Genome-Wide Association Studies of Multiple Keratinocyte Cancers

Luba M. Pardo; Wen-Qing Li; Shih-Jen Hwang; Joris A. C. Verkouteren; Albert Hofman; André G. Uitterlinden; Peter Kraft; Constance Turman; Jiali Han; Eunyoung Cho; Joanne M. Murabito; Daniel Levy; Abrar A. Qureshi; Tamar Nijsten

doi:10.1371/journal.pone.0169873

Abstract

There is strong evidence for a role of environmental risk factors involved in susceptibility to develop multiple keratinocyte cancers (mKCs), but whether genes are also involved in mKCs susceptibility has not been thoroughly investigated. We investigated whether single nucleotide polymorphisms (SNPs) are associated with susceptibility for mKCs. A genome-wide association study (GWAS) of 1,666 cases with mKCs and 1,950 cases with single KC (sKCs; controls) from Harvard cohorts (the Nurses' Health Study [NHS], NHS II, and the Health Professionals Follow-Up Study) and the Framingham Heart Study was carried-out using over 8 million SNPs (stage-1). We sought to replicate the most significant statistical associations (p-value≤ 5.5x10^-6) in an independent cohort of 574 mKCs and 872 sKCs from the Rotterdam Study. In the discovery stage, 40 SNPs with suggestive associations (p-value ≤5.5x10^-6) were identified, with eight independent SNPs tagging all 40 SNPs. The most significant SNP was located at chromosome 9 (rs7468390; p-value = 3.92x10^-7). In stage-2, none of these SNPs replicated and only two of them were associated with mKCs in the same direction in the combined meta-analysis. We tested the associations for 19 previously reported basal cell carcinoma-related SNPs (candidate gene association analysis), and found that rs1805007 (MC1R locus) was significantly associated with risk of mKCs (p-value = 2.80x10^-4). Although the suggestive SNPs with susceptibility for mKCs were not replicated, we found that previously identified BCC variants may also be associated with mKC, which the most significant association (rs1805007) located at the MC1R gene.

Citation: Pardo LM, Li W-Q, Hwang S-J, Verkouteren JAC, Hofman A, Uitterlinden AG, et al. (2017) Genome-Wide Association Studies of Multiple Keratinocyte Cancers. PLoS ONE 12(1): e0169873. https://doi.org/10.1371/journal.pone.0169873

Editor: Yongshuai Jiang, Harbin Medical University, CHINA

Received: July 1, 2016; Accepted: December 22, 2016; Published: January 12, 2017

Copyright: © 2017 Pardo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data can be obtained upon request. Requests should be directed towards the management team of the Rotterdam Study (secretariat.epi@erasmusmc.nl), which has a protocol for approving data requests. Because of restrictions based on privacy regulations and informed consent of the participants, data cannot be made freely available in a public repository. For requests concerning NHS NHS II and HPFS summary statistics please contact professor Abrar Qhreshi (abrarqureshi@brown.edu). For requests concerning FHS summary statistics please contact professor Joanne Murabito (murabito@bu.edu).

Funding: The study is funded by a Vidi Grant of ZonMw (nr. 91711315). The Rotterdam Study is supported by Erasmus Medical Center and Erasmus University, Rotterdam; the Research Institute for Diseases in the Elderly; The Netherlands Genomics Initiative (NGI); The Neth-erlands Organisation for Scientific Research (NOW); The Netherlands Organization for the Health Research and Development (ZonMw); The Ministry of Education, Culture and Science; The Ministry for Health, Welfare and Sports; The European Commission (DG XII), and the Municipality of Rotterdam. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The Framingham Heart Study is funded by National Institutes of Health contract N01-HC-25195. The analytical component of this project was funded by the Division of Intramural Research, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD. The study was also supported by NIH grants UM1 CA186107, P01 CA87969, R01 CA49449, UM1 CA176726, R01 CA67262, UM1 CA167552, CA122838, CA055075, CA100264, and CA093459.

Competing interests: Dr. Abrar A. Qureshi serves as a consultant for Abbvie, Amgen, Centers for Disease Control, Janssen, Merck, Novartis, Pfizer, with honorarium donated to charity. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Introduction

Basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) of the skin are known together as keratinocyte carcinomas (KC), since they both originate from keratinocytes of the epidermal layer of the skin, and share similar risk factors, treatments and prognosis [1]. KC is the most common cancer in adults of northern-European descent and is becoming a major health burden due to the high prevalence and increasing incidence in Western countries [2, 3]. A systematic review showed that patients with a primary BCC or SCC are likely to develop subsequent KCs with proportions as high as 44% in USA and 32% in The Netherlands [4]. However, it was recently shown that patients with only single KCs have a lower risk for subsequent KCs when compared with patients with a history of two or more KCs suggesting a differential risk profile of patients with single KCs than patients with a history of prior multiple KCs [3].

Environmental, tumour, and individual risk factors, including ultraviolet radiation (UVR), pale hair and skin, and male gender have been associated with an increased risk for multiple KCs (mKCs) [5–7]. There is also suggestive evidence for a genetic predisposition to mKCs, since genetic mutations in PTCH1 [1, 8] and PTCH2 [9] cause multiple BCCs [10] in individuals with nevoid BCC syndrome (NBCCS), a Mendelian disease. In addition, over 19 loci have been associated with sporadic BCC [11–13] and two with SCC. However, these previous studies included all prevalent non-melanoma skin cases and therefore it is not clear whether mKCs patients share the same genetic susceptibility variants as these with single KC.

In a recent study we found that common variants associated with BCC did not predict susceptibility for mBCC [14]. Other studies assessing genetic susceptibility in patients with mKCs are scarce. Here, we carried out a meta-analysis of GWAS on mKCs to investigate genetic susceptibility for mKCs comparing 1,241 mKCs to 2,822 single KCs (sKCs). We used patients with single KC as controls to increase the chance of identifying variants associated with susceptibility for having multiple KCs.

Materials and Methods

Study population

The nurses' health study (NHS), NHS II and the health professionals follow-up study (HPFS)–harvard cohorts.

