Risk of subsequent melanoma after melanoma in situ and invasive melanoma: a population-based study from 1973 to 2011.

Abstract

Patients with melanoma in situ are at an increased risk of subsequent melanoma compared with the general population, but the risk of subsequent melanoma after initial melanoma in situ versus after initial invasive melanoma is not known. We sought to compare the risk of subsequent melanoma in the cohort whose first cancer was melanoma in situ to the risk in the cohort whose first cancer was invasive melanoma. In this cohort study, we identified individuals whose first cancer was either melanoma in situ or invasive melanoma from the Surveillance, Epidemiology, and End Results (SEER) program between 1973 and 2011 and used Cox proportional hazards models for comparison. Compared with the invasive melanoma cohort, the melanoma in situ cohort was more likely to develop subsequent melanoma of any stage after 2 years, subsequent invasive melanoma after 10 years, and subsequent melanoma in situ at all the time points (P < .001, P = .003, P < .001, respectively). Underreporting of melanomas, particularly melanoma in situ cases, and missing cases of subsequent melanomas as a result of patient migration from the SEER registry areas could affect results. Given the increased long-term risk of subsequent melanoma in the melanoma in situ cohort, the patients with melanoma in situ diagnosis may benefit from a long-term surveillance for subsequent melanomas.