Answers are grounded in this paper and cite their sources.
Dear Editor, Seropositivity to cutaneous human papillomavirus (HPV) β‐types and β‐HPV DNA in eyebrow hairs has been associated with an increased risk of cutaneous squamous cell carcinoma (SCC).1 2 While immunocompromised individuals3 have an increased susceptibility to cutaneous HPV infection, factors predisposing to β‐HPV infection among immunocompetent individuals are largely unknown. We examined genetic factors associated with cutaneous β‐HPV infection in cancer‐free, immunocompetent individuals and assessed whether the same genetic factors were associated with SCC. Single‐nucleotide polymorphisms (SNPs) in the epidermodysplasia verruciformis (EVER)/transmembrane channel‐like (TMC) genes, regulating the anti‐HPV barrier,4 have been associated with β2‐HPV seropositivity among immunocompromised patients.5 Only one previous study evaluated the association of a single SNP in EVER2 (TMC8) with β‐HPV infection and SCC among immunocompetent individuals.6 We examined the associations of SNPs in EVER1 (TMC6)/EVER2 with β‐HPV infection in eyebrows, and β‐HPV seropositivity among cancer‐free immunocompetent individuals, as well as the association between EVER SNPs and β‐HPV DNA in SCC tumours among patients with SCC. Histologically confirmed cases of SCC (n =185) and controls (n =281) were enrolled in a previously conducted clinic‐based case–control study designed to evaluate the association between cutaneous HPV infection and SCC.1 HPV serology and eyebrow DNA data were available for 168 cases and 290 controls from the previous study.2 In the current study, additional data on the EVER genotypes were obtained from 142 cases and 265 controls. Of the 142 cases, both tumour DNA and EVER genotype data were available for 119 cases. All participants provided written informed consent and the study protocol was approved by the institutional review board. Human papillomavirus DNA was extracted from eyebrow and SCC tumour tissue with the Qiagen EZ1 DNA Tissue Kit (Qiagen, Venlo, the Netherlands), and HPV genotyping was performed using a type‐specific multiplex genotyping assay to detect DNA from 25 genus‐β HPV types (5, 8, 9, 12, 14, 15, 17, 19–25, 36–38, 47, 49, 75, 76, 80, 92, 93 and 96).7 Serum antibodies to the major capsid protein L1 for 16 cutaneous β‐HPV types (5, 8, 9, 15, 17, 20, 23, 24, 36, 38, 49, 75, 76, 92, 96 and 107) were measured using an enzyme‐linked immunosorbent assay and multiplex serology.8 Genomic DNA was extracted from eyebrow hairs, and genotyping of 21 TagSNPs in EVER genes, previously associated with cervical cancer9 or SCC,6 was performed by multiplex PCR and Luminex hybridization.9 Associations between SNPs and ‘any’ (at least one β‐HPV type vs. none), ‘single’ (one β‐HPV type vs. none) and ‘multiple’ (more than one β‐HPV type vs. one or fewer types) infections were examined. This approach was used for defining both HPV status in eyebrow hairs and HPV seropositivity. Due to the ubiquitous prevalence of β‐HPV,10 single HPV infection in eyebrow hairs or by serology may not be clinically relevant; hence, results for multiple infections among controls are presented. In contrast, SCC tumours have a lower viral DNA load than eyebrow hairs,2 therefore tumour DNA positivity was defined by the presence of a single HPV type. Among controls, SNPs were evaluated for associations with multiple β‐HPV infection overall and by species (β1: HPV types 5, 8, 12, 14, 19–21, 24, 25, 36, 47 and 93; β2: HPV types 9, 15, 17, 22, 23, 37, 38, 80 and 107), separately for eyebrow DNA positivity and seropositivity. Among cases of SCC, associations were examined between SNPs and β‐HPV DNA in SCC tumour tissue (at least one β‐HPV vs. none). All SNP associations were examined using a log‐additive model to estimate odds ratios (ORs) and 95% confidence intervals (CIs), adjusting for confounders. False discovery rates were calculated to account for multiple comparisons. Analyses were conducted using R (version 3.0.2; http://www.r-project.org/), SAS 9.1.3 (SAS Institute Inc., Cary, NC, U.S.A.) and PLINK (version 1.07) software,11 as appropriate. Compared with controls, patients with SCC were significantly older and more likely to be male (Table 1). Among controls, none of the SNPs was associated with seropositivity to multiple β‐HPV infections (Table 2). Two SNPs, rs16970829 and rs1048591, were significantly associated with increased risk of multiple β‐HPV infections in eyebrow hair (Table 2), after correcting for multiple testing, particularly with multiple β1‐HPV infections (rs16970829, OR 3·1, 95% CI 1·3–7·2; rs1048591, OR 3·8, 95% CI 1·6–9·1). The SNP rs16970829 was also positively associated with ‘any’ and ‘single’ β1‐HPV and with ‘any’ and ‘single’ β2‐HPV, while rs2290907 was associated with ‘any’ β‐HPV infection in eyebrow hair, among controls (data not shown). Among patients with SCC, rs16970829 and rs1048591 were significantly and inversely associated with β‐HPV DNA in SCC tumour tissue (OR 0·3, 95% CI 0·1–0·9 for both SNPs, Table 2). No SNPs were associated with SCC overall, after adjusting for confounders (data not shown). Characteristics of the cases of cutaneous squamous cell carcinoma (SCC) and healthy controls Values are n (%) unless stated otherwise. Some subtotals may