697 - Risk of non-melanoma skin cancer in patients with moderate-to-severe atopic dermatitis: a United States population-based study using claims data

Mark Lebwohl,Emma Yue,Whitney Krueger,Brian Berman,Christopher G Bunick,Todd Schlesinger,Ayman Grada

DOI: https://doi.org/10.1093/bjd/ljae266.071

IF: 11.113

2024-08-01

British Journal of Dermatology

Abstract:Abstract Introduction/Background Non-melanoma skin cancer (NMSC), including basal and squamous cell carcinoma, are associated with prolonged intermittent sun exposure (specifically ultraviolet radiation). Atopic dermatitis (AD) is a chronic inflammatory skin disease associated with multiple comorbidities.1,2 While AD is associated with an increased risk of skin cancers,3 data on the underlying risk of NMSC in patients with AD are inconsistent.1 Objective To evaluate NMSC incidence and risk in patients with AD compared with a non-AD matched control cohort and patients with rheumatoid arthritis (RA). To evaluate the effect of disease activity, analyses were repeated in patient subgroups with moderate-to-severe disease. Methods This retrospective observational study used US claims data from the Optum Clinformatics Data Mart. Eligible patients were aged ≥ 18 years with diagnosed AD (≥ 2 claims for AD or ≥ 1 claim for AD or eczema with asthma and/or hay fever, food allergies, or allergic rhinitis) between March 2017–March 2023. The cohort entry date was the date of the first qualifying disease diagnosis. Patients were classified as having moderate-to-severe disease if they received dupilumab for AD or advanced systemic therapy for RA during follow-up. Comparator groups included non-AD control cohorts (matched 1:1 to the AD and moderate-to-severe AD cohorts, respectively, by age [± 1 year], sex, and cohort entry date), patients diagnosed with RA (≥ 2 claims ≥ 7 days apart filed by a rheumatologist), and patients with moderate-to-severe RA. Crude NMSC incidence rates were reported. Relative risk was calculated using multivariable Cox proportional hazard models adjusted for baseline demographics, comorbidities, and medications. Results This analysis included data from 391,753 patients with AD (7136 with moderate-to-severe AD) and 97,445 patients with RA (35,846 with moderate-to-severe RA). The matched AD and non-AD cohorts each included 381,221 patients. The matched moderate-to-severe AD and non-AD cohorts each included 7134 patients. The mean (SD) age in years was 58.1 (18.8) for AD and non-AD cohorts after matching and 67.0 (13.6) for RA. The median (IQR) follow-up time in days was 1087 (487–1485) for AD, 1218 (548–1782) for RA, and 1013 (376–1562) for non-AD controls. The NMSC incidence per 100 patient-years (95% CI) was 2.12 (2.10, 2.15) for AD and 1.74 (1.72, 1.77) for matched non-AD controls, 2.11 (1.87, 2.35) for moderate-to-severe AD and 1.28 (1.10, 1.47) for matched non-AD controls, 2.33 (2.28, 2.38) for RA, and 2.03 (1.94, 2.12) for moderate-to-severe RA. The relative NMSC risk (adjusted hazard ratio [95% CI]) was greater in patients with AD vs matched non-AD controls (1.32 [1.30, 1.35]) and moderate-to-severe AD vs matched non-AD controls (1.36 [1.12, 1.65]. There was no significant difference in NMSC risk in patients with AD compared with RA (1.03 [1.00, 1.06]) or moderate-to-severe AD compared with moderate-to-severe RA (0.97 [0.87, 1.08]). On average, NMSC risk was > 6 times higher in patients with AD with a history of NMSC vs those without; history of other malignancies and organ transplantation were also associated with increased NMSC risk. NMSC risk was ≥ 35% lower in patients who were female vs male, and patients who were Asian, Hispanic, or Black vs White. Conclusions Patients with AD demonstrated a higher NMSC risk compared with non-AD matched controls, and a similar NMSC risk compared with patients with RA; patterns were consistent for patients with moderate-to-severe disease. NMSC risk was higher in patients with AD with a history of NMSC or other malignancies. As a limitation, patients with AD were commonly examined by dermatologists who were likely to look for and find NMSC. Characterizing the underlying NMSC risk among patients with AD may inform treatment benefit-risk assessments.

dermatology