Improving Melanoma Risk Stratification of Skin Color Type by Polygenic Scores of Skin Color and Melanoma

Background The Fitzpatrick skin type (FST) is commonly used in clinical settings to stratify melanoma risk across different skin types. However, it is subjective and does not capture intra‐type variability in risk. These limitations are particularly evident among lighter‐skinned individuals, who constitute the majority of those with European ancestry and generally have a higher melanoma risk. Methods We assessed whether supplementing FST with validated polygenic scores for skin color (PGS SC ) and melanoma (PGS M ) improves risk stratification in 479,607 UK Biobank participants. Melanoma prevalence in individuals by FST, PGS SC , and PGS M was estimated, and their differences were tested, adjusting for age, gender, and genetic background (top 10 principal components). Results FST distribution was as follows: Very fair (7.77%), fair (68.06%), light olive (18.56%), olive (1.87%), brown (2.96%), and black (0.78%), with corresponding melanoma prevalence of 1.94%, 1.31%, 0.70%, 0.45%, 0.15%, and 0% ( P −trend < 0.001). Notably, PGS SC significantly differentiated melanoma prevalence within each lighter‐skinned type (very fair, fair, and light olive), P −trend < 0.001. Among lighter‐skinned individuals, the melanoma prevalence ranged widely across the lowest to highest deciles of PGS SC (0.62%–2.12%). In addition, PGS M also differentiated risk, with prevalence ranging from 0.39%–2.08% between low‐ and high‐risk groups. Among darker skin types (olive and brown), only PGS M significantly differentiated melanoma risk, with prevalence ranging from 0.13% to 0.80% between low‐ and high‐risk groups. Conclusions Integrating PGS of skin color and melanoma with the conventional FST classification significantly improves melanoma risk stratification and addresses key limitations of the FST, which, while widely used, lacks sensitivity to intra‐type risk variation.