Study participants were included from three ongoing longitudinal cohorts: NHS, NHS II and HPFS. The NHS was established in 1976 when 121,701 married, female registered nurses aged 30–55 in the US were enrolled using a mailed questionnaire inquiring about their medical history and lifestyle practices. Between 1989 and 1990, blood samples were collected from 32,826 cohort members. NHS II began in 1989 when 116,430 female nurses aged 25–42 completed a mailed questionnaire. Between 1996 and 1998, blood samples were collected from 29,616 cohort members. The HPFS consisted of 51,529 male health professionals who completed their baseline questionnaire in 1986. Between 1993 and 1994, blood samples were collected from 18,159 cohort members. Information on lifestyle factors and medical history was collected biennially by mailed questionnaire. The follow-up rate exceeds 90% in each cohort. The study protocol was approved by the Institutional Review Board of Brigham and Women's Hospital and the Harvard School of Public Health.

We combined data from several case-control studies nested within the cohorts for type 2 diabetes (NHS and HPFS), coronary heart disease (NHS and HPFS), breast cancer (NHS and NHS II), colon cancer (NHS and HPFS), kidney stone (NHS, NHS II and HPFS), advanced prostate cancer (HPFS), endometrial cancer (NHS), gout (NHS and HPFS), glaucoma (NHS and HPFS), mammographic density (NHS), and pancreatic cancer (NHS and HPFS). The description of the studies is presented elsewhere [13].

mKCs case ascertainment. Participants reported diagnoses of cancers biennially. Medical records were reviewed to confirm the diagnoses. Medical records were not obtained for self-reported cases of BCC, but previous studies showed high validity of BCC self-reports [15, 16]. Information on the cumulative number of KCs was collected in 2004 (NHS), 2005 (NHS II) and 2008 (HPFS); details are presented elsewhere [5, 7]. A validation study among 200 cases who reported 5–10 and ≥11 KC showed a confirmation rate of 92% [7]. All the participants included in the analysis were Caucasians who reported at least one pathologically confirmed diagnosis of SCC or self-reported BCC in the cohort follow-up. For this study, cases were defined as individuals with more than one KC (mKCs) and controls were defined as those with single KC (sKCs).

The framingham heart study.

The Framingham Heart Study (FHS) is a community-based prospective study that began in 1948 to characterize cardiovascular disease and its risk factors. The Original Cohort was composed of 5,209 Framingham residents primarily of white European-ancestry. In 1971, 5,124 offspring of the Original Cohort and their spouses were recruited into the Offspring Cohort. In 2002, 4095 children of the offspring cohort were invited to the Third Generation Cohort. The study design and participant descriptions of the three cohorts have been published elsewhere [17–19].

mKCs case ascertainment. Participants have undergone routine research examinations every two to six years. Cancer cases were identified at the research examinations or by medical history updates for participants who did not attend an examination. Two independent reviewers examined the medical records of all cancer cases and used the World Health Organization ICD-O coding and in 2010 ICD-10 coding to classify all primary tumours. All skin cancer cases were verified with pathology reports. FHS participants with GWAS genotype information and with pathologically confirmed skin cancer (until December 31 2013, melanoma excluded) were included in the current study. Participants with more than one KC were defined as cases and these with single KCs were defined as controls.

The rotterdam study (RS).

The RS is a prospective population-based follow-up study of the determinants and prognosis of chronic diseases, including skin cancer, in the elderly [20]. The RS consists of a major cohort (RS-I) and two extensions (RS-II and RS-III). RS-I started in 1990 and included 7,983 participants living in the Ommoord district (Rotterdam, the Netherlands). RS-II began in 2000 and now includes 3,011 participants. RS-III was started in 2006 and now includes 3,932 participants. By the end of 2008, the RS comprised 14,926 subjects aged 45 years or over. The RS consists predominantly (90%) of participants of North-European ancestry. A detailed description of the design of the RS is presented elsewhere [20].The Medical Ethics Committee of the Erasmus Medical Center and the review board of the Dutch Ministry of Health, Welfare and Sports have ratified the RS. Written informed consent was obtained from each participant.

mKCs case ascertainment. To identify histopathologically confirmed BCCs, SCCs and melanomas, RS participants were linked with the nationwide registry of histo- and cytopathology in the Netherlands (PALGA; up to 31st December 2013) [21]. The case definition for KC has been described previously [14]. In the majority of reports extracted from PALGA it was possible to distinguish between participants with single or subsequent tumours. If the diagnosis or the number of unique KC remained unclear, the medical files were searched by hand and a consensus decision was made. The number of KC was recorded separately for BCC and SCC. Individuals with either a single BCC or SCC were considered as controls while multiple BCC and/or SCC were taken as mKCs cases.

Genotyping and imputation

Details of DNA collection, genotyping and quality control for the Harvard cohorts [13, 22], the FHS [23] and the RS [20] cohorts has been detailed elsewhere. A summary of the genotyping quality control for the all cohorts is presented in the supplementary file S1 Appendix.

Harvard GWAS.

Genotyping was performed on three platforms: Affymetrix (n = 1230: 539 controls, 691 cases), Illumina HumanHap (n = 845: 363 controls, 482 cases), and Illumina Omni Express (n = 645: 287 controls, 358 cases). The genotypes per platform were merged from the different cohorts (NHS, NH II and HPFS) [13] and thus, had men and women. Based on combined GWAS genotypes on each genotyping platform and the 1000 Genomes Project ALL Phase I Integrated Release Version 3 Haplotypes (2010–11 data freeze, 2012-03-14 haplotypes) as reference panel, we imputed the genotypes of markers in the 1000 Genomes Project using MACHv.1.0.18.c [24]. Only SNPs with imputation Rsq > 0.95 and minor allele frequency (MAF) >1% were included in meta-analysis.

FHS GWAS.

Genotyping was conducted using the Affymetrix 500K mapping array and the Affymetrix 50K gene-focused molecular imprinted polymer array. We imputed using 1000Genomes Phase I Version 3 as the reference panel using MACH-Minimac [24]. SNPs with MAF ≤1% and imputation quality value <0.3 were excluded.

RS GWAS.

Details of genotyping approach is presented elsewhere [20]. Briefly, cohorts RS-I and RS-II were genotyped with the Infinium II HumanHap550K Genotyping BeadChip version 3 (Illumina, San Diego, California USA) and the cohort RS-II was genotyped using the Illumina Human 610 Quad Arrays. We imputed the RS-I, RS-II and RS-III cohorts separately, using 1000Genomes (GIANT Phase I version 3) as the reference panel and using MACH-Minimac with default parameters [24]. Next, markers with a MAF ≤1% and an imputation quality score (Rsq) < 0.3 were removed.

Statistical analysis

Stage-1; discovery phase.

