Upadacitinib Kept Atopic Dermatitis Under Control for Most of a Year in Italian Real-World Study

Patients with moderate-to-severe atopic dermatitis treated with upadacitinib spent an average of 253.3 equivalent days—roughly 70 percent of their first treatment year—in a state of moderate disease control, according to a 52-week multicenter real-world study conducted across Italian dermatology centers. The same patients accumulated an average of 176.1 equivalent days in optimal disease control, a stricter state combining near-complete skin clearance with minimal itch. The findings, drawn from the ongoing LANDSCAPE clinical registry and published in Dermatology and Therapy, offer a longitudinal view of treatment benefit that conventional single-timepoint response rates cannot capture.

The study, designated IL-AD (Italian Landscape Atopic Dermatitis), took a retrospective look at prospectively collected registry data from 433 adults receiving a stable dose of upadacitinib, an oral selective JAK1 inhibitor approved for moderate-to-severe atopic dermatitis. Of these, 315 patients (72.7 percent) received 15 mg once daily and 118 (27.3 percent) received 30 mg once daily. Patients whose dose was modified between 15 mg and 30 mg during the observation period were excluded so that longitudinal outcomes could be consistently attributed to a stable dose. The initial dose was chosen by treating physicians as part of routine clinical practice, not by any study algorithm or random allocation.

At week 52, 85.7 percent of patients had achieved EASI75, meaning at least a 75 percent improvement in the Eczema Area and Severity Index, while 72.7 percent had reached EASI90, a 90 percent or greater improvement. The composite endpoints told a similar story: moderate disease control, defined as simultaneous EASI75 plus a peak pruritus numerical rating scale (pp-NRS) score of 4 or less, was achieved by 80.6 percent of patients at week 52. Optimal disease control, operationalized as EASI90 plus pp-NRS of 1 or less and described as a minimal-disease-activity-like endpoint, was reached by 62.6 percent. Response was already substantial by week 16, when EASI75 stood at 82.4 percent, EASI90 at 64.2 percent, moderate control at 77.1 percent, and optimal control at 54.3 percent.

The time-in-target methodology is the study’s most distinctive contribution. Because disease-control status was assessed only at scheduled visits—baseline, week 16, week 32, and week 52—the researchers coded each control state as a binary variable at each visit and then applied the trapezoidal rule across the scheduled intervals, assuming linear changes between consecutive assessments. The resulting figures represent estimated equivalent days in target rather than directly observed daily status, with a theoretical maximum of 364 days corresponding to 52 weeks of continuous achievement. The authors caution that this approach cannot capture short-term fluctuations, transient flares, or brief losses of control between visits, and that more frequent assessments would yield more precise estimates.

Sustained response, defined as achieving a target at all three follow-up visits (weeks 16, 32, and 52), was stricter still. Sustained EASI75 was achieved by 71.1 percent of patients and sustained EASI90 by 48.7 percent. Sustained moderate disease control was observed in 63.0 percent, and sustained optimal control/MDA in 38.3 percent. These endpoints complement the cumulative analysis by showing that a substantial proportion of patients maintained response across multiple visits rather than merely at the final timepoint. No statistically significant differences between the 15 mg and 30 mg groups emerged for any sustained endpoint.

Objective and patient-reported measures both improved markedly throughout the year. Mean EASI fell from 20.3 at baseline to 1.9 at week 52, mean IGA dropped from 3.2 to 0.5, pp-NRS declined from 7.2 to 1.4, and sleep NRS fell from 5.7 to 0.5. The parallel improvement in clinician-assessed severity and patient-reported itch and sleep supports the broad clinical benefit of the drug in routine practice, the authors note, and is consistent with sustained improvements reported in the pivotal Measure Up 1 and Measure Up 2 trials through 52 weeks.

Stratified by dose, patients on 15 mg spent significantly more equivalent days in moderate disease control than those on 30 mg (261.0 versus 232.8 days, p = 0.018), and at week 16 the 15 mg group showed higher rates of EASI75 (84.8 percent versus 76.3 percent) and moderate control (80.6 percent versus 67.8 percent). The authors explicitly warn against interpreting this as evidence that 15 mg is more effective. Dose allocation was not randomized but reflected routine clinical decision-making, making confounding by indication a serious concern. Previous exposure to JAK inhibitors was significantly more frequent in the 30 mg group (7.6 percent versus 1.9 percent), suggesting that subgroup included more treatment-experienced or difficult-to-treat patients. The cumulative number of prior systemic therapies per patient was not systematically recorded and could not be reliably reconstructed, precluding adjusted analyses. By week 52, no significant between-dose differences remained for any efficacy endpoint, and the authors characterize all dose-stratified comparisons as exploratory and hypothesis-generating.