not add up to the total value due to missing data. aBased on having ever smoked 100 cigarettes during the entire lifetime. bNo β‐human papillomavirus (HPV) DNA in eyebrow hairs or single β‐HPV type DNA used as reference. cNo seropositivity to any β‐HPV type or seropositivity to one β‐HPV type used as reference. Characteristics of the cases of cutaneous squamous cell carcinoma (SCC) and healthy controls Values are n (%) unless stated otherwise. Some subtotals may not add up to the total value due to missing data. aBased on having ever smoked 100 cigarettes during the entire lifetime. bNo β‐human papillomavirus (HPV) DNA in eyebrow hairs or single β‐HPV type DNA used as reference. cNo seropositivity to any β‐HPV type or seropositivity to one β‐HPV type used as reference. Associations of single‐nucleotide polymorphisms (SNPs) in epidermodysplasia verruciformis (EVER1/TMC6 and EVER2/TMC8) genes with multiple cutaneous human papillomavirus (HPV) infections in eyebrow hairs and serum, among controls and patients with squamous cell carcinoma (SCC) MA, minor allele; MAF, minor allele frequency. Significant values are in italics. SNPs were aintronic, b3′‐untranslated region, cnonsynonymous coding or dupstream of gene. All analyses used a log‐additive regression model, adjusted for age. eNo HPV infection or single HPV infection as reference group. *P = 0·02 false discovery rate, **P = 0·01 false discovery rate. fNo HPV infection with any type or single β1‐HPV infection vs. multiple β1‐HPV infection. gNo HPV infection with any type or single β2‐HPV infection vs. multiple β2‐HPV infection. hNo HPV DNA in SCCs as reference group. SNP locations are based on information from the Database of Single Nucleotide Polymorphisms (dbSNP).14 Associations of single‐nucleotide polymorphisms (SNPs) in epidermodysplasia verruciformis (EVER1/TMC6 and EVER2/TMC8) genes with multiple cutaneous human papillomavirus (HPV) infections in eyebrow hairs and serum, among controls and patients with squamous cell carcinoma (SCC) MA, minor allele; MAF, minor allele frequency. Significant values are in italics. SNPs were aintronic, b3′‐untranslated region, cnonsynonymous coding or dupstream of gene. All analyses used a log‐additive regression model, adjusted for age. eNo HPV infection or single HPV infection as reference group. *P = 0·02 false discovery rate, **P = 0·01 false discovery rate. fNo HPV infection with any type or single β1‐HPV infection vs. multiple β1‐HPV infection. gNo HPV infection with any type or single β2‐HPV infection vs. multiple β2‐HPV infection. hNo HPV DNA in SCCs as reference group. SNP locations are based on information from the Database of Single Nucleotide Polymorphisms (dbSNP).14 In this clinic‐based study, two SNPs (rs16970829 and rs1048591), in perfect linkage disequilibrium (R2 = 1·0), were positively associated with β‐HPV infection in eyebrow hairs of immunocompetent individuals. Interestingly, the same SNPs were inversely associated with β‐HPV DNA in SCC tumour tissues, among cases. The SNPs rs16970829 and rs104859, located in the TNRC6C gene, have been shown to bind to transcription factors (http://regulomedb.org/) and to regulate gene expression and protein synthesis,12 respectively. While our findings do not support a direct role for EVER SNPs in HPV‐associated SCC, they may be involved in the development of premalignant skin lesions that harbour β‐HPV, perhaps giving rise to SCC tumours that have lost β‐HPV gene expression during progression.13 Unlike the positive associations reported previously,6 we did not observe any associations between the variant genotype at rs7208422 and either SCC or seropositivity to HPV using a genotypic model (data not shown). In another study of immunocompromised patients, SNPs in EVER genes were associated with seropositivity to β2‐HPV among controls,5 in contrast to our null association between EVER SNPs and seropositivity to HPV in immunocompetent individuals. However, similarly to our study, SNPs in EVER genes were not associated with SCC.5 Our findings are limited by the small sample size and inadequate statistical power to detect modest SNP associations, and they may be biased if the true associations are affected by unknown or unexamined HPV types. Although the TagSNPs captured > 80% genetic variation in the EVER genes,9 rare genetic variants may have been missed. Despite these limitations, our results were based on the evaluation of multiple β‐HPV markers and provide critical evidence in support of genetic regulation of cutaneous β‐HPV infection and its possible indirect role in SCC. The authors thank Kristen Jonathan, Jill Weber, Carolyn Gerow and the University of South Florida/LCS staff for assistance with patient recruitment, and Monika Junk for technical assistance. Funding sources: The parent case–control study was supported by a Florida Biomedical grant awarded to D.E.R. (06NIR‐08). This research was also supported in part by the Miles for Moffitt Foundation Funds, and by the Tissue Core and Survey Core at H. Lee Moffitt Cancer Center & Research Institute, an NCI‐designated Comprehensive Cancer Center (P30‐CA076292). Conflicts of interest: none declared.
No takes yet. Share an insight, caveat, or question.
Hampras et al. (2015) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: