Abstract
Context:
ATM is critical in response to ionizing radiation-induced DNA damage.
Objective:
Variations in ATM are hypothesized to affect individual susceptibility to thyroid cancer. Our objective was to evaluate the association between ATM polymorphisms and thyroid cancer risk.
Design, Participants, and Methods:
Six ATM single nucleotide polymorphisms (SNP) were genotyped in two independent case-control series including 592 patients with differentiated thyroid carcinoma (DTC) and 885 healthy individuals. An unconditional logistic regression model was applied to calculate odds ratios (OR) and 95% confidence intervals (CI) for each SNP with respect to risk of DTC and the combination effect of SNP on cancer risk.
Results:
The risk-allele frequencies of all the SNP were similar in the two case-control populations. Under a dominant model of inheritance, the G allele of ATM rs189037 exhibited a protective effect against DTC (adjusted OR = 0.8; 95% CI, 0.6–1.0; P = 0.04), and the G allele of rs1800057 was associated with increased risk of DTC (adjusted OR = 1.9; 95% CI, 1.1–3.1; P = 0.02). A protective haplotype (A-G-C-T-C-A) was associated with decreased risk of DTC in non-Hispanic whites (adjusted OR = 0.2; 95% CI, 0.0–0.8; P = 0.03). A significant dose-response relationship was observed between the total number of risk alleles of ATM and DTC risk (P = 0.01). Carriers of a combination of six to seven and eight to 10 risk alleles were at 30% (adjusted OR = 1.3; 95% CI, 1.0–1.7) and 50% (adjusted OR = 1.5; 95% CI, 1.1–2.1) increased risk of DTC, respectively.
Conclusion:
Individual susceptibility to DTC may be attributable to polymorphisms of ATM, and the associations warrant confirmation in independent studies.
Thyroid cancer ranks among the nine most common cancers in women worldwide (1). In the United States, a marked increase in thyroid cancer incidence has occurred since the mid-1990s (2, 3). Papillary thyroid carcinoma (PTC), follicular thyroid carcinoma, and Hürthle cell carcinoma, collectively termed differentiated thyroid carcinoma (DTC), account for the vast majority of the increase in incidence.
Little is known about the causes of adult DTC. Previous studies found a strong dose-response relationship between moderate to high levels of radiation exposure and risk of childhood DTC, based on the observations of increased incidence of pediatric and adolescent DTC after the Chernobyl nuclear accident of 1986 and after head and neck radiation therapy for childhood illness (4). But a significant relationship was not observed in adults (4), and moreover, the majority of adults with DTC have no history of radiation exposure. Radiation is a carcinogen that is widely accepted to have no threshold of exposure below which there is no genotoxic effect (5). Therefore, it is reasonable to hypothesize that cumulative exposure to environmental radiation (cosmic, terrestrial, medical, etc.), to which we are constantly exposed, could be associated with development of DTC in adults. However, the statistical uncertainty due to the small excess risk expected at low levels of exposure and technical difficulty in accurately estimating cumulative exposure dose make it very unlikely that such an association could be revealed in an epidemiological study (5). Some indirect evidence for the hypothesis that cumulative environmental radiation exposure is associated with increased DTC risk comes from reports of increased cancer risk in air crew, who are exposed to increased levels of cosmic radiation because of the high altitudes at which they travel (5, 6). Should this hypothesis be true, individuals who are genetically predisposed to radiosensitivity would be expected to have an increased risk of DTC.
One proposed mechanism of DTC development involves DNA double-strand breaks (DSB), which are the primary genotoxic lesions induced by exposure to ionizing radiation. A single unrepaired DSB is sufficient to cause abnormal mitosis and results in loss of large DNA fragments. In response to this severe DNA damage, cells develop complex mechanisms to reverse and diminish the lesions, including cell cycle arrest, DNA repair and recombination, and apoptosis (7). ATM, the ataxia telangiectasia mutated protein, plays a central role in response to DSB. After radiation causing DNA damage, ATM is recruited to sites of DSB and autoactivated (8). The activated ATM phosphorylates a host of substrates, including MDM2, p53, NBS1, BRCA1, CHK2, and H2AX, and thereby mediates cell cycle control, DNA repair, and apoptosis (7, 8).
Defects of the ATM gene are responsible for ataxia telangiectasia, a rare inherited disorder characterized by progressive ataxia, radiosensitivity, cell-cycle checkpoint defects, genome instability, and a predisposition to cancer (8). Thus, it is possible that inherited functional polymorphisms in ATM could influence inherited radiosensitivity and the host capacity to repair DSB, leading to increased predisposition to DTC in affected individuals. Indeed, some population-based studies have reported significant associations between specific ATM alleles and increased risk of common cancers, including breast, lung, and prostate cancer (9–14). However, the impact of single nucleotide polymorphisms (SNP) of ATM on DTC risk has rarely been studied and merits further investigation. To date, only one study assessed thyroid cancer risk and observed decreased risk of PTC related to the A allele of ATM SNP rs1801516, regardless of history of radiation exposure, and increased risk of sporadic PTC related to mutant genotype of rs664677 (15).
To investigate the impact of ATM SNP on DTC risk, we genotyped functional SNP in 592 patients with DTC and 885 cancer-free individuals pooled from two independent hospital-based case-control studies in the United States and Brazil. The frequency distributions of these polymorphisms either alone or in combination were compared between cases and controls. The polymorphisms (namely, rs228589, rs189037, rs1800054, rs4986761, rs1800057, and rs1801516) were selected on the basis of previously published evidence that they likely have an effect on ATM protein function and/or are associated with cancer. rs228589 resides in the promoter region of ATM; it has been implicated in DNA repair capacity to repair DNA damage caused by polycyclic aromatic hydrocarbons (16). rs189037 also resides in the promoter of ATM; the alleles have different binding affinity to the transcription factor AP-2α and exhibit a different level of mRNA expression (17). rs1800054 (S49C) has been suggested by several epidemiological studies to be a susceptible polymorphism for breast cancer (12–14), although its functionality has not been reported. rs4986761 (S707P), rs1800057 (P1054R), and rs1801516 (D1853N) have been linked to clinical radiosensitivity (18–21); these SNP lie within highly conserved regions, predicting potential functional significance (22).
