JAK inhibitors have been approved rapidly across inflammatory dermatology, and prescribing has run ahead of any consolidated picture of how they compare. This systematic review pulled together 68 studies — 42 randomised trials, 14 open-label extensions and 12 real-world cohorts — covering 15,427 patients across atopic dermatitis, psoriasis, vitiligo, alopecia areata and hidradenitis suppurativa.
The efficacy figures are worth having as ranges rather than point estimates, because that is what the underlying data support. Oral JAK inhibitors achieved EASI-75 in 38 to 73% in atopic dermatitis, PASI-75 in 33 to 67% in psoriasis, and SALT-50 in 30 to 62% in alopecia areata. Topical formulations worked in atopic dermatitis and vitiligo but showed limited benefit in alopecia areata — a useful distinction, since the hair follicle sits deeper than a topical reliably reaches.
Safety followed the expected pattern. Upper respiratory infections in 5 to 14% and herpes zoster reactivation in 1 to 3%, both above placebo. Serious events — major cardiovascular events, venous thromboembolism, malignancy — were rare but did occur, which is the phrasing that matters: rare is not never, and these are the events that determine who should not receive the drug.
The comparative conclusions are the softest part and are labelled as such. Selective JAK1 inhibitors appeared to offer the best efficacy-to-safety balance in atopic dermatitis, and JAK1/2 inhibitors showed numerically higher efficacy in alopecia areata without reaching significance. Both rest on indirect comparison across trials that were never run against each other.
For practice, three things follow. Quote the response rate as a range and set expectations accordingly — a third to two-thirds is the honest answer, not a single figure. Screen and vaccinate for herpes zoster before starting where that is available, given a 1 to 3% reactivation rate in a drug taken for years. And take the cardiovascular risk assessment seriously rather than as paperwork, since the patients in whom rare serious events concentrate are identifiable in advance. In Indian practice, add latent tuberculosis screening, which this predominantly Western evidence base does not emphasise and which matters considerably here.
- Quote response as a range — roughly a third to two-thirds achieve the standard endpoint, depending on disease and drug.
- Topical JAK inhibitors work for atopic dermatitis and vitiligo but not reliably for alopecia areata.
- Screen for and vaccinate against herpes zoster before starting where available; reactivation ran 1 to 3%.
- Do a proper cardiovascular and thrombotic risk assessment; serious events are rare but concentrate in identifiable patients.
- Screen for latent tuberculosis before starting — a gap in the mostly Western evidence base that matters here.
The statistics, in plain English
The wide efficacy ranges — 38 to 73% for EASI-75, for instance — are ranges across studies, not confidence intervals, and they reflect genuine differences in drug, dose, disease severity and trial population rather than statistical uncertainty. That is why the review reports ranges rather than pooling: the studies are too different to average meaningfully. The comparative claims about JAK1 versus JAK1/2 selectivity come from indirect comparison, which means inferring how two drugs compare from how each performed against placebo in separate trials — a method that is vulnerable to differences between those trials and is much weaker than a head-to-head study.
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