Risk of subsequent primary melanoma in 68 002 five-year survivors of childhood and adolescent cancer in Europe: the PanCareSurFup cohort study

BACKGROUND: Survivors of childhood cancer are at risk of subsequent primary melanoma, but the magnitude of the risk after different childhood cancer types and beyond age 50 years remains inadequately characterized. We quantified risks in the largest cohort of childhood cancer survivors worldwide. METHODS: The PanCare Childhood and Adolescent Cancer Survivor Care and Follow-Up Studies cohort includes 68 002 five-year childhood cancer survivors across Europe (diagnosed 1940-2008). Subsequent primary melanoma risks were quantified using standardized incidence ratios (SIRs), absolute excess risks (AERs), cumulative incidence, and relative risks (RRs). RESULTS: Over 1 239 675 person-years, 197 subsequent primary melanoma were ascertained, whereas 84.9 were expected (SIR = 2.3, 95% confidence interval [CI] = 2.0 to 2.7). Although the standardized incidence ratio decreased with attained age (P for trend < .001), it remained increased at 2.1-fold (95% CI = 1.5 to 3.1) at age 50 years and older. By age 65 years, cumulative incidence was 1.2% (0.8% expected). Of all cancer types, heritable retinoblastoma survivors had the greatest risk (SIR = 16.5, 95% CI = 10.8 to 25.3; AER = 94.4, 95% CI = 59.9 to 148.9), with cumulative incidence of 3.3% at age 50 years and 5.5% at age 60 years. Survivors treated with radiotherapy had a 70% greater risk than those treated without (RR = 1.7, 95% CI = 1.1 to 2.7), increasing to 4.0-fold by age 50 years and older (RR = 3.9, 95% CI = 1.1 to 13.9). For chemotherapy, no statistically significant association was found (RR = 0.9, 95% CI = 0.5 to 1.6). CONCLUSIONS: Childhood cancer survivors, particularly those with heritable retinoblastoma and those treated with radiotherapy, remain at increased risk of melanoma beyond age 50 years. Current survivorship guidelines focus principally on treatment history; however, our findings suggest that long-term follow-up guidelines should additionally include childhood cancer type and attained age as risk-stratifying factors, particularly where individual treatment records are unavailable.