Subjects and Methods
Study population
Populations from two case-control studies of DTC were included in this pooled analysis.
The University of Texas MD Anderson Cancer (UTMDACC) study
This population consisted of 303 patients with DTC and 511 cancer-free individuals. The DTC patients were identified between November 1999 and October 2008 at UTMDACC, and a final diagnosis of DTC was confirmed by histopathology; the controls were from the control group accrued at the same institution between November 1996 and March 2005 for a molecular epidemiological study of squamous cell carcinoma of the head and neck. The recruitment criteria and procedure have been described in detail in previous publications (23, 24). In brief, identical inclusion criteria were applied to both cases and controls, including U.S. resident 18 yr of age and older, no blood transfusion history in the past 6 months, not on immunosuppressant medications, and no prior malignancy (except nonmelanoma skin cancer). A self-administrated questionnaire was used to collect demographic information including self-defined race/ethnicity.
State University of Campinas (SUC) study
This population consisted of 289 patients with DTC and 374 cancer-free individuals. The DTC patients were consecutively recruited from a teaching hospital, the Faculty of Medical Sciences at the State University of Campinas, for thyroid nodule evaluation between July 1999 and January 2010. The final diagnosis of DTC was confirmed by experienced pathologists at the same institution. The controls were recruited from the general population in the region of the teaching hospital between April 1998 and January 2010. These individuals were matched to the patients on the basis of genotypes, age, lifetime occupational history, smoking history, general health conditions, and previous diseases, as described in detail in previous publications (25, 26). Individuals with a history of past thyroid disease, radiation exposure, specific environmental or occupational exposure risks, or previous malignancy were excluded. An in-person questionnaire was used to collect demographic information, and the race/ethnicity was determined by a trained interviewer upon direct interview of case and control subjects.
Both studies were approved by their respective institutional review boards, and informed consent was obtained from each participant. Overall, 592 patients with DTC and 885 cancer-free individuals volunteered to participate and provided blood samples for genotyping analysis.
DNA extraction and genotyping analysis
We included in the analysis the missense SNP rs1800054, rs4986761, rs1800057, and rs1801516 and two functional promoter SNP that have been associated with breast cancer risk, rs228589 and rs189037 (27, 28). DNA samples from UTMDACC patients were extracted from blood samples using a commercial DNA extraction kit (QIAamp DNA blood mini kit; QIAGEN Inc., Hilden, Germany), according to the manufacturer's instructions. A PCR-restriction fragment length polymorphism (PCR-RFLP) assay was used for genotyping these ATM SNP (primer sequences available upon request). Genotyping was performed by laboratory personnel blinded to case-control status. Repeated analysis was performed in a randomly selected subset of 10% of the samples, and greater than 99% concordance with the initial results was observed.
DNA samples from SUC patients were obtained from peripheral blood leukocytes by phenol standard procedures. The polymorphisms were genotyped using the TaqMan system (Applied Biosystems, Foster City, CA). Fluorescence signals were detected using 7500 system sequence detection software (Applied Biosystems). TaqMan PCR and genotyping analyses were performed on an Applied Biosystems 9600 Emulation System. The reaction mixtures were amplified in 2 μl of genomic DNA (10 ng/ml), 2.5 μl of 2X TaqMan Universal Master Mix, 0.25 ml of 40X primer/probe mix, and 0.25 μl of double distilled H2O in a volume of 5 μl. PCR cycling conditions were as follows: one cycle at 60 C for 1 min as the initial step; one cycle at 95 C for 20 min; 40 cycles at 92 C for 3 min and at 60 C for 30 sec; and one cycle at 60 C for 1 min as the annealing step. The results were analyzed on a 9600 Emulation System using the allelic discrimination assay program. In a randomly selected subset of about 5% of the SUC samples, the genotyping results were confirmed using PCR-RFLP-based methods. Greater than 95% concordance with the initial results was observed.
Statistical analysis
Demographic characteristics were compared between cases and controls using the χ2 test. For each SNP, deviation from Hardy-Weinberg equilibrium was assessed by χ2 analysis in the control group. Allele frequencies and genotype distributions were compared between the cases and controls in the pooled population and also within the two individual populations. Akaike information criterion was applied to select the best model of inheritance, and unconditional logistic regression was used to estimate crude and age-, sex-, race/ethnicity-, and study center-adjusted odds ratios (OR) and 95% confidence intervals (CI) for genotype-specific risk, both in the pooled populations and stratified by study center, sex, and race/ethnicity.
To assess the effect of ATM SNP in combination, allele dosage for each SNP, defined as the number of specific risk alleles, ranging from 0 to 2, was determined, and the total number of risk alleles was fitted into the regression model as a continuous variable to estimate the dosage effect. Pairwise linkage disequilibrium between SNP was examined using the Lewontin standardized coefficient D′, calculated by Haploview software. A haplotype analysis was performed using the expectation maximization algorithm in the SNPStats program (http://bioinfo.iconcologia.net/SNPStats) (29), with the most frequent haplotype as the reference group.
All statistical tests were two-sided, and a P value less than 0.05 was considered to be statistically significant. All analyses were performed using SAS software, version 9.2 (SAS Institute Inc., Cary, NC), unless otherwise specified.
Results
Table 1 provides the demographic characteristics and clinical features of the study participants. Significant differences in sex, age, and race/ethnicity were observed between patients with DTC and cancer-free individuals in the pooled population and/or in individual populations. The majority of cases and controls were non-Hispanic whites (>70%). Patients with DTC recruited at UTMDACC and SUC were similar in terms of age at diagnosis of cancer and frequency of multifocal primary tumors, but SUC patients with DTC were more likely to be in the early stage of disease in terms of T and N status.
