This week in Allergy & Immunology
Nov 16, 2026
Upadacitinib versus abrocitinib in real-world atopic dermatitis: a propensity score-weighted comparative effectiveness and safety analysis.
In 290 registry patients with atopic dermatitis, upadacitinib and abrocitinib showed no significant difference in one-year effectiveness or drug survival, while gastrointestinal complaints were about three times more frequent with abrocitinib.
The Journal of Allergy and Clinical Immunology: In Practice · 2026 · PubMed
This week’s highlights
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: dupilumab paired with dust mite immunotherapy in severe eczema, two oral JAK1 inhibitors compared in everyday atopic dermatitis care, and wasp venom extracts compared after real-world re-stings.
Welcome to This Week in Allergy & Immunology. We're covering 10 papers this week: 3 in depth, 4 in brief, 3 quick signals.
Bożek and colleagues report three-year results in The Journal of Allergy and Clinical Immunology: In Practice from a prospective, randomized, open-label study at three centres [1]. They enrolled 65 adults with severe atopic dermatitis and house dust mite sensitization and assigned them to dust mite sublingual immunotherapy alone, dupilumab alone, the two combined, or cyclosporine, which was given for twelve months only. All four groups improved at one year. By month 36 the combination group had the lowest mean EASI score, 3.5, against 11 with dupilumab alone, 17 with immunotherapy alone and 35 with cyclosporine. Nineteen of twenty patients on the combination reached clear or almost clear skin on the investigator global assessment, compared with about three quarters to four fifths of patients on either monotherapy. The combination group also showed the largest descriptive rise in allergen-specific IgG4 and the largest shifts in type 2 biomarkers. The limitations are substantial: the groups were small, treatment was open-label, cyclosporine was stopped early by design, and patients were still on therapy, so the study cannot show durable benefit after withdrawal. The authors frame the combination as a candidate disease-modifying strategy that needs larger trials.
Also in atopic dermatitis, and also in The Journal of Allergy and Clinical Immunology: In Practice, van der Gang and colleagues compared upadacitinib with abrocitinib using the prospective multicentre BioDay registry [2]. They included 290 patients aged twelve and over who were starting a JAK1 inhibitor for the first time, 213 on upadacitinib and 77 on abrocitinib, and balanced the groups with propensity weighting. Both drugs worked fast: mean EASI fell from about 15 to around 6 with upadacitinib and to around 4 with abrocitinib by week four, and itch scores roughly fell by half, then stayed stable. At one year, about four in ten patients on upadacitinib and about a third of patients on abrocitinib were at the EASI target, counting discontinuations as failures, and just over half of patients on each drug were still taking it. None of the effectiveness outcomes differed significantly. Gastrointestinal complaints were about three times more frequent with abrocitinib. This is observational data with a small abrocitinib group, so residual confounding remains possible. It adds the first year-long real-world comparison, and the authors conclude that cost and preference may guide the choice.
Stefania and colleagues, again in The Journal of Allergy and Clinical Immunology: In Practice, examined field re-stings during vespid venom immunotherapy in a retrospective European multicentre cohort of 340 patients [3]. Patients were treated with Vespula extract, Polistes dominula extract or Vespa crabro extract. Diagnosis was often uncertain: the culprit insect could not be identified in more than three in ten patients, and polysensitization was common. About half of patients were stung again during treatment, and four in ten of those stings came from an unidentified vespid. Systemic reactions after a re-sting occurred in about 9 percent of patients on Vespula extract, 17 percent on Polistes extract and none on Vespa crabro extract. In a Bayesian analysis, Polistes treatment carried about 1.7 times the odds of a reaction compared with the pooled Vespinae extracts, with an 84 percent probability that the difference is real; the estimate was imprecise and still compatible with no difference. The design is retrospective, the events are few and the culprit was often unknown. The authors read it as tentative support for broader protection from Vespinae extracts when diagnostic profiles are equivocal.
Now, four in brief.
