Risk of Differentiated Thyroid Carcinoma and Polymorphisms within the Susceptibility Cancer Region 8q24

In the human genome a segment of 1Mb within 8q24, thereafter termed “susceptibility cancer region of 8q24” (SR8q24), was associated with increased risk of prostate, colon, breast, and ovarian cancer, in genome-wide association studies (GWAS). However, SR8q24 did not reach the genome-wide statistical significance threshold in previous GWAS conducted on differentiated thyroid carcinoma (DTC; refs. 1, 2). Then, several authors conducted hypothesis-driven case–control studies to verify specific single-nucleotide polymorphisms (SNP) within SR8q24 in relation to DTC. Positive associations were found among Polish (3) and British (4), whereas lack of associations was found in Spanish (5) and Texans (6). In summary, the findings were poorly reproducible also because a limited statistical power of some studies. Here, a study carried out on 1,250 patients with DTC and 1,250 healthy controls living in regions of central and southern Italy is presented. It has the adequate statistical power to evaluate six haplotype tagging SNPs to ascertain whether SR8q24 could constitute a genomic risk region also among Italians.

Cases and controls, all Caucasians, were volunteers collected at the University Hospital of Cisanello (Pisa, Italy), as described elsewhere (7). They gave their written informed consent and the study was cleared by the local Ethical Committee.

rs6983267 and rs1447295 were chosen because they were previously associated with DTC. The inclusion criteria for the other SNPs were: (i) haplotype tagging SNP of linkage disequilibrium blocks as described in ref. (8), (ii) associated with the risk of human cancer in at least five independent studies, and (iii) minor allele frequency (MAF) more than 0.05 in Caucasians. The rs10808556, rs7000448, rs13254738, and rs13281615 passed these criteria.

The study had the statistical power of 80% to detect an expected OR of 1.36 for rs6983267 (the one described by; ref. 3), considering a type I error α = 0.05 and a MAF = 0.4, in a recessive model. For SNPs with lower MAF, for example rs1447295, a minimal OR = 1.27 on per allele basis could be detected.

DNA was extracted from peripheral blood and genotyped by personnel blinded for the case/control status with the use of Predesigned TaqMan Genotyping Assays (Life Technologies, Inc.). Three percent of samples did not elicit an appropriate amount of DNA, thus the genotyping could be conducted on 1,212 cases and 1,208 controls. Two percent of DNA samples were repeated as quality control.

Hardy–Weinberg equilibrium (HWE) was tested in controls by the χ² test. Multiple logistic regression (MLR) was used to calculate the adjusted OR (OR_adj) and their 95% confidence intervals (CI), as described elsewhere (7). The Cochran-Armitage trend test was used to evaluate the best inheritance model. Bonferroni's correction was used and the novel statistical significance threshold was 8.3 × 10⁻³.

A minimal call rate of 98.7% was obtained (for rs13254738). The quality control was satisfactory (>99.5% of correspondence). The genotypes followed HWE. The characteristics of cases and controls are reported in Table 1. A positive association was observed between the smoking habit and the risk of DTC (OR, 1.63; 95% CI, 1.39–1.91; P < 10⁻⁶). The OR_adj, the 95% CIs, and their P_ass following MLR analysis of the additive models are reported in Table 2.

For rs13254738 and rs13281615, a marginal statistical significance association was observed for the follicular type (FTC), but it did not accomplish with the statistical threshold set after Bonferroni's correction. For rs10808556, the heterozygotes showed an increased risk (OR, 1.38; 95% CI, 1.13–1.68; P_ass = 0.001), whereas the homozygotes did not showed any association. The best model (dominant) showed a P_trend = 0.033 and the carriers of the C allele had an OR_adj = 1.28 (95% CI, 1.06–1.54; P_ass = 0.010). For rs1447295, the heterozygotes showed an OR, 1.35 (95% CI, 1.02–1.78; P_ass = 0.035). However, all the associations were not statistically significant following Bonferroni's correction.

In the present study, a positive association between the risk of DTC and smoking habit was found. This is conceivable given that cigarette smoke was associated with DTC predisposing conditions such as goiter, thyrotoxicosis, increased serum thyroglobulin, thyroid volume, and multiple nodules. Concerning SR8q24, no evidence of associations was found between specific haplotype tagging SNPs and the risk of DTC or papillary thyroid carcinoma (PTC). Thus, at least among Italians, SR8q24 seems not to be involved in the etiology of thyroid carcinomas. When looking more attentively to differences among studies, it should be noticed that the associations between SR8q24 and risk were detectable in populations with low incidences of DTC [in Poland and United Kingdom the age-standardized rates are 4.1 and 3.8 cases out of 100,000, respectively; International Agency for Research on Cancer (IARC), EUCAN database], whereas they were lacking in Texans and Italians where the incidence is much higher (10.8, data from Center for Disease Control and Prevention database, and 13.5, EUCAN database, respectively). Thus, it could be hypothesized that, when other environmental risk factors are prevailing, the small-size effect of SR8q24 is not measurable.

No potential conflicts of interest were disclosed.

Conception and design: M. Cipollini, D. Landi, R. Elisei, F. Gemignani, S. Landi

Development of methodology: M. Cipollini, F. Gnudi, F. De Paola

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): G. Figlioli, S. Bramante, L. Maiorano, A. Cecchini, L. Damicis, T. Frixa, D. Landi, L. Cancemi, C. De Santi, A. Cristaudo, A. Bonotti, C. Romei, R. Elisei, G. Pellegrini

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Cipollini, S. Garritano, F. Gnudi, F. De Paola, T. Frixa, R. Foddis, C. Romei, S. Landi

Writing, review, and/or revision of the manuscript: M. Cipollini, G. Figlioli, O. Melaiu, A. Cristaudo, R. Elisei, S. Landi

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S. Garritano, A. Cecchini, L. Damicis, T. Frixa, D. Landi, C. Romei

Study supervision: R. Elisei, R. Barale, S. Landi

This study was financially supported by a grant from the Istituto Toscano Tumori (ITT; year 2010) and AIRC (Associazione Italiana Ricerca Cancro) that supported the study with an investigator grant (year 2008). ITT and MIUR (Italian Ministry of the Research, PRIN) financed the costs of the reagents and AIRC and MIUR financed the costs of the fellowships.