The discovery samples (stage-1) consisted of the Harvard cohorts (NHS, NHS II, and HPFS) and the FHS cohort. The association analyses between the SNPs and mKCs were performed using an additive logistic regression model on subjects with more than one KC as cases and subjects with only one KC as controls. As the Harvard cohorts were genotyped on three different platforms [13], GWAS analyses were conducted for each platform, adjusting for age at first diagnosis of SCC/BCC, sex and four principal components of genetic variance (PCAs) using ProbABEL [25]. The association for each SNP from three platforms for the Harvard cohorts was combined in an inverse-variance-weighted meta-analysis using METAL [26].

The FHS GWAS was carried out using an additive generalized estimation equation (GEE) model [27] that takes into account the pedigree structure of the FHS study. The model was adjusted for age at first diagnosis, sex and four PCAs. These analysis were performed using the R package [27].

The quality control of the GWAS summary statistics from Harvard cohorts and the FHS GWAS summary statistics was performed using the EasyQC software [28]. After quality control there were 9,001,799 markers from Harvard cohorts and 8,246,930 markers from FHS. The cleaned files of both datasets (Harvard cohorts and FHS) were meta-analysed using the inverse variance approach implemented in METAL[26]. SNP heterogeneity was tested using I² and Cochran’s Q, both of which are implemented in METAL. The inflation factor lambda (genomic control) was close to 1.0 (λ = 1.08) and therefore no further adjustments for genomic control were done The SNPs that showed significant associations with mKCs (p-value ≤ 5.5x10^-6) were selected for stage-2 phase.

Stage-2 phase; replication and joint meta-analysis.

The stage-2 analysis of the top SNPs identified in the discovery phase was carried out in the RS cohort. A logistic regression with an additive model to test for associations between SNPs and mKCs was implemented adjusting for age at diagnosis, sex and four PCs. The significance of the association was tested using the likelihood ratio test (LRT) with one degree of freedom. To correct for multiple testing, we calculated the pair-wise linkage disequilibrium (LD; r²) between the top SNPs using SNAP [29] and the p-values were adjusted by dividing the nominal p-value by the number of independent tests (SNPs were considered independent with r²≤0.6).

We also carried out a GWAS on the RS cohort as described previously [14] and used the p-values from the LRT [25] of the three cohorts for a meta-analysis. The quality control of the GWAS summary statistics per cohort was done with EasyQC [28]. After QC there were 7,898,815 markers. The cleaned files of the RS were then meta-analyzed with the FHS and Harvard cohorts using the weighted Z-score method, implemented in METAL [26]. SNP heterogeneity was tested using I² and Cochran’s Q methods. The top SNPs were annotated to genes using Ensembl (http://browser.1000genomes.org/index.html).

Results

The age and sex distribution of cases (mKCs) and controls (sKCs) for the stage-1 (discovery) and stage-2 cohorts are presented in Table 1. The discovery cohorts consisted of 1,666 subjects with mKCs and 1,950 subjects with sKCs. The ascertainment of KC was done primarily using self-reports in the Harvard cohorts. There were differences in the proportion of males in the cohorts, since the NHS and NHS II are women’s cohorts and HPFS is men’s cohort. We also observed that cases were older than controls.

Download:

Table 1. Demographic characteristics of the population-based cohort.

https://doi.org/10.1371/journal.pone.0169873.t001

In the discovery stage, suggestive genome-wide associations (p-value ≤5.5x10^-6) were identified for 40 SNPs (Fig 1 and S1 Table). Due to the strong LD (r²>0.6) among these 40 SNPs (eight SNPs tagged 32 of the top SNPs), only eight of them were considered independent signals (S1 Table). The most significant hit was an intergenic SNP on the short arm of chromosome 9 (rs7468390, p-value = 3.92x10^-7), with an OR (95% CI) of 0.73 (0.64–0.82) for the C allele (Table 2). This is a common SNP in strong LD with 13 other SNPs with suggestive associations (S1 Table and S1 Fig). The region of LD of rs7468390 spans approximately 13 kb. Of the 40 SNPs with suggestive associations, 29 were intergenic, three mapped to non-coding RNA, four within regulatory regions and four to the NCKAP5 gene (S1 Table).

Download:

Fig 1. Manhattan plot of the GWAS associations for mKCs in the discovery sample (FHS and Harvard cohorts).

The observed -log10 p-values (Y-axis) of the association between the SNPs and susceptibility for mKC are shown. All SNP are represented by dots and displayed per chromosome (X-axis).

https://doi.org/10.1371/journal.pone.0169873.g001

Download:

Table 2. Top SNPs identified in the discovery samples (Harvard cohorts and FHS) and p-values of the stage 2 in the RS and joint-meta-analysis stages (Harvard cohorts, FHS and RS- all combined).

https://doi.org/10.1371/journal.pone.0169873.t002

For the stage-2, we tested for associations between the 40 top SNPs from the stage-1 and mKC in an independent sample of 574 mKCs and 872 sKCs from the RS using an adjusted p-value of 0.006 (corresponding to a p-value of 0.05 divided by eight independent SNPs/tests. None of the SNPs replicated at this threshold (S2 Table). A combined analysis of both the stage-1 and stage-2 datasets showed suggestive associations for the 40 significant SNPs, but none reached genome-wide significance, and there was significant heterogeneity in the estimates, most likely due to the different direction of the effects in the RS [30] (Table 2 and S2 Table).

Other than the above-described 40 SNPs, we found suggestive statistical signals in the combined meta-analysis (p-values ≤5.5x10^-6, S3 Table) for other SNPs. The most significant SNP was rs4761496 that mapped to an intergenic region of chromosome 12 (p-value = 4.5x10^-7). Other SNPs with suggestive associations mapped to protein coding genes including CSMD1 (rs11777268, rs116045237, rs17393453) and PRF1 (rs35401316), both of which have an indirect involvement with SCC [31, 32].

Since up to 80% of KCs are BCCs, we also looked at whether SNPs previously associated with susceptibility for BCC conferred susceptibility to mKCs. Six of the 19 SNPs tested were significantly associated with susceptibility for mKCs (Table 3). However, only rs1805007, which that mapped to MC1R was significant (p-value = 2.8x10^-4) after Bonferroni correction (adjusted p-value≤0.0026, 0.05/19). This gene is a well-known susceptibility locus for melanoma and KC. Other candidate loci identified in a recent GWAS for mBCC [14] were investigated but none were significantly associated with mKCs after Bonferroni correction (data not shown).

