Research in context
Evidence before this study
Tapering of low-dose glucocorticoids is a common clinical issue across many disciplines. We searched the PubMed database from its inception to Oct 30, 2019, using the terms (“rheumatoid arthritis“ AND [“prednisone” OR “glucocorticoid” OR “corticosteroid”] AND [“taper” OR “withdrawal” OR “step-down”]). We reviewed the results and identified 12 clinical trial articles that included glucocorticoid tapering. Two trials included an appropriate control for the effect of glucocorticoid tapering in stable disease activity conditions. Only one of these trials was a blinded study; this single-centre randomised, placebo-controlled withdrawal trial of 31 patients with established rheumatoid arthritis receiving long-term treatment, including conventional synthetic disease-modifying antirheumatic drugs (csDMARDs), suggested that continued low-dose (1–4 mg per day) prednisone leads to fewer patient-reported poor efficacy events than tapering (1 mg every 4 weeks) prednisone in a usual clinical care setting. Another single-centre open-label study in 58 patients with established rheumatoid arthritis receiving long-term prednisone with or without csDMARDs showed that receiving the continued prednisone regimen led to better clinical outcomes and no apparent negative effect on bone mineral density compared with randomisation to prednisone tapering. There were no reports from double-blind, randomised, placebo-controlled trials investigating a specific low-dose prednisone tapering schedule compared with continuing low-dose prednisone in patients receiving biological therapy who achieved rheumatoid arthritis disease control.
Added value of this study
This is the first large randomised, placebo-controlled, multicentre trial to investigate tapering of low-dose prednisone in patients with rheumatoid arthritis who achieved stable disease control with glucocorticoids and csDMARDs. Patients who achieved low disease activity with tocilizumab with or without csDMARDs and at least 6 months of prednisone-equivalent glucocorticoids 5–15 mg per day who continued prednisone 5 mg per day had superior disease activity control compared with those who tapered prednisone from 5 mg per day to 0 mg per day. Treatment success (defined as disease control, no flare, and no adrenal insufficiency necessitating replacement therapy) was achieved by 77% of patients in the continued-prednisone group and 65% of those who received the tapered regimen. More adverse events were observed in the tapering group than the continuing group over 24 weeks. There were no reports of clinically manifest adrenal insufficiency or deaths.
Implications of all the available evidence
The safe and effective use of glucocorticoids reported in this study might influence future treatment guideline updates on the optimal timing of low-dose glucocorticoid withdrawal in patients with rheumatoid arthritis and low disease activity receiving effective biological therapy. The 65% treatment success rate with prednisone tapering has the potential to inform shared decision making. Although no adrenal insufficiency was observed with the tapering scheme, information about monitoring and prophylaxis should be provided to physicians and patients. Our findings offer a potential framework for investigations into glucocorticoid tapering and discontinuation in other therapeutic areas in which there may be uncertainty regarding the risks and benefits of glucocorticoid withdrawal.