MDM2 Promoter SNP309 Is Associated with Risk of Occurrence and Advanced Lymph Node Metastasis of Nasopharyngeal Carcinoma in Chinese Population

Nasopharyngeal carcinoma (NPC) is an epithelial malignancy with striking racial and geographic distribution differences (1). High incidence rates are observed in the southern Chinese and other individuals of Chinese descent, including Singaporeans, Taiwanese, and Hong Kong Chinese (2–4). These incidence rates are ∼100-fold higher than in the Caucasian populations (1). The marked racial and geographic differences in NPC susceptibility have been considered as a multifactorial and polygenic event with EBV, environmental, and genetic components (1, 5–9). The identification of susceptibility genes contributing to NPC would assist in predicting individual and population risks of NPC development and would help to clarify pathogenesis of this malignancy.

Linkage analyses have suggested several chromosomal regions that may harbor NPC susceptibility genes. Lu et al. (10) were the first to map a susceptibility locus to chromosome 6p22 in affected sib pairs collected from southern China, supporting the involvement of the human leukocyte antigens in the pathogenesis of NPC. Recently, two genome-wide searches carried out in 20 Cantonese-speaking families from the Guangdong province and 18 families from the Hunan province in southern China provided support for susceptibility loci on chromosome 4p15.1-q12 and 3p21.31-21.2, respectively (11, 12). Association studies have also been done, and some of the identified loci, including the cyclin D1 (CCND1; ref. 13), DNA repair genes hOGG1 and XRCC1 (14), heat shock protein 70-2 (HSP70-2; ref. 15), and glutathione S-transferase M1 (GSTM1; refs. 16, 17), seem to contribute to disease susceptibility. Recently, we also reported an association between risk of NPC and the polymorphisms in the promoter of palate, lung and nasal epithelial clone (PLUNC) gene, which codes a protein that may function in the innate immune response in regions of mouth and nose (18). Despite these advances, the alleles that account for most of the genetic susceptibility to NPC remain undiscovered.

It is well known that the p53 pathway plays a key role in preventing carcinogenesis by causing cell cycle arrest or apoptosis (19–21). The activity of p53 may be inactivated through mutations in the p53 gene. It has also been shown that in cancers lacking p53 mutations, the p53 function is abolished or attenuated by other mechanisms, such as the overexpressed human homologue of the mouse double minute 2 (MDM2) protein (22). The MDM2 acts as a principal negative regulator of p53 through at least three different ways. First, by binding to transactivation domain of p53, it inhibits its transcriptional activity (23). Second, MDM2, which acts as an ubiquitin ligase, promotes p53 degradation (24). Third, on binding to p53, MDM2 favors the export of p53 because it contains a nuclear export signal (25). Additionally, MDM2 also seems to possess significant p53-independent functions, through interacting with other cellular proteins that are important in cell cycle control, including pRb, E2F/DP1, and p19ARF, although some of these functions are less well characterized (26, 27). It has been shown that transgenic mice with overexpression of MDM2 are predisposed to spontaneous tumor formation and show both the p53-dependent and the p53-independent tumorigenicity of MDM2 (28). In humans, the amplification and/or overexpression of the MDM2 have been extensively described in more than 40 different types of malignancies (29), and this overexpression has been extensively linked with a worse clinical prognosis, an increased metastasis, as well as a decreased response to therapeutic intervention in cancers (29). With regard to NPC, MDM2 is overexpressed in both EBV-infected cells and 31% to 56% of the tumors (30–33). Furthermore, it has been shown that the MDM2 mRNA overexpression was significantly related to neck lymph node metastasis of NPC (32). Based on above in vitro and in vivo potential functional relevance of MDM2 in the pathogenesis of NPC, we hypothesize that the MDM2 may be an excellent biological candidate susceptibility gene for the NPC; it is expected that the genetic polymorphisms within MDM2 could result in genotype-dependent difference in susceptibility to NPC.

