Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Mar 1.
Published in final edited form as: J Invest Dermatol. 2017 Oct 17;138(3):557–561. doi: 10.1016/j.jid.2017.09.034

Susceptibility Loci-Associated Cutaneous Squamous Cell Carcinoma Invasiveness

Wei Wang 1, Eric Jorgenson 2, Alice S Whittemore 1, Maryam M Asgari 2,3
PMCID: PMC6211552  NIHMSID: NIHMS956527  PMID: 29054604

Abstract

Genome-wide association studies have identified genetic loci associated with cutaneous squamous cell carcinoma (cSCC) risk, but single-nucleotide polymorphism associations with cSCC invasiveness have not been investigated. We examined associations between cSCC invasiveness and 23 reported single-nucleotide polymorphisms among 67,833 non-Hispanic white subjects. Additionally, we performed a genome-wide scan and identified one SNP with significantly different frequencies in 5,724 subjects with at least one invasive tumor and 1,943 subjects with in situ tumors only. We then compared genotype frequencies among the invasive and in situ groups with those of 60,166 control subjects. The genome-wide scan identified that the T allele in single-nucleotide polymorphism rs41269979 in the class II human leukocyte antigen region was more frequent in the invasive than the in situ group (P = 4.93 × 10−8). Single-nucleotide polymorphisms in five of the 23 previously associated loci showed odds ratio heterogeneity between the in situ and invasive groups: rs447510 in HLA-DQA1 (Phet = 2.93 × 10−3), rs12203592 in IRF4 (Phet = 3.94 × 10−4), rs1805007 in MC1R (Phet = 7.71 × 10−3), and two SNPs in DEF8 (rs4268748, Phet = 1.09 × 10−4 and rs8063761, Phet = 1.40 × 10−4). These findings may provide new insight into the genetic basis of cSCC invasiveness and may help identify individuals at higher risk for developing clinically aggressive cSCC.

INTRODUCTION

Cutaneous squamous cell carcinoma (cSCC) is the second most common cancer in the United States, with over 700,000 cases diagnosed annually, and its incidence is on the rise. Although cSCC is usually not life threatening, it carries a low but significant risk of metastasis (3.7%) and disease-specific mortality (2.1%) (Schmults et al., 2013). Aggressive clinical behavior in cSCCs is associated with patient risk factors, including advanced age, male sex, and immunosuppression, as well as tumor risk factors, the most salient of which is invasiveness (Skulsky et al., 2017). When confined to the epidermis (cSCC in situ or Bowen’s disease), cSCCs very rarely metastasize or cause disease-specific mortality and have an excellent prognosis (Burton et al., 2016; Karia et al., 2013).

cSCC in situ, characterized by full-thickness keratinocyte dysplasia on histology, often presents as a well-demarcated erythematous hyperkeratotic neoplasm on sun-exposed areas, with most arising on the head and neck (Morton et al., 2014). cSCCs in situ can progress to invasive cSCC in approximately 3–5% of cases but are often not the precursor of invasive disease (Smoller, 2006). In fact, actinic keratoses (and not SCC in situ) are the most common precursors of invasive cSCC, with invasive disease often arising directly from early basilar keratinocyte atypia (Fernandez Figueras, 2017; Smoller, 2006). Although cSCC in situ and invasive cSCCs share risk factors, including UV irradiation, immunosuppression, and possibly human papillomavirus (Morton et al., 2014), an underlying genetic propensity to in situ versus invasive diseases has not been previously explored.

We previously performed a genome-wide association study that identified 10 susceptibility loci associated with cSCC risk (Asgari et al., 2016), nine of which were recently validated in a large independent cohort (Chahal et al., 2016). However, to our knowledge, no prior studies have examined genetic susceptibility associated with clinically aggressive cSCC behavior, such as cSCC invasiveness. Given the association between invasiveness and clinically aggressive behavior, we identified 23 single-nucleotide polymorphisms (SNPs) associated with cSCC risk in the literature (Asgari et al., 2016; Chahal et al., 2016; Siiskonen et al., 2016) and performed a genome-wide scan to identify susceptibility loci for invasiveness of cSCCs.

