The risk of subsequent invasive melanoma after a primary in situ or invasive melanoma in a high incidence country (New Zealand)

BACKGROUND: Patients with invasive melanoma are at increased risk of developing subsequent invasive melanoma, but the risks for those with primary in situ melanoma are unclear. OBJECTIVES: To assess and compare the cumulative risk of subsequent invasive melanoma after primary invasive or in situ melanoma. To estimate the standardized incidence ratio (SIR) of subsequent invasive melanoma compared to population incidence in both cohorts. METHODS: Patients with a first diagnosis of melanoma (invasive or in situ) between 2001 and 2017 were identified from the New Zealand national cancer registry, and any subsequent invasive melanoma during follow‐up to the end of 2017 identified. Cumulative risk of subsequent invasive melanoma was estimated by Kaplan–Meier analysis separately for primary invasive and in situ cohorts. Risk of subsequent invasive melanoma was assessed using Cox proportional hazard models. SIR was assessed, allowing for age, sex, ethnicity, year of diagnosis and follow up time. RESULTS: Among 33 284 primary invasive and 27 978 primary in situ melanoma patients, median follow up time was 5.5 and 5.7 years, respectively. A subsequent invasive melanoma developed in 1777 (5%) of the invasive and 1469 (5%) of the in situ cohort, with the same median interval (2.5 years) from initial to first subsequent lesion in both cohorts. The cumulative incidence of subsequent invasive melanoma at 5 years was similar in the two cohorts (invasive 4.2%, in situ 3.8%); the cumulative incidence increased linearly over time in both cohorts. The risk of subsequent invasive melanoma was marginally higher for primary invasive compared to in situ melanoma after adjustment for age, sex, ethnicity and body site of the initial lesion (hazard ratio 1.11, 95% CI 1.02–1.21). Compared to population incidence, the SIR of invasive melanoma was 4.6 (95% CI 4.3–4.9) for the primary invasive and 4 (95% CI 3.7–4.2) for the primary in situ melanoma cohorts. CONCLUSIONS: The risk of subsequent invasive melanoma is similar whether patients present with in situ or invasive melanoma. Thus follow‐up surveillance for new lesions should be similar, although patients with invasive melanoma require more surveillance for recurrence.