SEARCH

SEARCH BY CITATION

Keywords:

  • lung cancer;
  • SNP309;
  • SNP285

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Conflicting reports exist regarding the contribution of SNP309 in MDM2 to cancer risk. Recently, SNP285 was shown to act as an antagonist to SNP309 by overriding the effect of SNP309 on SP1-mediated transcription. Moreover, SNP285 modified the relationship between SNP309 and risk of breast, ovarian and endometrial cancer. We assessed whether SNP285 confounded the effect of SNP309 in lung cancer in a cohort of 720 controls and 556 cases. Our cohort included both Caucasians and African Americans. Neither SNP309 nor SNP285 was associated with lung cancer risk or survival. In addition, removal of individuals who carried the variant C allele of SNP285 did not modify the association between SNP309 with either lung cancer risk or survival. Although an effect of SNP285 has been demonstrated in breast, ovarian and endometrial cancer, our findings do not support a role for this SNP in lung cancer and raise the possibility that the effect of SNP285 is restricted to cancers in women.

Lung cancer is the leading cause of cancer-related death in the United States. In 2012, it is estimated to account for 14% of total cancer cases and 28% of cancer-related deaths.1 Incidence and mortality rates of lung cancer are known to vary by race and ethnicity. African American men have a higher risk of both developing and dying from lung cancer as opposed to European American men.1 Many risk factors, including genetic variation, may contribute to these disparities.2

During the DNA damage response, TP53 helps to prevent cell proliferation through mechanisms such as cell cycle arrest and apoptosis. An excess of p53 could negatively effect and destroy normal cells in the absence of stress; therefore, several genes attenuate the p53 mechanism as a means to maintain homeostasis. Murine double minute 2, or MDM2, is the principal regulator of p53, primarily through its function as an E3 ligase. In addition, it can bind to the terminal transactivation domain of TP53.3 The production of MDM2 is activated by the cytoplasmic presence of p53 through a negative feedback loop and as such, both proteins are intricately dependent on each other for normal cell proliferation and growth.4

Single nucleotide polymorphisms (SNPs) are the most common type of germline genetic variation. SNP309, a T>G nucleotide change, is a SNP found at the 309th nucleotide in the inducible P2 promoter region of MDM2. SNP309 enhances the binding affinity of the transcriptional activator SP1 for the MDM2 P2 promoter.5 This higher affinity causes more MDM2 mRNA and protein to be produced. The increased levels of MDM2, coupled with normal production of p53, may cause lower than normal cytoplasmic levels of p53. This in turn can augment DNA damage and hasten cellular transformation.3

In epidemiological studies, associations between SNP309 and cancer risk or age at disease onset have been found in some studies, but not others.6–8 Previous work found SNP309 to be associated with an increased risk of lung cancer in Norwegian and Korean populations and also a decreased risk of incidence in Caucasians and Chinese.9–12 Others have found no association with SNP309 and lung cancer-specific risk in African Americans, Caucasians and Chinese populations.13, 14 SNP309 has also been associated with both better lung cancer-specific survival in Chinese and worse survival in a multiethnic cohort.15, 16 Despite many studies performed on SNP309, associations still remain unclear.

Recently, a novel SNP in the P2 promoter of MDM2 was identified. SNP 285, a G>C nucleotide change, lies 24 nucleotides upstream from SNP309.17 Interestingly, Knappskog and coworkers reported that SNP285 acts as an antagonist to SNP309 by overriding the effect of SNP309 on SP1-mediated transcription18 and causes an overall decrease in MDM2 protein production when the variant alleles of both SNPs are present.18

The minor allele frequency of SNP309 shows differences among population groups,19 whereas the variant allele of SNP285 is generally rare and only observed in Caucasians.19 Moreover, in breast, ovarian and endometrial cancer, epidemiological studies have found that SNP285 neutralizes the effect of SNP309 and that SNP309 confers cancer risk only when associated with ancestral, but not variant, SNP285.18, 20 As such, it is possible that the biological function of SNP285, and its contrasting population frequencies, could explain some of the epidemiological heterogeneity regarding SNP309 in the literature. Here, we investigated this hypothesis in lung cancer.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Study population

The NCI-University of Maryland lung cancer case–control study: Patients in the NCI/UMD case–control cohort resided in Baltimore City, MD, and the surrounding regions. Nonhospital controls were identified using the Department of Motor Vehicles records and matched based on age, race and gender to cases. Study design, inclusion and exclusion criteria were described previously.13 Written informed consent was obtained from all participants, and the study was approved by the Institutional Review Boards of the participating institutions. All participants filled out questionnaires concerning reproductive health, nutrition, smoking and alcohol consumption. Blood components were collected from participants, processed and stored at −80°C until needed. This study included 720 controls and 556 cases. The clinical and demographic characteristics of all individuals included in this study are outlined in Table 1.

