Risk of Next Melanoma in Patients With Familial and Sporadic Melanoma by Number of Previous Melanomas.

IMPORTANCE The risk of next melanoma in patients with 2 or more previous melanomas stratified by familial and sporadic cases separately has not yet been reported, although a few population-based studies have assessed the risk of second melanoma. OBJECTIVE To assess the risk of next melanoma in patients with multiple primary melanomas by number of previous melanomas, stratified by demographic and melanoma characteristics. DESIGN, SETTING, AND PARTICIPANTS Prospective population-based cohort study with follow-up from 1958 to 2010 using the Swedish Family-Cancer Database with information on cancer cases retrieved from the Swedish Cancer Registry. A total of 65,429 patients with invasive or in situ melanoma who received a diagnosis during 1958 through 2010 were observed for next melanoma incidence. MAIN OUTCOMES AND MEASURES Standardized incidence ratios (SIRs) expressing risk of next melanoma by calculating the incidence of next (second, third, fourth, and fifth) melanoma in melanoma patients who had received a diagnosis of 1, 2, 3, and 4, respectively, previous melanomas, compared with the risk of first melanoma in the Swedish population. RESULTS For patients with either familial or sporadic melanoma, we observed a stable 2- to 3-times elevated risk by increasing number of previous melanomas; for example, for 2 previous melanomas, the SIR was 2.8 (95% CI, 2.3-3.4) for patients with familial melanoma and 2.5 (95% CI, 2.3-2.7) for patients with sporadic melanoma. Overall risk of second melanoma was higher in patients with familial melanoma who received a diagnosis at younger than 40 years (SIR, 4.7 [95% CI, 3.9-5.6]), and we found a notable risk in young patients with familial melanoma during the first 5-year follow-up after first melanoma: SIR of 6.1 (95% CI, 4.0-9.0) for interval up to 1 year, 6.2 (95% CI, 3.2-11) for 2 to 3 years, and 19 (95% CI, 10-31) for 4 to 5 years. Risk was notable in young (<40 years) patients with sporadic melanoma within the first year of follow-up (SIR, 5.3 [95% CI, 4.3-6.4]) and afterward remained steadily elevated by 2 to 3 times. CONCLUSIONS AND RELEVANCE We found a stable 2- to 3-times elevated risk by number of previous melanomas for patients with either familial or sporadic melanoma. Notable risk in young patients with familial melanoma during first 5-year follow-up after first melanoma may suggest that it is important to refer these patients for clinical genetic testing.