Real-World Maintenance of Remission in Atopic Dermatitis Patients With Upadacitinib: A Multicenter Retrospective...

Upadacitinib (UPA) is a selective Janus kinase 1 inhibitor that improves moderate-to-severe atopic dermatitis, but real-world strategies to maintain remission after response remain unclear. We conducted a multicenter retrospective study in Japan of 219 patients aged ≥ 12 years who achieved protocol-defined remission, defined as Investigator's Global Assessment (IGA) 0/1 at two visits ≥ 4 weeks apart during UPA treatment. Patients were classified as on-label (continuous daily 15 or 30 mg), spacing/reduction (extension of the dosing interval and/or dose-reduction after remission), or discontinuation (stopping UPA after remission). The primary endpoint was maintenance of remission over 48 weeks; secondary endpoints were time to relapse, outcomes after treatment modification, remission at 72 weeks after ≥ 24 weeks of sustained remission, and safety. Because the analysis was restricted to responders who achieved remission, generalizability to all UPA-treated patients is limited. Seventy-five percent of patients achieved IGA 0/1 within about 4 months of starting UPA. Over the subsequent 48 weeks, 127/200 (63.5%, 95% CI 56.6-69.9) maintained IGA 0/1. When stratified by post-remission dosing strategy (n = 193; excluding 7 who switched to other systemic therapy), remission at 48 weeks was maintained in 93/145 (64.1%, 95% CI 55.8-71.9) in the on-label group, 24/32 (75.0%, 95% CI 56.6-88.5) in the spacing/reduction group, and 6/16 (37.5%, 95% CI 15.2-64.6) in the discontinuation group. Among patients with ≥ 24 weeks of sustained remission, week-72 IGA 0/1 maintenance rates were 70.7% (95% CI 60.7-79.1; n = 107), 100% (95% CI 63.1-100.0; n = 8), and 60.0% (95% CI 20.0-90.0; n = 5) in the on-label, spacing/reduction, and discontinuation groups, respectively; estimates for spacing/reduction and discontinuation were imprecise due to small sample sizes. Longer on-label treatment before tapering or discontinuation tended to be associated with a lower relapse risk. UPA showed a safety profile consistent with previous studies. Acne, herpes simplex, herpes zoster, and folliculitis were the most frequent adverse events, and most were mild and manageable, although one herpes zoster case required temporary interruption. In this responder-enriched real-world Japanese cohort, UPA induced remission in most patients within 4 months and allowed approximately two-thirds to maintain remission over 48 weeks. Observed outcomes suggest that maintaining on-label dosing and cautious spacing/reduction after sustained remission may help preserve disease control, whereas discontinuation may be associated with a higher risk of relapse; however, comparisons between dosing strategies are descriptive and hypothesis-generating. Prospective studies are needed to confirm these findings.