Abstract
The etiology of meningioma, the second-most common type of adult brain tumor in the United States, is largely unknown. Prior studies indicate that history of immune-related conditions may affect the risk of meningioma. To identify genetic markers for meningioma in genes involved with innate immunity, we conducted an exploratory association study of 101 meningioma cases and 330 frequency-matched controls of European ancestry using subjects from a hospital-based study conducted by the National Cancer Institute. We genotyped 1407 “tag” single nucleotide polymorphisms (SNPs) in 148 genetic regions chosen on the basis of an r2> 0.8 and minor allele frequency > 5% in Caucasians in HapMap1. Risk of meningioma was estimated by odds ratios and 95% confidence intervals. Seventeen SNPs distributed across twelve genetic regions (NFKB1 (3), FCER1G (3), CCR6 (2), VCAM1, CD14, TNFRSF18, RAC2, XDH, C1D, TLR1/TLR10/TLR6, NOS1, DEFA5) were associated with risk of meningioma with p<0.01. Although individual SNP tests were not significant after controlling for multiple comparisons, gene region-based tests were statistically significant (p<0.05) for TNFRSF18, NFKB1, FCER1G, CD14, C1D, CCR6, and VCAM1. Our results indicate that common genetic polymorphisms in innate immunity genes may be associated with risk of meningioma. Given the small sample size, replication of these results in a larger study of meningioma is needed.
Key words for Indexing: Meningioma, polymorphism, genetic region, innate immunity, brain, tumor, neoplasm, case-control
INTRODUCTION
Meningioma is the second most common type of brain/central nervous system (CNS) tumor in the United States, comprising approximately 30% of all brain/CNS tumors (1). Despite their largely benign histology, these tumors can cause serious morbidity by virtue of their intracranial location. Other than ionizing radiation and certain rare predisposing genetic syndromes, very little is known about the etiology of these tumors. Evidence from prior epidemiologic studies, although inconsistent, suggests a possible inverse association between risk of meningioma and personal history of allergic disease (2–3), raising the possibility that alterations in the immune system may contribute to the etiology of these tumors.
The innate immune system is phylogenetically ancient, and works closely with the adaptive immune system in an integrated process to ensure effective responses to a wide range of antigenic challenges, including tumors (4). Genetic polymorphisms as well as functional alterations in the innate immune system have been implicated in the pathogenesis of several intracranial conditions including glioma, meningioma, and neurodegenerative diseases (5–10).
Given the epidemiological and biological evidence suggesting a link between innate immune system alterations and CNS disorders, we evaluated common genetic germline variants in innate immune genes to estimate the effect of low penetrance alleles on risk of meningioma. Using data from non-Hispanic whites in a hospital-based case-control study conducted by the National Cancer Institute (NCI) between 1994 and 1998, we evaluated risk of meningioma (n=101) compared to non-cancer controls (n=330) with respect to 1407 tag SNPs in 148 innate immune genes and their surrounding regions.
MATERIALS AND METHODS
Study Population and Setting
A detailed description of study methods can be found elsewhere (11). Briefly, eligible patients were 18 years or older with a first intracranial meningioma (ICD-O-2 codes 9530–9538) diagnosed during 1994–98 at one of three hospitals specializing in brain tumor treatment (in Boston, Phoenix, and Pittsburgh) within the eight weeks preceding hospitalization. 197 patients with histologically confirmed meningioma agreed to participate (94% of all contacted).
Controls were chosen from individuals admitted to the same hospitals for injuries (25%), circulatory system disorders (22%), musculoskeletal disorders (22%), digestive disorders (12%), or a variety of other non-neoplastic conditions, and were frequency-matched in a one-to-one ratio to a total case series (glioma, meningioma and acoustic neuroma) based on age (18–29, 30–39, 40–49, 50–59, 60–69, 70–79, 80–99 years); race/ethnicity (non-Hispanic white, Hispanic, African-American, other), sex, hospital, and residential proximity to the hospital. 799 control patients (86% of all contacted) were enrolled. This analysis was restricted to individuals of European ancestry (89% of all study participants) who provided blood samples, and for whom an adequate amount of DNA was extracted from blood samples. The study protocol was approved by the Institutional Review Board of each participating institution, and written informed consent was obtained from each patient or proxy.
Laboratory methods
DNA extraction and genotyping
DNA was extracted from blood using a phenol–chloroform protocol. Samples were genotyped for 101 patients with meningioma and 330 controls of European ancestry. Genotyping was performed at the NCI Core Genotyping Facility (CGF: Advanced Technology Corporation, Gaithersburg, MD) using an Illumina GoldenGate OPA panel designed to tag 148 candidate innate immunity genes and their surrounding regions (20kb 5′ of the start of the first exon and 10kb 3′ of the end of the last exon of each candidate gene). The innate immunity panel was composed of genes selected from known innate immune pathways (oxidative response, pattern recognition molecules and antimicrobials, integrins and adhesion molecules, complement, chemokines with their receptors and signaling molecules, and response genes and tissue factors). Tag SNPs were chosen from the SNPs that were genotyped as part of the International HapMap (12) using TagZilla (13) algorithm with the following parameters: minor allele frequency (MAF) > 5% in HapMap Caucasian (CEU) samples; r2 > 0.8; and greater weighting for SNPs with a design score of 1.1 (SNPs with a design score of less than 0.4 were designated as “obligate excludes”).