Download:

Table 3. Association analysis of BCC-loci and mKC susceptibility from the combined analyses (Harvard cohorts, FHS and RS).

https://doi.org/10.1371/journal.pone.0169873.t003

Discussion

In this two-stage GWAS of mKCs, we did not identify genome-wide significant associations between SNPs and mKCs. Several SNPs with suggestive associations mapped to genes involved in cancer pathology, but the findings need to be confirmed in larger samples. A candidate SNP-based analysis of previous BCC/SCC variants showed significant associations between mKCs and only one SNP (rs1805007) at MC1R, known to be associated with BCC, SCC and melanoma was significant. This suggests that genetic susceptibility for mKCs may partly overlap with that for BCC, which is expected given that up to 80% of KC are BCCs.

The lack of replication of the suggestive associations in might be due to several factors. First, phenotypic heterogeneity due to a differential ascertainment of KCs (pathology-confirmed versus self-reports) in the cohorts could have led to some phenotypic heterogeneity, although common variants for KCs have been replicated in the NHS, NHS II and HPFS cohorts [13] as well as in RS [14]. In addition, the ratio BCC/SCC may be different for the American and the European populations. In the RS, BCCs accounted for 82% of all mKCs. For the USA cohorts, BCC/SCC ratios were not available, but a higher proportion of SCCs after prior KCs in the USA were shown previously [4]. In addition, this study was underpowered to detect variants with small to moderate effects (S2 Fig). Indeed, in the joint meta-analysis it was shown that the eight variants with the most significant associations in the discovery samples had the opposite direction in the RS, which led to significant heterogeneity (Table 2). This may have caused a drop in the significance of the associations in the meta-analysis [30]. Interestingly, we found other SNPs hits in the joint analysis that had the same direction in the three cohorts, although the sample size was not large enough to reach genome-wide significance (lowest p-value was 4.5x10^-7). Most likely, the lack of replication is a combination of both phenotype heterogeneity and low power to detect variants with moderate to low effects in the RS. Last but not least, one may argue that due to differences in the imputation quality thresholds between the Harvard cohorts and the RS and FHS, we may have missed GWAS hits. However, we did not expect a dramatic drop in power due to this reason because the Harvard cohorts, where a very stringent threshold was used to include SNPs for final meta-analysis (Rsq≥0.95) provided most of the markers (9,001,799 SNPS).

In the candidate SNP analysis nested within the GWAS, we found that MC1R, a gene previously associated with BCC was also associated with an increased risk for mKCs. This contrasts with a recent study from the RS where no association between known BCC-SNPs and susceptibility for mBCC was found [14]. Since the BCC cases were included in the replication dataset of this study, this shows that the previous findings were most likely due to a lack of power of the RS. Although only one of the 19 BCC-related SNPs was significant after Bonferoni correction, we found nominal associations for six of the previously identified BCC SNPs, suggesting that larger sample sizes will be necessary to validate these associations.

As shown previously [33], most variants identified through GWAS are expected to have low to moderate risks effects and therefore large consortia of participants with phenotype and GWA-SNP data are needed. While this is feasible for traits such weight or blood pressure for disease-related phenotypes this can be challenging. As mentioned above, all previous GWAS studies of BCC or non-melanoma skin cancer published so far did not separate cases with mKCs from those with sKC, and thus our series of mKCs cases could be considered as a rare phenotype. With our findings one may argue that there are no common variants with strong effects contributing to the genetic susceptibility for mKCs, although we only tested eight million common SNPs (frequencies higher than 2%). Whether the differential risk between patients with mKC and sKCs is due to genes or mostly due to environmental factors, or an interaction of genes and environmental factors remains to be elucidated. We did not test for SNP and environmental interactions that may be relevant in explaining susceptibility to mKCs, because we did not have all environmental risk factors assessed in all cohorts and the sample size was already small to detect SNP main effects. Heritability studies could help to determine to what extent genetic risk factors explain susceptibility for mKC. We found an heritability of 8% using GWAS data from the RS (data not shown), but the power was low to have a significant estimate. Determining the heritability for mKC as well as to identify individual susceptibility loci will require larger consortia of well characterized cases and controls. In addition, rare variants were not evaluated. Although such variants may not be clinically relevant to predict disease risk, they may reveal new pathways predisposing to mKCs and new targets for drug discovery, as in the example of vismodegib, a drug used to treat patients with NBCCS and sporadic, metastatic BCC [34].

Conclusion

We found suggestive associations of common variants that were not replicated. To identify new loci and to confirm the suggestive associations found in this study, larger mKCs cohorts will be required.

Supporting Information

S1 Appendix. Quality control of the genotyping for the NHSI-II-HPFS, FHS and RS GWAS and supplementary references.

https://doi.org/10.1371/journal.pone.0169873.s001

(DOCX)

S1 Fig. Regional plot of the most significant associations between SNPs in the short arm of chromosome 9 and mKCs.

The plots represents the LD patterns of the most significant SNP in the study (rs468390) and nearby SNPs from this study (+/- 500 kb). Pairwise r2 is represented in colours. The log p-values of the associations of the rs468390 SNP and markers from the study is presented in the left Y-axis and the recombination rates is presented in the right Y-axis. The physical position of the markers is presented in Mb. The figure was generated using LocusZoom.

https://doi.org/10.1371/journal.pone.0169873.s002

(TIF)

S2 Fig. Power calculation of the study design.

The power of the study was calculated using the program CaTS with sample size, p-value (1x10^-6) and a disease prevalence of 10% as fixed parameters. An 80% power was expected for markers with MAF>25% and Odd ratios of >1.3

https://doi.org/10.1371/journal.pone.0169873.s003

(TIF)

S1 Table. Summary statistics of the most significant statistical associations between SNPs and mKCs in the discovery sample.

The table presents the summary statistics of the eight regions with the most significant SNP associations and mKCs in the discovery sample.

https://doi.org/10.1371/journal.pone.0169873.s004

(XLSX)

S2 Table. Summary statistics of the most significant statistical associations from the discovery phase in the replication and joint meta-analysis phase.

The table presents the summary statistics of the most significant associations that were followed in the replication sample and also in the combined analysis (meta-analysis).

https://doi.org/10.1371/journal.pone.0169873.s005

(XLSX)

S3 Table. Summary statistics of the most significant statistical associations for other SNPs and mKCs in the joint meta-analysis phase.

The table presents the summary statistics of the most significant associations in the combined analysis (meta-analysis).

https://doi.org/10.1371/journal.pone.0169873.s006

(XLSX)

Acknowledgments

Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services.

