Machine Learning-Based Deep Phenotyping of Atopic Dermatitis: Severity-Associated Factors in Adolescent and Adult Patients
Laura Maintz,Thomas Welchowski,Nadine Herrmann,Juliette Brauer,Anna Sophie Kläschen,Rolf Fimmers,Matthias Schmid,Thomas Bieber,CK-CARE Study Group,Peter Schmid-Grendelmeier,Claudia Traidl-Hoffmann,Cezmi Akdis,Roger Lauener,Marie-Charlotte Brüggen,Claudio Rhyner,Eugen Bersuch,Ellen Renner,Matthias Reiger,Anita Dreher,Gertrud Hammel,Daria Luschkova,Claudia Lang
DOI: https://doi.org/10.1001/jamadermatol.2021.3668
2021-12-01
Abstract:Importance: Atopic dermatitis (AD) is the most common chronic inflammatory skin disease and is driven by a complex pathophysiology underlying highly heterogeneous phenotypes. Current advances in precision medicine emphasize the need for stratification. Objective: To perform deep phenotyping and identification of severity-associated factors in adolescent and adult patients with AD. Design, setting, and participants: Cross-sectional data from the baseline visit of a prospective longitudinal study investigating the phenotype among inpatients and outpatients with AD from the Department of Dermatology and Allergy of the University Hospital Bonn enrolled between November 2016 and February 2020. Main outcomes and measures: Patients were stratified by severity groups using the Eczema Area and Severity Index (EASI). The associations of 130 factors with AD severity were analyzed applying a machine learning-gradient boosting approach with cross-validation-based tuning as well as multinomial logistic regression. Results: A total of 367 patients (157 male [42.8%]; mean [SD] age, 39 [17] years; 94% adults) were analyzed. Among the participants, 177 (48.2%) had mild disease (EASI ≤7), 120 (32.7%) had moderate disease (EASI >7 and ≤ 21), and 70 (19.1%) had severe disease (EASI >21). Atopic stigmata (cheilitis: odds ratio [OR], 8.10; 95% CI, 3.35-10.59; white dermographism: OR, 4.42; 95% CI, 1.68-11.64; Hertoghe sign: OR, 2.75; 95% CI, 1.27-5.93; nipple eczema: OR, 4.97; 95% CI, 1.56-15.78) was associated with increased probability of severe AD, while female sex was associated with reduced probability (OR, 0.30; 95% CI, 0.13-0.66). The probability of severe AD was associated with total serum immunoglobulin E levels greater than 1708 IU/mL and eosinophil values greater than 6.8%. Patients aged 12 to 21 years or older than 52 years had an elevated probability of severe AD; patients aged 22 to 51 years had an elevated probability of mild AD. Age at AD onset older than 12 years was associated with increased probability of severe AD up to a peak at 30 years; age at onset older than 33 years was associated with moderate to severe AD; and childhood onset was associated with mild AD (peak, 7 years). Lifestyle factors associated with severe AD were physical activity less than once per week and (former) smoking. Alopecia areata was associated with moderate (OR, 5.23; 95% CI, 1.53-17.88) and severe (OR, 4.67; 95% CI, 1.01-21.56) AD. Predictive performance of machine learning-gradient boosting vs multinomial logistic regression differed only slightly (mean multiclass area under the curve value: 0.71 [95% CI, 0.69-0.72] vs 0.68 [0.66-0.70], respectively). Conclusions and relevance: The associations found in this cross-sectional study among patients with AD might contribute to a deeper disease understanding, closer monitoring of predisposed patients, and personalized prevention and therapy.