This Week in Rheumatology — Aug 28, 2026

Welcome to This Week in Rheumatology. This week we're covering 10 notable papers spanning lupus — from rare-manifestation consensus to cognition and epidemiology — drug safety and cardiovascular risk in gout and vasculitis, and biomarkers and imaging that help us stratify risk in myositis, scleroderma and inflammatory arthritis. Let's dive in.

Let's start with lupus, which dominates the week. The Lancet Rheumatology publishes the second part of a large international consensus effort from three expert groups — the European Reference Network on connective tissue diseases, the Systemic Lupus International Collaborating Clinics, and the European Lupus Society [1]. Having already covered 24 rare manifestations, Arnaud and colleagues convened 77 participants to build therapeutic strategies for 22 more: diffuse pulmonary haemorrhage, bullous and chilblain lupus, toxic epidermal necrolysis-like disease, interstitial nephritis and lupus podocytopathy, chorea, catatonia, small fibre neuropathy, protein-losing enteropathy, intestinal pseudo-obstruction, lupus hepatitis, myositis, Jaccoud's arthropathy, uveitis, interstitial cystitis and lupus mastitis. These are expert-based strategies, not evidence-based recommendations, and the taskforce is explicit about that — but if you have ever stared at a patient with lupus podocytopathy or protein-losing enteropathy and found nothing in the guidelines, this is now the document to reach for.

Staying with lupus but moving to the brain, Annals of the Rheumatic Diseases reports the ClearMEMory trial, a multisite phase 2 randomised placebo-controlled study of memantine titrated to 40 milligrams daily over 12 weeks in adults with lupus and objectively documented neuropsychological dysfunction [2]. Rhoads and colleagues randomised 56 patients, and 43 completed the primary assessment — 18 on memantine, 25 on placebo. Cognitive scores on the Repeatable Battery for the Assessment of Neuropsychological Status improved more with memantine, with a median gain of 8 points versus 5 on placebo, a statistically significant difference. Two thirds of the memantine group crossed the threshold considered clinically important, compared with just over a third on placebo — a number needed to treat of about three. Patient-reported global impression of change favoured memantine numerically but was not statistically significant, and discontinuation for adverse events was similar in both arms, around one in five. This is a small trial with substantial attrition, so it is a signal rather than a mandate, but it is the first randomised evidence that an NMDA receptor antagonist can move objective cognition in lupus, and it deserves a confirmatory trial.

The third lupus paper reframes who we should be suspecting of disease in the first place. In Arthritis and Rheumatology, Izmirly and colleagues pooled the Centers for Disease Control and Prevention National Lupus Registry network — five state registries plus an Indian Health Service registry, covering 2002 to 2018 — to produce age-specific incidence by sex, race and ethnicity [6]. Among women, incidence peaked in the thirties, and Black women aged 20 to 39 had by far the highest rates. Among men, the peak was in the sixties, not the reproductive years. Overall, one in ten patients with lupus was diagnosed before age 20 and about one in seven after age 60. Among women diagnosed before 20, incidence was highest in Asian individuals; after 60, highest in American Indian and Alaska Native individuals. The practical message is that a new inflammatory multisystem illness in a 65-year-old man should not have lupus struck from the differential simply on demographics.

Our second theme is drug safety, where two papers push in opposite directions on how anxious we should be. In Rheumatology, Wei and colleagues performed a one-stage individual participant data meta-analysis reconstructing patient-level survival data from published Kaplan-Meier curves across four studies and just over 19,000 patients with gout, comparing febuxostat with allopurinol [3]. Rather than a single hazard ratio, they modelled treatment-by-time interaction, and the effect was clearly time-dependent: cardiovascular risk was modestly lower with febuxostat in the first year, most strongly lower between one and two years — roughly a 40 percent relative reduction — still slightly lower between two and four years, and then no different beyond four years. All-cause mortality followed the same pattern. Now, a caution: this is reconstructed data from published curves, mixing randomised trials with propensity-matched observational cohorts, and reconstruction introduces its own noise. What it does not show is late excess harm from febuxostat, which sits closer to the reassuring end of the long-running controversy. The authors argue for time-aware monitoring rather than a blanket verdict on the drug.

On the more cautionary side, Arthritis and Rheumatology reports the first United States real-world quantification of avacopan hepatotoxicity in ANCA-associated vasculitis, using the Rheumatology Informatics System for Effectiveness registry [4]. Roberts and colleagues followed 238 patients prescribed avacopan between late 2021 and early 2025. Liver enzyme elevations meeting Food and Drug Administration thresholds were uncommon — a handful of patients each for transaminases, alkaline phosphatase and bilirubin, with incidence rates of roughly two to four events per 100 person-years. But two patients, just under one percent, had drug-induced liver injury with Roussel Uclaf causality scores of 9, meaning highly probable, and one had histologically confirmed severe hepatocellular injury. Worth noting for your own practice: median time to first liver enzyme check was 32 days, and the average interval between tests was about four months. Given that severe injury did occur, that monitoring cadence looks too loose. With the regulatory reassessment ongoing after the agency requested market withdrawal, this argues for early and then sustained liver monitoring, and an explicit conversation with patients about hepatic risk.