Using genome-wide genotype data from a large, well-characterized cohort of the Research Program on Genes, Environment and Health based at Kaiser Permanente Northern California, we examined genetic variants associated with risk of invasive cSCC. Participants were classified as control subjects (no cSCCs at baseline or during follow-up), patients having in situ SCCs (i.e., developed in situ tumors only during follow-up), or patients with invasive tumors (developed at least one invasive cSCC during follow-up). The overall goal was to examine the association between known cSCC risk loci and invasiveness and to identify potentially new genetic loci associated with cSCC invasiveness that could ultimately be used to better guide treatment, screening, and prevention efforts.

RESULTS

The clinical characteristics of Research Program on Genes, Environment and Health cohort members with at least one biopsy-proven cSCC included in the previously published genome-wide association study are presented in Table 1. Of the 7,667 subjects, 1,943 developed in situ cSCCs only, whereas 5,724 had at least one invasive cSCC. Subjects who developed invasive tumors tended to be slightly older than those with in situ tumors only (mean age = 72.6 ± 10.1 years for in situ vs. 71.0 ± 9.6 years for invasive, P < 0.05). They were also more often male (60.6% male for invasive vs. 49.1% for in situ, P < 0.05). There was no difference in immunosuppression status between subjects with in situ only vs. ever invasive tumors (P = 0.56).

Table 1.

Characteristics of subjects within Research Program on Genes, Environment and Health cohort with at least one incident cSCC by tumor invasiveness (n = 7,667)

Characteristic In Situ Group1 (n = 1,943) Invasive Group2 (n = 5,724) P-Value
Age in years, mean (SD) 71.0 (10.1) 72.6 (9.6) <0.05
Sex, n (%)
 Male 934 (48.1) 3,467 (60.6)
 Female 1,009 (51.9) 2,257 (39.4) <0.05
Immunosuppressed,3 n (%)
 Yes 25 (1.3) 86 (1.5)
 No 1,918 (98.7) 5,638 (98.5) 0.56
Open in a new tab

Abbreviations: cSCC, cutaneous squamous cell carcinoma; SD, standard deviation.

1

Defined as patients with in situ cSCC only.

2

Defined as patients with at least one invasive SCC.

3

Defined as a history of organ transplantation, chronic lymphocytic, or HIV infection.

When comparing subjects with invasive tumors versus subjects who developed in situ tumors only, an intergenic SNP, rs41269979, in the human leukocyte antigen (HLA) class II gene region reached genome-wide significance (P = 4.93 × 10−8) (Figure 1), with an odds ratio (OR) of 1.38 (95% confidence interval [CI] = 1.23–1.58) for the minor T allele. The minor T allele of rs41269979 had a lower frequency in the in situ group (0.12) than in the control group (0.15) or invasive group (0.16). In polytomous logistic regression analysis, this SNP was associated with a reduced risk of in situ cSCC (OR = 0.71, 95% CI = 0.64–0.79) but was not associated with invasive tumors (OR = 0.97, 95% CI = 0.92–1.03) (Table 2). In other words, the major C allele of rs41269979 is associated with an increased risk of in situ, but not invasive, cSCC. This intergenic SNP located between HLA-DQA1 and HLA-DQB1 lies about 12 kilo base pairs from rs4455710 (low R2 value of 0.12), an SNP previously shown to be associated with cSCC risk (Asgari et al., 2016). The minor T allele of this previously reported cSCC-associated rs4455710 SNP was also more strongly associated with increased risk of in situ tumors.

Figure 1. Manhattan plot for cSCC invasiveness.

Figure 1

Open in a new tab

Genome-wide analysis of invasivness is shown (comparing invasive group with in situ group): rs41269979, in the HLA class II gene region, reached genome-wide significance (P = 4.93 × 10−8).

Table 2.