Table 1.
Demographic and clinical characteristics of study participants, pooled and by study center
| Pooled |
SUC |
UTMDACC |
||||
|---|---|---|---|---|---|---|
| Controls (n = 885) | Cases (n = 592) | Controls (n = 374) | Cases (n = 289) | Controls (n = 511) | Cases (n = 303) | |
| Sex | ||||||
| Male | 379 (42.8) | 146 (24.7) | 134 (35.8) | 44 (15.2) | 245 (48.0) | 102 (33.7) |
| Female | 506 (57.2) | 446 (75.3)a | 240 (64.2) | 245 (84.8)a | 266 (52.1) | 201 (66.3)a |
| Age (yr) | ||||||
| <45 | 385 (43.5) | 313 (52.9) | 208 (55.6) | 156 (54.0) | 177 (34.6) | 157 (51.8) |
| ≥45 | 500 (56.5) | 279 (47.1)a | 166 (44.4) | 133 (46.0) | 334 (65.4) | 146 (48.2)a |
| Race/ethnicity | ||||||
| Non-Hispanic white | 686 (80.7) | 459 (77.7) | 286 (84.4) | 245 (84.8) | 400 (78.3) | 214 (70.9) |
| Other | 164 (19.3) | 132 (22.3) | 53 (15.6) | 44 (15.2) | 111 (21.7) | 88 (29.1)a |
| DTC subtype | ||||||
| Papillary | 522 (88.2) | 249 (86.2) | 273 (90.1) | |||
| Follicular | 60 (10.1) | 38 (13.2) | 22 (7.3) | |||
| Hürthle cell | 10 (1.7) | 2 (0.7) | 8 (2.6) | |||
| T status | ||||||
| T1 or T2 | 346 (64.9) | 193 (71.0) | 153 (58.6) | |||
| T3 or T4 | 187 (35.1) | 79 (29.0) | 108 (41.4) | |||
| N status | ||||||
| N0 | 353 (65.6) | 204 (75.3) | 149 (55.8) | |||
| N1 | 185 (34.4) | 67 (24.7) | 118 (44.2) | |||
| Multifocal primary tumor | ||||||
| No | 345 (63.0) | 164 (60.5) | 181 (65.3) | |||
| Yes | 203 (37.0) | 107 (39.5) | 96 (34.7) | |||
Data are expressed as number (percentage). SUC, State University of Campinas; UTMDACC, University of Texas MD Anderson Cancer Center.
P < 0.05, χ2 test.
No deviations from Hardy-Weinberg equilibrium were observed for genotype distribution of the six ATM SNP among controls. Inheritance modeling suggested a dominant model for all six SNP in the pooled population and each individual population, and the corresponding genotype-specific DTC risk estimates are shown in Table 2. Two SNP, rs189037 and rs1800057, were significantly associated with DTC risk. In the pooled population, the AG/GG genotypes of rs189037 were significantly protective against DTC, compared with the AA genotype (adjusted OR = 0.8; 95% CI, 0.6–1.0; P = 0.04); however, in non-Hispanic whites, the association was not significant (adjusted OR = 0.9; 95% CI, 0.7–1.1; P = 0.27). Also in the pooled population, the CG/GG genotypes of rs1800057 were associated with a significantly increased risk of DTC, compared with the CC genotype (adjusted OR = 1.9; 95% CI, 1.1–3.1; P = 0.02). This association remained significant in non-Hispanic whites (adjusted OR = 1.8; 95% CI, 1.0–3.1; P = 0.04) and was not statistically significant in the SUC case-control series (P = 0.05). No significant association was found between the remaining SNP of interest and DTC risk. When the analysis was restricted to patients with PTC, similarly, the CG/GG genotypes of rs1800057 were significantly associated with increased risk of PTC (adjusted OR = 2.0; 95% CI, 1.2–3.3; P = 0.01), and the AG/GG genotypes of rs189037 were inversely associated with PTC risk (adjusted OR = 0.8; 95% CI, 0.6–1.0; P = 0.05) (data not shown). In subgroup analysis stratified by age and sex, no significant interactions were observed (data not shown).
Table 2.
Genotype distributions and OR estimates for DTC, pooled, non-Hispanic whites only and by study center
| ATM SNP ID | Genotype | Inheritance modela | Pooled population |
Non-Hispanic Whites |
SUC population |
UTMDACC population |
||||
|---|---|---|---|---|---|---|---|---|---|---|
| No. of cases/controls | Adjusted OR (95% CI)b | No. of cases/controls | Adjusted OR (95% CI)b | No. of cases/controls | Adjusted OR (95% CI)b | No. of cases/controls | Adjusted OR (95% CI)b | |||
| rs228589 | TT | Dominant | 227/294 | 1.0 | 178/241 | 1.0 | 116/126 | 1.0 | 111/168 | 1.0 |
| TA/AA | 363/587 | 0.8 (0.6–1.0) | 280/441 | 0.9 (0.7–1.1) | 173/244 | 0.8 (0.6–1.1) | 190/343 | 0.8 (0.6–1.1) | ||
| rs189037 | AA | Dominant | 215/277 | 1.0 | 170/233 | 1.0 | 105/112 | 1.0 | 110/165 | 1.0 |
| AG/GG | 375/606 | 0.8 (0.6–1.0) | 288/452 | 0.9 (0.7–1.1) | 184/260 | 0.8 (0.5–1.1) | 191/346 | 0.8 (0.6–1.1) | ||
| rs1800054 | CC | 576/868 | 1.0 | 445/671 | 1.0 | 280/369 | 1.0 | 296/499 | 1.0 | |
| CG | 16/16 | 1.6 (0.8–3.4) | 14/14 | 1.5 (0.7–3.4) | 9/4 | 3.6 (0.9–14.1) | 7/12 | 1.0 (0.4–2.7) | ||
| rs4986761 | TT | 574/860 | 1.0 | 441/668 | 1.0 | 277/363 | 1.0 | 297/497 | 1.0 | |
| TC | 18/22 | 1.2 (0.6–2.4) | 18/15 | 1.8 (0.9–3.7) | 12/8 | 1.9 (0.7–5.1) | 6/14 | 0.8 (0.3–2.1) | ||
| rs1800057 | CC | Dominant | 550/844 | 1.0 | 425/655 | 1.0 | 271/355 | 1.0 | 279/489 | 1.0 |
| CG/GG | 36/32 | 1.9 (1.1–3.1) | 29/25 | 1.8 (1.0–3.1) | 17/11 | 2.3 (1.0–5.6) | 19/21 | 1.6 (0.9–3.1) | ||
| rs1801516 | GG | Dominant | 483/697 | 1.0 | 366/519 | 1.0 | 244/305 | 1.0 | 239/392 | 1.0 |
| GA/AA | 109/188 | 0.9 (0.7–1.1) | 93/167 | 0.8 (0.6–1.1) | 45/69 | 0.8 (0.5–1.2) | 64/119 | 0.9 (0.6–1.3) | ||
SUC, State University of Campinas; UTMDACC, University of Texas MD Anderson Cancer Center. P < 0.05 is shown in bold.
