• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Thorac Oncol. Author manuscript; available in PMC Jun 1, 2011.
Published in final edited form as:
PMCID: PMC3000124
NIHMSID: NIHMS253743

Methylation markers for small cell lung cancer in peripheral blood leukocyte DNA

Liang Wang, M.D.,1 Jeremiah A. Aakre, B.S.,2 Ruoxiang Jiang, B.S.,2 Randolph S. Marks, M.D.,4 Yanhong Wu, Ph.D.,1 Stephen N. Thibodeau, Ph.D.,1 V. Shane Pankratz, Ph.D.,2 and Ping Yang, M.D.3

Abstract

Introduction

Small-cell lung cancer (SCLC) is the most aggressive form of lung malignancy.

Methods

To identify and validate potential DNA methylation markers for risk assessment and disease detection, we examined peripheral blood leukocyte DNA specimens for methylation differences between SCLC cases and controls. We tested 1,505 CpG sites using the Illumina Beadchip assay and validated 9 CpG sites using pyrosequencing technology.

Results

In 44 matched SCLC case-control pairs, we identified significant differences at 62 CpG sites (false discovery rate ≤ 0.05) in 52 independent genes. Of those, we further determined 43 sites in 36 genes with a mean methylation level difference greater than 0.03 between the cases and controls. We then selected and validated 9 CpG sites for methylation differences in an independent set of 138 matched case-control pairs. The 9 validated CpG sites predicted a higher risk for cases than controls in 85.8% of all pairs of cases and controls, and two (in genes CSF3R and ERCC1) jointly contributed most of the discriminating ability.

Conclusions

Our replicated results demonstrated feasibility of applying large-scale methylation arrays for biomarker discovery and subsequent validation in peripheral blood DNA. The CpG sites identified in this study may potentially assist in risk prediction and diagnosis of SCLC.

Keywords: methylation, biomarker, small cell lung cancer, leukocyte

Introduction

Small-cell lung cancer (SCLC) constitutes approximately 13% of all newly diagnosed lung cancers. 1 In comparison to the more common non small cell lung cancer (NSCLC), SCLC has more rapid doubling time, higher growth fraction, earlier development of widespread metastases, and more dramatic initial response to chemotherapy and radiation. Despite high initial responses to therapy, most patients die from recurrent disease. Untreated SCLC has the most aggressive clinical course of any lung tumor, with a median survival of only 2 to 4 months after diagnosis. 1 Cigarette smoking is the strongest risk factor for the development of SCLC. Virtually all patients with SCLC are current or past smokers, and its risk is related to the duration and intensity of the smoking. 2, 3

Epigenetics is defined as the study of heritable changes in gene expression, which occur in the absence of a DNA sequence change and is believed to be important in the etiology of common human diseases 4, 5 including cancer. Increasing evidence has demonstrated that epigenetic modifications may result from various types of environmental insults and can lead to cancer development. In a recent review, 6 epigenetics is noted to stand at the epicenter of modern medicine because it unites nuclear reprogramming during development, environmentally induced changes in the body, and the cellular response to external stimuli. Unlike the DNA sequence, epigenetic changes distinguish one tissue type from another and environmental exposures alter the epigenetic program. The ability of genes to alter their expression is controlled by epigenetic factors such as DNA methylation. 7, 8

Differential methylation status in peripheral blood DNA has been linked to risk of several cancers, although little is known for lung cancer. 9-11 To determine whether global methylation in DNA derived from peripheral blood, an easily accessible tissue, is associated with head and neck squamous cell carcinoma, Hsiung et al. 9 assessed the LRE1 sequence methylation level in a population-based case-control study and found that hypomethylation of the sequence led to a significant 1.6-fold increased risk for the disease. Moore et al 10 reported in a case-control study an association of leukocyte DNA hypomethylation with increased risk of developing bladder cancer, independent of smoking and other assessed risk factors. More currently, Widschwendter et al 11 examined locus-specific methylation and found that particular methylation patterns in peripheral blood DNA may serve as surrogate markers for breast cancer risk. Therefore, methylation status in peripheral blood DNA specimens may provide a useful biomarker for disease risk assessment and potential early detection and differential diagnosis. Because the methylation status is reversible, further understanding the role of methylation in disease etiology may facilitate targeted therapy and pave the road for future chemo-prevention and arresting disease progression. In this study, we used an array-based genomic DNA methylation approach to identify potential quantitative biomarkers for diagnosis or risk assessment of SCLC using peripheral blood DNA in a case-control study.

