Abstract

Aims: High naevus counts, and UV photodamage are strong melanoma risk factors. However, whole-of-body measures fail to capture variability across body sites. Three-dimensional (3D) total body photography (TBP) and artificial intelligence (AI) allows the opportunity to automate the extraction of site-specific distributions of naevi and photodamage. This study combined 3D-TBP, AI, and unsupervised clustering in a high-risk cohort to identify distinct phenotypic patterns associated with melanoma.

Methods: Participants with a personal melanoma history (diagnosed >50 years) underwent 3D-TBP. Site-specific photodamage and naevus counts were assessed using density-based spatial clustering of applications with noise (DBSCAN) to identify body-site dependent phenotypic patterns. Melanoma prevalence (none, single, multiple) relative to phenotypic pattern was evaluated using population prevalence ratios (PPR).

Results: Analysis on 117 individuals revealed four phenotypic patterns of increasing severity: moderate v-neck photodamage with few (median=38) naevi (24%), moderate generalised photodamage with several (median=155) naevi (27%), moderate v-neck photodamage with many (median=203) naevi (17%), and severe generalised photodamage with few (median=37) naevi (32%). No individuals had severe photodamage and several-many naevi. Inter-pattern comparisons revealed that participants in the mildest phenotypic pattern were least likely to be affected with invasive melanomas (PPR 1.51, 95% CI: 1.01–2.26), whereas the most severe phenotypic pattern was more likely to be affected by multiple invasive melanomas (PPR 2.00, 95% CI: 1.06–3.77). The prevalence of MiS was consistent across patterns. Melanoma was more likely at sites of large naevi (>5mm, p<0.05) in moderate photodamage patterns, but were independent of naevi (>2mm) in severe photodamage patterns. From a control cohort unaffected by melanoma (n=114), only 18 (16%) matched to a high-risk phenotypic pattern.

Conclusions: 3D-TBP phenotyping of a high-risk, older Australian cohort revealed four distinct phenotypic patterns associated with invasive melanoma risk. Specifically, individuals with severe photodamage, in the presence of relatively few naevi had a significantly higher risk of developing multiple invasive melanomas. For individuals with moderate photodamage, the risk of invasive melanoma was positively associated with the number of naevi. Thus, comprehensive phenotypes may be more predictive for the diagnosis and site of invasive melanoma, which may assist with nuanced risk stratification and customised surveillance.