Atopic dermatitis (AD) represents the most prevalent chronic inflammatory skin disorder in children, with a globally rising incidence. Current epidemiological data indicate that 15–20% of children are affected by AD, and approximately 60% of these patients develop symptoms within the first 12 months of life. This condition imposes a substantial disease burden on the pediatric population (1).

The precise etiology of AD remains incompletely understood. Current management primarily relies on conventional topical therapies, such as corticosteroids and calcineurin inhibitors. However, for children with moderate-to-severe AD, these treatments often yield suboptimal efficacy, frequently necessitating adjunctive systemic interventions (e.g., corticosteroids or immunosuppressants). Nevertheless, long-term use of such systemic agents carries significant safety concerns, particularly in the developing pediatric population. These include potential impacts on growth and development and potential disease recurrence upon discontinuation (2). In contrast, dupilumab offers a promising alternative due to its favorable efficacy and safety profile.

Dupilumab is a monoclonal antibody targeting the interleukin (IL)-4 receptor alpha (IL-4Rα), thereby inhibiting signaling mediated by both IL-4 and IL-13. By blocking IL-4Rα, dupilumab suppresses downstream Th2 inflammatory mediators such as Immunoglobulin E (IgE) and eotaxin-1 (CCL11), modulates immune dysregulation, and interrupts the chronic inflammatory cascade (3). Following its initial approval in China in 2021 for patients aged ≥6 years with moderate-to-severe AD, dupilumab’s indication was extended in 2023 to include children as young as 6 months. It now stands as the only biologic agent approved in China for pediatric moderate-to-severe AD.

Multiple randomized controlled trials (RCTs) have demonstrated dupilumab’s efficacy in significantly improving AD signs, symptoms, and quality of life, alongside a favorable safety profile (4, 5). Nevertheless, real-world evidence regarding its effectiveness and safety in Chinese children remains scarce, with long-term data particularly limited, given its relatively recent introduction to this population.

Therefore, this study aims to explore the actual efficacy and safety profile of dupilumab in Chinese pediatric AD patients through systematic analysis of longitudinal clinical data. We further sought to characterize real-world response profiles and safety patterns, thereby providing region-specific evidence to support its clinical application in pediatric practice.

This single-center prospective observational cohort study was conducted at the China Air Force Medical Center from January 2022 to April 2025 to evaluate the real-world effectiveness and safety of dupilumab in pediatric patients with moderate-to-severe AD. The study protocol received approval from the hospital’s Ethics Committee (Approval Number: 2022-28-PJ01). Written informed consent was obtained from all legal guardians, and verbal assent was additionally provided by children aged ≥7 years.

The study protocol planned to include children aged 6 months to 12 years with moderate-to-severe AD, using modified Hanifin & Rajka criteria for children <4 years and Williams criteria for those ≥4 years for diagnosis. All participating physicians completed standardized training in AD diagnosis, encompassing Hanifin & Rajka/Williams criteria and severity grading. Diagnosis and severity classification for every patient underwent independent verification by a second senior physician. One discrepant case was resolved through expert panel arbitration, achieving full consensus.

Disease severity was defined by Eczema Area and Severity Index (EASI) scores: EASI ≥7 to <21 indicated moderate AD, while EASI ≥21 indicated severe AD. Patients with significant systemic comorbidities or active infections were excluded. Dosing regimens were strictly stratified by body weight: children weighing 5 kg to <15 kg received 200 mg subcutaneously every 4 weeks without a loading dose; those weighing ≥15 kg to <30 kg aged under 6 years received 300 mg subcutaneously every 4 weeks without a loading dose. For children aged 6–12 years weighing ≥15 kg to <30 kg, the initial dose was 600 mg (administered as two consecutive 300 mg subcutaneous injections), followed by 300 mg every 4 weeks; those weighing ≥30 kg to <60 kg received an initial 400 mg dose (two consecutive 200 mg injections) followed by 200 mg every 2 weeks; those weighing ≥60 kg received an initial 600 mg dose (two consecutive 300 mg injections) followed by 300 mg every 2 weeks. Efficacy assessments included dynamic monitoring of EASI, itch severity scores, and quality-of-life metrics, with systematic surveillance of treatment-emergent adverse events (TEAEs) throughout the median 33-week follow-up period.

