Abstract: 【Abstract】 Objective To analyze the 1-year drug survival rates of dupilumab and Janus kinase （JAK） 1 inhibitors in the treatment of moderate-to-severe atopic dermatitis （AD）, and to identify potential predictors of treatment discontinuation. Methods This was a single-center retrospective cohort study. Clinical data were collected from patients with moderate-to-severe AD who initiated treatment with dupilumab or JAK1 inhibitors （upadacitinib or abrocitinib） at the Department of Dermatology, Chengdu Second People′s Hospital between May 2022 and October 2023, and were included in a 1-year drug survival analysis. Propensity score matching （PSM） with a 1∶1 ratio was used to balance baseline characteristics between the two groups. Survival analysis and Cox regression analysis were performed to assess the 1-year drug survival rates and predictors of treatment discontinuation in the dupilumab group and the JAK1 inhibitor group, respectively, and the incidence of adverse events was compared between the two groups before and after matching. Results The dupilumab group included 246 patients with moderate-to-severe AD （163 males ［66.3%］ and 83 females ［33.7%］）, with the age being 42.95 ± 21.50 years; the JAK1 inhibitor group included 102 patients （65 males ［63.7%］ and 37 females ［36.3%］）, with the age being 47.49 ± 18.69 years. During the 1-year follow-up, 101 （41.1%） patients discontinued dupilumab, while 35 （34.3%） discontinued JAK1 inhibitors. Kaplan-Meier analysis showed that the 1-year overall drug survival rates were 58.0% in the dupilumab group and 65.1% in the JAK1 inhibitor group, with no significant difference between the two groups by the log-rank test （χ2 = 1.92, P = 0.166）. After PSM, 93 patients were included in each group, and the baseline characteristics were comparable; the 1-year overall drug survival rates were 56.6% in the dupilumab group and 61.7% in the JAK1 inhibitor group, with no significant difference observed （χ2 = 0.68, P = 0.409）. Regarding predictors of treatment discontinuation, Cox regression analysis identified obesity as an independent risk factor for overall treatment discontinuation in the dupilumab group （HR = 5.290, 95% CI: 1.407 - 19.880, P = 0.014）; non-response at 1 month （HR = 3.953, 95% CI: 1.077 - 14.514, P = 0.038） and adulthood （compared with adolescence, HR = 0.211, 95% CI: 0.067 - 0.666, P = 0.008） were independent predictors of treatment discontinuation due to ineffectiveness in the dupilumab group. No predictors of treatment discontinuation were identified in the JAK1 inhibitor group. For safety outcomes, the incidence of adverse events was significantly higher in the JAK1 inhibitor group （19.6% ［20/102］） than in the dupilumab group in the original cohort （8.9% ［22/246］; χ2 = 7.73, P = 0.005）; after PSM, there was no significant difference in the incidence of adverse events between the JAK1 inhibitor group and the dupilumab group （17.2% ［16/93］ vs. 9.7% ［9/93］; χ2 = 2.26, P = 0.132）. Conclusions Dupilumab and JAK1 inhibitors demonstrated comparable 1-year overall drug survival rates in the treatment of moderate-to-severe AD, while the reasons for treatment discontinuation and adverse event profiles differed between the two therapies. Obesity was identified as a predictor of overall treatment discontinuation in the dupilumab group, while non-response at 1 month and adulthood （compared with adolescence） were predictors of dupilumab discontinuation due to ineffectiveness.

Key words: Dermatitis, atopic, Dupilumab, Janus kinase inhibitors, Upadacitinib, Abrocitinib, Drug survival, Real-world study