Material and Methods

Sample recruitment

The methods of identifying and enrolling SCLC patients and controls were published previously. 12-14 In brief, newly diagnosed cases of lung cancer are identified by a daily electronic pathology reporting system. Once identified, patients are consented and enrolled, their medical records abstracted, and interviews conducted. Overall participation and blood sample donation rates were 87% and 73%, respectively. For controls, we selected community residents, identified by having had a general medical examination and a leftover blood sample from routine clinical tests, excluding individuals diagnosed with major organ failure (e.g., heart, brain, lung, kidney, or liver) on or prior to their visit. SCLC cases were identified from among all lung cancer cases, and controls were selected such that the distributions of age, sex, and smoking history were comparable between the cases and controls. Ninety-five percent of the study subjects were White, representing an U.S. Midwestern population in and surrounding Minnesota. The Mayo Clinic Institutional Review Board approved this study.

DNA modification by Sodium Bisulfite

We extracted DNA from 5ml of whole blood utilizing an automated platform following QIAGEN kit (Germantown, MD USA). Because whole blood DNA was predominantly derived from leukocytes and freely circulating DNA in whole blood is negligible, we referred to the whole blood DNA as leukocyte DNA. We modified the genomic DNA specimens using an EZ DNA Methylation kit from Zymo Research Corporation (Orange, CA) that combined bisulfite conversion and DNA clean up. This modification kit is based on the three-step reaction that takes place between cytosine and sodium bisulfite where cytosine is converted into uracil. We used 1μg of genomic DNA from peripheral blood DNA for the modification under recommendation from the manufacturer. Treated DNA specimens were stored at -20°C and were assayed within two weeks.

Methylation profiling analysis

We labeled and hybridized the modified DNA specimens with equal numbers of samples from each group, balanced across the entire Beadchip, to avoid confounding study results with processing variance. We imaged the arrays using a BeadArray Reader scanner, which represented each methylation data point as fluorescent signals from the M (methylated) and U (unmethylated) alleles. The proportion methylated (β-value) at each CpG site was calculated using BeadStudio Software (Illumina) after subtracting background intensity, computed from negative controls, from each analytical data point.

Pyrosequencing methylation assays

Primers were designed using Pyrosequencing Assay Design Software (Biotage AB, Uppsala, Sweden). Sequences of the primers are listed in Table 1. The PCR was carried out on 10ng of bisulfite treated DNA using TaqGold DNA polymerase (Applied Biosystems) under the following conditions: 10 min at 95°C, followed by 50 cycles of 35 sec at 95°C, 35 sec at 57.5°C, and 1 min at 72°C. Pyrosequencing reactions were performed on Biotage PyroMark MD System (Biotage AB, Uppsala, Sweden) according to the manufacturer's protocols by the sequential addition of single nucleotides in a predefined order. Raw data were analyzed using Pyro Q-CpG 1.0.9 analysis software (Biotage AB, Uppsala, Sweden). The CpG methylation level (ranging from 0 to 1) was represented by percentage of methylated C among the sum of methylated and unmethylated C.

Table 1
Primers for pyrosequencing methylation assay

Data analysis

We summarized and compared demographic characteristics between cases and controls using chi-square tests for nominal variables or rank sum tests for the quantitative variables. We also summarized the percent methylated measurements by their mean and standard deviation within the two study groups and used analysis of covariance approaches to compare the degree of methylation between study groups for each CpG site while adjusting for pack years of smoking. Because of the non-normality of the methylation values, we used rank-based analyses, which are analogous to rank-sum tests when there are no covariates. After obtaining the p-values for each of the CpG sites in the testing set, we employed false discovery rate (FDR) approaches and computed a q-value for each p-value 15. CpG sites with q-values of less than 0.05 were considered to be significant.

In the validation phase, we compared the methylation levels of the 10 selected CpGs between cases and controls in the validation set using the rank-based procedures outlined above. We also used logistic regression approaches to simultaneously assess the association between all nine validation CpGs and case-control status. We further refined this multivariable model via stepwise model selection with the p-value to enter and remain in the model set at 0.05, to determine a CpG set that simultaneously contributes to the discrimination between SCLC cases and controls. As part of these logistic regression analyses, we measured the degree of concordance between model predictions and observed case-control status by extracting estimates of the area under the receiver operating characteristic (ROC) curve. This quantity, often referred to as the c-statistic, examines all possible case control pairs and measures the proportion of the time the statistical model predicts higher risk for the case 16. All analyses were conducted using the SAS software system (Cary NC).