Association of 24 selected SNPs with risk of in situ and invasive cSCC

SNP_Minor Allele1 hg19 Position Gene INFO2 Control Subjects Freq. Patients P-Value4
In Situ Group Invasive Group
Freq. OR (95% CI)3 Freq. OR (95% CI)3
From genome-wide scan
rs41269979_T chr6: 32620835 HLA-DQA1, HLA-DQB1 0.88 0.15 0.12 0.71 (0.64–0.79) 0.16 0.97 (0.92–1.03) 1.85 × 10−7
From prior GWAS
rs192481803_T6 chr2: 35336564 intergenic 0.82 0.01 0.01 0.81 (0.52–1.27) 0.01 0.91 (0.69–1.19) 0.66
rs62246017_A5 chr3: 71483084 FOXP1 0.94 0.33 0.34 1.09 (1.01–1.17) 0.35 1.11 (1.07–1.17) 0.51
rs6791479_T5 chr3: 189205032 TPRG1, TP63 1.00 0.43 0.44 1.08 (1.01–1.15) 0.46 1.17 (1.12–1.22) 0.02
rs35407_A6 chr5: 33946571 SLC45A2 0.74 0.04 0.02 0.49 (0.38–0.63) 0.02 0.53 (0.46–0.62) 0.58
rs16891982_C5 chr5: 33951693 SLC45A2 1.00 0.05 0.02 0.49 (0.39–0.61) 0.02 0.52 (0.45–0.60) 0.62
rs17247181_T7 chr5: 65255672 ERBB2IP 0.99 0.10 0.10 1.03 (0.92–1.15) 0.10 1.00 (0.93–1.07) 0.60
rs12203592_T5,6 chr6: 396321 IRF4 0.99 0.16 0.23 1.45 (1.34–1.58) 0.26 1.71 (1.63–1.80) 3.94 × 10−4
rs4455710_T5 chr6: 32608858 HLA-DQA1 0.99 0.38 0.43 1.30 (1.22–1.38) 0.40 1.16 (1.11–1.21) 2.93 × 10−3
rs9689649_C7 chr6: 162101178 PARK2 0.97 0.24 0.24 0.99 (0.92–1.07) 0.23 0.98 (0.93–1.03) 0.69
rs117132860_A6 chr7: 17134708 AHR 0.70 0.02 0.02 1.04 (0.79–1.36) 0.03 1.31 (1.12–1.53) 0.13
rs9643297_G7 chr8: 134483695 ST3GAL1 0.91 0.31 0.30 0.99 (0.92–1.07) 0.30 0.99 (0.94–1.04) 0.91
rs10810657_T6 chr9: 16884586 BNC2, CNTLN 0.99 0.41 0.38 0.90 (0.84–0.97) 0.38 0.90 (0.86–0.94) 0.91
rs74664507_T5 chr9: 16913836 BNC2, CNTLN 0.95 0.44 0.41 0.89 (0.83–0.95) 0.41 0.89 (0.85–0.93) 0.92
rs57994353_C6 chr9: 139356987 SEC16A 0.78 0.26 0.29 1.20 (1.10–1.30) 0.27 1.08 (1.02–1.14) 0.03
rs1126809_A5,6 chr11: 89017961 TYR 0.98 0.27 0.32 1.21 (1.13–1.30) 0.32 1.20 (1.15–1.26) 0.92
rs74899442_C6 chr11: 115890279 CADM1, BUD13 0.76 0.004 0.004 0.99 (0.57–1.72) 0.004 0.86 (0.59–1.24) 0.66
rs1800407_T6 chr15: 28230318 OCA2 0.83 0.07 0.08 1.14 (1.00–1.30) 0.08 1.18 (1.08–1.28) 0.70
rs12916300_C5 chr15: 28410491 HERC2 1.00 0.27 0.23 0.93 (0.85–1.00) 0.22 0.85 (0.81–0.89) 0.07
rs1805007_T6 chr16: 89986117 MC1R 0.81 0.08 0.10 1.39 (1.22–1.57) 0.11 1.67 (1.55–1.80) 7.71 × 10−3
rs4268748_C5 chr16: 90026512 DEF8 0.87 0.26 0.29 1.21 (1.12–1.31) 0.32 1.44 (1.37–1.51) 1.09 × 10−4
rs8063761_T7 chr16: 90027626 DEF8 0.91 0.31 0.34 1.19 (1.11–1.28) 0.37 1.39 (1.33–1.46) 1.40 ×10−4
rs6059655_A5,6 chr20: 32665748 RALY 0.99 0.07 0.10 1.28 (1.14–1.43) 0.11 1.38 (1.28–1.47) 0.24
rs754626_G7 chr20: 36017340 SRC 0.61 0.18 0.19 1.10 (0.99–1.22) 0.18 1.01 (0.94–1.08) 0.19
Open in a new tab

Abbreviations: chr, chromosome; CI, confidence interval; cSCC, cutaneous squamous cell carcinoma; Freq., frequency; GWAS, genome-wide association study; INFO, IMPUTE-2 information measure for imputation accuracy; OR, odds ratio; SNP, single-nucleotide polymorphism.

