Dear Editor
Breslow thickness is considered to be the most important prognostic tumor feature in melanoma patients and is associated with age, sex, and phenotypic risk factors for melanoma such as number of nevi. However, its association with inherited genetic variants in recently identified melanoma risk loci is largely unknown.
In a Western Australian Melanoma Health Study (WAMHS) study, published in the British Journal of Dermatology, IRF4 rs12203592, OCA2 rs1800401 and TP53 rs1042522 were significantly associated (P < 0.05) with Breslow thickness; however, these associations did not pass false discovery.1 Within the large, international population-based Genes, Environment, and Melanoma (GEM) Study, we previously investigated the association of 47 SNPs in melanoma risk loci with multiple primary melanoma occurrence and found IRF4 rs12203592 to predict histologic markers of the hypothesized dual melanoma pathways.2, 3 In the present study, we pooled WAMHS and GEM data to investigate the association of those melanoma risk SNPs genotyped in both WAMHS and GEM with Breslow thickness adjusting for patient characteristics and potential confounding variables.
Details concerning study population and genotyping have been published previously for GEM3 and WAMHS.1 Each study’s institutional review board approved the study. Genotype information for each SNP included in this study (n = 22) is presented in Table S1 (see Supporting Information). In GEM, 2,458 patients of European origin were diagnosed with an incident first primary melanoma in the year 2000 (Table 1). Patient characteristics were collected via phone interviews and self-completed questionnaires. DNA was collected from buccal swab kits. SNPs were genotyped using the MassArray iPLEX platform (Sequenom, San Diego, CA) with quality control measures described previously.4
Table 1.
Characteristics of patients diagnosed with a first primary melanoma in the GEM Study (n = 2,458) and WAMHS (n = 1,215)a
| Characteristic | GEM Study n (%)b |
WAMHS n (%)b |
|---|---|---|
| Age at diagnosis, years | ||
| Median (IQR) | 55 (14) | 62 (17) |
| Sex | ||
| Male | 1274 (52) | 705 (58) |
| Female | 1184 (48) | 510 (42) |
| Breslow thickness, mm | ||
| Median (IQR) | 0.73 (0.84) | 0.60 (0.70) |
| 0.01 to 1.00 | 1598 (65) | 866 (72) |
| 1.01 to 2.00 | 479 (19) | 182 (15) |
| 2.01 to 4.00 | 224 (9) | 87 (7) |
| >4.00 | 124 (5) | 45 (4) |
| Missing | 33 (1) | 35 (3) |
| Number of back nevi | ||
| 0–10 | 1390 (57) | 611 (50) |
| >10 | 1017 (41) | 579 (48) |
| Missing | 51 (2) | 25 (2) |
| Hair color | ||
| Dark hair (dark brown, black) | 766 (31) | 398 (33) |
| Light hair (light brown, blonde) | 1454 (59) | 662 (54) |
| Red | 212 (9) | 128 (11) |
| Missing | 26 (1) | 27 (2) |
| Eye color | ||
| Dark eyes (brown, black) | 476 (19) | 174 (14) |
| Light eyes (blue, grey, green, hazel) | 1957 (80) | 1033 (85) |
| Missing | 25 (1) | 8 (1) |
| Ability to tan | ||
| Deep/moderate tan | 1419 (58) | 698 (57) |
| Mild/no tan | 981 (40) | 509 (42) |
| Missing | 58 (2) | 8 (1) |
GEM, Genes, Environment and Melanoma; IQR, Interquartile range; WAMHS, Western Australia Melanoma Health Study.
Limited to whites of European origin diagnosed with a first primary melanoma.
Median (IQR) reported for continuous variables; n (%) reported for categorical variables.
In WAMHS, 1,215 incident cases of cutaneous melanoma in patients of European origin with genotyping information were included for this study (Table 1). All cases were diagnosed between the years 2006 and 2010 and recruited through the Western Australian Cancer Registry (WACR). Patient characteristics were obtained from questionnaires administered by phone interview, and histopathology data were obtained from WACR records. DNA samples were extracted from peripheral blood samples and genotyping was performed on an Illumina OmniXpressExome-v1 chip, using standard quality control measures (Illumina, San Diego CA).
