Risk Factors for Incident Nevus-Associated vs De Novo Invasive Melanoma.

Importance: No prospective epidemiologic studies have investigated genetic vs environmental factors on the risks of nevus-associated melanoma (NAM) and de novo melanoma.

Objective: To determine whether the risk factor profile differs for nevus-associated and de novo invasive melanoma, and whether the associations differ according to sex.

Design, setting, and participants: This population-based prospective cohort study (the QSkin Study) conducted in Queensland, Australia, included participants aged 40 to 69 years at baseline. Participants were recruited in 2011 and followed up until December 2022. All analyses were conducted between October 2024 and January 2025.

Exposures: Self-reported information from the baseline survey on phenotypic factors (hair color, tanning ability, nevus density, family history), sun exposure-related factors (sunburns, history of skin cancers and actinic lesions), and polygenic risk scores for melanoma and nevus development was collected.

Main outcome and measures: The primary outcome was incident invasive melanoma (nevus-associated and de novo).

Results: A total of 17 752 males and 21 049 females were included and 859 were analyzed. During a median (IQR) follow-up of 11.4 (11.2-11.7) years, 209 participants developed an invasive nevus-associated melanoma (129 in males and 80 in females) and 650 developed an invasive de novo melanoma (362 in males and 288 in females). Many of the known phenotypic and sun exposure-related risk factors for melanoma were similarly associated with nevus-associated and de novo melanoma, but high nevus density and high genetic propensity for melanoma development had significantly higher hazard ratios (HRs) for NAM than de novo melanoma (HR for many moles vs no moles, 6.86 [95% CI, 3.82-12.33] vs 3.21 [95% CI, 2.23-4.63]; P = .001; HR for melanoma polygenic risk score tertile 3 vs tertile 1, 6.46 [95% CI, 3.42-12.20) vs 2.98 [95% CI, 2.21-4.02]; P = .006). No significant differences in the risk factor profile for NAM were found for sex, but the HR for older age was significantly higher among males with de novo melanoma than in females. The site distribution of NAM differed for males and females, occurring mostly commonly on the trunk in males and on the limbs in females.

Conclusions and relevance: Results of this study identified distinct risk factor profiles for NAM and de novo melanomas, particularly polygenic risk and nevus propensity. Males and females tended to develop NAM on different body sites, which may have implications for early detection strategies.