The best model of inheritance for each SNP was determined using Akaike information criterion, and only the genotype distribution and OR under the best model are provided.
OR adjusted for age, sex, race/ethnicity, and study center.
Only rs228589 and rs189037, both of which are in the promoter region, were in strong linkage disequilibrium (D′ > 0.90). Haplotype analyses identified a protective haplotype (A-G-C-T-C-A) that was marginally associated with decreased risk of DTC (P = 0.05), and this protective effect was likely confined to non-Hispanic whites (P = 0.03) (Table 3). The global score test indicated that haplotype distributions were significantly different between non-Hispanic white cases and controls (P = 0.04).
Table 3.
Associations between ATM haplotypes and risk of DTC
| Haplotype | SNPa |
Frequency |
P | Adjusted OR (95% CI)b | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | Controls (n = 855) | Cases (n = 592) | ||||
| Overall | |||||||||||
| 1 | T | A | C | T | C | G | 0.43 | 0.46 | 1.00 | ||
| 2 | A | G | C | T | C | G | 0.39 | 0.37 | 0.19 | 0.9 (0.8–1.1) | |
| 3 | T | A | C | T | C | A | 0.09 | 0.09 | 0.39 | 0.9 (0.7–1.2) | |
| 4 | T | A | C | T | G | G | 0.02 | 0.03 | 0.15 | 1.5 (0.9–2.4) | |
| 5 | T | G | C | T | C | G | 0.02 | 0.02 | 0.14 | 0.6 (0.4–1.1) | |
| 6 | T | A | C | C | C | G | 0.01 | 0.01 | 0.63 | 1.2 (0.6–2.4) | |
| 7 | A | G | C | T | C | A | 0.02 | 0.01 | 0.05 | 0.4 (0.1–1.0) | |
| Othersc | 0.07 | 0.09 | 0.77 | 0.9 (0.5–1.7) | |||||||
| Non-Hispanic whites | |||||||||||
| 1 | T | A | C | T | C | G | 0.44 | 0.46 | 1.00 | ||
| 2 | A | G | C | T | C | G | 0.37 | 0.36 | 0.55 | 0.9 (0.8–1.1) | |
| 3 | T | A | C | T | C | A | 0.10 | 0.10 | 0.72 | 0.9 (0.7–1.3) | |
| 4 | T | A | C | T | G | G | 0.03 | 0.03 | 0.10 | 1.6 (0.9–2.7) | |
| 5 | T | G | C | T | C | G | 0.01 | 0.01 | 0.22 | 0.6 (0.3–1.3) | |
| 6 | T | A | C | C | C | G | 0.01 | 0.00 | 0.19 | 1.6 (0.8–3.4) | |
| 7 | A | G | C | T | C | A | 0.01 | 0.02 | 0.03 | 0.2 (0.0–0.8) | |
| Othersc | 0.091 | 0.076 | 0.78 | 1.1 (0.5–2.3) | |||||||
P < 0.05 is shown in bold.
SNP are as follows: 1, rs228589; 2, rs189037; 3, rs1800054; 4, rs4986761; 5, rs1800057; and 6, rs1801516.
OR adjusted for age and sex.
Rare haplotypes with frequencies <0.01.
A significant dose-response relationship was observed between the total number of risk alleles of ATM and DTC risk (P < 0.01 for crude and P = 0.01 for adjusted model). We further trichotomized the subjects into groups with three to five, six to seven, and eight to 10 risk alleles. As shown in Table 4, compared with subjects carrying three to five risk alleles, those carrying six to seven and eight to 10 risk alleles had 30 and 50% increased risk of DTC, respectively.
Table 4.
Association between the combined risk alleles of the ATM gene and risk of DTC and PTC
| No. of risk alleles | Controls (n = 885) | DTC cases (n = 592) |
PTC cases (n = 522) |
|||||
|---|---|---|---|---|---|---|---|---|
| n | % | n | % | Adjusted OR (95% CI)a | n | % | Adjusted OR (95% CI)a | |
| 3–5 | 261 | 30.0 | 140 | 23.9 | 1.0 | 123 | 23.8 | 1.0 |
| 6–7 | 429 | 49.4 | 299 | 51.0 | 1.3 (1.0–1.7) | 261 | 50.5 | 1.3 (1.0–1.7) |
| 8–10 | 179 | 20.6 | 147 | 25.0 | 1.5 (1.1–2.1) | 133 | 25.7 | 1.6 (1.2–2.2) |
| Ptrend < 0.01 | Ptrend < 0.01 | |||||||
P < 0.05 is shown in bold.