The authors are grateful to the study participants, the staff from the Rotterdam Study and the participating general practitioners and pharmacists. We thank staff from Internal Medicine for generation and maintenance of GWAS data (P. Arp, M. Jhamai, M. Verkerk, L. Herrera, M. Peters; and imputation (K. Estrada, M V. Struchalin). We also thank E. den Broek, L. Overbeek and S Koljenovic for her help in the linkage with the pathology data from PALGA. We would like to thank the participants and staff of the Nurses’ Health Study, Nurses’ Health Study II, and Health Professionals Follow-up Study for their valuable contributions as well as the following state cancer registries for their help: AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KY, LA, ME, MD, MA, MI, NE, NH, NJ, NY, NC, ND, OH, OK, OR, PA, RI, SC, TN, TX, VA, WA, and WY. The authors assume full responsibility for analyses and interpretation of these data.

Author Contributions

Conceptualization: LMP WQL SJH TN AAQ.
Data curation: LM WQL SJH.
Formal analysis: LMP WQL SJH.
Funding acquisition: TN.
Investigation: LMP WQL SJH.
Methodology: LMP WQL SJH TN AAQ.
Project administration: LMP WQL SJH AAQ TN.
Resources: AH AGU PK CT JH EC JMM DL.
Supervision: LMP WQL SJH AAQ TN.
Visualization: LMP WQL SJH.
Writing – original draft: LM WQL SJH.
Writing – review & editing: JACV AH AGU PK CT EC JMM DL.