Results

Characteristics of study subjects

By matching design, no difference in age, sex, and smoking status was found between the cases and controls in both the testing and validation sets. Basic descriptive information of the cases and controls are provided in Table 2. For the testing set, five cases were dropped due to DNA quality issues and the remaining 39 cases and 44 controls were used in the analysis. There was a greater than 3-year difference between the cases and controls in the mean pack-years of cigarette smoking (60.1 vs. 56.5). However, median pack-years were similar (51 vs. 52), and the test comparing the two groups did not reach statistical significance (p=0.525). To be conservative, the number of pack-years was adjusted in all DNA methylation analyses.

Table 2
Characteristics of patients with SCLC and healthy controls

Differentially methylated CpG sites

Since the majority of the SCLC patients received radiation treatment or chemotherapy before blood was drawn, we examined the correlations between the time on treatment (as a proxy for treatment intensity) and the degree of methylation to determine if the CpG methylation levels might be affected by treatment in the 39 SCLC patients. Among the 1,505 CpG sites, we found that the length of time on treatment was significantly correlated with the methylation levels of 173 CpGs (p<0.05). While some of these associations may be false positives, but to be conservative, we excluded all 173 CpGs from the analyses. Among the remaining 1332 CpG sites, 922 were located within CpG islands and 410 were in non-CpG islands. We observed significant differences between SCLC cases and controls at 62 sites in 52 independent genes (FDR<=0.05). Interestingly, only 25 of the 62 sites were in CpG islands, which was significantly lower than the expected 42.9 sites (62×922/1332) (p<0.001, Chi square test). The odds of a significant CpG not being in a CpG island was greater than three times higher than the odds of being in a CpG island (OR=3.56, 95%CI: 2.11-6.00). Furthermore, only 6 of the 62 sites showed an increased level of methylation in SCLC patients, including two in the ITK gene, two in the RUNX3 gene, one in each of the CTLA4 and PLG genes. Because some methylation differences were small and difficult to reliably detect, we further excluded the CpG sites with an absolute mean β difference of less than 0.03, resulting in 43 significant CpG sites of primary interest in 36 independent genes (Table 3).

Table 3
Differential methylations between 39 SCLC cases and 44 healthy controls in testing set

Validation of selected CpG sites by pyrosequencing methylation assay

Based on three major parameters (FDR q values, number of significant CpGs/gene, and mean difference between groups), we selected 10 CpG sites including 9 significant CpGs (FDR < 0.05) for validation and 1 non-significant CpG (FDR > 0.05). These CpG sites were located in 10 different genes (IL10, PECAM1, S100A2, MMP9, ERCC1, EMR3, SLC22A18, TRIP6, CSF3R and CAV1), with CAV1 serving as a negative control. We designed a new assay for each of the 10 CpG sites using pyrosequencing technology 17, 18. Figure 1 shows methylation levels of a CpG site, 85bp upstream to the transcription start site in the gene, IL10, in three different samples.

Figure 1
Methylation analysis of IL10_P85_F CpG by Pyrosequencing technology.

We then tested the 10 CpG sites for methylation differences, again in peripheral blood DNA specimens from a validation set between 138 SCLC cases and 138 matched controls (Table 2, right panel). The nine testing-set-positive CpG sites again demonstrated significant differences (all p-values ≤0.0003, Table 4), while the negative control CpG site only differed between the validation set of the cases and controls in an absolute percent methylated by less than 1%. This small difference did not reach statistical significance.

Table 4
Differential methylations between 138 SCLC cases and 138 matched controls: Validation study

CpG Methylation patterns and risk prediction of SCLC using logistic regression models

Based on the nine validated CpG sites accounting for age, sex, and smoking history, our model had an area under the ROC curve of 0.858 (Figure 2), suggesting the model correctly classified SCLC cases as being at a higher risk than controls for 85.8% of case-control pairs. Further stepwise selection identified two of the nine sites, one in CSF3R and the other in ERCC1, contributing independent information to discriminate cases from controls. Specifically, for each five-percent decrease in the methylation level of ERCC1, there was an approximately four-fold (OR=3.9, 95% CI: 2.0-6.1, p<0.001) increase in the odds ratio of SCLC; for each five-percent methylation decrease of CSF3R, there was a 1.5-fold higher odds ratio of SCLC (OR=1.5, 95% CI: 1.1-2.0, p=0.008).