OR adjusted for age, sex, race/ethnicity, and study center.
Discussion
In this study, ATM SNP rs189037 and rs1800057 showed association with risk of DTC; under a dominant inheritance model, the rare allele of rs189037 was associated with decreased DTC risk, and the rare allele of rs1800057 was associated with increased DTC risk. Our data also suggest a combination effect of ATM SNP on DTC risk. We estimated that individuals with eight or more risk alleles of ATM have 1.5 times the risk of developing DTC of individuals with fewer than six risk alleles. Together, these results suggest that SNP in ATM modulate individual susceptibility to DTC and support the argument that DTC susceptibility is likely to be determined by multiple variations in low-penetrance genes that affect a large segment of the general population (30).
ATM is a good candidate for a radiosensitivity gene because there is substantial evidence associating ATM mutations with enhanced radiosensitivity (31). The rs1800057 missense variant, resulting in the amino acid change P1054R, was previously found to exhibit increased radiosensitivity after in vitro exposure to ionizing radiation (32). In our present study, the rs1800057 variant was significantly associated with increased risk of DTC, supporting our hypothesis of increased risk of DTC in individuals who are genetically predisposed to radiosensitivity. Our finding of a dose-response relationship between ATM risk alleles and DTC risk provides additional support for our hypothesis, although further evaluation of functional significance and validation in larger studies are required.
The rs1800057 variant has been implicated in several types of cancer. In two independent case-control series of Caucasian individuals, carriers of the rs1800057 G allele (1054R) had a significant approximately 2-fold increase in the risk of prostate cancer (9, 10). This is in accordance with the direction and magnitude of association in our study, in which OR estimates were 1.9 (95% CI, 1.1–3.1) in the pooled population and 1.8 (95% CI, 1.0–3.1) in the non-Hispanic white subgroup. Rudd et al. reported a significant 1.7-fold increase in the risk of chronic lymphocytic leukemia associated with the 1054R variant in a study that included 992 cases and 2707 controls of Caucasian descent (33). Similarly, in a case-control study of colorectal cancer sharing the same controls, the 1054R variant was associated with a 1.4-fold increase in the risk of colorectal cancer, and moreover, a further kin-cohort analysis on the 14,704 first-degree relatives of the 2575 colorectal cancer patients found a stronger association (OR = 1.8) (34). In general, these findings by our group and other groups suggest that ATM rs1800057 confers a moderate increase in the risk of cancer. However, a recent large-scale study found no association between rs1800057 and breast cancer in a Caucasian population (35). ATM not only repairs DSB induced by radiation exposure, but also activates p53 and subsequently promotes p53-dependent responses including senescence. Each mechanism acts independently, and the latter has been implicated as a general tumor suppression mechanism of ATM in carcinogenesis in various tumors including colon, lung, bladder, and breast cancer (36–38). This could be a likely explanation of the observation that cancer predisposition linked to ATM is not site specific.
For rs189037, a weak inverse association between the G allele and risk of DTC was observed, but not in the non-Hispanic white subgroup, showing a heterogeneity of the effect across race/ethnicity. However, because of mixed ethnic background and limited sample size in the other race/ethnicity groups, it is hard to estimate the DTC risk associated with rs189037 in race/ethnicity groups other than non-Hispanic whites. The rs189037 A allele has been significantly associated with increased risk of female breast cancer and oral cancer (27, 39) and with poorer prognosis of lung cancer (40). It is worth noting that these prior studies having significant results were all conducted in Asian populations. The A allele frequency in our controls (0.56) was higher than that in those Asian populations (0.34–0.42) (27, 39, 40). This supports race/ethnicity-related genetic differences and may partly explain the discrepancy in associations. Further research is anticipated to address whether inherent differences of rs189037 across race/ethnicity have an impact on DTC risk.
In this study, two SNP and one haplotype of ATM were associated with DTC risk at a nominal significance level of 0.05. For these results, of major concern is the false-positive probability due to multiple comparisons. To assess whether the significant association was true or by chance, we applied the false-positive reporting probability introduced by Wacholder et al. (41). In brief, the likelihood of a false-positive result is evaluated on the observed OR and 95% CI and a prior probability. The result could serve as a reference in interpreting the statistically significant findings. In this study, a prior probability of 0.01–0.1 is presumed given the evidence of an impact of studied ATM polymorphisms on gene function and cancer risk. Given a prior probability of 0.1 or 0.01, the probabilities that the significant findings are false-positive are: 1) 25 and 79%, respectively, for association between rs1800057 and DTC risk; 2) 42 and 89%, respectively, for association between rs189037and DTC risk; and 3) 33 and 84%, respectively, for association between haplotype A-G-C-T-C-A and DTC risk in non-Hispanic whites.
The results from this study must be interpreted with caution. First, population stratification could lead to spurious association if disease prevalence and allele frequencies differ by subgroups, although this probability is unlikely in this study because: 1) the majority of cases and controls were from the same race/ethnicity group, and we also evaluated the association in subgroup analysis in non-Hispanic whites only; and 2) the minor allele frequencies were similar between the independent UTMDACC and SUC case-control series, and the genotype-specific risks from these two populations were similar as well. Second, the possibility of selection bias, which is common in a hospital-based case-control study, should be considered. For instance, the UTMDACC study had a higher proportion of patients with DTC at a more advanced stage compared with the SUC study; however, DTC risk estimates associated with ATM SNP were similar for both low and high T status as well as low and high N status (data not shown). Third, the association results are not generalizable to race/ethnicity populations other than non-Hispanic whites. Fourth, the sample size limits the statistical power, especially for SNP with low minor allele frequency and in stratification and haplotype analysis, although to the best of our knowledge, this is the largest case-control study to explore the impact of functional ATM SNP on DTC susceptibility. Fifth, the SNP selected in this study are not a complete list of ATM SNP. Because the exact mechanism by which they act on DTC development is largely unknown, the possibility should not be excluded that they may be in linkage disequilibrium with one or more causal loci responsible for DTC susceptibility. Sixth, different genotyping methods may have influenced the results, although this possibility is less likely as we employed internal validation and cross-validation (Taqman vs. PCR-RFLP-based assay) in genotyping analysis.