References

1. Madan V, Lear JT, Szeimies RM. Non-melanoma skin cancer. Lancet. 2010;375(9715):673–85. Epub 2010/02/23. pmid:20171403
- View Article
- PubMed/NCBI
- Google Scholar
2. Hollestein LM, de Vries E, Aarts MJ, Schroten C, Nijsten TEC. Burden of disease caused by keratinocyte cancer has increased in The Netherlands since 1989. Journal of the American Academy of Dermatology. 2014;71(5):896–903. pmid:25190484
- View Article
- PubMed/NCBI
- Google Scholar
3. Wehner MR, Linos E, Parvataneni R, Stuart SE, Boscardin W, Chren M. TIming of subsequent new tumors in patients who present with basal cell carcinoma or cutaneous squamous cell carcinoma. JAMA Dermatology. 2015;151(4):382–8. pmid:25588079
- View Article
- PubMed/NCBI
- Google Scholar
4. Flohil SC, van der Leest RJ, Arends LR, de Vries E, Nijsten T. Risk of subsequent cutaneous malignancy in patients with prior keratinocyte carcinoma: a systematic review and meta-analysis. Eur J Cancer. 2013;49(10):2365–75. Epub 2013/04/24. pmid:23608733
- View Article
- PubMed/NCBI
- Google Scholar
5. Qureshi AA, Wei-Passanese EX, Li T, Han J. Host risk factors for the development of multiple non-melanoma skin cancers. J Eur Acad Dermatol Venereol. 2013;27(5):565–70. Epub 2012/03/01. PubMed Central PMCID: PMC4196674. pmid:22360801
- View Article
- PubMed/NCBI
- Google Scholar
6. Verkouteren JA, Smedinga H, Steyerberg EW, Hofman A, Nijsten T. Predicting the Risk of a Second Basal Cell Carcinoma. J Invest Dermatol. 2015. Epub 2015/06/30.
- View Article
- Google Scholar
7. Wei-Passanese EX, Han J, Lin W, Li T, Laden F, Qureshi AA. Geographical Variation in Residence and Risk of Multiple Nonmelanoma Skin Cancers in US Women and Men. Photochemistry and Photobiology. 2012;88(2):483–9. pmid:22211791
- View Article
- PubMed/NCBI
- Google Scholar
8. Hahn H, Wicking C, Zaphiropoulos PG, Gailani MR, Shanley S, Chidambaram A, et al. Mutations of the Human Homolog of Drosophila patched in the Nevoid Basal Cell Carcinoma Syndrome. Cell. 1996;85(6):841–51. pmid:8681379
- View Article
- PubMed/NCBI
- Google Scholar
9. Smyth I, Narang MA, Evans T, Heimann C, Nakamura Y, Chenevix-Trench G, et al. Isolation and characterization of human patched 2 (PTCH2), a putative tumour suppressor gene inbasal cell carcinoma and medulloblastoma on chromosome 1p32. Hum Mol Genet. 1999;8(2):291–7. Epub 1999/02/05. pmid:9931336
- View Article
- PubMed/NCBI
- Google Scholar
10. Farndon PA, Del Mastro RG, Evans DG, Kilpatrick MW. Location of gene for Gorlin syndrome. Lancet. 1992;339(8793):581–2. Epub 1992/03/07. pmid:1347096
- View Article
- PubMed/NCBI
- Google Scholar
11. Stacey SN, Gudbjartsson DF, Sulem P, Bergthorsson JT, Kumar R, Thorleifsson G, et al. Common variants on 1p36 and 1q42 are associated with cutaneous basal cell carcinoma but not with melanoma or pigmentation traits. Nat Genet. 2008;40(11):1313–8. Epub 2008/10/14. pmid:18849993
- View Article
- PubMed/NCBI
- Google Scholar
12. Stacey SN, Helgason H, Gudjonsson SA, Thorleifsson G, Zink F, Sigurdsson A, et al. New basal cell carcinoma susceptibility loci. Nat Commun. 2015;6:6825. Epub 2015/04/10. PubMed Central PMCID: PMC4403348. pmid:25855136
- View Article
- PubMed/NCBI
- Google Scholar
13. Zhang M, Song F, Liang L, Nan H, Zhang J, Liu H, et al. Genome-wide association studies identify several new loci associated with pigmentation traits and skin cancer risk in European Americans. Human Molecular Genetics. 2013;22(14):2948–59. pmid:23548203
- View Article
- PubMed/NCBI
- Google Scholar
14. Verkouteren JA, Pardo LM, Uitterlinden AG, Hofman A, Nijsten T. Common Variants Affecting Susceptibility to Develop Multiple Basal Cell Carcinomas. J Invest Dermatol. 2015. Epub 2015/03/25.
- View Article
- Google Scholar
15. Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson L, Rosner B, et al. Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol. 1986;123(5):894–900. Epub 1986/05/01. pmid:3962971
- View Article
- PubMed/NCBI
- Google Scholar
16. Hunter DJ, Colditz GA, Stampfer MJ, Rosner B, Willett WC, Speizer FE. Diet and risk of basal cell carcinoma of the skin in a prospective cohort of women. Ann Epidemiol. 1992;2(3):231–9. Epub 1992/05/01. pmid:1342273
- View Article
- PubMed/NCBI
- Google Scholar
17. Castelli WP. The Framingham Heart Study—the future. Med Sect Proc. 1987:127–38. Epub 1987/01/01. pmid:3452822
- View Article
- PubMed/NCBI
- Google Scholar
18. Feinleib M, Kannel WB, Garrison RJ, McNamara PM, Castelli WP. The Framingham Offspring Study. Design and preliminary data. Prev Med. 1975;4(4):518–25. Epub 1975/12/01. pmid:1208363
- View Article
- PubMed/NCBI
- Google Scholar
19. Splansky GL, Corey D, Yang Q, Atwood LD, Cupples LA, Benjamin EJ, et al. The Third Generation Cohort of the National Heart, Lung, and Blood Institute's Framingham Heart Study: design, recruitment, and initial examination. Am J Epidemiol. 2007;165(11):1328–35. Epub 2007/03/21. pmid:17372189
- View Article
- PubMed/NCBI
- Google Scholar
20. Hofman A, Brusselle GO, Murad S, van Duijn C, Franco O, Goedegebure A, et al. The Rotterdam Study: 2016 objectives and design update. Eur J Epidemiol. 2015;30(8):661–708. pmid:26386597
- View Article
- PubMed/NCBI
- Google Scholar
21. Casparie M, Tiebosch AT, Burger G, Blauwgeers H, van de Pol A, van Krieken JH, et al. Pathology databanking and biobanking in The Netherlands, a central role for PALGA, the nationwide histopathology and cytopathology data network and archive. Cell Oncol. 2007;29(1):19–24. Epub 2007/04/13. pmid:17429138
- View Article
- PubMed/NCBI
- Google Scholar
22. Lindström S, Loomis S, Chen C, Hongyan H, Huang J, Aschard H, et al. A comprehensive survey of genetic variation in 20,691 subjects from four large cohorts. Submitted. 2015.
23. Newman AB, Walter S, Lunetta KL, Garcia ME, Slagboom PE, Christensen K, et al. A Meta-analysis of Four Genome-Wide Association Studies of Survival to Age 90 Years or Older: The Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2010;65A(5):478–87.
- View Article
- Google Scholar
24. Howie B, Fuchsberger C, Stephens M, Marchini J, Abecasis GR. Fast and accurate genotype imputation in genome-wide association studies through pre-phasing. Nat Genet. 2012;44(8):955–9. Epub 2012/07/24. PubMed Central PMCID: PMC3696580. pmid:22820512
- View Article
- PubMed/NCBI
- Google Scholar
25. Aulchenko YS, Struchalin MV, van Duijn CM. ProbABEL package for genome-wide association analysis of imputed data. BMC Bioinformatics. 2010;11:134. Epub 2010/03/18. PubMed Central PMCID: PMC2846909. pmid:20233392
- View Article
- PubMed/NCBI
- Google Scholar
26. Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010;26(17):2190–1. Epub 2010/07/10. PubMed Central PMCID: PMC2922887. pmid:20616382
- View Article
- PubMed/NCBI
- Google Scholar
27. Chen MH, Liu X, Wei F, Larson MG, Fox CS, Vasan RS, et al. A comparison of strategies for analyzing dichotomous outcomes in genome-wide association studies with general pedigrees. Genet Epidemiol. 2011;35(7):650–7. Epub 2011/08/06. PubMed Central PMCID: PMC3197768. pmid:21818773
- View Article
- PubMed/NCBI
- Google Scholar
28. Winkler TW, Day FR, Croteau-Chonka DC, Wood AR, Locke AE, Magi R, et al. Quality control and conduct of genome-wide association meta-analyses. Nat Protoc. 2014;9(5):1192–212. Epub 2014/04/26. PubMed Central PMCID: PMC4083217. pmid:24762786
- View Article
- PubMed/NCBI
- Google Scholar
29. Johnson AD, Handsaker RE, Pulit SL, Nizzari MM, O'Donnell CJ, de Bakker PI. SNAP: a web-based tool for identification and annotation of proxy SNPs using HapMap. Bioinformatics. 2008;24(24):2938–9. Epub 2008/11/01. PubMed Central PMCID: PMC2720775. pmid:18974171
- View Article
- PubMed/NCBI
- Google Scholar
30. Thompson JR, Attia J, Minelli C. The meta-analysis of genome-wide association studies. Briefings in Bioinformatics. 2011;12(3):259–69. pmid:21546449
- View Article
- PubMed/NCBI
- Google Scholar
31. Huang SJ, Hijnen D, Murphy GF, Kupper TS, Calarese AW, Mollet IG, et al. Imiquimod enhances IFN-gamma production and effector function of T cells infiltrating human squamous cell carcinomas of the skin. J Invest Dermatol. 2009;129(11):2676–85. Epub 2009/06/12. PubMed Central PMCID: PMC2841955. pmid:19516264
- View Article
- PubMed/NCBI
- Google Scholar
32. Sun PC, Uppaluri R, Schmidt AP, Pashia ME, Quant EC, Sunwoo JB, et al. Transcript Map of the 8p23 Putative Tumor Suppressor Region. Genomics. 2001;75(1–3):17–25. pmid:11472063
- View Article
- PubMed/NCBI
- Google Scholar
33. Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, et al. Finding the missing heritability of complex diseases. Nature. 2009;461(7265):747–53. pmid:19812666
- View Article
- PubMed/NCBI
- Google Scholar
34. Sekulic A, Migden MR, Oro AE, Dirix L, Lewis KD, Hainsworth JD, et al. Efficacy and safety of vismodegib in advanced basal-cell carcinoma. N Engl J Med. 2012;366(23):2171–9. Epub 2012/06/08. pmid:22670903
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Madan V, Lear JT, Szeimies RM. Non-melanoma skin cancer. Lancet. 2010;375(9715):673–85. Epub 2010/02/23. pmid:20171403
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Hollestein LM, de Vries E, Aarts MJ, Schroten C, Nijsten TEC. Burden of disease caused by keratinocyte cancer has increased in The Netherlands since 1989. Journal of the American Academy of Dermatology. 2014;71(5):896–903. pmid:25190484
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Wehner MR, Linos E, Parvataneni R, Stuart SE, Boscardin W, Chren M. TIming of subsequent new tumors in patients who present with basal cell carcinoma or cutaneous squamous cell carcinoma. JAMA Dermatology. 2015;151(4):382–8. pmid:25588079
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Flohil SC, van der Leest RJ, Arends LR, de Vries E, Nijsten T. Risk of subsequent cutaneous malignancy in patients with prior keratinocyte carcinoma: a systematic review and meta-analysis. Eur J Cancer. 2013;49(10):2365–75. Epub 2013/04/24. pmid:23608733
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Qureshi AA, Wei-Passanese EX, Li T, Han J. Host risk factors for the development of multiple non-melanoma skin cancers. J Eur Acad Dermatol Venereol. 2013;27(5):565–70. Epub 2012/03/01. PubMed Central PMCID: PMC4196674. pmid:22360801
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Verkouteren JA, Smedinga H, Steyerberg EW, Hofman A, Nijsten T. Predicting the Risk of a Second Basal Cell Carcinoma. J Invest Dermatol. 2015. Epub 2015/06/30.
View Article
Google Scholar

