This week in Rheumatology

Nov 19, 2026

Latest episode · 10 min
Editor’s pick

Prediction-aided tapering in rheumatoid arthritis patients treated with biOlogicals (PATIO): a randomised controlled trial.

In 162 patients with rheumatoid arthritis tapering biologics, adding a validated flare prediction model did not reduce flares or medication use compared with disease activity-guided tapering alone.

Annals of the Rheumatic Diseases · 2026 · PubMed

This week’s highlights

  1. 01Prediction-aided tapering in rheumatoid arthritis patients treated with biOlogicals (PATIO): a randomised controlled trial.

    In 162 patients with rheumatoid arthritis tapering biologics, adding a validated flare prediction model did not reduce flares or medication use compared with disease activity-guided tapering alone.

    Fadaei S et al. · Annals of the Rheumatic Diseases · 2026

    PMID 43045565

  2. 02Validation of rheumatoid arthritis low inflammatory activity: a novel clinical target quantifying inflammatory disease activity.

    A low inflammatory activity state based on modified disease activity scores was reached more often than traditional remission and predicted radiographic progression similarly and cardiovascular disease and death more strongly.

    Baker JF et al. · Annals of the Rheumatic Diseases · 2026

    PMID 43020702

  3. 03Efficacy and safety of upadacitinib in Takayasu arteritis (SELECT-TAK): a multicentre, double-blind, randomised, placebo-controlled, phase 3 trial.

    In 59 patients with Takayasu arteritis, upadacitinib numerically delayed relapse versus placebo but did not meet the prespecified significance threshold for its primary endpoint, with no new safety signals.

    Nakaoka Y et al. · Lancet Rheumatology · 2026

    PMID 43044624

  4. 04Herpes zoster in rheumatoid arthritis and in relation to biologic/targeted synthetic disease modifying anti-rheumatic drugs: a Swedish national study.

    Swedish national data show herpes zoster is more frequent in rheumatoid arthritis than the general population, similar with TNF and non-TNF biologics, and roughly doubled with JAK inhibitors versus TNF inhibitors.

    Kapetanovic MC et al. · Rheumatology · 2026

    PMID 43021635

  5. 05Real-world avacopan use in ANCA-associated vasculitis: glucocorticoid exposure and healthcare resource utilization from US claims data.

    In United States claims data, patients persisting on avacopan for ANCA-associated vasculitis had lower glucocorticoid doses and fewer emergency visits and hospital admissions over 12 months, without a comparator group.

    Sattui SE et al. · Rheumatology · 2026

    PMID 43033240

  6. 06ERN-RITA/PReS recommendations for the diagnosis, classification, monitoring, and treatment of childhood idiopathic (primary) angiitis of the central nervous system.

    First international consensus recommendations for childhood central nervous system vasculitis define new terminology, imaging principles and remission and relapse definitions, although no level 1 evidence exists in any domain.

    Constantin T et al. · Annals of the Rheumatic Diseases · 2026

    PMID 43020703

  7. 07GLADEL-PANLAR clinical practice guidelines for the management of systemic lupus erythematosus: 2025 update.

    Updated Latin American lupus guidelines support organ-based algorithms with earlier biologic and targeted therapy use, glucocorticoid minimisation and a treat-to-target goal of remission or low disease activity.

    Pons-Estel GJ et al. · Nature Reviews Rheumatology · 2026

    PMID 43020872

  8. 08Study of 34 prospective pregnancies in women receiving IL-1 inhibitors for autoinflammatory diseases and comparison to unexposed pregnancies (GR2 study).

    In 34 prospectively followed pregnancies exposed to IL-1 inhibitors, mostly anakinra, maternal and obstetric outcomes did not differ from matched unexposed pregnancies and infant development was normal.

    M D et al. · RMD Open · 2026

    PMID 43039653

  9. 09Decoding treatment non-response in spondyloarthritis.

    A review proposes distinguishing apparent from true treatment non-response in spondyloarthritis and describes four true patterns: primary non-response, secondary loss of response, divergent domain responses and paradoxical reactions.