1

Allele with frequency ≤ 0.5 in control group.

2

IMPUTE-2 information measure for imputation accuracy (Marchini and Howie, 2010).

3

OR per minor allele with 95% CI, adjusted for the first 10 principal components, sex, age, immunosuppressive status, array (European vs. non-European), and reagent type (Kit-A vs. Kit-O).

4

P for heterogeneity from test of H0: β_in situ = β_invasive. P value can reflect a statistically significant difference between the ORs, even if CIs overlap.

5

Reported by Asgari et al. (2016).

6

Reported by Chahal et al. (2016).

7

Reported by Siiskonen et al. (2016).

Of the 23 published cSCC-associated SNPs, five showed significant evidence for heterogeneity between the in situ and invasive groups. Four were more strongly associated with invasive cSCC than with in situ cSCC, including the T allele of rs12203592 in IRF4 (OR = 1.71, 95% CI = 1.63–1.80 for invasive cSCC; OR = 1.45, 95% CI = 1.34–1.58 for in situ cSCC; P for heterogeneity = 3.94 ×10−4), T allele of rs1805007 in MC1R (OR = 1.67, 95% CI = 1.55–1.80 for invasive cSCC; OR = 1.39, 95% CI = 1.22–1.57 for in situ cSCC; P for heterogeneity = 7.71 × 10−3), two SNPs in DEF8 (C allele of rs4268748 with OR = 1.44, 95% CI = 1.37–1.51 for invasive cSCC and OR = 1.21, 95% CI = 1.12–1.31 for in situ cSCC), and T allele of rs8063761 (OR = 1.39, 95% CI = 1.33–1.46 for invasive cSCC and OR = 1.19, 95% CI = 1.11–1.28 for in situ cSCC; P for heterogeneity = 1.40 × 10−4). The two SNPs in DEF8 are about 1 kilo base pair apart and are in tight linkage disequilibrium (R2 = 0.81). The T allele of SNP rs4455710 in HLA-DQA1 was more strongly associated with increased risk of in situ tumors (OR = 1.30, 95% CI = 1.22–1.38 for in situ cSCC; OR = 1.16, 95% CI = .11–1.21 for invasive cSCC; P for heterogeneity = 2.93 × 10−3) (Table 2).

DISCUSSION

We report a genome-wide significant hit in the HLA class II gene region at locus rs41269979 whose minor allele is associated with risk of invasive cSCC, when compared with situ cSCC. More specifically, the major C allele of rs41269979 was associated with an increased risk of in situ cSCC but not associated with invasive cSCC, when both groups were compared with the control subjects. Similarly, we found a previously reported cSCC-associated SNP rs4455710 in HLA-DQA1 whose minor T allele was more strongly associated with increased risk of in situ tumors. HLA-DQA1 is a class II major histocompatibility gene that helps the immune system distinguish the body’s own proteins from proteins made by foreign invaders, such as viruses and bacteria. Mutations in HLA genes have been shown to be an important process for some tumor types, notably SCC in other organs, such as head and neck and lung SCCs (Shukla al., 2015). Associations with mutations in HLA-DQA1 suggest that immune evasion by cSCCs may be the mechanisms for in situ SCC formation or could suggest that functional mechanisms that impair recognition of foreign invaders, such as human papillomavirus, may be involved in situ cSCC risk. Additional studies focusing on the functional role of HLA mutations in cutaneous oncogenesis are needed.