During the baseline period, demographic characteristics, atopic comorbidities, prior treatment history, and disease severity measures were collected. These included the EASI, Investigator’s Global Assessment (IGA), Peak Pruritus Numerical Rating Scale (P-NRS), and quality-of-life scores (≤4 years: Infant’s Dermatitis Quality of Life Index, IDQOL; >4 years: Children’s Dermatology Life Quality Index, CDLQI). Efficacy and safety were evaluated according to a standardized visit schedule at Weeks 2, 4, 8, 12, and 16, followed by assessments every 8–12 weeks thereafter until treatment discontinuation (maximum follow-up: Week 96). At each visit, EASI, IGA, P-NRS, Atopic Dermatitis Control Tool (ADCT) scores, and TEAEs were documented. The primary efficacy endpoint was the percentage change from baseline in EASI score at Week 16. Secondary endpoints included safety metrics (TEAE incidence graded per CTCAE v5.0) and duration of therapy distribution.

By April 30, 2025, 59 pediatric AD patients treated with dupilumab were enrolled. One patient discontinued treatment within 72 hours post-initial dose due to a significant rash exacerbation (EASI increased from baseline 28.9 to 40.2) and was included only in the safety analysis but excluded from efficacy assessments. The remaining 58 patients completed ≥16 weeks of standardized treatment and comprised the final efficacy analysis cohort.

All statistical analyses were performed using SPSS version 27.0 (IBM Corporation, NY, United States) and R version 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria). Quantitative data are expressed as median (interquartile range, IQR), with between-group comparisons analyzed using the Mann-Whitney U test. Categorical data are reported as proportions (%), with intergroup comparisons analyzed using the chi-square test; a P-value < 0.05 was considered statistically significant.

This prospective real-world cohort study evaluated dupilumab in pediatric patients aged 6 months to 12 years with moderate-to-severe AD, with follow-up extending to 96 weeks. As the first biologic agent targeting the IL-4/IL-13 pathway, dupilumab demonstrates substantial efficacy and favorable safety in AD management. In China, where dupilumab has not been widely adopted—particularly for pediatric AD—existing limited domestic studies align with international reports regarding effectiveness. Nevertheless, comprehensive data on long-term treatment remain scarce. Our findings further validate the sustained efficacy of dupilumab in Chinese children with AD (EASI-75 response rate: 68.97% at Week 16; maintained at >58.82% during extended therapy), while also reporting unique safety observations including early-phase paradoxical eczema flares (incidence: 5.08%) and a late-onset pustular dermatitis case. These results augment the real-world safety evidence for dupilumab in clinical practice settings.

Our findings demonstrated superior efficacy outcomes at Week 16 relative to pivotal RCTs, evidenced by an EASI-75 response rate of 68.97% which surpassed rates reported in the Lancet Phase 3 trial for children aged ≥6 months (53.0%), a Japanese Phase 3 study (43.3%), and a multinational European/North American clinical trial (46.0%) (4, 6, 7). Cross-comparisons within the real-world evidence landscape revealed: The EASI-75 response rate in our study exceeded that of a small Czech cohort (75.0% at Week 16, n=24) and short-term improvements in a US cohort (mean EASI improvement: 39.6% ± 29.9% at 12–24 weeks, n=89), but fell below a large Spanish study (79.4% at Week 16, n=243). It tended to be comparable to the IGA 0–1 achievement rate in an Israeli cohort at 60 weeks (65.0%, n=230). Within Chinese domestic evidence, our 69.0% response rate was numerically higher than previous Chinese data (63.2%, n=57). Regarding long-term efficacy, the observed >16-week treatment group exhibited an EASI-75 maintenance rate >58%, aligning with sustained improvement patterns in an Italian cohort at 52 weeks (86.8%, n=91) and a Spanish cohort at 52 weeks (85.8%). Notably, subjective symptoms showed remarkable improvement: The reduction in Peak Pruritus Numerical Rating Scale (P-NRS) reached 68.2%, surpassing both the objective sign improvement rate reported in the US cohort (39.6% ± 29.9%) and pruritus relief levels in most European/American RCTs (44.9%–49.4%). This discrepancy may correlate with ethnicity- and geography-dependent itch perception characteristics (8–13). Collectively, these findings demonstrate that dupilumab’s efficacy profile in Chinese pediatric AD patients aligns with the global RCT and real-world evidence framework, while its outstanding subjective symptom control offers a clinically valuable therapeutic option for this population.