Table 1. Characteristics of cases and controls
inline image

Genotyping

DNA was extracted using the QIAamp DNA Mini Kit according to the manufacturer's instructions. SNP309 was genotyped as described previously.13 All samples for SNP285 (rs117039649), and additional SNP309 (rs2279744) samples accrued since our previous study, were genotyped using custom TaqMan® SNP Genotyping Assays (Applied Biosystems®, Life Technologies, Carlsbad, CA). Samples and 10% duplicates were randomized and blinded for both cases and controls. There was 100% concordance for both interplate and intraplate duplicates for MDM2 SNP309 and SNP285. The completion rate for SNP309 and SNP285 was 94%. In addition, 151 samples (cases and controls) that were genotyped for SNP309 in our previous study were regenotyped using the Taqman protocol to confirm the comparability of the two methods. There was 99% concordance between the two methods.

Statistical analysis

Tests for deviations from Hardy–Weinberg equilibrium were evaluated by calculating the expected genotype frequencies based on known allele frequencies and then comparing them to observed genotype frequencies using chi-square tests. Linkage disequilibrium (LD) was assessed using Haploview among controls.

Population means were compared between cases and controls using two-sided Student's t-tests, whereas categorical variables were compared using chi-square tests or Fisher's exact test where indicated (Table 1). Associations between the genotypes and lung cancer risk were estimated by odds ratios (ORs) using an unconditional logistic regression model. These models were adjusted for sex (male/female), age at diagnosis (continuous), smoking status (current/former/never) and pack-years of smoking (continuous). Associations between MDM2 SNP309 and SNP285 with lung cancer-specific survival were assessed using Cox proportional hazards models. Hazard ratios (HRs) and 95% confidence intervals were estimated using univariable and multivariable models controlled for race, stage and histology. Proportional hazards assumptions were verified by visual inspection of log–log plots and with a nonzero slope test of the Schoenfeld residuals. Survival curves were plotted according to Kaplan and Meier.

Tests for linear trend were conducted by including genotypes in the model as a continuous ordinal variable (ptrend). Given the apparent gender and race disparities in lung cancer incidence,1 we also performed a series of analyses stratified by these parameters for both SNPs. Analyses were performed using STATA version 11 software (STATA Corp, College Station, TX). All statistical tests were two sided.

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

SNP309 and SNP285 distributions

There were no statistically significant differences between cases and controls in terms of gender and age (Table 1). However, smoking status and pack-years of smoking were higher in cases than controls (p < 0.001 and p < 0.001, respectively). There were no significant differences between the frequency of SNP309 (p = 0.063) and SNP285 (p = 0.609) between cases and controls. The frequency of either SNP285 or SNP309 did not vary between hospital and population controls; therefore, the controls were treated as a single group. Neither SNP deviated from Hardy–Weinberg proportions. Our analysis showed that individuals carrying the ancestral T allele of SNP309 also carried the ancestral G allele of SNP285 (G/G; Supporting Information Table 1), indicating that the alleles are in LD. In addition, the GG genotype of SNP285 was only observed in cases.

Association between SNP309 and SNP285 with lung cancer risk

In our univariable analysis, SNP309 was associated with an increased risk of lung cancer (OR = 1.47; 95% CI = 1.01–2.16). However, this association was not significant in the multivariable analysis. When considered separately, SNP285 was not associated with risk (OR = 1.09; 95% CI = 0.62–1.90; Table 2).

Table 2. Association between SNP309 and SNP285 and lung cancer risk
inline image

To determine whether SNP285 might confound the effect of SNP309 in lung cancer, we removed those individuals who carried SNP285-C from our analysis. However, the risk estimates for SNP309 in the univariable and multivariable analyses did not change and the association between SNP309 and lung cancer risk remained nonsignificant (Table 2), consistent with a null effect of SNP285 as it pertains to lung cancer risk.

