Biologic class in psoriasis and the risk of later psoriatic arthritis

Design

retrospective real-world cohort, two university dermatology-rheumatology centres, 2008-2025

Population

393 adults with psoriasis on a biologic DMARD for at least six months; 257 with single-class exposure, 137 analysed by first biologic

Primary outcome

incident psoriatic arthritis during follow-up

Effect

86/393 (22%) developed PsA; non-TNF inhibitors adjusted OR 0.16-0.25 and HR 0.17-0.30 versus TNF inhibitors; no difference by first biologic received

Two university dermatology-rheumatology centres followed 393 adults treated with a biologic DMARD for psoriasis for at least six months between 2008 and 2025, to the point of a psoriatic arthritis (PsA) diagnosis or last visit. Eighty-six (22%) developed PsA.

Among the 257 with exposure to a single class, PsA was diagnosed more often on TNF inhibitors than on interleukin-17, interleukin-23 or interleukin-12/23 inhibitors. After adjustment, all the non-TNF classes carried substantially lower odds and hazards. The same pattern held when patients were grouped by the class used longest. It did not hold when they were grouped by the first biologic received, which is the analysis least vulnerable to later switching — an inconsistency the authors report and do not explain away.

This is retrospective and unrandomised, and the most likely alternative explanation is channelling: a patient with early joint symptoms is more likely to be started on, or moved to, a TNF inhibitor, so the drug follows the arthritis rather than causing it. That said, if you are choosing between classes for a psoriasis patient who has enthesitis, a family history of PsA, or nail or scalp disease, this is a real consideration to put alongside skin response and access. It is not yet a reason to switch a settled patient, and the concept of intercepting PsA needs a prospective trial.

• Ask every psoriasis patient starting a biologic about heel pain, dactylitis and inflammatory back pain
• Note nail and scalp involvement and family history — these mark the group most at risk of PsA
• Where classes are otherwise equivalent, this is a point in favour of IL-17 or IL-23 inhibition
• Do not switch a patient who is stable on a TNF inhibitor on the strength of a retrospective cohort
• In India, access and cost often decide class before anything else — say so to the patient rather than implying free choice

The statistics, in plain English

Adjusted odds ratios of 0.16 to 0.25 and hazard ratios of 0.17 to 0.30 look like large protective effects, but they come from an observational comparison in which the choice of drug was made by clinicians who could see the patient's joints. The first-biologic analysis, restricted to 137 patients, found no difference between classes — and that discordance is the most informative result here, because it is the analysis least distorted by switching. Twenty-two per cent developing PsA is higher than population estimates, as expected in a biologic-treated group with severe skin disease.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.