In atopic dermatitis remission, oral agents relapse early when stopped; biologics hold

Design

Systematic review and network meta-analysis

Population

55 studies; network of 9 studies, 13 regimens, 2,585 observations

Primary outcome

Sustained EASI-75 16-52 weeks after withdrawal, dose reduction or interval extension

Effect

Early relapse after oral withdrawal; lebrikizumab lowest risk of EASI-75 loss in the network

A systematic review and network meta-analysis (British Journal of Dermatology, 3 October 2026) brought together 55 studies on what happens to atopic dermatitis after remission when systemic treatment is withdrawn, reduced or given less often. The network comparison drew on 9 studies, 13 regimens and 2,585 observations, with sustained EASI-75 response between 16 and 52 weeks as the outcome.

Stopping oral treatments was followed by early relapse. Biologic responses were largely sustained, even after withdrawal or longer dosing intervals. In the network, lebrikizumab withdrawal and interval extension carried the lowest risk of losing EASI-75 compared with stepping down other agents.

Patients often ask to reduce treatment once their skin clears, and some must pause for pregnancy or vaccination. This analysis suggests the drug class should shape how that is done. With an oral agent, abrupt withdrawal is likely to bring disease back quickly, so plan close review or a gradual step-down. With a biologic, extending the interval is a reasonable first step.

The comparisons rest on few trials per regimen and mostly on industry withdrawal studies, so the ranking between individual biologics should be held loosely.

• Discuss the risk of early relapse before stopping an oral systemic agent in remission
• Consider extending the dosing interval of a biologic as a first step-down rather than stopping
• Book early review after any oral agent is stopped
• Plan pauses for pregnancy or live vaccines with the relapse pattern of the drug in mind

Why it matters

Step-down plans for eczema should differ by drug class, which most protocols do not yet reflect.

Don't overread it

The ranking between individual biologics comes from few trials and indirect comparisons.

The statistics, in plain English

A network meta-analysis compares regimens that were never tested head to head by linking them through common comparators. That widens uncertainty, and with only nine studies in the network, rankings between drugs are less reliable than the broad oral-versus-biologic pattern.

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.