Figure 2
ROC curve in the validation set of 138 cases and 138 controls using the 9 CpGs selected from the set.

Discussion

In applying newly-developed methylation BeadChip technology, we performed methylation profiling analysis of 1,505 tested CpG sites in 807 genes and identified 62 CpGs whose methylation status were significantly associated with SCLC. Forty-three of these 62 had an absolute mean difference greater than 0.03 between the cases and controls, and nine of these were further confirmed using pyrosequencing methylation assays in additional cases and controls. These results suggest methylation status of peripheral blood DNA, a stable and easily accessible material, might be reliably used for risk assessment and diagnosis of SCLC.

Among the nine validated genes, at least five were reported to be associated with lung cancer. For example, the IL10 expression by tumor-associated macrophages showed a significant role in the progression and prognosis of non-small cell lung cancer (NSCLC). 19 Patients whose tumors had a positive S100A2 expression had a significantly lower overall survival and disease-specific survival rate. 20 Another study showed S100A2 being upregulated in an early stage of NSCLC patients, indicating its important role for molecular diagnosis of NSCLC at an early stage; therefore, prognostically more favorable. 21 High ERCC1 expression was associated with short survival. 22, 23 Interestingly, results of these studies came from cancer tissues and were consistent with our findings from peripheral blood DNA, indirectly supporting that hypomethylation of these genes may increase gene expression. To date, it is not known whether the abnormal methylation in peripheral blood DNA from SCLC patients is inherited or acquired. Further investigation is clearly needed.

Imprinting genes in peripheral blood leukocytes may also serve as an indicator of cancer risk. 24 In this study, we observe an imprinting gene, SLC22A18, whose methylation levels in the tested CpG demonstrated a strong association with SCLC in both the testing and validation sets (adjusted p <0.0001, Tables 2 and and4).4). The gene is located in the imprinted gene domain of 11p15.5, an important tumor-suppressor gene region; alterations in this region have been associated with the Beckwith-Wiedemann syndrome and multiple malignancies including lung cancer. 25-27 The finding that the SLC22A18 methylation level is associated with SCLC further demonstrates the importance of the imprinting region in cancer etiology.

Global hypomethylation has been reported in cancer tissues (somatic) as well as constitutive tissues (germline) 9, 10, 28-30. Lack of DNA methylation at specific CpG sites has also been reported recently to be associated with breast cancer risk 11. Of the 62 CpG sites that were identified in this study, 56 are hypomethylated and only 6 are hypermethylated. The hypomethylation may be caused by different mechanisms. First, tumor formation requires activation of oncogenes and tumor response genes, which can be achieved through promoter hypomethylation of related genes. Second, this hypomethylation may be part of global hypomethylation commonly seen in cancer patients and may be not associated with activation of any specific genes. The fact that methylation changes preferentially occur in non-CpG islands indirectly support this possibility. If global hypomethylation causes genomic instability 29 and hence increases risk to cancer, the current finding from peripheral blood DNA may represent, at least in part, constitutive hypomethylation of SCLC patients and explain why these individuals are susceptible to developing cancer.

Among the six hypermethylated CpG sites, two are from the gene RUNX3. The gene, a tumor suppressor, is inactivated in a variety of cancers including SCLC 31-34. The promoter hypermethylation of the gene is believed to be a major mechanism for the gene inactivation in tumor tissues. One of the two CpG sites (RUNX3_P247_F) is located within a CpG island (promoter region) and shows an 8.72% difference between SCLC cases and controls. Although not tested, we predict a lower activity of the RUNX3 in peripheral blood leukocytes of SCLC patients. If reduced activity of the gene is also seen in lung tissue, it may be responsible for the increased risk of SCLC.

