If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Vitiligo, the most common cutaneous depigmentation disorder, affects up to 2% of the population and is characterized by immune-mediated destruction of melanocytes, causing depigmentation of the skin, hair, and oral mucosa (
). We investigated the role of these vitiligo-associated genetic loci in risk of melanoma, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC), using genome-wide association study data on 6,628 melanoma cases, 12,945 BCC cases, 6,579 SCC cases, and more than 274,000 controls of European ancestry.
We performed the analysis in three studies. Melanoma analysis (study 1) was based on a two-stage meta-analysis. 23andMe (Mountain View, CA) provided free access to aggregated genetic and phenotypic information for study 1 (Supplementary Figure S1a online). Study 2 participants were from a hospital-based case-control study of melanoma recruited at the MD Anderson Cancer Center (
; Supplementary Figure S1b, S1c online). 23andMe research participants provided written informed consent, in accordance with the company’s human subjects protocol (reviewed and approved by Ethical and Independent Review Services, an Association for the Accreditation of Human Research Protection Programs (AAHRPP)–accredited Institutional Review Board). Study protocols for study 2 were approved by the Institutional Review Board at MD Anderson, and written informed consent was obtained from all participants. Further information on methodology is presented in Supplementary Material online.
A combined analysis of melanoma study 1 and study 2, totaling 6,628 melanoma cases and 287,591 controls (Supplementary Table S1 online), identified four vitiligo-susceptibility loci reaching a Bonferroni-adjusted P-value threshold (P < 8.6 × 10–4 for 58 single-nucleotide polymorphisms), with three loci also reaching genome-wide significance (P < 5 × 10-8): RALY-EIF252-ASIP-AHCY-ITCH, IRF4, TYR, and MC1R (Table 1, Supplementary Table S2 online).
Table 1Loci reaching statistical significance after Bonferroni correction for melanoma risk
Meta-analysis odds ratios calculated with the random-effects method.
1.25 × 10–23
1.29 × 10–8
6.23 × 10–7
2.88 × 10–33
SNPs that met Bonferroni-adjusted significance (P < 8.6×10−4) in the overall meta-analysis are listed. In addition, we report genetic locus, nearest genes, major allele, minor allele, MAF in study 1 controls, average imputation r2 (a measure of imputation quality) for study 1, and OR with P-value for each stage, calculated with respect to the minor allele. Study 1 included 4,842 melanoma cases and 286,565 controls from 23andMe. Study 2 included 1,804 melanoma cases and 1,026 controls from the MD Anderson Cancer Center. The combined fixed-effect meta-analysis totaling 6,628 melanoma cases and 287,591 controls. Statistics for effect heterogeneity (Phet and I2) are included in Supplementary Table S2. All subjects lived in the USA and were of European ancestry.
) (Supplementary Table S1). Sixteen loci for BCC and seven loci for SCC reached the Bonferroni-adjusted P-value threshold (Supplementary Table S3 online). Interestingly, all four melanoma-associated loci were also significantly associated with the risk of BCC and SCC. Three additional SCC-associated loci were also found to be significantly related to the risk of BCC.
Within those loci with P-value ≤ 0.05, 100% (11 of 11), 80.0% (25 of 27), and 100% (19 of 19) indicated an inverse correlation between risk of vitiligo and risk of melanoma, BCC, and SCC, respectively (Figure 1, Supplementary Tables S2 and S3). Furthermore, among the loci that reached a Bonferroni-adjusted P-value threshold, we observed very consistent inverse relationships between vitiligo and risk of melanoma, BCC, and SCC (100%, 93.75%, 100% consistency). To identify possible causal genes, we interrogated publicly available Genotype-Tissue Expression Project (GTEx) expression quantitative trait loci (eQTL) v7 data sets for three tissues (Supplementary Table S4 online). Of the 58 loci, 20 were identified as eQTLs for whole blood. It was intriguing to find inverse associations for rs12203592 and IRF4 gene expression levels within whole blood and sun-exposed skin tissue. Further information on methods and imputation quality can be found in Supplementary Material and Supplementary Table S5 online.
Vitiligo is an autoimmune disease characterized by loss of skin pigmentation and affects approximately 0.5% to 2% of the population worldwide (
). Interestingly, although individuals with vitiligo were initially presumed to be more susceptible to UV mutagenesis and skin cancer because of their lack of melanoma, previous studies have instead highlighted an inverse relationship between vitiligo and both melanoma and NMSC (
). A large retrospective study of 10,040 patients with vitiligo reported a lower risk of melanoma and NMSC in such patients. However, this association remains controversial, as another study reported a higher risk of skin cancer among patients with vitiligo who underwent phototherapy (
). In addition, most studies did not describe how vitiligo might influence the risk of melanoma and NMSC.
We observed a consistent inverse relationship between risk of vitiligo and skin cancers in the RALY-EIF252-ASIP-AHCY-ITCH, IRF4, TYR, and MC1R genes. A haplotype near ASIP has been associated with skin sensitivity to sun, red, and blonde hair, and was shown to confer a significant risk of melanoma and BCC (
). IRF4 protein activates melanin synthesis by tyrosinase, an enzyme found in melanocytes. The genetic variant rs12203592 T allele impairs transcription factor binding and results in decreased protein expression of IRF4 and tyrosinase (
). Moreover, it is interesting to find that immune-related single-nucleotide polymorphisms located in the HLA-DRB1/DQA1, CTLA4, and PTPN22 genes, which encode important immunoregulative proteins, showed consistent inverse relationships between vitiligo and skin cancers, especially BCC. Overall, the inverse relationship might indicate that different or opposed biological pathways mediate vitiligo and skin cancer (
These associations suggest a possible genetic relationship between vitiligo and skin cancers. In our analysis of each of the three types of skin cancers, we did not exclude those with a history of the other skin cancer types. However, given the large sample size, those with such a history represented a small percentage of cases and controls, probably resulting in a limited bias in genetic risk estimates for each specific type of skin cancers. Further studies are warranted to understand the underlying mechanisms.
Conflict of Interest
The authors state no conflict of interest.
We would like to thank the research participants and employees of 23andMe for making this work possible. The 23andMe Research Team: Michelle Agee, Babak Alipanahi, Adam Auton, Robert K. Bell, Katarzyna Bryc, Sarah L. Elson, Pierre Fontanillas, Nicholas A. Furlotte, David A. Hinds, Karen E. Huber, Aaron Kleinman, Nadia K. Litterman, Jennifer C. McCreight, Matthew H. McIntyre, Joanna L. Mountain, Elizabeth S. Noblin, Carrie A. M. Northover, Steven J. Pitts, J. Fah Sathirapongsasuti, Olga V. Sazonova, Janie F. Shelton, Suyash Shringarpure, Chao Tian, Joyce Y. Tung, Vladimir Vacic, and Catherine H. Wilson.