A third safety-adjacent paper, also in Rheumatology, addresses thrombosis in high-risk immune thrombocytopenia — patients with antiphospholipid antibodies or with thrombocytopenia secondary to an autoimmune disease [8]. Tarasconi and colleagues systematically reviewed 13 studies covering 308 such patients on second-line therapy. About one in ten had a thrombotic event over follow-up ranging from six to 42 months. Stratified by treatment, the rate was around one in six on thrombopoietin receptor agonists, and broadly similar on rituximab, conventional immunosuppressants, and after splenectomy. The uniformity across mechanistically unrelated treatments suggests the driver is the prothrombotic immune milieu rather than any single drug — so screening for antiphospholipid antibodies and planning thromboprophylaxis matters more than agonist avoidance alone.

Our final theme is risk stratification through antibodies, imaging and structure. In Rheumatology, Pan and colleagues analysed 2,126 patients with idiopathic inflammatory myopathies from a single Chinese centre over two decades [7]. Anti-Ro52 antibodies were present in close to half OF PATIENTS, making them the most common myositis-associated antibody, and were enriched in anti-MDA5-positive dermatomyositis and antisynthetase syndrome. The clinical significance was subgroup-specific: in anti-MDA5-positive disease, anti-Ro52 tracked with interstitial lung disease, rapidly progressive interstitial lung disease and worse survival, with the worst prognosis when anti-Ro52 and anti-SSA coexisted. In antisynthetase syndrome, anti-Ro52 predicted pulmonary involvement but not survival, and elsewhere it added little. So a positive anti-Ro52 is not a uniform red flag — it is a red flag specifically in MDA5 disease.

Also in Rheumatology, Iacovantuono and colleagues applied automated lung texture radiomics to high-resolution computed tomography in 313 patients with systemic sclerosis-associated interstitial lung disease [10]. Over a median follow-up of about three and a half years, roughly a quarter of patients died, and pulmonary vessel volume independently predicted mortality — most strongly in the upper zones. For progression, baseline features were largely similar between progressors and non-progressors, with one exception: whole-lung honeycombing independently predicted progression, raising the odds by about half per unit increase. Two different radiomic signals for two different outcomes, potentially additive to functional and clinical predictors.

Finally, two papers on the musculoskeletal system. Annals of the Rheumatic Diseases publishes seven-year high-resolution peripheral quantitative computed tomography data from Temiz and colleagues in 482 patients — 221 seropositive rheumatoid arthritis, 86 seronegative, and 175 psoriatic arthritis [5]. Seropositive rheumatoid arthritis had the lowest trabecular bone density at every timepoint and lost about 16 milligrams of hydroxyapatite per cubic centimetre of total volumetric density over five years; psoriatic arthritis showed stable bone quality with relative cortical preservation, and seronegative disease sat in between. Crucially, sustained remission preserved bone structure and strength while high disease activity tracked with deterioration — another structural argument for treating to target, particularly in seropositive disease. And in a smaller but practically useful report, Ruscitti and colleagues describe 24 adults from the Italian GIRRCS network with acute parvovirus B19-associated inflammatory musculoskeletal disease [9]. Most were women, presentation was polyarticular in about six in ten and mostly seronegative, and of the 21 with follow-up, all but one achieved remission without any disease-modifying escalation, at a median of 21 days, though the range stretched to 135 days. One patient did require a biologic. The lesson is patience before committing a new polyarthritis to lifelong therapy — but also that a minority OF PATIENTS will not simply resolve.

If you only have time for one paper this week, make it the avacopan hepatotoxicity study in Arthritis and Rheumatology [4]. It changes something you can act on tomorrow: the frequency and timing of liver monitoring in every ANCA-associated vasculitis patient you have on this drug, and how you consent them.

Here are the key takeaways from this week in Rheumatology. First, for patients on avacopan, tighten early liver monitoring and revisit the risk conversation, because severe drug-induced liver injury does occur even when enzyme elevations are uncommon. Second, memantine improved objective cognition in a small randomised lupus trial with a number needed to treat around three — promising, but needing confirmation before routine use. Third, stop excluding lupus on demographics: incidence peaks in the sixties for men, and a meaningful share of patients present in childhood or after 60. Fourth, in myositis, anti-Ro52 carries prognostic weight chiefly in anti-MDA5-positive dermatomyositis, especially alongside anti-SSA. Fifth, sustained remission measurably preserves bone microarchitecture, most importantly in seropositive rheumatoid arthritis. And sixth, in high-risk immune thrombocytopenia, thrombotic risk is a property of the patient more than the drug, so assess antiphospholipid status and prophylaxis regardless of which second-line agent you choose.

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 think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free 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. Always review the original source materials.