The left-skewed distribution of Breslow thickness was normalized using a log transformation. Within GEM, IRF4 rs12203592*T, CCND1 rs11263498*T, and MX2 rs45430*G were nominally associated (P < 0.05) with increased Breslow thickness in linear regression models adjusted for age, sex and GEM study center (Table 2). Within WAMHS, IRF4 rs12203592*T was also significantly associated with increased Breslow thickness adjusting for age and sex (P = 0.005), consistent with previous WAMHS results.1 In a pooled analysis with both studies, IRF4 rs12203592*T was associated with increased Breslow thickness (β = 0.09, P = 5.47 × 10−5) adjusting for age, sex, and study center, and was the only SNP to pass the false discovery threshold (P = 0.0026) adjusted for multiple comparisons using the method described by Lin et al.5 The association between rs12203592*T and Breslow thickness was similar (β = 0.08, P = 8.66 × 10−4) in a pooled analysis further adjusting for phenotypic characteristics associated with rs12203592 (number of nevi, hair color, eye color, ability to tan) (results not shown).
Table 2.
Association of genotypes with Breslow thickness in patients diagnosed with a first primary melanoma in the GEM Study (n = 2,458) and WAMHS (n = 1,215)a
| Gene neighborhood | SNP | a/A | GEM Studyb | WAMHSc | Pooledd | I2 (%) | ||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
||||||||
| Mean change in log of Breslow thickness (95% CI) | % change in Breslow thickness per minor allelee | Mean change in log of Breslow thickness (95% CI) | % change in Breslow thickness per minor allelee | Mean change in log of Breslow thickness (95% CI) | % change in Breslow thickness per minor allelee | P-value | ||||
| ARNT | rs7412746 | C/T | 0.008 (−0.04, 0.05) | 0.79 | 0.004 (−0.07, 0.07) | 0.42 | 0.007 (−0.03, 0.05) | 0.70 | 0.723 | 0 |
| PARP1 | rs3219090 | A/G | −0.01 (−0.07, 0.04) | −1.47 | −0.06 (−0.14, 0.01) | −6.32 | −0.03 (−0.07, 0.01) | −2.78 | 0.19 | 16 |
| NID1 | rs3768080 | G/A | 0.003 (−0.08, 0.02) | −2.93 | 0.005 (−0.06, 0.07) | 0.53 | −0.02 (−0.06, 0.02) | −1.94 | 0.32 | 0 |
| TERT;CLPTM1L | rs4975616 | G/A | −0.01 (−0.06, 0.04) | −1.03 | −0.02 (−0.09, 0.05) | −1.68 | −0.01 (−0.05, 0.03) | −1.14 | 0.57 | 0 |
| SLC45A2 | rs35391 | T/C | 0.10 (−0.10, 0.30) | 10.24 | 0.23 (−0.10, 0.58) | 26.80 | −0.14 (−0.04, 0.31) | 14.73 | 0.12 | 0 |
| IRF4 | rs12203592 | T/C | 0.08 (0.02, 0.14) | 8.28 | 0.11 (0.03, 0.18) | 11.35 | 0.09 (0.05, 0.14) | 9.77 | 5.47 × 10−5 | 0 |
| IRF4 | rs872071 | A/G | 0.002 (−0.04, 0.05) | 0.20 | −0.05 (−0.12, 0.02) | −4.67 | −0.01 (−0.05, 0.02) | −1.46 | 0.45 | 29 |