    Fakih O et al. · Nature Reviews Rheumatology · 2026

    PMID 43026360

  10. 10Shared and distinct inflammatory signatures in gout flare and acute CPP crystal arthritis: systemic inflammation persists in gout despite achieving serum urate target.

    Gout and calcium pyrophosphate flares share circulating inflammatory profiles, while caspase-8 is higher in gout and remains elevated after serum urate target is achieved, suggesting persistent low-grade inflammation.

    Kischkel B et al. · Annals of the Rheumatic Diseases · 2026

    PMID 43039800

Read the weekly briefing

This AudioScholar briefing is generated by artificial intelligence from PubMed abstracts for healthcare professionals and is not medical advice.

This week: a JAK inhibitor narrowly misses its primary goal in Takayasu arteritis, a flare prediction model fails to improve biologic tapering in rheumatoid arthritis, and shingles risk climbs highest with JAK inhibitors in Swedish national data.

Welcome to This Week in Rheumatology. We're covering 10 papers this week: 3 in depth, 4 in brief, 3 quick signals.

In the Annals of the Rheumatic Diseases, Fadaei and colleagues report PATIO, a pragmatic, multicentre, open-label randomised superiority trial in 162 patients with rheumatoid arthritis in sustained low disease activity on stable biologic therapy [1]. Both arms tapered biologics using disease activity-guided dose optimisation. In one arm, a previously validated dynamic model that estimated flare risk every three months supported the tapering decisions. Over 18 months, the prediction-aided group had about one flare per patient, against about 0.7 in the standard group, and the difference was not statistically significant. Any flare occurred in 57 percent of the prediction-aided group and 44 percent of the standard group, and major flares in 22 and 15 percent. Disease activity over time, biologic dose reduction, infections and patient-reported outcomes were similar. The main limitation is adherence: tapering advice was not followed at close to a quarter of visits, mostly when the model advised continuing to taper. The trial was open-label and tested one specific model. What it adds is a prospective randomised test of a prediction tool that had performed well on validation, and it did not improve flare rates or medication use. The authors conclude that prediction models need robust accuracy and prospective evaluation before implementation.

Also in the Annals, Baker and colleagues address a related question: which disease state should guide treatment decisions in rheumatoid arthritis [2]. They define and validate a target called rheumatoid arthritis low inflammatory activity, based on previously validated modified disease activity scores. The study drew on three clinical trials with 960 patients and the Veterans Affairs Rheumatoid Arthritis Registry with over 3,400 patients. Thresholds were set against traditional remission at month six in a large trial, for example a modified DAS28 below 1.4. In every dataset, more patients reached this low inflammatory state than reached traditional remission. In the Veterans Affairs registry the figures were 44 percent and 32 percent. The new state predicted radiographic progression about as well as traditional remission, and it was a stronger predictor of cardiovascular disease and death. Patients in this state who still had high traditional scores had more anxiety and depression, and less seropositive and erosive disease. The main limitation is that this is a validation of thresholds in existing data. The target has not yet been tested prospectively in a treat-to-target strategy trial. It adds a candidate point-of-care target that the authors argue could make management more precise.

In Lancet Rheumatology, Nakaoka and colleagues report SELECT-TAK, a double-blind, randomised, placebo-controlled phase 3 trial of upadacitinib 15 milligrams daily in Takayasu arteritis [3]. The trial enrolled 59 patients at 31 centres in six countries, about three quarters of them Asian, all in remission after at least one failed glucocorticoid taper, and all following a structured steroid taper from week four. The primary endpoint, time to first relapse, was not met. The analysis numerically favoured upadacitinib but did not reach the prespecified significance threshold. Relapse occurred in 11 of 30 patients on upadacitinib and 18 of 29 on placebo, and median time to relapse was about 108 weeks against about 20 weeks. There were nine serious adverse events with upadacitinib and four with placebo, with no deaths and no new safety signals. The key limitation is size: a rare disease trial with under sixty patients had little power to detect a moderate effect. It adds the first placebo-controlled phase 3 data for a JAK inhibitor in this disease, and the authors call for further study in refractory Takayasu arteritis.

Now, four in brief.