In summary, the data support the concept that individual susceptibility to DTC may be attributable to polymorphisms of ATM, a critical gene in response to radiation-induced DSB. These findings argue for comprehensive genotyping of ATM in larger population-based studies and functional testing of ATM variations.
Acknowledgments
The authors thank Margaret Lung, Kathryn Patterson, and Jenny Vo for their help with subject recruitment; Chong Zhao, Yingdong Li, Raquel Barbieri Bueno, and Mariana Bonjiorno Martins for DNA extraction and genotyping analysis; and Stephanie P. Deming for manuscript editing.
This work was supported by an American Thyroid Association research grant [Principal Investigator (PI), E. M. Sturgis], National Institutes of Health Grant U01 D019765-01 (PI, Adel K. El-Naggar), National Institutes of Environmental Health Sciences Grant R01 ES-11740 (PI, Q. Wei), and Cancer Center Support Grant CA016672 (PI, John Mendelsohn). L.X. is currently supported by Halliburton Employees Fellow in Cancer Prevention funds at The University of Texas MD Anderson Cancer Center.
L.S.W. is a researcher of the National Council for Scientific and Technological Development (CNPq-Brazil).
Disclosure Summary: The authors declare no conflicts of interest.
Footnotes
- ATM
- Ataxia telangiectasia mutated (gene)
- CI
- confidence interval
- DSB
- double-strand break
- DTC
- differentiated thyroid carcinoma
- OR
- odds ratio
- PCR-RFLP
- PCR-restriction fragment length polymorphism
- PTC
- papillary thyroid carcinoma
- SNP
- single nucleotide polymorphism.
References
- 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. 2011. Global cancer statistics. CA Cancer J Clin 61:69–90 [DOI] [PubMed] [Google Scholar]
- 2. Davies L, Welch HG. 2006. Increasing incidence of thyroid cancer in the United States, 1973–2002. JAMA 295:2164–2167 [DOI] [PubMed] [Google Scholar]
- 3. Howlader N, Noone AM, Krapcho M, Neyman N, Aminou R, Waldron W, Altekruse SF, Kosary CL, Ruhl J, Tatalovich Z, Cho H, Mariotto A, Eisner MP, Lewis DR, Chen HS, Feuer EJ, Cronin KA, Edwards BK. 2011. SEER Cancer Statistics Review, 1975–2008. Bethesda, MD: National Cancer Institute [Google Scholar]
- 4. Ron E, Lubin JH, Shore RE, Mabuchi K, Modan B, Pottern LM, Schneider AB, Tucker MA, Boice JD., Jr 1995. Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies. Radiat Res 141:259–277 [PubMed] [Google Scholar]
- 5. Prasad KN, Cole WC, Hasse GM. 2004. Health risks of low dose ionizing radiation in humans: a review. Exp Biol Med (Maywood) 229:378–382 [DOI] [PubMed] [Google Scholar]
- 6. Boice JD, Jr, Blettner M, Auvinen A. 2000. Epidemiologic studies of pilots and aircrew. Health Phys 79:576–584 [DOI] [PubMed] [Google Scholar]
- 7. Jackson SP. 2002. Sensing and repairing DNA double-strand breaks. Carcinogenesis 23:687–696 [DOI] [PubMed] [Google Scholar]
- 8. Derheimer FA, Kastan MB. 2010. Multiple roles of ATM in monitoring and maintaining DNA integrity. FEBS Lett 584:3675–3681 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Meyer A, Wilhelm B, Dörk T, Bremer M, Baumann R, Karstens JH, Machtens S. 2007. ATM missense variant P1054R predisposes to prostate cancer. Radiother Oncol 83:283–288 [DOI] [PubMed] [Google Scholar]
- 10. Angèle S, Falconer A, Edwards SM, Dörk T, Bremer M, Moullan N, Chapot B, Muir K, Houlston R, Norman AR, Bullock S, Hope Q, Meitz J, Dearnaley D, Dowe A, Southgate C, Ardern-Jones A, Easton DF, Eeles RA, Hall J. 2004. ATM polymorphisms as risk factors for prostate cancer development. Br J Cancer 91:783–787 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Zheng YL, Kosti O, Loffredo CA, Bowman E, Mechanic L, Perlmutter D, Jones R, Shields PG, Harris CC. 2010. Elevated lung cancer risk is associated with deficiencies in cell cycle checkpoints: genotype and phenotype analyses from a case-control study. Int J Cancer 126:2199–2210 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Buchholz TA, Weil MM, Ashorn CL, Strom EA, Sigurdson A, Bondy M, Chakraborty R, Cox JD, McNeese MD, Story MD. 2004. A Ser49Cys variant in the ataxia telangiectasia, mutated, gene that is more common in patients with breast carcinoma compared with population controls. Cancer 100:1345–1351 [DOI] [PubMed] [Google Scholar]
