Personal history of non-melanoma skin cancer diagnosis and death from melanoma in women

Abstract

Melanoma incidence is increasing. We evaluated risk of melanoma death after diagnosis of non-melanoma skin cancer (NMSC). We followed 77,288 female American nurses from the Nurses’ Health Study from 1986 to 2012. We used Cox proportional hazards models to determine the hazard ratio (HR) of lethal and non-lethal melanoma diagnosis and melanoma death, according to personal NMSC history. Among melanoma cases, we examined the HR of melanoma death and the odds ratio (OR) of melanoma with a Breslow thickness ≥0.8 mm or Clark's levels of IV and V according to history of NMSC. We documented 930 melanoma cases without NMSC history and 615 melanoma cases with NMSC history over 1.8 million person-years. The multivariate-adjusted HR (95% confidence interval) of melanoma death associated with personal history of NMSC was 2.89 (1.85–4.50). Women with history of NMSC were more likely to develop non-lethal melanoma than lethal melanoma (HR (95% CI): 2.31 (2.05–2.60) vs. 1.74 (1.05–2.87)). Among melanoma cases, women with history of NMSC had a non-significant decreased risk of melanoma deaths (0.87 (0.55–1.37)), Breslow thickness ≥0.8 mm (0.85 (0.59–1.21)) and Clark's levels IV and V (0.81(0.52–1.24)). Women with NMSC history were less likely to be diagnosed with a lethal melanoma than a non-lethal melanoma, but overall rate of melanoma diagnosis was increased in both subtypes, leading to the increased risk of melanoma death. Our findings suggest the continued need for dermatologic screening for patients after NMSC diagnosis, given increased melanoma risk. Early detection among NMSC patients may decrease deaths from melanoma.