Infections are the commonest adverse events on systemic psoriasis therapy, and choosing between an expanding list of biologics and oral small molecules requires knowing how they differ. The Spanish BIOBADADERM registry analysed 4,820 patients across 8,826 treatment cycles and 20,829 patient-years, calculating adjusted incidence rate ratios by propensity score analysis with adalimumab as comparator.
The reassuring headline first: serious infections were uncommon across all treatments, at 5.33% overall, with infliximab carrying the highest rate. The commonest serious infections were pneumonia, then COVID-19 pneumonia, then cellulitis. Among non-serious infections, respiratory tract infections led, followed by COVID-19 and urinary tract infections.
The differences between agents are where the clinical value sits. Risankizumab and ustekinumab carried significantly lower overall infection risk than adalimumab. Candida infection was significantly commoner with bimekizumab, brodalumab, secukinumab and ixekizumab — the interleukin-17 pathway agents, which is exactly what the biology predicts, since interleukin-17 signalling is central to mucocutaneous antifungal defence. Guselkumab, secukinumab, ixekizumab and ustekinumab were associated with fewer respiratory infections, and ixekizumab with less COVID-19, while dimethyl fumarate carried more.
That gives a usable framework for drug selection in a patient with a specific vulnerability. Recurrent candidiasis, poorly controlled diabetes, or dentures argue against an interleukin-17 agent and towards an interleukin-23 agent such as risankizumab or guselkumab. A patient with bronchiectasis or recurrent chest infection has a different calculus.
The limits are those of any registry. This is observational with propensity adjustment, so channelling — the tendency to give safer-seeming drugs to frailer patients, or newer drugs to those who failed older ones — cannot be fully removed, and newer agents have less accumulated follow-up than adalimumab. The Candida finding is the most secure, because it is large, consistent and biologically predicted.
- Avoid interleukin-17 pathway agents in patients with recurrent candidiasis or poorly controlled diabetes.
- Consider risankizumab or ustekinumab where overall infection risk is the dominant concern.
- Serious infections were uncommon overall at 5.33% — reassure patients accordingly.
- Warn patients on interleukin-17 agents about oral and genital candidiasis and how to recognise it.
- Interpret between-drug differences cautiously; newer agents have less follow-up than the adalimumab comparator.
The statistics, in plain English
Propensity score analysis matches patients on measured characteristics to imitate randomisation, and in a registry it addresses confounding by indication only as far as the recorded variables allow. Channelling bias is the specific worry here: prescribers choose newer agents for patients who have failed older ones, and safer-seeming agents for frailer patients, and neither tendency is fully captured in a registry field. The Candida finding is the most trustworthy result in the paper, not because its statistics are better but because it was predicted in advance by the biology of interleukin-17 signalling — a finding that matches a mechanism is less likely to be an artefact of who received what.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for dermatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free