This Week in Rheumatology — Sep 17, 2026

Welcome to This Week in Rheumatology. This week we're covering 10 notable papers spanning targeted therapy across spondyloarthritis, lupus and fibromyalgia, the science of measuring and monitoring disease over time, and a cluster of population-level studies on comorbidity and mortality. Let's dive in.

We start with drugs and what they do to tissue. In Arthritis and Rheumatology, Maksymowych and colleagues report the imaging substudies from the phase 3 BE MOBILE 1 and 2 trials of bimekizumab, the dual interleukin-17A and 17F inhibitor, in both non-radiographic and radiographic axial spondyloarthritis [1]. By week 52, patients randomised to bimekizumab showed substantial falls in sacroiliac joint inflammation on MRI, with a mean drop of nearly eight points on the SPARCC score in non-radiographic disease and around five points in radiographic disease, alongside more modest improvement in spinal inflammation. What makes this interesting beyond inflammation is the structural picture: erosion scores fell in both populations, while backfill and fat lesion scores rose slightly, which is the imaging signature of erosions filling in and repairing rather than progressing. Ankylosis scores barely moved. Out to two years on radiographs, progression was essentially negligible, with a mean change in the modified Stoke score of around three tenths of a point in radiographic disease. These are observed-case, uncontrolled data beyond week 16, so they cannot prove structural disease modification, but they are reassuring and consistent with what we have come to expect from potent interleukin-17 blockade.

Staying with targeted therapy, Lupus Science and Medicine publishes a real-world retrospective comparison of telitacicept, the dual BLyS and APRIL inhibitor, added to standard of care in 112 patients with biopsy-proven class three or four lupus nephritis [2]. Using propensity-score weighting to balance the groups, Tian and colleagues found that at 24 weeks the complete renal response rate was about 68 percent with telitacicept versus about 39 percent with standard care alone, with partial response also favouring the combination, and with lower glucocorticoid doses and lower composite disease activity scores. Telitacicept came out as an independent predictor of complete response, with roughly a fourfold increase in the likelihood of achieving it. This is retrospective, single-setting and unblinded, so it is hypothesis-generating rather than definitive, but it adds to the growing signal that B-cell-directed biologics belong in induction rather than only in maintenance.

The contrast this week is a negative trial, and an instructive one. In Lancet Rheumatology, Goebel and colleagues report a phase 2A randomised, placebo-controlled study of rozanolixizumab, a neonatal Fc receptor blocker that lowers circulating IgG, in 63 adults with severe fibromyalgia across seven sites in north west England [3]. The rationale was the emerging evidence that pathogenic IgG autoantibodies contribute to fibromyalgia symptoms. The primary endpoint, change in the Brief Pain Inventory interference score at 12 weeks, crossed the trial's pre-specified one-sided ten percent threshold but was not statistically significant by conventional two-sided testing, and the authors are explicit that the drug did not demonstrate broad efficacy and that the improvement was not meaningful at a group level. Safety was unremarkable, with headache the commonest adverse event and no serious treatment-emergent events on active drug. The authors' interpretation is that without a biomarker to select patients who actually carry pathogenic autoantibodies, any true subgroup effect is diluted. For now, nothing changes in fibromyalgia practice, but the autoantibody hypothesis is not dead; it needs a companion test.

That problem of defining what a drug must actually achieve is the subject of a consensus paper in Annals of the Rheumatic Diseases, where Distler and colleagues propose a conceptual framework for disease modification in systemic sclerosis [4]. Their argument is that a disease-modifying agent in scleroderma should target core pathogenic cells or pathways rather than downstream regulatory mechanisms, and should show sustained effects on activity and prevention of damage across the vascular, inflammatory and fibrotic domains, across organs and disease stages. It is a framework rather than data, but it is the kind of document that shapes what trial endpoints look like over the next decade, and it is worth reading before you next consent a patient for a scleroderma study.

Our second theme is measurement — how we decide whether treatment is working and how often we need to look. JAMA carries a comprehensive review of rheumatoid arthritis in adults by Smolen and colleagues, which is a useful refresher on where the field has landed [5]. The headline framing is that the target is at least a fifty percent improvement in disease activity by three months and remission or low disease activity by six months; methotrexate remains first-line, escalated to 20 to 25 milligrams weekly within four to eight weeks, with short-term glucocorticoids tapered and stopped within three months. About 40 percent of newly diagnosed patients reach remission on that initial strategy, and adding a biologic or a JAK inhibitor lifts overall remission or low disease activity rates to around 80 percent, with the familiar caveat that JAK inhibitors are reserved for patients who are not at high cardiovascular, thromboembolic or malignancy risk.

