ARGS neoepitope release in explant cultures and was most evident in a distinct subset of patients where elevated 374ARGS neoepitope levels were present in explants prior to treatment. Notably, suppression of 374ARGS neoepitope release was sustained for weeks after a single treatment of human cartilage explants and in cynomolgus monkeys, consistent with a high affinity antibody/antigen interaction and slow ADAMTS-5 turnover that could translate to less frequent clinical dosing. As a result an ADAMTS-5 selective monoclonal antibody (GSK2394002) was progressed for preclinical development as a potential OA disease modifying therapeutic. While desired pharmacology was observed in explant cartilage and disease models, ADAMTS-5 is also expressed in cardiovascular (CV) tissue. Recently it was reported that knockout mice show developmental heart valve defects and altered vascular and fibroblast proteoglycan processing. Along with these observations, an initial histologic cardiac signal in cynomolgus monkeys (subendocardial hemorrhage) prompted a careful assessment of CV function in a set of dedicated safety pharmacology studies. In these studies, increased mean arterial pressure (MAP) was observed within days after >3 mg/kg dosing and ST segment elevation was noted on 24 hour continuous ECG monitoring >30 mg/kg. Surprisingly, these effects were sustained for up to 8 months following a single dose and MAP was increased upon repeated low dose administration suggesting the effects are cumulative and, unlike the systemic pharmacodynamic marker signal, once induced they may not be readily reversible. Although a potential mechanistic link between ADAMTS-5 activity and versican processing is demonstrated, the cause of these cardiovascular observations is currently unknown. In a parallel general toxicology study, there were no histologic or biochemical correlates to the CV findings and no evidence of off-target mechanisms could be identified using in vitro tests examining cardiac conductance, ion channels, or binding selectivity in protein arrays. Conclusions: The findings support a hypothesis set forth from knockout mouse studies that ADAMTS-5 is the major aggrecanase involved in cartilage degradation and confirms a link between a native biological pathway and pharmacology which translates to human tissues and non-human primate models, subsequently pointing to a specific OA patient population to target. However, our results suggest selective and potent ADAMTS-5 engagement/inhibition with GSK2394002, while demonstrating desired effects on cartilage-related endpoints, results in modulation of CV functions that pose considerable challenges for clinical development.

1. The effects of intravenous injections of adenosine agonists and antagonists on pulmonary ventilation were investigated in conscious male Sprague-Dawley rats. 2. Adenosine agonists reduced pulmonary ventilation, whereas an adenosine antagonist increased pulmonary ventilation. 3. The adenosine-induced decrease in ventilation could be partially blocked by pretreatment with either an adenosine or opioid antagonist and completely blocked by a combined pretreatment with adenosine and opioid antagonists. 4. Thus, endogenous adenosine appears to have an inhibitory role in the control of pulmonary ventilation in conscious rats, and the mechanism of action appears to involve both adenosine and opioid receptors.