[22] View Article

[23] Google Scholar

[ref7] 7. Wei-Passanese EX, Han J, Lin W, Li T, Laden F, Qureshi AA. Geographical Variation in Residence and Risk of Multiple Nonmelanoma Skin Cancers in US Women and Men. Photochemistry and Photobiology. 2012;88(2):483–9. pmid:22211791
View Article
PubMed/NCBI
Google Scholar

[25] View Article

[26] PubMed/NCBI

[27] Google Scholar

[ref8] 8. Hahn H, Wicking C, Zaphiropoulos PG, Gailani MR, Shanley S, Chidambaram A, et al. Mutations of the Human Homolog of Drosophila patched in the Nevoid Basal Cell Carcinoma Syndrome. Cell. 1996;85(6):841–51. pmid:8681379
View Article
PubMed/NCBI
Google Scholar

[29] View Article

[30] PubMed/NCBI

[31] Google Scholar

[ref9] 9. Smyth I, Narang MA, Evans T, Heimann C, Nakamura Y, Chenevix-Trench G, et al. Isolation and characterization of human patched 2 (PTCH2), a putative tumour suppressor gene inbasal cell carcinoma and medulloblastoma on chromosome 1p32. Hum Mol Genet. 1999;8(2):291–7. Epub 1999/02/05. pmid:9931336
View Article
PubMed/NCBI
Google Scholar

[33] View Article

[34] PubMed/NCBI

[35] Google Scholar

[ref10] 10. Farndon PA, Del Mastro RG, Evans DG, Kilpatrick MW. Location of gene for Gorlin syndrome. Lancet. 1992;339(8793):581–2. Epub 1992/03/07. pmid:1347096
View Article
PubMed/NCBI
Google Scholar

[37] View Article

[38] PubMed/NCBI

[39] Google Scholar

[ref11] 11. Stacey SN, Gudbjartsson DF, Sulem P, Bergthorsson JT, Kumar R, Thorleifsson G, et al. Common variants on 1p36 and 1q42 are associated with cutaneous basal cell carcinoma but not with melanoma or pigmentation traits. Nat Genet. 2008;40(11):1313–8. Epub 2008/10/14. pmid:18849993
View Article
PubMed/NCBI
Google Scholar

[41] View Article

[42] PubMed/NCBI

[43] Google Scholar

[ref12] 12. Stacey SN, Helgason H, Gudjonsson SA, Thorleifsson G, Zink F, Sigurdsson A, et al. New basal cell carcinoma susceptibility loci. Nat Commun. 2015;6:6825. Epub 2015/04/10. PubMed Central PMCID: PMC4403348. pmid:25855136
View Article
PubMed/NCBI
Google Scholar

[45] View Article

[46] PubMed/NCBI

[47] Google Scholar

[ref13] 13. Zhang M, Song F, Liang L, Nan H, Zhang J, Liu H, et al. Genome-wide association studies identify several new loci associated with pigmentation traits and skin cancer risk in European Americans. Human Molecular Genetics. 2013;22(14):2948–59. pmid:23548203
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref14] 14. Verkouteren JA, Pardo LM, Uitterlinden AG, Hofman A, Nijsten T. Common Variants Affecting Susceptibility to Develop Multiple Basal Cell Carcinomas. J Invest Dermatol. 2015. Epub 2015/03/25.
View Article
Google Scholar

[53] View Article

[54] Google Scholar

[ref15] 15. Colditz GA, Martin P, Stampfer MJ, Willett WC, Sampson L, Rosner B, et al. Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am J Epidemiol. 1986;123(5):894–900. Epub 1986/05/01. pmid:3962971
View Article
PubMed/NCBI
Google Scholar

[56] View Article

[57] PubMed/NCBI

[58] Google Scholar

[ref16] 16. Hunter DJ, Colditz GA, Stampfer MJ, Rosner B, Willett WC, Speizer FE. Diet and risk of basal cell carcinoma of the skin in a prospective cohort of women. Ann Epidemiol. 1992;2(3):231–9. Epub 1992/05/01. pmid:1342273
View Article
PubMed/NCBI
Google Scholar

[60] View Article

[61] PubMed/NCBI

[62] Google Scholar

[ref17] 17. Castelli WP. The Framingham Heart Study—the future. Med Sect Proc. 1987:127–38. Epub 1987/01/01. pmid:3452822
View Article
PubMed/NCBI
Google Scholar

[64] View Article

[65] PubMed/NCBI

[66] Google Scholar

[ref18] 18. Feinleib M, Kannel WB, Garrison RJ, McNamara PM, Castelli WP. The Framingham Offspring Study. Design and preliminary data. Prev Med. 1975;4(4):518–25. Epub 1975/12/01. pmid:1208363
View Article
PubMed/NCBI
Google Scholar

[68] View Article

[69] PubMed/NCBI

[70] Google Scholar

[ref19] 19. Splansky GL, Corey D, Yang Q, Atwood LD, Cupples LA, Benjamin EJ, et al. The Third Generation Cohort of the National Heart, Lung, and Blood Institute's Framingham Heart Study: design, recruitment, and initial examination. Am J Epidemiol. 2007;165(11):1328–35. Epub 2007/03/21. pmid:17372189
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref20] 20. Hofman A, Brusselle GO, Murad S, van Duijn C, Franco O, Goedegebure A, et al. The Rotterdam Study: 2016 objectives and design update. Eur J Epidemiol. 2015;30(8):661–708. pmid:26386597
View Article
PubMed/NCBI
Google Scholar

[76] View Article

[77] PubMed/NCBI

[78] Google Scholar

[ref21] 21. Casparie M, Tiebosch AT, Burger G, Blauwgeers H, van de Pol A, van Krieken JH, et al. Pathology databanking and biobanking in The Netherlands, a central role for PALGA, the nationwide histopathology and cytopathology data network and archive. Cell Oncol. 2007;29(1):19–24. Epub 2007/04/13. pmid:17429138
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref22] 22. Lindström S, Loomis S, Chen C, Hongyan H, Huang J, Aschard H, et al. A comprehensive survey of genetic variation in 20,691 subjects from four large cohorts. Submitted. 2015.

