Tapering biologics held disease control in rheumatoid arthritis — mostly

Once rheumatoid arthritis is quiet, a reasonable question follows: can the patient take less of the expensive, immune-suppressing drug that got them there? Treatment guidelines already say tapering may be considered when disease activity has been low for a sustained period, but they note the evidence behind that advice is thin [s2]. A new randomised trial set out to put a number on it [s1].

What the trial tested

The study enrolled patients with rheumatoid arthritis who were on a stable dose of a biologic or targeted synthetic disease-modifying drug — the b/tsDMARD class that includes TNF inhibitors and JAK inhibitors — and who had held low disease activity, by the DAS28-ESR score, for at least six months [s1]. A total of 348 patients were randomly assigned to one of two strategies over 24 weeks: stepwise dose tapering (n=171) or continuing the full dose (n=177) [s1].

This was a multicentre, open-label, non-inferiority trial [s1]. That design is worth understanding, because it changes what a positive result means. The trial was not asking whether tapering is better; it was asking whether tapering is not meaningfully worse than staying on full dose. The threshold was set in advance: tapering would be declared non-inferior only if the drop in the share of patients keeping low disease activity stayed within 10 percentage points [s1]. The primary endpoint was the maintenance of DAS28-ESR low disease activity at week 24 [s1].

The numbers

In the per-protocol analysis, low disease activity was maintained in 89.6% of the continuation group and 88.2% of the tapering group — a risk difference of −1.43 percentage points, with a 95% confidence interval of −8.09 to 5.23 [s1]. Because the lower bound of that interval stayed inside the prespecified −10-point margin, tapering met the criterion for non-inferiority [s1]. The full analysis set, which counts every randomised patient, pointed the same way [s1].

Two details temper the clean headline. First, among patients whose disease did worsen, most regained low disease activity after a protocol-guided step back up in dose, without having to stop treatment permanently — a reassuring sign that a flare on a taper is usually recoverable [s1]. Second, not every measure was a tie. Some secondary disease-activity scores favoured staying on full dose, including DAS28-CRP at week 12 and both DAS28-ESR and the SDAI composite at week 24 [s1]. Adverse events were uncommon in both groups [s1]. On its own terms, then, the trial delivered the result it was designed to find, with the usual caveat that a non-inferiority "win" and a true tie are not the same thing [s1].

How to read a non-inferiority result

The margin is doing a lot of work here, and readers should see it. A non-inferiority trial can return a "success" even when the tapering group does slightly worse, as happened on the point estimate — 1.4 fewer percentage points holding low disease activity, and a confidence interval that reaches to 8 points worse before crossing the line [s1]. Whether that trade is acceptable is a judgement, not a statistic: a small, usually reversible increase in flare risk weighed against less drug, lower cost, and reduced immune suppression.

The design limits also matter. The trial was open-label, so patients and clinicians knew who was tapering, which can colour subjective components of a disease-activity score [s1]. It ran for 24 weeks — long enough to catch early flares, short enough that it says little about whether control holds over years, which is the timescale that matters for a chronic disease [s1]. And it tested a specific population: people already in sustained low disease activity, not those with active or recently active disease, for whom none of this applies.

What it means, and what to watch

The practical reading is narrow and useful. For a patient who has held low disease activity for at least six months, a structured, monitored taper kept most people's control intact over six months, and flares were generally recovered by stepping the dose back up [s1]. That is consistent with the direction guidelines already lean, and it strengthens the evidence they called thin [s2].

It is not a green light to stop treatment, and it is not advice to act on alone — the trial tested a supervised, protocol-driven reduction with defined rules for re-escalation, not patient-led dose-cutting [s1]. The open questions are the ones 24 weeks cannot answer: how control holds over several years of tapering, which patients can taper furthest, and whether the small gap that favoured full dose on some measures widens with time.

This article is informational and does not constitute medical advice.

Sources

• [s1] Outcomes of Tapering Biologic and Targeted Synthetic DMARDs in Rheumatoid Arthritis: A Multicenter Randomized Non-Inferiority Trial. Arthritis & Rheumatology, 6 October 2026. https://doi.org/10.1002/art.70363
• [s2] EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2022 update. Annals of the Rheumatic Diseases, January 2023. https://doi.org/10.1136/ard-2022-223356