Several studies have also previously suggested that there was an interaction between MDM2 SNP309 and the classic codon 72 (Arg/Pro) TP53 SNP. In our cohort, we had 901 samples,21 where we had overlapping codon 72, SNP309 and SNP285 genotyping data. However, we did not observe an interaction between the codon 72 SNP and the MDM2 SNPs in our cohort (data not shown).

Association between SNP309 and SNP285 with lung cancer survival

SNP309 was significantly associated with prolonged survival in Caucasians (HR = 0.61; 95% CI = 0.42–0.87), but as was observed with risk, this association was not independent (HR = 0.76; 95% CI = 0.50–1.16). Removal of all individuals with at least one copy of the SNP285 C allele did not influence the calculated HR for SNP309 in the univariable or multivariable analysis (Table 3; Supporting Information Fig. 1), indicating that the effect of SNP309 in our study was independent of SNP285. We saw no association between SNP285 and lung cancer-specific survival (Table 3).

Table 3. Association between SNP309 and SNP285 with lung cancer-specific survival
inline image

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Our study sought to determine whether the recently discovered SNP285 neutralized the effect of SNP309 in a lung cancer cohort composed of Caucasians and African Americans. This hypothesis is supported by functional data from a previously published study18 and is consistent with the observation that most of the positive results between SNP309 and lung cancer risk have been in Asian populations, where the C allele of SNP285 is not found. Therefore, we hypothesized that the negative/null reports in Caucasians were confounded by the presence of SNP285. We found, in agreement with previous reports,17, 18 that SNP285 is in complete linkage with the ancient SNP309 and that the CC genotype was observed only in cases. However, when we removed carriers of the SNP285 C allele from our analyses, we did not find that the association between SNP309 and lung cancer risk was modified, thereby suggesting that at least in lung cancer, SNP285 does influence the relationship between SNP309 and cancer risk. This is in contrast to breast, endometrial and ovarian cancer, where SNP309 was found to be associated with increased risk of cancer upon removal of individuals with SNP285 from the analysis.18, 20

As observed in our risk analysis, we did not find evidence for an independent association between SNP309 with survival, although the association was again significant in the univariable model where the G allele was associated with a reduced risk of death. Given that increased MDM2 is associated with good outcome,22 this result is in agreement with the literature. Most importantly, however, we did not find that SNP285 modifies the relationship between SNP309 and lung cancer survival.

In summary, our data suggest that SNP285 does not contribute to lung cancer risk either directly or by modifying the effects of SNP309 and, therefore, does not explain the considerable heterogeneity in the literature concerning SNP309 and its effects in human lung cancer. However, because two functional SNPs have been recently discovered in the P2 region of MDM2, it is possible that additional functional SNPs are present. Furthermore, it is interesting that the studies that found a role for SNP285 in cancer have been in cancers in women (no association was observed in prostate cancer20), thus raising the possibility that interactions between SNP309 and SNP285 are somewhat restricted to this gender. This is an interesting hypothesis given the existing literature regarding the gender-specific nature of the relationship between SNP309 and cancer. Previous studies found that women with the G allele of MDM2 SNP309 have an earlier onset of cancer,9, 23, 24 with early onset classified based on premenopausal and postmenopausal status. Although many of these cancers are not typically related to hormone signaling, there is biological and functional evidence to support the association. For example, the estrogen receptor has been shown to act as a cotranscriptional activator of SP1.25 Furthermore, in ovarian and breast cancer, the association between the G allele of MDM2 and risk of cancer was observed only in estrogen-positive disease.23 In our study, most of our cohort was diagnosed after 60 years of age, and as such, our ability to investigate this hypothesis was restricted. However, in light of both our findings, and those from previous studies, it remains an attractive hypothesis. Future studies should address this question and also consider other novel SNPs in conjunction with those that have previously been studied.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors thank Glenwood Trivers, Donna Perlmutter, Leoni Leondaridis and the Surgery and Pathology Departments from the participating Hospitals; Audrey Salabes and John Cottrell for their contribution to patient accrual and Karen Yarrick for administrative assistance.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_27573_sm_SuppFig1.ppt134KSupporting Information Figure 1.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.