[ref23] 23. Newman AB, Walter S, Lunetta KL, Garcia ME, Slagboom PE, Christensen K, et al. A Meta-analysis of Four Genome-Wide Association Studies of Survival to Age 90 Years or Older: The Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2010;65A(5):478–87.
View Article
Google Scholar

[85] View Article

[86] Google Scholar

[ref24] 24. Howie B, Fuchsberger C, Stephens M, Marchini J, Abecasis GR. Fast and accurate genotype imputation in genome-wide association studies through pre-phasing. Nat Genet. 2012;44(8):955–9. Epub 2012/07/24. PubMed Central PMCID: PMC3696580. pmid:22820512
View Article
PubMed/NCBI
Google Scholar

[88] View Article

[89] PubMed/NCBI

[90] Google Scholar

[ref25] 25. Aulchenko YS, Struchalin MV, van Duijn CM. ProbABEL package for genome-wide association analysis of imputed data. BMC Bioinformatics. 2010;11:134. Epub 2010/03/18. PubMed Central PMCID: PMC2846909. pmid:20233392
View Article
PubMed/NCBI
Google Scholar

[92] View Article

[93] PubMed/NCBI

[94] Google Scholar

[ref26] 26. Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010;26(17):2190–1. Epub 2010/07/10. PubMed Central PMCID: PMC2922887. pmid:20616382
View Article
PubMed/NCBI
Google Scholar

[96] View Article

[97] PubMed/NCBI

[98] Google Scholar

[ref27] 27. Chen MH, Liu X, Wei F, Larson MG, Fox CS, Vasan RS, et al. A comparison of strategies for analyzing dichotomous outcomes in genome-wide association studies with general pedigrees. Genet Epidemiol. 2011;35(7):650–7. Epub 2011/08/06. PubMed Central PMCID: PMC3197768. pmid:21818773
View Article
PubMed/NCBI
Google Scholar

[100] View Article

[101] PubMed/NCBI

[102] Google Scholar

[ref28] 28. Winkler TW, Day FR, Croteau-Chonka DC, Wood AR, Locke AE, Magi R, et al. Quality control and conduct of genome-wide association meta-analyses. Nat Protoc. 2014;9(5):1192–212. Epub 2014/04/26. PubMed Central PMCID: PMC4083217. pmid:24762786
View Article
PubMed/NCBI
Google Scholar

[104] View Article

[105] PubMed/NCBI

[106] Google Scholar

[ref29] 29. Johnson AD, Handsaker RE, Pulit SL, Nizzari MM, O'Donnell CJ, de Bakker PI. SNAP: a web-based tool for identification and annotation of proxy SNPs using HapMap. Bioinformatics. 2008;24(24):2938–9. Epub 2008/11/01. PubMed Central PMCID: PMC2720775. pmid:18974171
View Article
PubMed/NCBI
Google Scholar

[108] View Article

[109] PubMed/NCBI

[110] Google Scholar

[ref30] 30. Thompson JR, Attia J, Minelli C. The meta-analysis of genome-wide association studies. Briefings in Bioinformatics. 2011;12(3):259–69. pmid:21546449
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref31] 31. Huang SJ, Hijnen D, Murphy GF, Kupper TS, Calarese AW, Mollet IG, et al. Imiquimod enhances IFN-gamma production and effector function of T cells infiltrating human squamous cell carcinomas of the skin. J Invest Dermatol. 2009;129(11):2676–85. Epub 2009/06/12. PubMed Central PMCID: PMC2841955. pmid:19516264
View Article
PubMed/NCBI
Google Scholar

[116] View Article

[117] PubMed/NCBI

[118] Google Scholar

[ref32] 32. Sun PC, Uppaluri R, Schmidt AP, Pashia ME, Quant EC, Sunwoo JB, et al. Transcript Map of the 8p23 Putative Tumor Suppressor Region. Genomics. 2001;75(1–3):17–25. pmid:11472063
View Article
PubMed/NCBI
Google Scholar

[120] View Article

[121] PubMed/NCBI

[122] Google Scholar

[ref33] 33. Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, et al. Finding the missing heritability of complex diseases. Nature. 2009;461(7265):747–53. pmid:19812666
View Article
PubMed/NCBI
Google Scholar

[124] View Article

[125] PubMed/NCBI

[126] Google Scholar

[ref34] 34. Sekulic A, Migden MR, Oro AE, Dirix L, Lewis KD, Hainsworth JD, et al. Efficacy and safety of vismodegib in advanced basal-cell carcinoma. N Engl J Med. 2012;366(23):2171–9. Epub 2012/06/08. pmid:22670903
View Article
PubMed/NCBI
Google Scholar

[128] View Article

[129] PubMed/NCBI

[130] Google Scholar

Figures

Abstract

Introduction

Materials and Methods

Study population

The nurses' health study (NHS), NHS II and the health professionals follow-up study (HPFS)–harvard cohorts.

The framingham heart study.

The rotterdam study (RS).

Genotyping and imputation

Harvard GWAS.

FHS GWAS.

RS GWAS.

Statistical analysis

Stage-1; discovery phase.

Stage-2 phase; replication and joint meta-analysis.

Results

Discussion

Conclusion

Supporting Information

S1 Appendix. Quality control of the genotyping for the NHSI-II-HPFS, FHS and RS GWAS and supplementary references.

S1 Fig. Regional plot of the most significant associations between SNPs in the short arm of chromosome 9 and mKCs.

S2 Fig. Power calculation of the study design.

S1 Table. Summary statistics of the most significant statistical associations between SNPs and mKCs in the discovery sample.

S2 Table. Summary statistics of the most significant statistical associations from the discovery phase in the replication and joint meta-analysis phase.

S3 Table. Summary statistics of the most significant statistical associations for other SNPs and mKCs in the joint meta-analysis phase.

Acknowledgments

Author Contributions

References