The use of glucocorticoids in rheumatoid arthritis--no 'rational' approach yet.

The relationship between glucocorticoids and bone mineral density in rheumatoid arthritis is complex. Further study into the optimal dosing, timing and duration of glucocorticoid use in rheumatoid arthritis is necessary.

In the previous issue of Arthritis Research and Th erapy, Ibanez and colleagues [1] report on the 'rational' use of gluco corticoids (GCs) in the management of early arthritis. Th is article concludes that GCs cause minimal varia tion in bone mineral density (BMD) at multiple skeletal sites, and in fact may increase BMD at the ultra distal forearm, a juxta-articular site. Although this article is notable for examining the eff ect of GCs on BMD at fi ve anatomic sites, a 'rational' use of GCs for rheumatoid arthritis (RA) is still elusive. To grasp the complex relation ship between GCs and both localized (juxta-articular, bony erosions) and systemic (osteoporosis) bone loss in RA, we need to fi rst step back and appreciate the interplay of the immune system and bone metabolism.
Th e osteoclast plays a central role at the site of infl amed joints and is critical in the pathogenesis of joint erosions in RA [2]. Receptor activator of nuclear factor-kappa B ligand (RANKL), expressed by TH1 and TH17 T cell subsets, is a potent inducer of osteoclast diff erentiation. Additionally, an array of pro-infl ammatory cytokines such as TNF, IL-1, IL-6 and IL-17 can stimulate RANKL expression [3] (Figure 1). GCs, in turn, directly aff ect both osteoblast and osteoclast activity, and indirectly exert many eff ects on bone metabolism, leading to an increased fracture risk [4] (Figure 2).
How should the relationship between the bone biology in RA and the eff ects of GCs translate into the use of GCs in clinical practice? Th e COBRA trial provides a rationale for the use of GCs in combination with other disease modifying anti-rheumatic drugs (DMARDs) to significantly reduce RA disease activity [5]. A subsequent review demonstrates that GCs (mean cumulative dose of 2,300 mg prednisone equivalent over the fi rst year), when used in combination with traditional DMARD therapy, can decrease the rate of radiographic progression in RA [6]. Th e eff ect of GCs on bone mass, among non-RA patients, has been evaluated in a small randomized, placebo-controlled trial demonstrating that serum markers of bone formation are rapidly decreased among healthy post-menopausal women treated with just 5 mg of prednisone daily for 6 weeks [7]. In early RA patients treated with prednisolone 7.5 mg per day and traditional DMARD therapy (compared to traditional DMARD therapy alone), markers of bone formation, markers of bone resorption, and lumber spine BMD, but not femoral BMD, were decreased [8]. However, in a randomized, placebo-controlled trial of 95 early RA patients, GCs decreased the degree of localized hand bone loss [9]. Th ese studies suggest that low dose GCs may reduce markers of bone formation leading to generalized osteoporosis, but they also counteract RA-associated infl amma tion and slow the rate of bone loss proximal to sites of active disease. Other factors, such as GC resistance and genetic polymorphisms that predispose to either GC sensitivity or resistance, may play a role in the BMD variation seen at various anatomic sites [10,11].
Despite the disease-modifying properties of GCs seen in RA patients, the risk of GC-induced osteoporosis and its associated morbidity often give the rheumatologist pause when determining whether to use GCs, in what dosing, for how long, at what time in the disease process, and in which types of RA patients (seropositive versus sero negative). Ibanez and colleagues have begun to advance our knowledge of GC use in RA. In this cohort of early RA patients, they found a signifi cant decrease in BMD at all sites except the ultradistal and distal forearm. In the multivariate analysis there was no signifi cant relation ship between cumulative GC use and BMD variation at multiple sites, except at the ultradistal (increased BMD) and mid-forearm (decreased BMD) [1].  At fi rst glance, the seemingly simple lack of BMD variation among a cohort of early RA patients treated with GCs is a compelling argument for GC use -but in this small 2-year study of early RA, the dosing, duration and long-term eff ects remain unknown. Th e apparent increase in ultradistal forearm BMD among patients treated with GCs warrants further exploration. However, this study of early RA examines only 116 patients treated with a median cumulative GC dose of 22 mg/month and 45 mg/month among those who actually received GCs. Is it the relatively small cumulative dose of GCs used in this study that explains the lack of signifi cant BMD variation or the fact that only 67% of the cohort actually received GCs, with 17.3% of patients on GC therapy at the end of the study?

Conclusion
While Ibanez and colleagues have furthered our understanding of the eff ects of GCs on BMD variation through their detailed analysis of fi ve skeletal sites, we are far from a 'rational' use of GCs in the management of RA.