We report three other SNPs that are associated with an invasive cSCC tumor phenotype. The minor T risk allele at rs12203592 in IRF4 has been shown to modulate an enhancer element that controls IRF4 expression (Visser et al., 2015) and has been previously shown to be associated with increased skin cancer susceptibility (Wu et al., 2016). IRF4 is associated with immune system development and response, and its transcription status in the epidermis is strongly correlated with the allelic status of rs12203592 (Visser et al., 2015). The T allele of rs12203592 leads to reduced chromatin loop formation and diminished IRF4 expression in melanocytes. Thus, this SNP could have a functional role in leading to more aggressive and invasive cSCC susceptibility.

Two intronic SNPs in DEF8 at chromosome 16q24, rs4268748 and rs8063761, have been associated with cSCC risk by previous genome-wide association studies. Given the high linkage disequilibrium between the two SNPs, they are likely markers of the same underlying risk variant for cSCC. It is unclear why the risk alleles of these two SNPs are more strongly associated with risk of more aggressive cSCC. DEF8 is involved in a molecular complex that regulates lysosome positioning and secretion (Fujiwara et al., 2016). Both rs4268748 and rs8063761 are associated with expression levels of CDK10, a gene critical for cell cycle progression, in both sun-exposed and non0sun-exposed skin in GTEx (GTEx Consortium, 2015). DEF8 lies in close proximity (39 kilo base pairs) from MC1R, and the functional significance of these associations needs further study. A previously reported cSCC-associated SNP in MC1R, rs1805007, also showed stronger association with susceptibility to more aggressive cSCC. The MC1R SNP and the DEF8 SNPs are in moderate linkage disequilibrium (r2 = 0.22). After additional adjustment for rs4268748, the association between rs1805007 and cSCC aggressiveness was attenuated (OR = 0.93, 95% CI = 0.8–1.09 and P = 0.37, in comparing subjects with invasive versus in situ tumors only). The functional significance of the observed associations between the SNPs in this region and more aggressive cSCC needs further study.

Strengths of this study include the large sample size, high-quality genotyping data (Kvale et al., 2015), long-term follow-up on a stable community-based population, and the availability of comprehensive pathology reports on all health plan members, allowing for classification of cSCCs by invasiveness. To our knowledge, the association of genetic loci with tumor invasiveness in cSCCs has not been previously reported. Limitations include the study population, which was limited to non-Hispanic whites, limiting generalizability. However, when examining the racial-ethnic constitution of Kaiser Permanente Northern California members who develop cSCC, we found that 96% arise among non-Hispanic whites, limiting our power to detect associations in other racial-ethnic groups. Differential misclassification of control subjects may be a rare possibility if in situ or invasive cSCCs were not biopsied, but that is unlikely in a prepaid health care plan setting that provides ready access to care, where pathological verification of a malignancy is the standard of care. In addition, our study focused on SNPs associated with tumor invasiveness, comparing in situ to invasive SCCs, and was underpowered to examine histologic features associated with aggressive invasive SCCs such as degree of histologic differentiation of invasive tumors, which can be explored in future studies with larger datasets.

In conclusion, our data suggest that risk alleles in IRF4 and DEF8 predispose to invasive SCC, whereas risk alleles in HLA-DQA1 predispose to in situ cSCCs. Further studies are needed to replicate these results in independent populations. Our findings may provide insight into the complex pathways that underlie risk of invasive cSCCs. Future functional studies could help uncover the function of genes, such as DEF8, and the role of immune function and HLA in cSCC risk.

MATERIALS AND METHODS

Study population

Details of the study population and selection criteria for the Research Program on Genes, Environment and Health genome-wide association study cSCC cohort have been published (Asgari et al., 2016). Genotypes used in this genome-wide association study have been registered with dbGAP (study accession: phs000674.v1.p1). Potentially eligible study subjects were Research Program on Genes, Environment and Health members who at cohort entry reported non-Hispanic white race-ethnicity, were 18 years of age or older, had no diagnostic codes for rare genetic disorders associated with increased SCC risk, and whose pathology records documented a cSCC (excluding anogenital and mucosal SCCs) between study entry (2003–2007) and end of follow-up (December 31, 2012). Subjects were censored if they had a gap in membership of longer than 90 days or departed from the Kaiser Permanente Northern California health care system, whichever occurred first. The methodology for capturing cSCCs from health plan pathology records screened up to each subject’s censoring date have previously been reported (Asgari et al., 2016). Data on cSCC invasiveness (in situ vs. invasive) was abstracted from the electronic pathology report. Pathology reports where full-thickness epidermal dysplasia with no invasion beyond the epidermis was noted and the final diagnosis included terms such as in situ or Bowen’s were classified as in situ cSCCs. Tumors with histologic invasion of atypical keratinocytes beyond the epidermis noted in the pathology report were classified as invasive cSCCs. This study was approved by the Kaiser Foundation Research Institute Institutional Review Board and was conducted per the Declaration of Helsinki principles. Written, informed consent was obtained from all study participants.