Recently, a functional single nucleotide polymorphism (SNP) in the MDM2 intronic promoter (named SNP309) was identified. The cells carrying the SNP309 GG genotype were found to increase the binding affinity of the transcriptional activator Sp1 and subsequently resulted in higher expression levels of MDM2 mRNA and protein compared with those carrying the SNP309 TT genotype (34). As a consequence, the degradation of p53 in cells carrying the GG genotype was enhanced, resulting in a decreased response to DNA-damaging agents and acceleration in tumorigenesis of both sporadic and hereditary cancers compared with those carrying the TT genotype (34). Therefore, it is hypothesized that this functional SNP in the MDM2 promoter may contribute to individual's tumor susceptibility. Indeed, several epidemiologic studies have shown SNP309 to be associated with the risk of esophageal squamous cell carcinoma and lung cancer (35, 36). The role of the SNP309 in NPC, however, has never been specifically investigated. In the present study, we examined whether the functional polymorphism in MDM2 promoter, SNP309, have any bearing on the risk or severity of NPC in the Chinese population.

Study subjects. This study consisted of two populations of patients with NPC and controls resided in Guangxi and Guangdong province, respectively, both of which located in southern China. The Guangxi population consisted of 593 patients with NPC and 480 controls. All subjects were unrelated ethnic Han Chinese and residents in Nanning city and the surrounding regions. All patients with NPC were newly diagnosed and pathologically confirmed, which were consecutively recruited at the Guangxi Cancer Hospital (Nanning, China) between September 2003 and July 2005. The response rate for case patients was 95%. Patients that received chemotherapy or radiotherapy before surgery or had other type of cancer were excluded from this study. Tumor staging was done according to the tumor-node-metastasis classification by the 1997 American Joint Committee on Cancer system (37). All tumor-node-metastasis classifications were determined by senior pathologists of the hospital based on the postoperative histopathologic examination. The controls were randomly selected from a community cancer screening program for early detection of cancer conducted in the same regions during the same time period as the NPC cases were collected. The selection criteria for the controls included no individual history of cancer. The response rate for control subjects was 91%. At recruitment, informed consent was obtained from each subject, and personal information on demographic factors, medical history, and tobacco and alcohol use were collected via structured questionnaire. The Guangdong population, which contained 239 NPC patients and 286 controls, has been described in detail previously (18). As described previously, the data on tumor staging, smoking, and drinking were not available for this sample set. This study was done with the approval of the Ethical Committee of Chinese National Human Genome Center.

Polymorphism genotyping. We extracted genomic DNA from peripheral blood leukocytes of 5 mL whole blood using standard phenol/chloroform protocols. DNA samples were diluted to 10 ng/μL and distributed to 96-well plates; each 96-well plate contained 94 samples and 2 no-DNA control water. Then, the MDM2 SNP309 polymorphism was genotyped in our case-control populations using PCR direct sequencing. The primers 5′-GTCCCCTCTATCGCTGGTTC-3′ and 5′-AGCAAGTCGGTGCTTACCTG-3′ were used for amplifying and sequencing the target region containing the SNP309 site. PCR conditions were identical to those for the SNP discovery described previously (18, 38) except for an annealing temperature of 56.5°C for the SNP309 site. Genotyping was done in a blind manner that the performers did not know the subjects' case and control status. For quality control, a 15% masked, random sample of cases and controls was tested twice by different people and all results were 100% concordance.

Statistical analysis. Genotype and allele frequencies for the SNP309 were determined by gene counting, and departures from Hardy-Weinberg equilibrium were tested using the χ² test. Subjects were considered smokers if they smoked up to 1 year before the date of cancer diagnosis for cases or up to the date of interview for controls. Information was collected on the amount of cigarettes smoked per day, the age at which the subjects started smoking, and the age at which ex-smokers stopped smoking. Lighter or heavier smokers were categorized by the approximated 50th percentile pack-year value among controls [i.e., ≤19 or >19 pack-years; (cigarettes per day / 20) × (years smoked)]. An alcohol drinker was defined as someone who consumed alcoholic beverages at least once per week for ≥6 months. Comparisons of age, sex, smoker, smoking level, and drinker distributions between patient and control groups were done by use of the χ² test. Differences of mean age and mean smoking level between the groups were analyzed by use of an unpaired t test. Multivariate logistic regression analyses were done to evaluate whether there was association between the MDM2 SNP309 and risk and severity of NPC. The P value, odds ratio (OR), and 95% confidence interval (95% CI) were calculated and adjusted for age, gender, and tobacco and alcohol use where appropriate. Potential modification of the effect of the SNP309 genotypes on risk and severity of NPC was assessed for the above factors by the addition of interaction terms in the logistic model and by separate analyses of subgroups of subjects determined by these factors. Because age at diagnosis was normally distributed, t tests were used to examine associations of SNP309 genotypes with age at diagnosis. A P value of <0.05 was used as the criterion of statistical significance, and all statistical tests were two sided. These analyses were done using SPSS software (version 9.0, SPSS, Inc.). To estimate the contribution of the genotypes containing G allele (GT + GG genotype) to an increase in susceptibility to NPC in the population, the population attributable fraction was calculated (39).