At Week 16, most patients achieved significant clinical response (EASI-75). Subgroup analysis revealed no statistically significant difference in response rates between the standard-duration and extended-duration groups (P>0.05), suggesting prolonged treatment may not confer additional clinical benefit—a finding contrasting with Paller et al.’s proposition that “extended therapy enhances outcomes” (14). The observed differences in this study may stem from heterogeneity in the enrolled population. Given the relatively small sample size, the statistical power may have been insufficient to detect potential minor clinical benefits. It must be emphasized that “statistical non-significance” does not necessarily equate to “absence of clinical benefit.” Future studies should expand cohorts and incorporate biomarkers to further explore these effects.

During extended follow-up, we observed reduced EASI responses in 3 patients (5.17%) after Week 16 (mean decline: 25.39%). Symptomatic fluctuations during later treatment phases are often attributed to anti-drug antibodies (ADAs); however, existing evidence indicates pediatric ADA incidence is merely 2%–5.3%, typically manifesting as transient low-titer reactions (<1,000), with spontaneous antibody decay occurring during long-term therapy. Consequently, ADA-induced efficacy fluctuations remain rare (15). Notably, reduced frequency of adjunctive topical therapies (particularly emollients) post-symptom improvement was observed in our cohort, potentially contributing to symptom recurrence during dupilumab treatment. Multiple evidence-based studies confirm synergistic effects between dupilumab and topical emollients, aligning with AD’s multifactorial pathogenesis (16). Moreover, American Academy of Dermatology guidelines emphasize “biologics plus topical therapy” as the optimal strategy for sustained disease control (17). Therefore, in clinical practice, symptom fluctuations should not be reflexively ascribed to ADAs, warranting premature discontinuation or dose adjustment. Instead, standardized therapeutic management prioritizing synergistic integration of foundational skincare with biologics is essential to ensure treatment continuity and efficacy.

In this study, treatment discontinuation due to inadequate response occurred in 3 patients (5.08%), lower than the 9.5% non-response rate reported in adult studies (18). This discrepancy may stem from distinct immunophenotypic profiles between pediatric and adult AD. The pathogenesis of AD involves complex immune pathways; certain patients may exhibit dominant non-Th2-mediated inflammation (e.g., Th17 or JAK-STAT activation), potentially limiting dupilumab response (19, 20). Consequently, future research should investigate whether quantitative profiling of pretreatment inflammatory biomarkers (e.g., IL-4, IL-13, IgE) could guide personalized therapeutic strategies, thereby enhancing efficacy while reducing unnecessary drug exposure and associated adverse effects.

Regarding safety, the overall incidence of TEAEs in our study (25.42%) was lower than rates reported in international RCTs (64%–66.7%) (4, 6). Consistent with prior literature, ocular events constituted the primary TEAEs (10.17%), with a moderately higher incidence compared to Western studies—a discrepancy potentially attributable to distinct ocular surface anatomy in East Asian children. We also observed higher AETE risk in children with severe AD, consistent with the studies by Akinlade and Nahum et al.: baseline AD severity is an independent risk factor for dupilumab-associated conjunctivitis (mechanism remains unclear) (21, 22). Although the results are associated with this, it should be noted: smaller sample size may lead to insufficient statistical power (Type II error risk), thus requiring cautious interpretation of this phenomenon.

