Relationship between cyclooxygenase 8473T>Cpolymorphism and the risk of lung cancer: a case-control study

Although cigarette smoking is the major cause of lung cancer, only a fraction of smokers develop lung cancer during their lifetime, and this suggests that genetic factors are of importance in determining an individual's susceptibility to lung cancer [1].

Cyclooxygenase (COX, also known as prostanglandin endoperoxide synthase) is the rate-limiting enzyme for the production of prostaglandins (PGs) from free arachidonic acid. Two isoforms of COX have been identified. COX-1 is constitutively expressed in most normal tissues and it is considered to be a housekeeping enzyme that is responsible for various physiologic functions. In contrast, COX-2 is rarely expressed in normal tissues, but it is rapidly induced in response to cytokines, growth factors and tumor promoters [2. 3]. Overexpression of COX-2 has been reported in a variety of human malignancies, including lung cancer [4–6]. Up-regulation of COX-2 can promote tumor growth and metastasis by stimulating cell proliferation, angiogenesis, and invasiveness in addition to inhibiting apoptosis and immune surveillance [7–10]. Moreover, a number of epidemiologic studies have recently suggested that COX-2 inhibitor may have a protective effect on the development of lung cancer [11, 12]. Taken together, these observations indicate that COX-2 plays an important role in the development of lung cancer. In addition, several studies have shown that the frequency and level of COX-2 up-regulation was higher in adenocarcinoma (AC) than in squamous cell carcinoma (SCC), and small cell carcinoma (SM) showed virtually negligible expression [6, 13, 14]. This suggests that the COX-2 gene may have a pronounced association with the development of AC.

Single nucleotide polymorphisms are the most common form of human genetic variation, and they may contribute to the individual susceptibility to cancer. Many studies have demonstrated that some of these variants affect either the expression or activities of enzymes; therefore, they are associated with the risk of cancer [15, 16]. Thus, we hypothesized that functional SNPs in the COX-2 gene may have an impact on COX-2 expression or activity, and so they modulate the susceptibility to lung cancer. To test this hypothesis, a case-control study was conducted to evaluate the association between COX-2 polymorphisms and the risk of lung cancer. Among the identified COX-2 polymorphisms [17–19], we have focused on the -1186T>G and -765G>C in the promoter region and the 8473T>C in the 3'-untranslated region (3'-UTR) since these polymorphisms in the transcription or posttranscriptional regulatory regions have been reported to be associated with various human diseases [17–21]. In the present study, we evaluated with the association of the 8473T>C polymorphism with lung cancer because the -1186T>G and -765G>C polymorphisms were not detected in a preliminary study that consisted of 27 lung cancer cases and 27 healthy controls.

The lung cancer risk related to the COX-2 8473T>C genotypes was also shown in Table 2. When the 8473 TT genotype was used as the reference group, the 8473 TC and the 8473 CC genotypes were associated with a non-significant decreased risk for overall lung cancer, respectively [adjusted odds ratio (OR) = 0.87; 95% confidence interval (CI) = 0.68–1.11; and adjusted OR = 0.72; 95% CI = 0.41–1.25, respectively). When the analyses were stratified by the tumor histology, the 8473 TC genotype and the combined 8473 TC + CC genotype were associated with a significantly decreased risk of AC, respectively (adjusted OR = 0.64; 95% CI = 0.45–0.91, P = 0.01; and adjusted OR = 0.64; 95% CI = 0.46–0.90, P = 0.01, respectively). No significant association was found between the 8473T> C polymorphism and the risk of SCC or SM.

We investigated the influence of COX-2 polymorphisms on the risk of lung cancer by conducting a hospital-based case-control study. The 8473T>C polymorphism in the 3'-UTR region of the COX-2 gene was associated with a significantly decreased risk of AC of the lung. The protective effects were evident in the males, the older individuals and the smokers. These findings suggest that the COX-2 8473T>C polymorphism could be used as a marker for the genetic susceptibility to AC of the lung. Of three major histological types of lung cancer, the proportion of AC is increasing world-wide. Thus, identifying the genetic factors that are responsible for the susceptibility to AC is indispensable for establishing novel and efficient ways to prevent this disease.

Because the COX-2 -1186T>G and -765G>C polymorphisms were not detected in the preliminary study that included 27 healthy controls and 27 lung cancer cases, we analyzed only the association of the COX-2 8473T>C polymorphism with the risk of lung cancer. The frequency of the COX-2 8473C allele among the healthy controls in the current study was 0.244, which was significantly lower than that (0.535) of healthy Norwegians [18], but it was significantly higher than that (0.187) of healthy Chinese [26].