Initially, we estimated the effect of SNP309 on NPC susceptibility in Guangxi population consisted of 593 patients with NPC and 480 controls (Table 1). Overall, controls were comparable with cases with regard to the status of smoking. However, more men (χ² = 8.8; P = 0.003), ethnic Han Chinese (χ² = 4.4; P = 0.04), and higher mean age (P = 0.003, t test) were presented in the controls compared with cases, whereas more drinkers (χ² = 10.7; P = 0.001) and higher smoker level (P = 0.03, t test) were in the cases compared with controls. An absolute majority (97.0%) of cases was classified as poorly differentiated squamous cell carcinoma. According to the tumor-node-metastasis system, 5.4%, 42.5%, 34.2%, and 17.9% of patients had stage I, II, III, and IV disease, respectively.

The genotyping results are shown in Table 2. The observed genotype frequencies for the MDM2 SNP309 in both patients (P = 0.49) and controls (P = 0.087) in the Guangxi population conformed to the Hardy-Weinberg equilibrium. The frequencies of the TT, GT, and GG genotype among patients were significantly different from those among controls (χ² = 6.035; df = 2; P = 0.049), and this difference was mainly caused by a higher frequency of the GG genotype among patients compared with controls (33.3% versus 28.9%). Based on logistic regression analysis with adjustment for age, sex, and status of smoking and drinking, the subjects bearing SNP309 GG genotype had a significantly increased susceptibility to NPC compared with ones bearing TT genotype (OR, 1.51; 95% CI, 1.08-2.12; P = 0.016). The heterozygous GT genotype also presented a higher risk for the NPC (OR, 1.36; 95% CI, 0.99-1.86), although the association was not statistically significant (P = 0.056). When the GT genotype was combined with the GG genotype, the subjects bearing the SNP309 G allele (GT + GG genotype) had a significantly increased susceptibility to NPC compared with ones bearing TT genotype (OR, 1.43; 95% CI, 1.04-1.91; P = 0.019). When cases were limited to those with poorly differentiated squamous cell carcinoma (n = 575), the general pattern of results was similar (data not shown).

The associations between the SNP309 and susceptibility to NPC in the Guangxi population were further examined with stratification by age, sex, nationality, smoking and drinking status, and smoking level (Table 3). Although the susceptibility to NPC associated with the SNP309 GT + GG genotype seemed to be more pronounced in subjects who were males, older (>47 years), drinkers, smokers, and heavier smokers (>19 pack-years), these differences could be attributed to chance (all P > 0.30, test for homogeneity), indicating that these potential confounding factors had no modification effect on the risk of NPC related to SNP309 genotypes.

We then confirmed this finding (i.e., increased susceptibility of SNP309 GT + GG genotype to NPC) in a further sample set, the Guangdong population (Table 1). This population contained 239 cases and 286 controls, and as described previously (18), there was no significant difference between patients and control subjects in terms of mean age distribution. However, more men were presented in the cases compared with controls (68.2% versus 52.8%; χ² = 12.2; P < 0.01). Overall, the SNP309 genotypic frequencies observed in this group were very comparable with those from the Guangxi population and conformed to the Hardy-Weinberg equilibrium (P = 0.21 in patients and P = 0.094 in controls; Table 2). Again, there was an excess of GT + GG genotype in patients than in controls (82.1% versus 75.1%; OR, 1.53; 95% CI, 1.00-2.36; P = 0.049). Similarly, in the stratification analyses, sex and age had no modification effect on the risk of NPC related to SNP309 genotypes (Table 3). When Guangxi and Guangdong population were combined, the OR for SNP309 GT + GG genotype developing NPC was 1.45 (95% CI, 1.12-1.85; P = 0.0028) compared with TT genotype. By Adams's formula (39), the population attributable fraction calculated by relative risk (OR, 1.45; 95% CI, 1.12-1.85) combined with the SNP309 GT + GG genotype frequency in the overall population (78.2%; Table 2) indicates that 26.0% (95% CI, 8.6-39.9%) of elevation in risk of NPC can be attributed to the susceptible effect of the SNP309 GT + GG genotype.