- 13. Cox A, Dunning AM, Garcia-Closas M, Balasubramanian S, Reed MW, Pooley KA, Scollen S, Baynes C, Ponder BA, Chanock S, Lissowska J, Brinton L, Peplonska B, Southey MC, Hopper JL, McCredie MR, Giles GG, Fletcher O, Johnson N, dos Santos Silva I, Gibson L, Bojesen SE, Nordestgaard BG, Axelsson CK, Torres D, Hamann U, Justenhoven C, Brauch H, Chang-Claude J, Kropp S, Risch A, Wang-Gohrke S, Schürmann P, Bogdanova N, Dörk T, Fagerholm R, Aaltonen K, Blomqvist C, Nevanlinna H, Seal S, Renwick A, Stratton MR, Rahman N, Sangrajrang S, Hughes D, Odefrey F, Brennan P, Spurdle AB, Chenevix-Trench G, Beesley J, Mannermaa A, Hartikainen J, Kataja V, Kosma VM, Couch FJ, Olson JE, Goode EL, Broeks A, Schmidt MK, Hogervorst FB, Van't Veer LJ, Kang D, Yoo KY, Noh DY, Ahn SH, Wedrén S, Hall P, Low YL, Liu J, Milne RL, Ribas G, Gonzalez-Neira A, Benitez J, Sigurdson AJ, Stredrick DL, Alexander BH, Struewing JP, Pharoah PD, Easton DF. 2007. A common coding variant in CASP8 is associated with breast cancer risk. Nat Genet 39:352–358 [DOI] [PubMed] [Google Scholar]
- 14. Stredrick DL, Garcia-Closas M, Pineda MA, Bhatti P, Alexander BH, Doody MM, Lissowska J, Peplonska B, Brinton LA, Chanock SJ, Struewing JP, Sigurdson AJ. 2006. The ATM missense mutation p.Ser49Cys (c. 146C>G) and the risk of breast cancer. Hum Mutat 27:538–544 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Akulevich NM, Saenko VA, Rogounovitch TI, Drozd VM, Lushnikov EF, Ivanov VK, Mitsutake N, Kominami R, Yamashita S. 2009. Polymorphisms of DNA damage response genes in radiation-related and sporadic papillary thyroid carcinoma. Endocr Relat Cancer 16:491–503 [DOI] [PubMed] [Google Scholar]
- 16. Wang Y, Cheng J, Li D, Duan H, Yang H, Bin P, Dai Y, Huang C, Liang X, Leng S, Chen W, Zheng Y. 2011. Modulation of DNA repair capacity by ataxia telangiectasia mutated gene polymorphisms among polycyclic aromatic hydrocarbons-exposed workers. Toxicol Sci 124:99–108 [DOI] [PubMed] [Google Scholar]
- 17. Chen T, Dong B, Lu Z, Tian B, Zhang J, Zhou J, Wu H, Zhang Y, Wu J, Lin P, Zhang J, Xu H, Mo X. 2010. A functional single nucleotide polymorphism in promoter of ATM is associated with longevity. Mech Ageing Dev 131:636–640 [DOI] [PubMed] [Google Scholar]
- 18. Angèle S, Romestaing P, Moullan N, Vuillaume M, Chapot B, Friesen M, Jongmans W, Cox DG, Pisani P, Gérard JP, Hall J. 2003. ATM haplotypes and cellular response to DNA damage: association with breast cancer risk and clinical radiosensitivity. Cancer Res 63:8717–8725 [PubMed] [Google Scholar]
- 19. Cesaretti JA, Stock RG, Lehrer S, Atencio DA, Bernstein JL, Stone NN, Wallenstein S, Green S, Loeb K, Kollmeier M, Smith M, Rosenstein BS. 2005. ATM sequence variants are predictive of adverse radiotherapy response among patients treated for prostate cancer. Int J Radiat Oncol Biol Phys 61:196–202 [DOI] [PubMed] [Google Scholar]
- 20. Iannuzzi CM, Atencio DP, Green S, Stock RG, Rosenstein BS. 2002. ATM mutations in female breast cancer patients predict for an increase in radiation-induced late effects. Int J Radiat Oncol Biol Phys 52:606–613 [DOI] [PubMed] [Google Scholar]
- 21. Pugh TJ, Keyes M, Barclay L, Delaney A, Krzywinski M, Thomas D, Novik K, Yang C, Agranovich A, McKenzie M, Morris WJ, Olive PL, Marra MA, Moore RA. 2009. Sequence variant discovery in DNA repair genes from radiosensitive and radiotolerant prostate brachytherapy patients. Clin Cancer Res 15:5008–5016 [DOI] [PubMed] [Google Scholar]
- 22. Sommer SS, Buzin CH, Jung M, Zheng J, Liu Q, Jeong SJ, Moulds J, Nguyen VQ, Feng J, Bennett WP, Dritschilo A. 2002. Elevated frequency of ATM gene missense mutations in breast cancer relative to ethnically matched controls. Cancer Genet Cytogenet 134:25–32 [DOI] [PubMed] [Google Scholar]
- 23. Xu L, Doan PC, Wei Q, Liu Y, Li G, Sturgis EM. 2012. Association of BRCA1 functional single nucleotide polymorphisms with risk of differentiated thyroid carcinoma. Thyroid 22:35–43 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Ho T, Li G, Lu J, Zhao C, Wei Q, Sturgis EM. 2009. Association of XRCC1 polymorphisms and risk of differentiated thyroid carcinoma: a case-control analysis. Thyroid 19:129–135 [DOI] [PubMed] [Google Scholar]
- 25. Guilhen AC, Bufalo NE, Morari EC, Leite JL, Assumpcao LV, Tincani AJ, Ward LS. 2009. Role of the N-acetyltransferase 2 detoxification system in thyroid cancer susceptibility. Clin Cancer Res 15:406–412 [DOI] [PubMed] [Google Scholar]
- 26. Barbieri RB, Bufalo NE, Secolin R, Silva AC, Assumpção LV, Maciel RM, Cerutti JM, Ward LS. 2012. Evidences that polymorphisms in detoxification genes modulate the susceptibility for sporadic medullary thyroid carcinoma. Eur J Endocrinol 166:241–245 [DOI] [PubMed] [Google Scholar]
- 27. Wang HC, Chang WS, Tsai RY, Tsai CW, Liu LC, Su CH, Cheng HN, Tsou YA, Sun SS, Lin CC, Bau DT. 2010. Association between ataxia telangiectasia mutated gene polymorphisms and breast cancer in Taiwanese females. Anticancer Res 30:5217–5221 [PubMed] [Google Scholar]