The current study, however, has some limitations. First, we were unable to determine what mechanisms caused the differential methylation between the cases and controls. The possible mechanisms include differences in genetics, treatment, nutritional status, and subpopulation of leukocytes. A comprehensive analysis that addresses these potential mechanisms will be necessary for future study. Second, we tested only one to two CpG sites per gene that are predefined by the manufacturer for inclusion on the methylation array that we used (Illumina Inc.). The tested CpGs are not necessarily most representative for a particular gene. More detailed analysis is needed to fully cover all CpGs in related genes. Nevertheless, our results demonstrate methylation differences between SCLC patients and controls are present and can be reliably detected in peripheral blood leukocyte DNA. SCLC is traditionally not treated with surgical resection; thus, it is difficult to obtain adequate tissue samples for molecular analysis. DNA from peripheral blood leukocytes has been widely used in genetic analysis. The successful use of the easily accessible specimen in this study will significantly expand the research application from genetics (such as genome-wide association studies) to epigenetics (such as epigenome-wide association studies). We believe that our methylation panel is potentially useful as a second-tier disease prediction and a non-invasive detection tool among high-risk individuals, particularly smokers with equivocal findings from CT screening.

Acknowledgments

We would like to thank Susan Ernst, M.A., for her technical assistance and Dr. Zhifu Sun for his helpful suggestions.

Sources of Funding: This work was partly supported by the National Institutes of Health (NIH) research grants: R01 CA 80127; R01 CA 84354; and R03 CA 77118.

Footnotes

Conflicts of Interest: There are no conflicts of interest by any of the authors. The corresponding author, Ping Yang, certifies that all authors have agreed to all the content in the manuscript, including the data as presented. Additionally, the corresponding author had full access to all the data in the study and final responsibility for the decision to submit for publication.