We also assessed the effect of SNP309 GT + GG genotype on severity of NPC (as measured by tumor-node-metastasis staging system) in the Guangxi population. The distributions of the SNP309 genotypes were not statistically significantly different among the subgroups with different clinical stage, or different T and M classification of the cancer (data not shown). However, we did note a trend toward SNP309 GT + GG genotype in the patients with more frequent involvement of lymph node (P = 0.030, test for trend; Table 4). After adjustment for age, sex, and status of smoking and drinking, multivariate regression analyses revealed that patients with the GT + GG genotype, compared with the TT genotype, had an OR of 1.84 (95% CI, 1.09-3.05; P = 0.019) for being more frequent involvement of lymph node (N₂ + N₃ versus N₀ + N₁; Table 4). In the stratification analyses, sex, age, and smoking and drinking had no modification effect on the risk of more frequent involvement of lymph node related to the GT + GG genotype (all P > 0.20, test for homogeneity within each strata). The effect of SNP309 genotypes on severity of NPC was not reassessed in the Guangdong population because the tumor-node-metastasis staging data were not available in this sample set.

In 593 NPC patients of Guangxi population, the average age at diagnosis (±SD, years) were 47.1 ± 10.7 years for subjects with the GG genotype, 45.9 ± 11.9 years for those with the GT genotype, and 47.3 ± 10.9 years for those with the TT genotype and these differences were not significant (P = 0.23 for GG versus TT and P = 0.36 for GT versus TT). Similarly, in 239 NPC patients of Guangdong population, the average age at diagnosis were 49.3 ± 11.0, 45.8 ± 11.6, and 46.4 ± 10.5 years, respectively, for subjects with the GG, GT, and TT genotype, respectively, and these differences were also not statistically significant (P = 0.19 for GG versus TT and P = 0.71 for GT versus TT).

In this study, we investigated the associations of the functional SNP309 in the intronic promoter of MDM2 with risk of occurrence and progression of NPC in populations in southern China. We found that the SNP309 GT + GG genotypes were associated with a significantly increased risk of occurrence of NPC in both independent populations. We also observed that this polymorphism was significantly associated with the advanced neck lymph node metastasis of NPC in Guangxi population. Our data, together with the earlier described functional significance of the SNP309 (34) and the recent evidence for association of SNP309 with the risk of esophageal squamous cell carcinoma (35) and lung cancer (36), suggest that the SNP309 GT + GG genotypes were potent genetic risk factor for both onset and advanced neck lymph node metastasis of NPC. Given the role of MDM2 in the development of cancers, one might expect individuals who carry the SNP309 GT + GG genotypes, and thus have increased expression of MDM2 and subsequently attenuated p53 function over a lifetime, may be at a higher susceptibility to developing NPC and a higher risk of advanced lymph node metastasis after establishment of this malignancy. To our knowledge, this is the first report of the genetic associations between MDM2 and occurrence and progression of NPC, confirming the initial hypothesis that the MDM2 oncoprotein may play a role in the pathogenesis of this malignancy.

The genetic association between MDM2 SNP309 and occurrence of NPC is biologically plausible. It is well known that the p53 pathway plays a key role in preventing carcinogenesis (19–21). The p53 function can be abolished or attenuated by overexpressed MDM2 protein (22). Transgenic mice with overexpression of MDM2 are predisposed to spontaneous tumor formation and show both the p53-dependent and the p53-independent tumorigenicity of MDM2 (28). Furthermore, amplification and/or overexpression of the MDM2 gene have been extensively described in several forms of human cancer (29). With regard to NPC, it has been shown that MDM2 is overexpressed in both EBV-infected cells and 31% to 56% of the tumors (30–33).