Genotyping

DNA extracted from saliva samples from eligible subjects was genotyped using custom Affymetrix (Santa Clara, CA) Axiom arrays (Hoffman et al., 2011a, 2011b). Extensive genotype quality control was conducted and has been described previously in detail (Kvale et al., 2015).

Statistical analysis

To adjust for population stratification, we determined ancestry principal components using the smartpca program in the EIGEN-SOFT4.2 software package (Patterson et al., 2006), as has been described elsewhere. PLINK, version 1.9 (https://www.cog-genomics.org/plink2) (Chang et al., 2015) was used to perform a genome-wide scan of SNPs associated with cSCC invasiveness, comparing subjects with invasive tumors versus subjects who developed in situ tumors only. Odds ratios and P values from Armitage-Cochran trend tests were reported, adjusted for age, sex, immunosuppressive status, first 10 principal components of ancestry, genotyping array, and reagent kit. In addition, we used STATA (StataCorp, College Station, TX) to perform C logistic regressions to examine and compare the associations between SNPs and the risk of developing in situ and invasive cSCC, using control subjects as the reference group. All P values are two sided. All study procedures were approved by the institutional review board of the Kaiser Foundation Research Institute.

Acknowledgments

This work was supported by the National Institutes of Health (R01 CA166672).

Abbreviations

CI

confidence interval

cSCC

cutaneous squamous cell carcinoma

HLA

human leukocyte antigen

OR

odds ratio

SNP

single nucleotide polymorphism

Footnotes

CONFLICT OF INTEREST

MMA received grant funding to her institution from Pfizer and Valeant, but these grants were not relevant to the topic of this manuscript.