Additionally, by blocking the IL-4/IL-13 signaling pathway, dupilumab inhibits eosinophil tissue migration, leading to transient retention in peripheral blood. This mechanism may contribute to elevated eosinophil counts and related adverse events (23). However, no hypereosinophilia (≥1.5×10⁹/L) or associated clinical symptoms were observed in our study, consistent with reports by Paller and Wollenberg et al. (incidence: 3.4%, transient nature) (4, 24). Although the vast majority of elevations are benign, asymptomatic events requiring no intervention, long-term therapy warrants vigilance for rare complications (e.g., hypereosinophilic syndrome). Monitoring eosinophil counts during treatment remains essential; such simple laboratory surveillance can effectively prevent severe adverse events and enable proactive risk management.

Notably, this study is the first to systematically document two unique cutaneous reactions associated with dupilumab treatment in pediatric AD: paradoxical eczema flares and incident pustular dermatitis. Three patients (5.08%) exhibited transient paradoxical rash exacerbation during initial treatment, a rarely reported phenomenon in pediatric AD. The mechanism may involve Th1/Th17 immune deviation secondary to Th2 pathway inhibition, whereby IL-4/IL-13 blockade disinhibits the IL-23/IL-17 axis, triggering keratinocyte IL-36β overexpression (25). For the acute pustular rash with fever and arthralgia occurring 12 hours after the 10th dose, the proposed mechanism requires consideration of dupilumab’s microbial modulation effects: The agent achieves rapid Staphylococcus aureus clearance (26), but delayed restoration of commensal microbiota equilibrium may promote Malassezia colonization dominance (27). Concurrently, IL-13 suppression downregulates antimicrobial peptide expression, compromising cutaneous innate immunity. This dual microbial-immune imbalance state, combined with persistent immune deviation (IL-23/IL-17 pathway activation inducing IL-36γ overexpression), could synergistically trigger acute pustular inflammation. As no skin biopsy was performed, the microbial infiltrate profile and IL-36 pathway activation remain unverified, constituting a significant study limitation.

While mechanistic underpinnings require further investigation, these findings suggest that acute rash exacerbation post-initial dupilumab administration should raise suspicion for transient paradoxical reactions. Furthermore, during extended therapy, vigilant monitoring is warranted for immunomodulation-induced shifts in inflammatory thresholds, which may elevate risk exposure to rare adverse events.

This was a single-center study, which may limit the generalizability of the findings. The limited sample size could also affect the ability to detect rare events. these factors could explain the differences with other studies. Additionally, the lack of detailed comorbidity severity data (e.g., asthma control levels, allergic rhinitis classifications) restricted further understanding of how underlying disease complexity might influence dupilumab treatment outcomes. Future prospective studies should comprehensively evaluate dupilumab application in children by incorporating objective comorbidity severity measures and expanding sample sizes, with a focus on addressing unresolved challenges in real-world pediatric AD management, including: tracking the long-term impacts of dupilumab beyond 96 weeks of therapy on child growth, development, and other organ systems via multicenter large-scale cohorts; elucidating the mechanisms by which ADA seropositivity contributes to reduced efficacy through ADA testing and trough concentration analysis; developing predictive models that integrate inflammatory cytokine profiles (e.g., IL-4, IL-13, IL-22) and microbiome signatures (e.g., Malassezia colonization), with prospective validation of their utility in predicting treatment response and specific adverse events; ultimately enabling safer and more effective personalized treatment strategies for children with AD.

Dupilumab demonstrated significant efficacy and favorable safety in Chinese pediatric patients (aged 6 months to 12 years) with moderate-to-severe AD. At Week 16, the EASI-75 response rate reached 68.97%, with a long-term maintenance rate exceeding 58.82%. Unlike previous reports, this study showed no additional improvement in response rates for children treated >16 weeks compared to those receiving 16 weeks of standard therapy; larger sample sizes are needed to confirm this observation. Furthermore, this study is the first to systematically document early-treatment paradoxical rash exacerbation (incidence: 5.08%) and rare de novo pustular rash (1.69%), highlighting the clinical relevance of these special events. Ocular adverse events constituted the most common mild-to-moderate side effects (13.56%). These findings provide crucial locally generated evidence for dupilumab’s clinical application in Chinese children with AD and supplement surveillance data on unique adverse drug reactions in this pediatric population.​