In the current study, the COX-2 8473T>C polymorphism was significantly associated with the risk of AC, but it was not associated with SCC or SM. Although the reasons for the observed histology-specific difference in the risk conferred by the COX-2 polymorphism remain to be elucidated, this difference may be attributable to the differences in the pathways of carcinogenesis among the different histological types of lung cancer. Therefore, certain genotypes could confer a greater susceptibility to a particular histological type of lung cancer [22–24, 27, 28]. Several studies have shown that the frequency and level of COX-2 up-regulation was higher in the ACs than in the SCCs, and the SMs showed a virtually negligible expression [6, 13, 14]. These observations suggest that the COX-2 gene may have a pronounced association with the development of AC, and they are comparable with our finding that the COX-2 polymorphism plays an important role in determining the genetic susceptibility to AC.

A few studies have investigated the association between the COX-2 8473T>C polymorphism and the risk of lung cancer [18, 26]. Campa et al. [18] have reported that the 8473 TC and the 8473 CC genotypes were associated with a significantly increased risk of lung cancer in a Norwegian population. In contrast to the Norwegian study, Hu et al. [26] have reported that the combined 8473 TC + CC genotype was associated with a significantly decreased risk of lung cancer in a Chinese population. In concordance with the Chinese study, the combined 8473 TC + CC genotype was associated with a significantly decreased risk of AC in our current study. Although it is hard to decipher the reasons for the oppositely directed association of the COX-2 8473T>C polymorphism for lung cancer in the different studies, several genetic and environmental factors that are relevant to this polymorphism might have caused the different results. Moreover, since a protective genotype in one population may increase the cancer risk in another population due to other linked polymorphisms that exhibit stronger effects on the susceptibility to cancer, the different genetic background of the study subjects should be also considered. In addition, inadequate study design such as nonrandom sampling, a limited sample size and the pitfalls arising from unknown confounders should be also considered.

Whether the COX-2 8473T>C polymorphism itself alters mRNA stability and/or translation efficiency or if it is in linkage disequilibrium with the other functional polymorphisms remains to be investigated. The 3'-UTR region of COX-2 contains highly conserved adenosine- and uridine-rich elements that are composed of the Shaw-Kamens sequence (AUUUA), which is also known as AU rich elements [29]. This motif is present within the 3'-UTRs of many protooncogenes and cytokine genes, and it is involved in the regulation of COX-2 production by acting both as a message instability determinant and a translation inhibitory element [30–32]. Therefore, the 8473T>C polymorphism located within the functional region of 3'-UTR of the COX-2 gene may affect the message stability and/or translation efficiency, and so this results in differential COX-2 expression.

Several polymorphisms have been identified in the promoter region of the COX-2 gene [17–19]. The -765G>C polymorphism, which is located at just 5' to a STAT1 binding site, has been found to influence COX-2 expression and it contributes to the risk of various human diseases [17, 21, 33]. Although the frequency of carrying the -765 GC or CC genotype has been reported to be 25% to 50% in Caucasians [17, 21, 33], the -765G>C polymorphism was not detected in the preliminary study that included 27 healthy controls. These samples included 54 chromosomes, and this provides for at least a 95% confident level to detect alleles with frequencies of > 5%. Thus, it is very likely that if this polymorphism exists, it may not play a major role in the genetic susceptibility to lung cancer in the Korean population [34, 35]. Kang et al. [19] have recently reported that the -1186T>G polymorphism in the nuclear factor kappa-B binding region was associated with the risk of bladder cancer in a Korean population. In this study, however, the -1186T>G polymorphism was very rare (the frequency of the -1186G allele was 1.0%), and this not associated with the risk of lung cancer. In the current study, the -1186T>G polymorphism was not detected in the preliminary study that included 27 healthy controls and 27 lung cancer cases, suggesting that this polymorphism may not play a major role in the genetic susceptibility to lung cancer in the Korean population.

During multistage carcinogenesis, COX-2 can enhance tumor invasion and metastasis [7–10]; therefore, the COX-2 8473T>C polymorphism may have an influence on the progression of disease. In the present study, however, no significant difference was observed in the genotype distribution according to the stage of lung cancer (data not shown).

We found that the 8473T>C polymorphism in the 3'-UTR region of the COX-2 gene was associated with the risk of lung cancer, and particularly for the risk of AC. This result indicates that this polymorphism could be used as one marker for defining the risk of lung cancer. However, it is possible that our findings, particularly those findings from stratified analyses, can be attributed to chance because of multiple comparisons and/or the relatively small numbers of subjects in the subgroups. Therefore, additional studies with larger sample sizes will be required to confirm these findings. Future studies on the other COX-2 sequence variants and their biologic function are also needed to understand the role of the COX-2 polymorphisms in determining the risk of lung cancer. Moreover, since genetic polymorphisms often vary significantly between the different ethnic groups, further studies are warranted to clarify the association of the COX-2 polymorphisms with the risk of lung cancer in diverse ethnic populations.