Glucocorticoids (GCs) are potent drugs to treat rheumatoid arthritis but exert adverse skeletal effects. Compound A (CpdA) is a selective GC receptor modulator with an improved risk/benefit profile in mouse models of inflammation and bone loss. Here we tested whether CpdA also exerts bone-sparing effects under proinflammatory circumstances using the collagen-induced arthritis model, a murine model of rheumatoid arthritis. CpdA decreased disease activity, paw swelling, and the paw temperature by 43%, 12%, and 7%, respectively, but was less potent than dexamethasone (DEX), which reduced these parameters by 72%, 22%, and 10%, respectively. Moreover, T cells isolated from CpdA- and DEX-treated animals were less active based on proliferation rates after challenge with type II collagen and produced smaller amounts of interferon-γ and TNF as compared with T cells from PBS-treated mice. Histological assessment of the joints confirmed the weaker potency of CpdA as compared with DEX in preventing infiltration of inflammatory cells, induction of osteoclastogenesis, and destruction of articular cartilage. Due to the lack of GC-susceptible arthritis models, we were not able to fully address the bone-sparing potential of CpdA in inflammatory conditions. Nevertheless, the bone formation marker procollagen type 1 N-terminal peptide, a surrogate marker for GC-mediated suppression of bone formation, was significantly decreased by DEX in arthritic mice but not by CpdA. Our data indicate that CpdA moderately suppresses inflammation, whereas the concurrent effects on bone remain unknown. In light of its narrow therapeutic range, CpdA may be more useful as a molecular tool for dissecting GC actions rather than a therapeutic agent.