Safety data were collected cumulatively over the 52 weeks without stratification by visit date, and the timing of adverse event onset was not available. Overall, 117 of 433 patients (27.0 percent) experienced at least one adverse event, with nearly identical rates in the two dose groups (27.0 percent versus 27.1 percent). The most frequently reported events were hypercholesterolemia (8.3 percent), acne or JAKne (7.9 percent), creatine phosphokinase increase (4.2 percent), and hypertriglyceridemia (3.5 percent). Less common events included upper respiratory tract infections (2.3 percent), herpes zoster (2.1 percent), recurrent herpes simplex (1.8 percent), neutropenia (1.8 percent), transaminase increase (1.4 percent), and asthenia (1.2 percent). No adverse event led to permanent discontinuation of upadacitinib, and no statistically significant differences between dose groups were observed for the overall rate or any individual category.

The study’s limitations are substantial and acknowledged in detail. The observational, non-randomized design prevents causal inference. Excluding patients who underwent dose modification may have introduced selection bias: patients escalating from 15 to 30 mg often do so because of inadequate control, so their exclusion may have preferentially retained good responders and inflated effectiveness estimates, particularly in the 15 mg group, while excluding patients who reduced from 30 to 15 mg due to adverse events or satisfactory control could bias estimates in the opposite direction. The magnitude and direction of this bias cannot be quantified. Multiple comparisons were performed without adjustment for multiplicity, and the pragmatic optimal-control definition, lacking quality-of-life measures such as DLQI, should be regarded as MDA-like rather than a fully multidimensional definition. The 52-week follow-up also cannot address longer-term durability.

Despite these caveats, the authors argue that cumulative disease control fills a genuine gap in how treatment benefit is measured in a chronic, relapsing condition. Fixed-timepoint endpoints such as EASI75 or EASI90 at week 16 or week 52 cannot distinguish a patient who transiently improves from one who remains controlled for most of the year. A recent post hoc analysis of the Measure Up 1, Measure Up 2, and Heads Up trials found that upadacitinib-treated patients spent 9.8 to 13.4 times as many days with EASI90 response, and 7.0 to 10.3 times as many days with worst-pruritus NRS 0/1 response, as placebo patients through 16 weeks. The IL-AD analysis extends this time-in-target concept to routine clinical practice over a full year, using composite targets that combine skin clearance with itch control. The authors conclude that time-in-target metrics may complement conventional endpoints by better reflecting the persistence of therapeutic benefit, while longer follow-up is needed to determine whether this level of disease control persists beyond the first treatment year.

Subject of Research: Real-world 52-week effectiveness and cumulative disease control of upadacitinib in moderate-to-severe atopic dermatitis

Article Title: Cumulative Disease Control with Upadacitinib in Moderate-to-Severe Atopic Dermatitis—A 52-Week Multicenter Real-World Study: IL-AD (Italian Landscape Atopic Dermatitis)

Article References: Potestio, L., Bianco, M., Mastorino, L., Gaiani, F. M., Barei, F., di Vico, F., Bighetti, S., Luyo Saboya, R. A., Maurelli, M., Di Brizzi, E. V., Manzo Margiotta, F., Cristallo, M., Cavaliere, A. F., Magliulo, M., Carugno, A., Podo Brunetti, A., Ingurgio, R. C., Carpentieri, G., Messina, F., … Orsini, D. (2026). Cumulative Disease Control with Upadacitinib in Moderate-to-Severe Atopic Dermatitis—A 52-Week Multicenter Real-World Study: IL-AD (Italian Landscape Atopic Dermatitis). Dermatology and Therapy. https://doi.org/10.1007/s13555-026-01958-w

Image Credits: AI Generated

DOI: 10.1007/s13555-026-01958-w

Keywords: atopic dermatitis, upadacitinib, JAK1 inhibitor, disease control, EASI75, EASI90, minimal disease activity, time-in-target, real-world evidence, pruritus, LANDSCAPE registry, drug safety