Natural hair color and skin cancers: A two-sample Mendelian randomization study.

ABSTRACT

Previous observational studies have indicated an association between hair color and the risk of melanoma and keratinocyte skin cancer (KSC); however, different hair colors show inconsistent effects on skin cancers. Here, we conducted a two-sample Mendelian randomization (MR) study to evaluate the causal relationship between natural hair color and skin cancers by using 211 single nucleotide polymorphisms as genetic instruments from a genome-wide meta-analysis of 360,270 individuals of European ancestry. Light hair colors (red, blonde, and light brown) were associated with high levels of cutaneous melanoma (CM) and KSC (CM-inverse variance weighted [IVW] odds ratio [OR]-red 1.034, 95% confidence interval [CI] 1.025-1.044, P < 0.001; OR-blonde 1.008, 95% CI 1.003-1.014, P = 0.003; OR-light brown 1.006, 95% CI 1.002-1.011, P = 0.009; KSC-IVW OR-red 1.078, 95% CI 1.053-1.103, P < 0.001; OR-blonde 1.024, 95% CI 1.009-1.040, P = 0.002; OR-light brown 1.018, 95% CI 1.004-1.033, P = 0.01). However, dark brown hair showed an inverse causal relationship with skin cancers (CM IVW OR 0.987, 95% CI 0.984-0.990, P < 0.001; KSC IVW OR 0.979, 95% CI 0.970-0.988, P < 0.001). Black hair was associated with a decreased risk of KSC (IVW OR 0.954, 95% CI 0.913-0.997, P = 0.036) but showed no causal relationship with CM. The present study provides strong MR evidence of a causal association between hair color and skin cancer. Secondary MR analyses enhances result robustness by replicating findings, exploring gender-specific effects, and providing a more comprehensive understanding of the complex relationship between hair color and skin cancers. More large-scale MR studies or randomized controlled trials are required to further investigate the mechanisms of the association between hair color and skin cancers.