It is also biologically plausible for the association between MDM2 SNP309 and advanced neck lymph node metastasis of NPC. Accumulating evidence shows that the overexpression of MDM2 is associated with increased metastasis, decreased response to therapy, and poor prognosis (29). With regard to NPC, it has been shown that the MDM2 mRNA overexpression was significantly related to neck lymph node metastasis but not to T stage of NPC (32). A recent report proposes a mechanism for the increase in metastasis. Hypoxia, which is common in many tumors, up-regulates MDM2 in a p53-independent manner. This overexpression leads to an increased metastatic efficiency in cell lines, perhaps by rendering tumor cells less sensitive to stress-induced cell death (40). Additionally, MDM2 increases levels of vascular endothelial growth factor (41), which facilitates extravasation of tumor cells and can lead to growth of metastases (42). MDM2 expression also increases basement membrane degradation in urothelial carcinoma (43), allowing the cancer to spread. Thus, MDM2 expression increases the risk of metastasis, contributing to the overall poorer prognosis for cancers expressing MDM2. However, the role of MDM2 in cancer metastasis and its underlying mechanisms are not fully understood, and more investigations with larger sample sizes and longer durations of follow-up are needed.

Given the functional relevance of SNP309 in modulation of MDM2 expression level and subsequent efficiency of the p53 pathway, it would be expected that this polymorphism has a phenotype that affects the age at onset of cancers. Indeed, Bond et al. (34) reported that individuals with the Li-Fraumeni syndrome with SNP309 GG genotype developed cancers some 10 to 12 years earlier than those individuals with a TT genotype. However, contrary to the expectation, we failed to find evidence of association between SNP309 genotype and age of NPC onset in patients from both Guangdong and Guangxi populations. Consistent with our results, very recent studies in several tumor types, including colorectal cancer, uterine leiomyosarcoma, and squamous cell carcinoma of the head and neck (44, 45), also showed no association of age of tumor onset with SNP309 genotype. Because of limited statistical power, subtle changes may have been missed. In addition, these findings could be the result of the different tumorigenesis mechanisms of different tumor types. Consequently, we urge that the role of SNP309 in age of tumor onset be investigated in additional studies with larger sample sizes.

Previous studies revealed that the long-term cigarette smoking was associated with the risk of NPC (1, 6). Furthermore, in the recent two reports, a significant interaction between the MDM2 SNP309 and smoking was observed (35, 36). In the present study, however, we did not find that the SNP309 interact with tobacco smoking, although the risk of NPC associated with the GT + GG genotypes seemed to be more pronounced in subjects who were smokers and heavier smokers with >19 pack-years (Table 3), suggesting that smoking may not have modification effect on the onset and advanced lymph node metastasis of NPC related to SNP309 genotypes. This discrepancy may be the result of different molecular mechanisms of NPC and other types of cancers. Alternatively, the limited power of statistics due to the small sample size of smokers in the present study might also account for the null interaction between smoking and SNP309. Additional studies investigating the interaction between the smoking and SNP309 should be required in the future.

Recently, the MDM2 SNP309 has been extensively studied in many cancers but the results are conflicting. Several studies have reported an association of SNP309 G allele with the increased risk of esophageal squamous cell carcinoma and lung cancer (35, 36). In contrast, some reports showed no association between SNP309 and risk of breast and lung cancer (46, 47); a very recent study revealed reverse association about the risk allele (i.e., TT genotype was reported to be associated with lung cancer risk; ref. 48). The conflicting results could be attributable to the different ethnicities of study populations and/or different tumorigenesis of different cancers. Additionally, other factors in the studies, such as small sample size or inadequate adjustment for confounding factors, could also cause the inconsistent results. Consequently, additional well-designed case-control studies in a wide spectrum of cancers with ethnically diverse populations are warranted to understand the roles of MDM2 polymorphism in the etiology of cancers.

In reviewing the results of this study, one must also keep several potential limits in mind. First, as a hospital-based study, our NPC cases were enrolled from the hospitals and the controls were selected from the community population; inherent selection bias cannot be completely excluded. However, by further adjustment and stratification in data analyses, the potential confounding factors might have been minimized. Second, several association studies have addressed to identify the genes that may relate to the susceptibility to NPC (13–18). Most of the results, however, could not be replicated in subsequent studies in other populations. Although the highly significant association between MDM2 and onset and metastasis of NPC derived from a biologically based a priori hypothesis, our initial findings should be independently verified in other populations with high incidence rate of NPC, such as other southern Chinese, Singaporeans, and Taiwanese.

In conclusion, our data provide strong evidence that the MDM2 SNP309 GT + GG genotypes may be genetic risk factor for the occurrence and progression of NPC in Chinese patients. If confirmed by other studies, knowledge of genetic factor contributing to the pathogenesis of the NPC as presented here may have implications for the cancer screening and treatment of this disorder.