Melanoma and additional primary cancers

The aim of this study was to examine the occurrence of additional, unrelated primary cancers in patients with melanoma. Data from the hospital-based, melanoma registry of a specialist unit (the Melanoma Unit at Charing Cross Hospital, London, UK) were compared with the incidence rates in a population-based cancer registry (the Thames Cancer Registry). In total, 2076 patients with the histological diagnosis of melanoma established between 1960 and 1997 who were registered with the Melanoma Unit at Charing Cross Hospital, were included in the study. Patterns in time and in the tumour type of the additional cancers were analysed in the cohort. The relative risk of subsequent cancers was evaluated, the number of expected cancers being calculated by applying incidence rates in the population of south-east England to the person-years of follow-up in the cohort. Sixty-six (3%) of the 2076 evaluable patients had a history of, or developed, 69 histologically verified additional cancers, the commonest being colorectal, breast and lymphoma. Twenty-six additional cancers preceded the diagnosis of melanoma by 1-42 years, 16 were diagnosed within 12 months and 27 followed the diagnosis of melanoma, 1-28 years later. Seven cancers occurred after chemotherapy for metastatic melanoma: two colorectal, one bladder, one renal, two myelodysplasias and one acute myeloid leukaemia. The relative risk of additional cancers developing after the diagnosis of melanoma during the 8537 evaluable years-at-risk was 0.64 (95% confidence interval 0.41-0.96; P = 0.0006). Thus, the risk of additional cancers following treatment for metastatic melanoma in this cohort is small. The potential influence of other factors on the occurrence of additional cancers observed overall in this study requires further investigation.