| MTAP | rs7023329 | G/A | 0.02 (−0.02, 0.07) | 2.46 | 0.001 (−0.07, 0.07) | −0.14 | 0.02 (−0.02, 0.05) | 1.61 | 0.41 | 0 |
| MTAP | rs10811629 | G/A | 0.004 (−0.04, 0.05) | 0.38 | 0.01 (−0.06, 0.08) | 1.00 | 0.006 (0.03, 0.04) | 0.56 | 0.78 | 0 |
| CCND1 | rs11263498 | T/C | −0.07 (−0.12, −0.02) | −6.98 | −0.02 (−0.09, 0.05) | −2.12 | −0.05 (−0.09, −0.01) | −5.34 | 0.007 | 26 |
| TYR | rs1042602 | A/C | 0.01 (−0.04, 0.06) | 0.98 | 0.02 (−0.05, 0.10) | 2.51 | 0.01 (−0.03, 0.05) | 1.22 | 0.55 | 0 |
| TYR | rs10765198 | C/T | 0.007 (−0.04, 0.06) | 0.72 | −0.01 (−0.08, 0.06) | −1.15 | 0.002 (−0.04, 0.04) | 0.17 | 0.94 | 0 |
| OCA2 | rs1800407 | A/G | −0.002 (−0.08, 0.08) | −0.20 | −0.01 (−0.13, 0.10) | −1.10 | −0.006 (−0.07, 0.06) | −0.64 | 0.85 | 0 |
| HERC2 | rs1129038 | G/A | 0.02 (−0.03, 0.08) | 2.24 | 0.04 (−0.40, 0.12) | 4.24 | 0.03 (−0.02, 0.07) | 2.86 | 0.23 | 0 |
| HERC2 | rs12913832 | A/G | 0.02 (−0.04, 0.08) | 2.07 | 0.04 (−0.04, 0.12) | 4.28 | 0.03 (−0.02, 0.07) | 2.72 | 0.25 | 0 |
| ASIP | rs17305657 | C/T | −0.03 (−0.11, 0.05) | −3.06 | 0.04 (−0.07, 0.15) | 4.24 | −0.006 (−0.07, 0.06) | −0.63 | 0.85 | 11 |
| ASIP | rs4911414 | T/G | −0.01 (−0.07, 0.03) | −1.96 | 0.03 (−0.04, 0.10) | 3.35 | −0.002 (−0.04, 0.04) | −0.16 | 0.94 | 32 |
| PIGU | rs910873 | A/G | −0.01 (−0.09, 0.06) | −1.37 | 0.04 (−0.06, 0.15) | 4.44 | 0.006 (−0.05, 0.07) | 0.56 | 0.85 | 0 |
| PIGU | rs17305573 | C/T | 0.006 (−0.07, 0.09) | 0.61 | 0.04 (−0.06, 0.15) | 4.44 | 0.02 (−0.04, 0.08) | 2.00 | 0.54 | 0 |
| MX2 | rs45430 | G/A | −0.07 (−0.12, −0.03) | −7.20 | −0.03 (−0.10, 0.04) | −3.05 | −0.06 (−0.10, −0.02) | −5.77 | 0.004 | 0 |
| PLA2G6 | rs132985 | T/C | 0.02 (−0.03, 0.07) | 2.19 | 0.006 (−0.06, 0.08) | 0.61 | 0.02 (−0.02, 0.06) | 1.73 | 0.39 | 0 |
Bold type indicates P-values < 0.05 (two-sided). a, minor allele; A, major allele; CI, confidence interval; MAF, minor allele frequency; SNP, single nucleotide polymorphism.
Limited to whites of European origin diagnosed with a first primary melanoma.
Adjusted for age (continuous), sex, and study center.
Adjusted for age (continuous) and sex.
Pooled analysis combining GEM and WAMHS for SNPs genotyped in both studies; adjusted for age (continuous), sex and study center (nine GEM centers plus WAMHS coded as one center).
As the outcome, Breslow thickness, was log-transformed, the values here are presented as 100*(eEstimated Beta Coefficient -1) which may be interpreted as the percent change in the estimated mean of Breslow thickness per minor allele.