References

1. Stupp R, Monnerate C, Turrisi AT. small cell lung cancer: State of the art and future perspectives. Lung Cancer. 2004;45:105–117. [PubMed]
2. Brownson RC, Chang JC, Davis JR. Gender and Histologic Type Variations in Smoking-Related Risk of Lung Cancer. Epidemiology. 1992;3(1):61–64. [PubMed]
3. de Andrade M, Ebbert JO, Wampfler JA, et al. Environmental Tobacco Smoke Exposure in Women with Lung Cancer. Lung Cancer. 2004;43:127–134. [PubMed]
4. Jiang YH, Bressler J, Beaudet AL. Epigenetics and human disease. Annual Reviews of Genomics and Human Genetics. 2004;5:479–510. [PubMed]
5. Bjornsson HT, Fallin MD, Feinberg AP. An integrated epigenetic and genetic approach to commom human disease. Trends Genet. 2004;20:350–358. [PubMed]
6. Feinberg AP. Epigenetics at the epicenter of modern medicine. J Natl Cancer Inst. 2008;299:1345–1350. [PubMed]
7. Dolinoy DC, Jirtle RL. Environmental epigenomics in human health and disease. Environ Mol Mutagen. 2008;49:4–8. [PubMed]
8. Reik W. Stability and flexibility of epigenetic gene regulation in mammalian development. Nature. 2007;447:425–432. [PubMed]
9. Hsiung DT, Marsit CJ, Houseman EA, et al. Global DNA methylation level in whole blood as a biomarker in head and neck squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 2007;16:108–114. [PubMed]
10. Moore LE, Pfeiffer RM, Poscablo C, et al. Genomic DNA hypomethylation as a biomarker for bladder cancer susceptibility in the Spanish Bladder Cancer Study: A case-control study. Lancet Oncology. 2008;9:359–366. [PMC free article] [PubMed]
11. Widschwendter M, Apostolidou S, Raum E, et al. Epigenotyping in peripheral blood cell DNA and breast cancer risk: a proof of principle study. PLoS ONE. 2008;3:e2656. [PMC free article] [PubMed]
12. Yang P, Sun Z, Krowka MJ, et al. Alpha-1 Antitrypsin Deficiency Carriers, Tobacco Smoke, Chronic Obstructive Pulmonary Disease, and Lung Cancer Risk. Archives of Internal Medicine. 2008;168:1097–1103. [PMC free article] [PubMed]
13. Yang P, Allen MS, Aubry MC, et al. Clinical Features of 5,628 Primary Lung Cancer Patients: Experience at Mayo Clinic from 1997-2003. Chest. 2005;128:452–462. [PubMed]
14. Yang P, Wentzlaff KA, Katzmann JA, et al. Alpha1-antitrypsin deficiency allele carriers in lung cancer patients. Cancer Epidemiol Biomarkers Prev. 1999;8:461–465. [PubMed]
15. Storey JD, Tibshirani R. Statistical significance for genomewide studies. Proc Natl Acad Sci U S A. 2003;100:9440–9445. [PMC free article] [PubMed]
16. Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: A fundamental evaluation tool in clinical medicine. Clin Chem. 1993;39:561–577. [PubMed]
17. Tost J, Gut IG. Analysis of gene-specific DNA methylation patterns by pyrosequencing technology. Methods Mol Biol. 2007;373:89–102. [PubMed]
18. Tost J, Gut IG. DNA methylation analysis by pyrosequencing. Nat Protoc. 2007;2:2265–2275. [PubMed]
19. Zeni E, Mazzetti L, Miotto D, et al. Macrophage expression of interleukin-10 is a prognostic factor in nonsmall cell lung cancer. Eur Respir J. 2007;4:627–632. [PubMed]
20. Wang H, Zhang Z, Li R, et al. Overexpression of S100A2 protein as a prognostic marker for patients with stage I non small cell lung cancer. Int J Cancer. 2008;116:285–290. [PubMed]
21. Bartling B, Rehbein G, Schmitt WD, et al. S100A2-S100P expression profile and diagnosis of non-small cell lung carcinoma: Impairment by advanced tumour stages and neoadjuvant chemotherapy. Eur J Cancer. 2007;43:1935–1943. [PubMed]
22. Ota S, Ishii G, Goto K, et al. Immunohistochemical expression of BCRP and ERCC1 in biopsy specimen predicts survival in advanced non-small-cell lung cancer treated with cisplatin-based chemotherapy. Lung Cancer. 2009;64:98–104. [PubMed]
23. Okuda K, Sasaki H, Dumontet C, et al. Expression of excision repair cross-complementation group 1 and class III beta-tubulin predict survival after chemotherapy for completely resected non-small cell lung cancer. Lung Cancer. 2008;62:105–112. [PubMed]
24. Cui H, Cruz-Correa M, Giardiello M, et al. Loss of IGF2 imprinting: A potential marker of colorectal cancer risk. Science. 2003;299:1753–1755. [PubMed]
25. Kohda M, Hoshiya H, Katoh M, et al. Frequent loss of imprinting of IGF2 and MEST in lung adenocarcinoma. Mol Carcinog. 2001;31:184–191. [PubMed]
26. Kondo H, Suzuki H, Ueda R, et al. Frequent loss of imprinting of the h19 gene is often associated with its overexpression in human lung cancers. Oncogene. 1995;10:1193–1198. [PubMed]
27. Scott RH, Douglass J, Baskcomb L, et al. Constitutional 11p15 abnormalities, including heritable imprinting center mutations, cause nonsyndromic Wilms tumor. Nat Genet. 2008;40:1329–1334. [PubMed]
28. Chalitchagorn K, Shuangshoti S, Hourpai N, et al. Distinctive pattern of LINE-1 methylation level in normal tissues and the association with carcinogenesis. Oncogene. 2004;23:8841–8846. [PubMed]
29. Daskalos A, Nikolaidis G, Xinarianos G, et al. Hypomethylation of retrotransposable elements correlates with genomic instability in non-small cell lung cancer. Int J Cancer. 2009;124:81–87. [PubMed]
30. Ogino S, Nosho K, Kirkner GJ, et al. A cohort study of tumoral LINE-1 hypomethylation and prognosis in colon cancer. J Natl Cancer Inst. 2008;100:1734–1738. [PMC free article] [PubMed]
31. Suzuki M, Shigematsu H, Shames DS, et al. DNA methylation-associated inactivation of TGFbeta-related genes DRM/Gremlin, RUNX3, and HPP1 in human cancers. Br J Cancer. 2005;93:1029–1037. [PMC free article] [PubMed]
32. Sato K, Tomizawa Y, Iijima H, et al. Epigenetic inactivation of the RUNX3 gene in lung cancer. Oncol Rep. 2006;15:129–135. [PubMed]
33. Li QL, Kim HR, Kim WJ, et al. Transcriptional silencing of the RUNX3 gene by CpG hypermethylation is associated with lung cancer. Biochem Biophys Res Commun. 2004;314:223–228. [PubMed]
34. Yanada M, Yaoi T, Shimada J, et al. Frequent hemizygous deletion at 1p36 and hypermethylation downregulate RUNX3 expression in human lung cancer cell lines. Oncol Rep. 2005;14:817–822. [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...