XPA, haplotypes, and risk of basal and squamous cell carcinoma

Katie L.Miller 1 2 Margaret R.Karagas 5 Peter Kraft 2 4 David J.Hunter 2 4 Paul J.Catalano 0 3 Steven H.Byler 6 Heather H.Nelson 1 0 Department of Biostatistics 1 Department of Environmental Health 2 Department of Epidemiology 3 Department of Biostatistics and Computational Biology, Dana Farber Cancer Institute , Boston, MA, USA 4 Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School , Boston, MA, USA 5 Section of Biostatistics and Epidemiology, Department of Community and Family Medicine and Norris Cotton Cancer Center, Dartmouth Medical School , Lebanon, NH, USA 6 Department of Genetics and Complex Diseases, Harvard School of Public Health , Boston, MA, USA # The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: 1670 - Nucleotide excision repair (NER) is instrumental in removing DNA lesions caused by ultraviolet (UV) radiation, the dominant risk factor for keratinocyte carcinoma, including basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). We evaluated whether BCC or SCC risk was influenced by the A23G single nucleotide polymorphism (SNP) in Xeroderma pigmentosum group A (XPA), which codes for an essential protein in NER. We also investigated whether haplotypes of XPA, determined by seven haplotype-tagging SNPs, better define susceptibility to keratinocyte carcinoma. Incident cases of BCC and SCC from New Hampshire were identified through dermatologists and pathology laboratories. Population-based controls were frequency-matched to cases by gender and age. Cases of BCC (886) and of SCC (682) were compared with controls (796). Models controlled for age, gender, pigmentation factors and severe sunburns and were restricted to Caucasians. Using GG as the reference, the A allele was less frequent among cases of BCC (ORAG 0.82, 95% CI (0.66, 1.01); ORAA 0.74, 95% CI (0.53, 1.03); trend test P 0.03) and SCC (ORAG 0.85, 95% CI (0.67, 1.07); ORAA 0.74, 95% CI (0.52, 1.05); trend test P 0.05) than controls. Risk from 3 severe sunburns was elevated for those with the GG genotype only, and this interaction was nearly significant for BCC (P 0.07). XPA genotype also modified a relationship between SCC and the amount of pigmentation (P 0.02). Using a haplotype analysis identifying seven common XPA haplotypes indicated that the A23G polymorphism alone captured the differences in susceptibility to keratinocyte carcinoma. The common G allele of the A23G polymorphism was associated with an increased risk of BCC and SCC and this polymorphism appeared to be the determining polymorphism in XPA that alters cancer susceptibility. Abbreviations: BCC, basal cell carcinoma; CI, confidence interval; htSNP, haplotype tagging SNP; KC, keratinocyte; SNP, single nucleotide polymorphism; SCC, squamous cell carcinoma; UTR, untranslated region; XPA, xeroderma pigmentosum group A. Ultraviolet (UV) radiation from sunlight is the dominant risk factor for cancers of keratinocytes (KCs), including basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) of the skin. UV induces DNA lesions, such as pyrimidine dimers and 6,4-photoproducts, which may lead to cancer if not repaired. The nucleotide excision repair (NER) pathway is necessary to remove these DNA lesions. A rare autosomal recessive condition, Xeroderma pigmentosum (XP), demonstrates the importance of this pathway. XP occurs when a gene (e.g., predominantly from Xeroderma pigmentosum group A through G) involved in NER contains a mutation on both copies of the gene that, when translated, results in a protein that is not capable of repairing photolesions. This leads to extreme photosensitivity and an estimated 1000-fold increased risk of KC with a much earlier age at onset (1,2). Null mutations in NER genes that result in XP are rare; however, these same genes are known to be highly polymorphic (36). Little is known about how these more common polymorphisms affect the risk of KC on a population level. Xeroderma pigmentosum group A (XPA) is a gene that is necessary for NER. Null mutations in this gene lead to the most severe form of XP (7). In NER, XPA has a central role in interacting with a number of proteins, including RPA, TFIIH, and the ERCC1-XPF protein complex (8,9). A common polymorphism in XPA has been reported by several groups (1012). The A23G polymorphism, also referred to as the XPA ( 4) G-to-A polymorphism, is located in the 50-untranslated region (UTR) and is four nucleotides upstream of the start codon. Polymorphisms in this area proximal to the start codon, referred to as the Kozak sequence, could have implications for the binding of the 40S ribosomal subunit and as a result influence protein levels in the cell (13,14). One or more copies of the G allele resulted in significantly higher DNA repair capacity as measured by the host cell reactivation assay (15). Also, a reduced repair phenotype has been found to increase susceptibility to KC as well as other cancers (1517). Epidemiologic studies have observed an increased risk of lung cancer with the A allele (15,1820); however, XPA polymorphisms have not been studied in relation to KC risk. Given the essential role of XPA in repairing UV lesions, we examined whether the A23G polymorphism is related to risk of BCC and SCC. Also, we investigated gene-environment interaction between UV exposure and this common polymorphism. Further, we conducted a haplotype analysis in order to determine whether additional polymorphisms are needed to identify those who are susceptible to KC. Materials and methods Study population Newly diagnosed cases of histologically confirmed BCC and SCC in New Hampshire were identified through the collaboration of dermatologists, dermatopathologists, and pathology laboratories throughout the state and bordering regions from July 1, 1993 to June 30, 1995 (series 1) and July 1, 1997 to March 30, 2000 (series 2) (21). Eligible cases were between 25 and 74 years of age, had a listed telephone number, and spoke English. Only living cases that were mentally competent and not too ill to participate were included. All eligible SCC cases and a ratio of approximately two to one BCC cases in series 1 and one to one ratio in series 2 were selected to take part in the study. The BCC cases were randomly sampled in order to ensure representativeness of age, sex and anatomic site for all incident BCCs within New Hampshire. A complete description of ascertainment of BCC and SCC cases has been described previously (21). Population lists of New Hampshire residents obtained from the New Hampshire State Department of Transportation files were used to identify potential controls ages 2564 years. Enrollment lists from the Center for Medicaid and Medicare Services provided a source of controls ages 65 74 years. Controls were frequency-matched to the combined case groups on age and gender. A personal interview (usually conducted in the participants home) covered d (...truncated)