The relationship between in situ and invasive melanoma: An epidemiological study of melanoma in New Zealand, 2001-2017

Abstract

Background: Melanoma is diagnosed as either in situ or invasive disease. The relationship between in situ and invasive melanoma is unclear. If every in situ lesion is an early stage of invasive melanoma, diagnosis and removal of in situ melanoma should reduce the incidence of invasive melanoma and ultimately mortality from this disease. However, if disease progression is more complex, the excision of an early lesion may not prevent more advanced disease and may represent overdiagnosis. Aim: To compare the epidemiology of in situ melanoma and invasive melanoma in order to obtain a better understanding of the natural history of these diseases and their relationship to each other. Method: A systematic review (Part I) was carried out to describe the methods and results of published epidemiological studies that have compared the incidence, trends and characteristics of in situ and invasive melanoma. A population-based cohort study (Part II) of people identified from the New Zealand Cancer Registry (NZCR) who had been diagnosed with either in situ or invasive melanoma between 2001 and 2017 was conducted. The in situ and invasive melanoma data were compared in terms of incidence of the two disease, trends and key patient’s characteristics (age at diagnosis, sex, body site, ethnicity and geography). In addition, the risk of invasive melanoma among those with in situ melanoma was assessed with survival analysis and observed and expected invasive melanoma were compared in this subgroup. Results: The systematic review noted a wide use of age-standardised incidence rate and annual percentage change to describe the incidence trend of in situ and invasive melanoma. Various univariate and multivariate analysis methods were employed to determine the similarities or differences between in situ and invasive melanoma by patient characteristics. The systematic review found a dramatic increase in the incidence of in situ melanoma with no obvious decline in that of of invasive melanoma globally. Patients with in situ melanoma tended to be younger than those with invasive melanoma, and there were differences in in situ and invasive melanoma by ethnicity for incidence trends, but no consistent patterns in the two conditions were observed by sex or body site. In New Zealand, the incidence of in situ melanoma was found to have increased annually by 3.77% whereas that of invasive melanoma was relatively stable over the study period (annual increase 0.04%). In situ and invasive melanoma were similar in terms of patient sex and ethnicity, but differed by body site. It was difficult to compare in situ and invasive melanoma by age at diagnosis because this was highly influenced by body site and sex. The observed risk of invasive melanoma among patients with in situ melanoma was four times higher than that expected among the general population. The cumulative risk of invasive melanoma among patients with in situ melanoma was 5.6% at 5 years and 9.46% at 10 years. Conclusion: The relationship between in situ and invasive melanoma is complex. Not every in situ lesion was a precursor of invasive melanoma, but some did progress to invasive lesions.