Interferon regulatory factor-4 (IRF4) is a transcription factor that regulates B- and T-cell differentiation and is a histological marker detected in hematological malignancies and melanoma.6, 7 rs12203592 is a functional, intronic variant affecting IRF4 promoter activity and IRF4 expression in immune cell lines and melanocytes.6, 7 In immune cells, rs12203592*T is associated with increased IRF4 expression, which has been shown to activate the TERT promoter and lead to increased telomerase activity.8 Telomerase activity is associated with increased ulceration, mitotic rate and Breslow thickness of melanomas; however, its association with IRF4 expression in melanomas has not been reported.9
Several studies have investigated the association of genetic variants in well-characterized melanoma risk loci (e.g., MC1R, CDKN2A, VDR) with Breslow thickness; however, the prognostic importance of inherited variants in recent GWAS-identified loci is uncertain.1 In a hospital-based study in Spain (n = 493), rs12203592*T was associated with worse melanoma survival, but was not associated with Breslow thickness.10 The association between IRF4 rs12203592 and Breslow thickness in this study is a novel finding.
Strengths of this study include its large sample size and international, population-based recruitment. Potential limitations are that not all GWAS-identified SNPs in melanoma risk loci were genotyped in both GEM and WAMHS and different genotyping platforms were used in each study. Another potential limitation is inter-study heterogeneity; however, the I2 value for the majority of SNPs, including rs12203592, was zero (Table 2).
While future studies are warranted, these results indicate that an inherited functional variant in IRF4 may independently predict Breslow thickness - the most important prognostic indicator for melanoma.
Supplementary Material
Acknowledgments
Funding source(s): This work was supported by the National Cancer Institute (R01CA112243 to N.E.T, U01CA83180 and R01CA112524 to M.B., R01CA098438 to C.B.B, and P30CA016086, P30CA014089, and P30CA008748); National Institute of Environmental Health Sciences (P30ES010126); and University of Sydney Medical Foundation Program grant to B.K.A
GEM Study Group
Coordinating Center, Memorial Sloan Kettering Cancer Center, New York, NY (USA): Marianne Berwick (PI, currently at the University of New Mexico, Albuquerque, NM), Colin Begg, Ph.D. (co-PI), Irene Orlow, Ph.D., M.S. (co-Investigator), Klaus J. Busam, M.D. (Dermatopathologist), Pampa Roy, Ph.D. (Senior Laboratory Technician), Himali Patel, M.S. (Senior Laboratory Technician), Sarah Yoo (Senior Laboratory Technician), Anne Reiner, M.S. (Biostatistician), Siok Leong, M.S. (Research Assistant), Sergio Corrales Guerrero, M.S. (Research Technician), Vikram Mavinkurve, M.S. (Senior Research Technician); University of New Mexico, Albuquerque, NM: Marianne Berwick, M.P.H., Ph.D. (PI), Li Luo, Ph.D. (Biostatistician), Susan Paine, M.P.H. (Data Manager). Study Centers: The University of Sydney and The Cancer Council New South Wales, Sydney, Australia: Anne Cust, Ph.D. (PI), Bruce K. Armstrong M.D. Ph.D. (former PI), Anne Kricker Ph.D., (former Co-PI); Menzies Institute for Medical Research University of Tasmania, Hobart, Australia: Alison Venn (current PI), Terence Dwyer (PI, currently at University of Oxford, United Kingdom), Paul Tucker (Dermatopathologist); British Columbia Cancer Research Centre, Vancouver, Canada: Richard P. Gallagher, M.A. (PI), Cancer Care Ontario, Toronto, Canada: Loraine D. Marrett, Ph.D. (PI), Lynn From, M.D. (Dermatopathologist); CPO, Center for Cancer Prevention, Torino, Italy: Roberto Zanetti, M.D (PI), Stefano Rosso, M.D., M.Sc. (co-PI); University of California, Irvine, CA: Hoda Anton-Culver, Ph.D. (PI); University of Michigan, Ann Arbor, MI: University of Michigan, Ann Arbor, MI: Stephen B. Gruber, M.D., M.P.H., Ph.D. (PI, currently at University of Southern California, Los Angeles, CA), Shu-Chen Huang, M.S., M.B.A. (co-Investigator, joint at USC-University of Michigan); University of North Carolina, Chapel Hill, NC: Nancy E. Thomas, M.D., Ph.D. (PI), David W. Ollila, M.D. (co-Investigator), Pamela A. Groben, M.D. (Dermatopathologist), David C. Gibbs, B.S. (Research Assistant, current MD/PhD candidate at Emory University, Atlanta, GA); University of Pennsylvania, Philadelphia, PA: Timothy R. Rebbeck, Ph.D. (former PI), Peter A. Kanetsky, M.P.H., Ph.D. (PI, currently at H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL); UV data consultants: Julia Lee Taylor, Ph.D. and Sasha Madronich, Ph.D., National Centre for Atmospheric Research, Boulder, CO.
