(082) Use of Systemic and Topical Glucocorticoids and Risk of Skin Cancers in a Prospective Cohort of French Women.

Background: Evidence suggests that glucocorticoid use may increase skin cancer risk through immunosuppression. However, results from epidemiological studies have been inconsistent.

Objectives: To evaluate the associations between systemic and topical glucocorticoid use and the risk of basal cell carcinoma, squamous-cell carcinoma, and cutaneous melanoma.

Methods: E3N is a prospective cohort of 98,995 French women born between 1925 and 1950. From 2004, data from biennial self-reports on health and lifestyle information were linked with drug reimbursement data allowing identification of glucocorticoid use for each participant. Women were followed from 2004 until the date of first diagnosis of skin cancer or of any other cancer, date of the last returned questionnaire, or end of study period (November 17, 2014). We defined “ever” use of systemic or topical glucocorticoids as at least two reimbursements since July 1, 2004. Exposure was further classified by type of glucocorticoid, time since/age at first use, cumulative dose/number of reimbursements, and, for topical glucocorticoids, potency level. Cox regression models with age as the time scale and stratified by birth cohort were used to estimate hazard ratios (HRs) of skin cancers comparing users to non-users. Exposure was considered as a time-varying parameter and lagged by 1 year. Models were adjusted for phenotypic characteristics, family history of skin cancer, educational level, body mass index, smoking status, alcohol consumption, inflammatory/auto-immune diseases and use of concomitant drugs.

Results: Among 59,569 women free of cancer at the start of follow-up (mean age, 62 years at baseline) followed for a median of 9 years, 2552 cases of skin cancer were diagnosed (1672 basal cell carcinoma, 261 squamous-cell carcinomas, 434 melanomas and 185 of unknown type). Multivariable analyses showed no clear evidence of an association between ever use of any glucocorticoids and risk of basal cell carcinoma [HR=1.13 (1.00-1.28 ) for systemic use and HR=1.06 (CI 0.93-1.21) for topical use], squamous-cell carcinoma [HR=1.11 (0.82-1.51) for systemic use and HR=0.95 (95% CI 0.68-1.32) for topical use] and melanoma [HR=1.02 (0.81-1.30) for systemic use]. However, an inverse association was found between ever use of topical glucocorticoids and risk of melanoma [HR=0.75 (0.57-0.99)]. The glucocorticoids-skin cancer associations did not differ by glucocorticoid type, time since/age at first use, cumulative dose/number of reimbursements and potency level.

Conclusions: This study found no clear evidence of a positive association between the use of glucocorticoids and the risk of skin cancer. However, we found a negative association between the use of topical glucocorticoids and the risk of melanoma.