That treat-to-target arithmetic gets sharpened by a study in Annals of the Rheumatic Diseases from Terabe and colleagues, who applied group-based trajectory modelling to 843 patients starting JAK inhibitors in a multicentre registry [6]. Four trajectories emerged over 52 weeks: rapid responders made up about two thirds of the cohort, partial responders with residual activity about a fifth, a group with high baseline activity that still responded about an eighth, and persistent non-responders around four percent. Week-52 remission rates ranged from nothing at all in the non-responders to just under half in the best group. The practically useful part is the landmark analysis: the absolute CDAI score at week 4 or week 12 substantially improved prediction of a favourable long-term trajectory, whereas the slope of change added almost nothing. The thresholds were a CDAI of about 12 at four weeks and about 9 at twelve weeks. In other words, where the patient is matters more than how fast they got there — which argues for anchoring your switch decisions to an absolute number at a fixed timepoint.

And if disease is quiet, how often does the patient need to come in? In Rheumatology, Sellies and colleagues tested digital home monitoring in 84 young people aged 6 to 20 with inactive juvenile idiopathic arthritis, replacing one routine three-monthly visit with remotely completed quality-of-life and JAMAR questionnaires reviewed by a clinician [7]. Sixty-three patients had their interval prolonged, of whom seven flared, and the flare rate across the study met the pre-specified non-inferiority criterion against historical and matched comparators. The analysis of flares specifically after a prolonged interval was underpowered, so caution is warranted, but satisfaction was high and more than nine in ten participants wanted home monitoring used more often. Reminders were necessary — the technology does not run itself.

Our final theme is the long view: who develops these diseases, and what happens to them over decades. Arthritis and Rheumatology publishes a Danish nationwide study by Donskov and colleagues covering more than 51,000 patients with polymyalgia rheumatica identified across both primary and secondary care over two decades [8]. The age-standardised incidence was about 112 per 100,000, and the trends diverged strikingly by sector: diagnoses in primary care fell by roughly half while secondary care diagnoses rose by about thirty percent, suggesting a shift in where these patients are managed rather than a true fall in disease. Giant cell arteritis developed at a rate of about five per thousand patient-years, and crucially most cases appeared within the first year, at a median of about ten weeks after the polymyalgia diagnosis. The most sobering figure is steroid persistence: 28 percent of patients were still on prednisolone at five years and 22 percent at ten, at daily doses of four to five milligrams. That is a strong argument for considering steroid-sparing therapy far earlier than most of us currently do.

Two mortality studies round this out. Also in Rheumatology, Bardan and colleagues linked Norwegian national registry data on nearly 3,000 adults with juvenile idiopathic arthritis to roughly ten matched comparators each [9]. Adults with juvenile arthritis carried more hypertension, ischaemic heart disease, chronic kidney disease, type 1 diabetes, coeliac disease and autoimmune thyroid and skin conditions, and all-cause mortality was modestly but clearly higher, at about 2.4 versus 1.8 deaths per thousand person-years. Importantly, comorbidity and mortality did not differ between those with and without recent DMARD exposure, which does not support the idea that the drugs are driving the excess. And from Western Australia, Pelkas and colleagues used a linked population registry to follow patients with Sjögren disease over three decades [10]. Mortality risk was increased by roughly 80 percent in primary Sjögren disease, roughly doubled when it accompanied rheumatoid arthritis, and roughly tripled when it accompanied another connective tissue disease, with survival in primary disease around 77 percent at 25 years. The excess deaths were driven by infection, ischaemic heart disease, lymphoid malignancy and interstitial lung disease — a list that maps neatly onto what we should be screening for and vaccinating against.

If you only have time for one paper this week, make it the Danish polymyalgia rheumatica cohort [8]. It quantifies something we all suspect but rarely confront — that roughly a quarter of our polymyalgia patients are still on glucocorticoids five to ten years on — and it tells you exactly when to be vigilant for giant cell arteritis.

Here are the key takeaways from this week in Rheumatology. Bimekizumab reduced sacroiliac and spinal inflammation on MRI with falling erosion scores and minimal radiographic progression over two years in axial spondyloarthritis. Telitacicept added to standard induction nearly doubled complete renal response at 24 weeks in proliferative lupus nephritis in a real-world weighted comparison, while lowering steroid exposure. Rozanolixizumab did not show broad efficacy in severe fibromyalgia, and the IgG hypothesis now needs a patient-selection biomarker before it is tested again. In rheumatoid arthritis on JAK inhibitors, the absolute disease activity score at week 4 or week 12 predicts the long-term trajectory better than the rate of change does. Home monitoring can safely extend visit intervals in inactive juvenile idiopathic arthritis, with high patient satisfaction. And across polymyalgia rheumatica, adult juvenile idiopathic arthritis and Sjögren disease, the long-term burden is cardiovascular, infectious and malignant — screen accordingly, and spare the steroids where you can.

That's your roundup for This Week in Rheumatology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.