Footnotes
Conflict of Interest Disclosures: None declared.
References
- 1.Cadby G, Ward S, Cole J, Moses E, Millward M, Palmer L. The association of host and genetic melanoma risk factors with Breslow thickness in the Western Australian Melanoma Health Study. British Journal of Dermatology. 2014;170(4):851–7. doi: 10.1111/bjd.12829. [DOI] [PubMed] [Google Scholar]
- 2.Gibbs DC, Orlow I, Bramson JI, Kanetsky PA, Luo L, Kricker A, et al. Association of Interferon Regulatory Factor-4 Polymorphism rs12203592 With Divergent Melanoma Pathways. Journal of the National Cancer Institute. 2016;108(7):djw004. doi: 10.1093/jnci/djw004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gibbs DC, Orlow I, Kanetsky PA, Luo L, Kricker A, Armstrong BK, et al. Inherited genetic variants associated with occurrence of multiple primary melanoma. Cancer Epidemiology Biomarkers & Prevention. 2015;24(6):992–7. doi: 10.1158/1055-9965.EPI-14-1426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Orlow I, Roy P, Reiner AS, Yoo S, Patel H, Paine S, et al. Vitamin D receptor polymorphisms in patients with cutaneous melanoma. International journal of cancer Journal international du cancer. 2012;130(2):405–18. doi: 10.1002/ijc.26023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lin DY. An efficient Monte Carlo approach to assessing statistical significance in genomic studies. Bioinformatics. 2005;21(6):781–7. doi: 10.1093/bioinformatics/bti053. [DOI] [PubMed] [Google Scholar]
- 6.Visser M, Palstra R-J, Kayser M. Allele-specific transcriptional regulation of IRF4 in melanocytes is mediated by chromatin-looping of the intronic rs12203592-enhancer to the IRF4-promoter. Human molecular genetics. 2015:ddv029. doi: 10.1093/hmg/ddv029. [DOI] [PubMed] [Google Scholar]
- 7.Do TN, Ucisik-Akkaya E, Davis CF, Morrison BA, Dorak MT. An intronic polymorphism of IRF4 gene influences gene transcription in vitro and shows a risk association with childhood acute lymphoblastic leukemia in males. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease. 2010;1802(2):292–300. doi: 10.1016/j.bbadis.2009.10.015. [DOI] [PubMed] [Google Scholar]
- 8.Hrdličková R, Nehyba J, Bose HR. Regulation of telomerase activity by interferon regulatory factors 4 and 8 in immune cells. Molecular and cellular biology. 2009;29(3):929–41. doi: 10.1128/MCB.00961-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Carvalho L, Lipay M, Belfort F, Santos I, Andrade J, Haddad A, et al. Telomerase activity in prognostic histopathologic features of melanoma. Journal of Plastic, Reconstructive & Aesthetic Surgery. 2006;59(9):961–8. doi: 10.1016/j.bjps.2006.01.022. [DOI] [PubMed] [Google Scholar]
- 10.Potrony M, Rebollo-Morell A, Giménez-Xavier P, Zimmer L, Puig-Butille JA, Tell-Marti G, et al. IRF4 rs12203592 functional variant and melanoma survival. International Journal of Cancer. 2017;140(8):1845–9. doi: 10.1002/ijc.30605. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.