- 28. Koren M, Kimmel G, Ben-Asher E, Gal I, Papa MZ, Beckmann JS, Lancet D, Shamir R, Friedman E. 2006. ATM haplotypes and breast cancer risk in Jewish high-risk women. Br J Cancer 94:1537–1543 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Solé X, Guinó E, Valls J, Iniesta R, Moreno V. 2006. SNPStats: a web tool for the analysis of association studies. Bioinformatics 22:1928–1929 [DOI] [PubMed] [Google Scholar]
- 30. Sturgis EM, Li G. 2009. Molecular epidemiology of papillary thyroid cancer: in search of common genetic associations. Thyroid 19:1031–1034 [DOI] [PubMed] [Google Scholar]
- 31. Stankovic T, Kidd AM, Sutcliffe A, McGuire GM, Robinson P, Weber P, Bedenham T, Bradwell AR, Easton DF, Lennox GG, Haites N, Byrd PJ, Taylor AM. 1998. ATM mutations and phenotypes in ataxia-telangiectasia families in the British Isles: expression of mutant ATM and the risk of leukemia, lymphoma, and breast cancer. Am J Hum Genet 62:334–345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Gutiérrez-Enríquez S, Fernet M, Dörk T, Bremer M, Lauge A, Stoppa-Lyonnet D, Moullan N, Angèle S, Hall J. 2004. Functional consequences of ATM sequence variants for chromosomal radiosensitivity. Genes Chromosomes Cancer 40:109–119 [DOI] [PubMed] [Google Scholar]
- 33. Rudd MF, Sellick GS, Webb EL, Catovsky D, Houlston RS. 2006. Variants in the ATM-BRCA2-CHEK2 axis predispose to chronic lymphocytic leukemia. Blood 108:638–644 [DOI] [PubMed] [Google Scholar]
- 34. Webb EL, Rudd MF, Sellick GS, El Galta R, Bethke L, Wood W, Fletcher O, Penegar S, Withey L, Qureshi M, Johnson N, Tomlinson I, Gray R, Peto J, Houlston RS. 2006. Search for low penetrance alleles for colorectal cancer through a scan of 1467 non-synonymous SNPs in 2575 cases and 2707 controls with validation by kin-cohort analysis of 14 704 first-degree relatives. Hum Mol Genet 15:3263–3271 [DOI] [PubMed] [Google Scholar]
- 35. Fletcher O, Johnson N, dos Santos Silva I, Orr N, Ashworth A, Nevanlinna H, Heikkinen T, Aittomäki K, Blomqvist C, Burwinkel B, Bartram CR, Meindl A, Schmutzler RK, Cox A, Brock I, Elliott G, Reed MW, Southey MC, Smith L, Spurdle AB, Hopper JL, Couch FJ, Olson JE, Wang X, Fredericksen Z, Schürmann P, Waltes R, Bremer M, Dörk T, Devilee P, van Asperen CJ, Tollenaar RA, Seynaeve C, Hall P, Czene K, Humphreys K, Liu J, Ahmed S, Dunning AM, Maranian M, Pharoah PD, Chenevix-Trench G, Beesley J, Bogdanova NV, Antonenkova NN, Zalutsky IV, Anton-Culver H, Ziogas A, Brauch H, Ko YD, Hamann U, Fasching PA, Strick R, Ekici AB, Beckmann MW, Giles GG, Severi G, Baglietto L, English DR, Milne RL, Benítez J, Arias JI, Pita G, Nordestgaard BG, Bojesen SE, Flyger H, Kang D, Yoo KY, Noh DY, Mannermaa A, Kataja V, Kosma VM, García-Closas M, Chanock S, Lissowska J, Brinton LA, Chang-Claude J, Wang-Gohrke S, Broeks A, Schmidt MK, van Leeuwen FE, Van't Veer LJ, Margolin S, Lindblom A, Humphreys MK, Morrison J, Platte R, Easton DF, Peto J. 2010. Missense variants in ATM in 26,101 breast cancer cases and 29,842 controls. Cancer Epidemiol Biomarkers Prev 19:2143–2151 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Bartkova J, Horejsí Z, Koed K, Krämer A, Tort F, Zieger K, Guldberg P, Sehested M, Nesland JM, Lukas C, Ørntoft T, Lukas J, Bartek J. 2005. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 434:864–870 [DOI] [PubMed] [Google Scholar]
- 37. Gorgoulis VG, Vassiliou LV, Karakaidos P, Zacharatos P, Kotsinas A, Liloglou T, Venere M, Ditullio RA, Jr, Kastrinakis NG, Levy B, Kletsas D, Yoneta A, Herlyn M, Kittas C, Halazonetis TD. 2005. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 434:907–913 [DOI] [PubMed] [Google Scholar]
- 38. Kwong LN, Weiss KR, Haigis KM, Dove WF. 2008. Atm is a negative regulator of intestinal neoplasia. Oncogene 27:1013–1018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Bau DT, Chang CH, Tsai MH, Chiu CF, Tsou YA, Wang RF, Tsai CW, Tsai RY. 2010. Association between DNA repair gene ATM polymorphisms and oral cancer susceptibility. Laryngoscope 120:2417–2422 [DOI] [PubMed] [Google Scholar]
- 40. Dong J, Hu Z, Shu Y, Pan S, Chen W, Wang Y, Hu L, Jiang Y, Dai J, Ma H, Jin G, Shen H. 7 July 2011. Potentially functional polymorphisms in DNA repair genes and non-small-cell lung cancer survival: a pathway-based analysis. Mol Carcinog doi: 10.1002/mc.20819 [DOI] [PubMed] [Google Scholar]
- 41. Wacholder S, Chanock S, Garcia-Closas M, El Ghormli L, Rothman N. 2004. Assessing the probability that a positive report is false: an approach for molecular epidemiology studies. J Natl Cancer Inst 96:434–442 [DOI] [PMC free article] [PubMed] [Google Scholar]