References

  1. Asgari MM, Wang W, Ioannidis NM, Itnyre J, Hoffmann T, Jorgenson E, et al. Identification of susceptibility loci for cutaneous squamous cell carcinoma. J Invest Dermatol. 2016;136:930–7. doi: 10.1016/j.jid.2016.01.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Burton KA, Ashack KA, Khachemoune A. Cutaneous squamous cell carcinoma: a review of high-risk and metastatic disease. Am J Clin Dermatol. 2016;17:491–508. doi: 10.1007/s40257-016-0207-3. [DOI] [PubMed] [Google Scholar]
  3. Chahal HS, Lin Y, Ransohoff KJ, Hinds DA, Wu W, Dai HJ, et al. Genome-wide association study identifies novel susceptibility loci for cutaneous squamous cell carcinoma. Nat Commun. 2016;7:12048. doi: 10.1038/ncomms12048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015;4:7. doi: 10.1186/s13742-015-0047-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Fernandez Figueras MT. From actinic keratosis to squamous cell carcinoma: pathophysiology revisited. J Eur Acad Dermatol Venereol. 2017;31(Suppl 2):5–7. doi: 10.1111/jdv.14151. [DOI] [PubMed] [Google Scholar]
  6. Fujiwara T, Ye S, Castro-Gomes T, Winchell CG, Andrews NW, Voth DE, et al. PLEKHM1/DEF8/RAB7 complex regulates lysosome positioning and bone homeostasis. JCI Insight. 2016;1(17):e86330. doi: 10.1172/jci.insight.86330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. GTex Consortium. Human genomics. The Genotype-Tissue Expression (GTEx) pilot analysis: multitissue gene regulation in humans. Science. 2015;348(6235):648–60. doi: 10.1126/science.1262110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hoffmann TJ, Kvale MN, Hesselson SE, Zhan Y, Aquino C, Cao Y, et al. Next generation genome-wide association tool: design and coverage of a high-throughput European-optimized SNP array. Genomics. 2011a;98:79–89. doi: 10.1016/j.ygeno.2011.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hoffmann TJ, Zhan Y, Kvale MN, Hesselson SE, Gollub J, Iribarren C, et al. Design and coverage of high throughput genotyping arrays optimized for individuals of East Asian, African American, and Latino race/ethnicity using imputation and a novel hybrid SNP selection algorithm. Genomics. 2011b;98:422–30. doi: 10.1016/j.ygeno.2011.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Karia PS, Han J, Schmults CD. Cutaneous squamous cell carcinoma: estimated incidence of disease, nodal metastasis, and deaths from disease in the United States, 2012. J Am Acad Dermatol. 2013;68:957–66. doi: 10.1016/j.jaad.2012.11.037. [DOI] [PubMed] [Google Scholar]
  11. Kvale MN, Hesselson S, Hoffmann TJ, Cao Y, Chan D, Connell S, et al. Genotyping informatics and quality control for 100,000 subjects in the Genetic Epidemiology Research on Adult Health and Aging (GERA) cohort. Genetics. 2015;200:1051–60. doi: 10.1534/genetics.115.178905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Marchini J, Howie B. Genotype imputation for genome-wide association studies. Nat Rev Genet. 2010;11:499–511. doi: 10.1038/nrg2796. [DOI] [PubMed] [Google Scholar]
  13. Morton CA, Birnie AJ, Eedy DJ. British Association of Dermatologists’ guidelines for the management of squamous cell carcinoma in situ (Bowen’s disease) 2014. Br J Dermatol. 2014;170:245–60. doi: 10.1111/bjd.12766. [DOI] [PubMed] [Google Scholar]
  14. Patterson N, Price AL, Reich D. Population structure and eigenanalysis. PLoS Genet. 2006;2(12):e190. doi: 10.1371/journal.pgen.0020190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Schmults CD, Karia PS, Carter JB, Han J, Qureshi AA. Factors predictive of recurrence and death from cutaneous squamous cell carcinoma: a 10-year, single-institution cohort study. JAMA Dermatol. 2013;149:541–7. doi: 10.1001/jamadermatol.2013.2139. [DOI] [PubMed] [Google Scholar]
  16. Shukla SA, Rooney MS, Rajasagi M, Tiao G, Dixon PM, Lawrence MS, et al. Comprehensive analysis of cancer-associated somatic mutations in class I HLA genes. Nat Biotechnol. 2015;33:1152–8. doi: 10.1038/nbt.3344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Siiskonen SJ, Zhang M, Li WQ, Liang L, Kraft P, Nijsten T, et al. A genome-wide association study of cutaneous squamous cell carcinoma among European descendants. Cancer Epidemiol Biomarkers Prev. 2016;25:714–20. doi: 10.1158/1055-9965.EPI-15-1070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Skulsky SL, O’Sullivan B, McArdle O, Leader M, Roche M, Conlon PJ, et al. Review of high-risk features of cutaneous squamous cell carcinoma and discrepancies between the American Joint Committee on Cancer and NCCN Clinical Practice Guidelines In Oncology. Head Neck. 2017;39:578–94. doi: 10.1002/hed.24580. [DOI] [PubMed] [Google Scholar]
  19. Smoller BR. Squamous cell carcinoma: from precursor lesions to high-risk variants. Mod Pathol. 2006;19(Suppl 2):S88–92. doi: 10.1038/modpathol.3800509. [DOI] [PubMed] [Google Scholar]
  20. Visser M, Palstra RJ, Kayser M. Allele-specific transcriptional regulation of IRF4 in melanocytes is mediated by chromatin looping of the intronic rs12203592 enhancer to the IRF4 promoter. Hum Mol Genet. 2015;24:2649–61. doi: 10.1093/hmg/ddv029. [DOI] [PubMed] [Google Scholar]
  21. Wu M, Cao Y, Liu YL, Ma LL, Yang XH. Meta-analysis of the correlation between interleukin-6 promoter polymorphism-174G/C and interferon regulatory factor 4 rs12203592 polymorphism with skin cancer susceptibility. Am J Ther. 2016;23(6):e1758–67. doi: 10.1097/MJT.0000000000000429. [DOI] [PubMed] [Google Scholar]

RESOURCES