Huang and colleagues, in The Journal of Allergy and Clinical Immunology: In Practice, clustered 764 patients with type 2 severe asthma from the Australasian Severe Asthma Network [4]. Three subtypes emerged. The largest group was a well-controlled allergic subtype, mostly women. A second group of older patients carried non-type 2 features such as sputum neutrophilia, fixed airflow limitation and more comorbidities, with about two and a half times the severe exacerbation rate. A third group had several raised type 2 biomarkers, poor control, a similar rise in exacerbations and roughly triple the hospitalization rate. A classifier reproduced the clusters with about 84 percent agreement. Validation was internal only and follow-up lasted a year, so these subtypes still need external testing.
In Annals of Allergy, Asthma & Immunology, Riedl and colleagues used a United States claims database covering 224,958 adults with chronic spontaneous urticaria [5]. Claims for mental health conditions rose by about fifteen points after diagnosis, to just over half of patients, and claims for sleep conditions also rose. About half of patients met proxy criteria for uncontrolled disease, and that group had more sleep and mental health claims and more related health care use. This is descriptive claims data with a proxy definition of control and no comparison population, so it shows association with diagnosis and disease activity, not cause.
Chiarella and colleagues, in the World Allergy Organization Journal, surveyed 322 health care professionals from 71 countries about biologics in pregnancy [6]. About a quarter of respondents had prescribed a biologic during pregnancy, most often omalizumab, then dupilumab. Patient concern about the pregnancy was the leading barrier. Twelve of 23 clinicians who stopped a biologic reported worsening symptoms afterwards, and only about four in ten prescribers enrolled patients in a pregnancy registry. Seven respondents reported adverse maternal or fetal outcomes, but a self-reported survey without denominators cannot link those events to the drugs.
In Pediatric Allergy and Immunology, Genç Özbay and colleagues retrospectively reviewed multiplex testing in 1,400 children and found 89 sensitized to Gal d 5, the egg yolk albumin [7]. Only about one in six of those children had clinically relevant egg yolk allergy, and reactions to birds or poultry meat were uncommon. Higher Gal d 5 IgE was associated with egg yolk allergy, but its diagnostic accuracy was moderate, and egg yolk IgE performed numerically better without a significant difference. The number of allergic children is small, which limits the precision of the proposed cut-off.
Three quick signals.
Schauberger and Phillips, in Annals of Allergy, Asthma & Immunology, review nucleic acid therapies across the specialty [8]. They cite a CRISPR gene-editing therapy that reduced hereditary angioedema attacks by 87 percent against placebo, and donidalorsen as the first approved RNA-targeted prophylaxis for the disease. This is a narrative review; delivery beyond the liver and long-term safety remain open.
Nalabothu and colleagues, in Current Allergy and Asthma Reports, review drug allergy pharmacogenetics [9]. Screening for a few drug and HLA pairs has reduced abacavir hypersensitivity and severe cutaneous reactions to carbamazepine and allopurinol in targeted populations. The authors treat genotype as a risk modifier, given incomplete predictive value and underrepresented populations. It is a clinical perspective, not new data.
Murphy and Ege, in the Journal of Allergy and Clinical Immunology, trace thirty years of research on the farm effect [10]. The protective signal centres on microbial exposure in animal sheds and unprocessed cow's milk, and the field is now moving towards prevention trials. As a narrative review, it summarizes mostly observational evidence.
Stepping back across this week's papers, the questions closest to the clinic were mostly answered with designs that cannot settle them. Comparative effectiveness came from registry weighting and retrospective cohorts, the one randomized study was small and open-label, and the safety signals in pregnancy came from recollection rather than records. Several of the more ambitious therapeutic claims sit in narrative reviews, where proof of concept is ahead of controlled, long-term data in allergic patients.
If you only have time for one paper this week, make it the real-world comparison of upadacitinib and abrocitinib [2]. It gives a year of head-to-head data on two oral JAK1 inhibitors and reframes the choice between them as one of tolerability, cost and preference rather than efficacy.
That's your roundup for This Week in Allergy & Immunology. 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 Allergy & Immunology every week
By email, Spotify, Apple or any podcast app.
Spot something worth flagging?