ATM rs189037 (G > A) polymorphism increased the risk of cancer: an updated meta-analysis

The occurrence of cancer is increasing because of the population aging, smoking, physical inactivity, et al [1]. It is a cellular abnormality, uncontrolled growth caused by numerous damages or mutations in the genetic material due to hereditary or environmental factors, which is immune to many signals that control cell growth and death [2]. The genetic factors takes more proportion on the causation of cancer than the lifestyle or environmental factors [3]. Many candidate genes or variations have been identified to contribute to the susceptibility of the cancer.

Ataxia-telangiectasia mutated (ATM) gene is located on the chromosome 11q22–23 with the full length 150 kb [4]. It comprises 66 exons and encodes a 12 kb transcript. The encoded protein belongs to the PI3/PI4-kinase family. As a Ser/Thr protein kinase, ATM protein plays an important role in DNA damage-induced signaling and initiation of cell cycle check-point signaling by phosphorylating [5]. After activated by DNA double strand breaks, it can be involved in recognizing broken or damaged DNA stands and assisting DNA repair by recruiting enzymes to recover the damaged strands [6]. Three deleterious missense variants of ATM gene were associated with an increased risk of cancer [7]. Moreover, ATM mutation and ATM protein loss included characteristics of old age, distal location of tumor, large tumor size, and histologic intestinal type in the human gastric cancer tissue [5]. The missense variants in ATM gene were also associated with the prostate cancer predisposition [8]. The loss of ATM function can give rise to ataxia telangiectasia, a pleiotropic disease with the whose hallmarks, such as neurodegeneration, cancer-proneness, premature aging, radio-sensitivity, et al [9]. It can control genome stability, modulate oxidative stress response, autophagy, and cancer stem cell survival as tumor suppressor gene [10].

The variation of ATM gene can affect the normal function of the protein and increase the risk of cancer. Rs189037 (G > A) is located at the 5’UTR of ATM gene and is one of the critical polymorphism that may be related to the occurrence of different cancers and tumor diffusing capacity [11‐15]. However, no consistent conclusion has been determined, and there remains discord between the findings in the literature, which may be attributable to a number of factors varying between studies including the types of cancer, the sample sizes, the genetic backgrounds of study subjects, and the potential presence of confounding bias [16].

When there is considerable variation in the results of studies on medical topics that have been studied extensively, meta-analysis can be used as a method to identify a common effect [17]. Such an analysis was conducted by Kang et al. (2014) to assess whether the ATM rs189037 polymorphism was associated with the risk of papillary thyroid carcinoma [18]. But only one case-control study was focused on rs189037. Bhowmik et al. analyzed the association of rs189037 with the risk of lung cancer and head and neck cancer in 2015 [19]. A total of 9 case-control studies were considered for this quantitative analysis. The third 2017 meta-analysis including ten case-control studies (4731 cases and 5142 controls) also reported the association between rs189037 and lung cancer susceptibility [12]. It seems superfluous to perform the meta-analysis of rs189037 and its association with cancer risk, whereas that the two latest meta-analyses only focused on the lung cancer and there are additional studies reporting its role in the other cancer types, such as breast cancer, papillary thyroid carcinoma, leukemia [14, 15, 20]. Therefore, we have performed a new meta-analysis of the ATM rs189037 polymorphism and the risk of different cancer types that includes more recent research.

We searched the database and identified 219 articles. According to the established inclusion criteria, a total of 15 publications were finally screened and included in our meta-analysis [13‐15, 20, 28‐38]. We collected 15 case-control studies, which contained 8660 patients with cancer (i.e., cases) and 9259 unaffected participants (i.e., controls). The individuals with the different genetic backgrounds were included (e.g., East Asian, Latino, and Caucasian). The main characteristics of the included studies were summarized in Table 1. Based on the results of the NOS scale, 12 studies were regarded as high quality and 3 studies were regarded as low quality. The genotype and allele frequencies of rs189037 SNP and HWE in controls were presented in Table 2. Of the 15 studies, no study deviated significantly from HWE.

Previously, three meta-analyses have reported the putative association between rs189037 and the occurrence of cancer [12, 18, 19]. Generally, our results were consistent with the previous studies. It seems that our meta-analysis is redundant, but there are some highlights compared with the previously published studies. Firstly, our analysis included the newly published studies since the previous meta-analyses were performed. A total of 15 studies were included, which could comprehensively represent rs189037 better compared with the previous meta-analyses. Additionally, the subgroup analyses were carried out by ethnicity, source of controls, and types of cancer to explore the potential origins of heterogeneity and to measure the study stability. Thus, to some degree, our meta-analysis could give a more accurate, comprehensive finding that there is an association between rs189037 SNP and lung cancer, breast cancer, and oral cancer, but not leukemia, thyroid carcinoma, glioma, and colorectal cancer.

However, the relatively small sample sizes of Latino and Caucasian populations limited our ability to isolate stable effects for these subgroups. Only one study reported the association of rs189037 with differentiated thyroid cancer in Latino including 592 cases and 885 controls [31]. For Caucasian, there is also just one study about the risk of papillary thyroid carcinoma including 83 cases and 324 controls [34]. Thus, we cannot obtain the comprehensive results of the association between rs189037 and cancer risk in Latino and Caucasian population because of the limited sample size.

Rs189037 is in the promoter region of ATM gene and markedly changes the folding architectures. The secondary structure of rs189037 G/A alleles was significant changed using RNAfold prediction [38]. It has been confirmed to be associated with carcinogenesis [38, 40]. The G allele of rs189037 SNP is an independent risk factor for radiation-induced pneumonitis in Chinese thoracic cancer patients [41]. Moreover, rs189037 and other polymorphism in DNA repair genes can serve as candidate prognostic markers of the survival of non-small-cell lung cancer patinets [42]. The combined analysis showed that this SNP was associated with the poor prognosis. In addition, Piaceri et al. reported that the rs189037 was associated with the longevity in Italian centenarians [43]. Taken into account that the A allele of rs89037 increased the risk of cancer in our meta-analysis, we need to do more efforts to explore its influence on the expression of ATM protein.