Staying with JAK inhibitors, Kapetanovic and colleagues, in Rheumatology, used Swedish national registers from 2014 to 2024 covering almost 78,000 patients with rheumatoid arthritis, with matched population controls [4]. Herpes zoster was more common in biologic-naive rheumatoid arthritis than in the general population, rose further and to a similar degree with TNF inhibitors and other biologics, and was highest with JAK inhibitors. Compared with TNF inhibitors, JAK inhibitors roughly doubled the adjusted risk, while non-TNF biologics showed no difference. The data are observational, and shingles was defined through diagnosis codes and antiviral prescriptions, but the national scale makes the gradient firm.

Turning to vasculitis, Sattui and colleagues, also in Rheumatology, analysed United States Medicare and registry claims for 167 adults starting avacopan for ANCA-associated vasculitis [5]. Among the 94 patients who stayed on treatment, mean daily prednisone dose fell from about 13 milligrams to about 5 milligrams over a year, and the proportion of patients admitted to hospital fell from just over half to about a fifth. Similar trends appeared in those aged 65 and over. This is an uncontrolled before-and-after claims analysis restricted to persistent users, so it cannot separate drug effect from natural disease course, and the authors call for comparator studies.

In the Annals, Constantin and colleagues present the first international recommendations for childhood idiopathic angiitis of the central nervous system, developed by a 42-member panel from 15 countries through ERN-RITA and PReS [6]. The panel produced 83 consensus statements, including new idiopathic versus disease-associated terminology, a paediatric vessel wall imaging principle, a risk stratification model, genome-wide sequencing as the preferred genetic test, and formal definitions of remission, relapse and refractory disease. The systematic review found no level 1 evidence in any domain, so these are consensus-based recommendations.

In Nature Reviews Rheumatology, Pons-Estel and colleagues publish the 2025 GLADEL-PANLAR update on systemic lupus erythematosus management, revising the 2018 recommendations [7]. The guidelines incorporate newly approved therapies and recent trial evidence into organ-based algorithms that support earlier use of biologics and targeted therapies where appropriate, selective combination therapy and glucocorticoid minimisation, within a treat-to-target framework aiming for remission or low disease activity. They are framed for Latin American health systems, where access to treatment varies, and their strength rests on the underlying trial evidence.

Three quick signals.

In RMD Open, the French GR2 prospective cohort followed 34 pregnancies exposed to anakinra or canakinumab for autoinflammatory diseases [8]. Maternal and obstetric outcomes did not differ from matched unexposed pregnancies, and infant development was normal over follow-up. The numbers are small and canakinumab exposure was rare, so the data are firmest for anakinra.

In Nature Reviews Rheumatology, Fakih and colleagues review treatment non-response in spondyloarthritis [9]. They separate apparent from true non-response and describe four patterns: primary non-response, secondary loss of response, divergent responses across domains, and paradoxical reactions. This is a conceptual framework, not new outcome data.

In the Annals, Kischkel and colleagues compared plasma inflammatory proteins in 71 gout and 86 calcium pyrophosphate flares [10]. Flare profiles were largely shared, with TNFSF14 proposed as a common target, while caspase-8 was higher in gout and stayed elevated even after serum urate target was reached. These are exploratory biomarker findings.

Stepping back across this week's papers, much of the evidence comes from small samples, uncontrolled designs or expert consensus. The two randomised trials both missed their primary endpoints, one through limited power in a rare disease and one through a tool that did not hold up prospectively. Several recommendations and real-world analyses rest on little or no high-level evidence, so the gap between what guidance proposes and what trials have confirmed remains wide in rare and paediatric rheumatic disease.

If you only have time for one paper this week, make it the PATIO trial [1]. It is a randomised test of whether a validated flare prediction model improves biologic tapering in rheumatoid arthritis, and it shows that good performance on validation did not translate into better outcomes in practice.

That's your roundup for This Week in Rheumatology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And think of one colleague, in any specialty, who never has time to keep up with the literature. Tell them about AudioScholar: a ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Use it to choose which papers to read in full.

Get Rheumatology every week

By email, Spotify, Apple or any podcast app.

Listen on Apple PodcastsListen on Spotify

Spot something worth flagging?