Quality control (QC) specimens included replicate samples from 3 non-study participants and blinded duplicate samples from 21 participants interspersed among cases and controls. 77 of 1536 SNPs originally chosen failed in assay manufacture or provided only monoallelic calls. For the study analyses, any SNPs that did not satisfy Hardy-Weinberg Equilibrium (HWE) at p < 0.001 (n=10) were excluded. Additional genotype assays were excluded for low completion rate < 90% (n=26), or poor concordance rates < 95% (n=16). Percent agreement between the three non-study replicates for the remaining 1407 SNPs was 100% for all SNPs. Concordance for study duplicates ranged from 95% to 100% (mean 98.8%), and the genotyping success rate ranged from 92% to 100% (mean 99.8%). Principal component analysis of these data revealed that population stratification was negligible in this dataset. A list of genes, chromosomal location and position for the SNPs included in the analysis is available online (5).
Statistical analysis
Unconditional multivariate logistic regression models were used to estimate odds ratios (ORs) and calculate 95% confidence intervals (CI) for the main effects of each individual SNP, using the homozygous wild-type genotype as the referent category. ORs were estimated separately for heterozygotes and rare homozygous allele groups, adjusting for study matching factors (age, sex, hospital, and residential distance from hospital). A likelihood ratio test of linear trend was conducted for each SNP using a three-level ordinal variable corresponding to the number of minor alleles for that SNP.
We used the rank truncated product (RTP) to adjust SNP and gene region-based findings for multiple testing, while accounting for correlations among SNPs induced by linkage disequilibrium (14). The RTP, which is based on the product of the most significant p-values within a gene or pathway, is an appropriate test in scenarios with a small set of true effects among a large number of null effects. Permutation based p-values for the RTP statistics were computed based on 20,000 permutations of case-control status under the null hypotheses of no association with genotype.
RESULTS
101 cases of meningioma and 330 controls were successfully genotyped for 1,407 SNPs in 148 genetic regions. Demographic characteristics did not differ appreciably between all participants and those that were genotyped, with genotyped cases being slightly more likely to be female and younger (Table 1). The observed distribution of p-values of trend for all 1,407 SNPs did not differ significantly from the expected (uniform) null distribution, making the possibility of systematic bias in the study unlikely.
Table 1.
Descriptive characteristics of non-Hispanic white participants with and without genotyping: National Cancer Institute Adult Brain Tumor Study, 1994–1998
| Characteristics | Cases (n = 197)
|
Controls (n = 799)
|
||
|---|---|---|---|---|
| All* (n = 163) | Genotyped (n = 101) | All* (n = 715) | Genotyped (n = 330) | |
| Male, % | 22.1 | 20.8 | 46.7 | 46.4 |
| Mean age, years | 56.1 | 54.3 | 50.4 | 49.4 |
| Education, % | ||||
| Less than high school | 8.0 | 6.9 | 10.8 | 10.9 |
| High school of general equivalency diploma or 3 y of college | 66.3 | 68.3 | 61.0 | 59.1 |
| Complete college or GRA or professional school | 25.2 | 23.8 | 25.9 | 28.5 |
| Unknown | 0.6 | 1.0 | 2.4 | 1.5 |
Limited to individuals of non-Hispanic Caucasian background.
Sixty-eight SNPs in 36 genetic regions were significantly associated with risk of meningioma at p<0.05 and seventeen SNPs in 12 genetic regions were associated at p<0.01 (Table 2). After correcting for multiple testing, none of the SNPs from the single SNP analysis remained statistically significant at p<0.05. However, gene-region based tests identified seven regions as statistically significant: TNFRSF18 (p = 0.003), NFKB1 (p = 0.008), FCER1G (p = 0.009), CD14 (p=0.01), C1D (p=0.03), CCR6 (p=0.03), and VCAM1 (p=0.03).
Table 2.
Tag-SNPs associated with risk of meningioma at p-trend<0.01 in hospital-based case-control study of meningioma, National Cancer Institute Adult Brain Tumor Study, 1994–1998
| Region Gene | SNP ID | Genotype | Control n (%) | Case n (%) | Odds Ratio (OR) | (95% CI) |
|---|---|---|---|---|---|---|
| NFKB1 | ||||||
| NFKB1 | rs230540 | TT | 126 (38.3) | 49 (48.5) | 1.00 | |
| CT | 162 (49.2) | 48 (47.5) | 0.67 | (0.41 – 1.11) | ||
| CC | 41 (12.5) | 4 (4.0) | 0.16 | (0.05 – 0.51) | ||
| Ptrend | 0.001 | |||||
| rs3755867 | AA | 140 (42.4) | 52 (51.5) | 1.00 | ||
| AG | 156 (47.3) | 47 (46.5) | 0.74 | (0.45 – 1.22) | ||
| GG | 34 (10.3) | 2 (2.0) | 0.11 | (0.02 – 0.48) | ||
| Ptrend | 0.002 | |||||
| rs1585213 | CC | 117 (35.6) | 42 (41.6) | 1.00 | ||
| CT | 164 (49.9) | 55 (54.5) | 0.86 | (0.52 – 1.42) | ||
| TT | 48 (14.6) | 4 (4.0) | 0.16 | (0.05 – 0.50) | ||
| Ptrend | 0.004 | |||||
| VCAM1 | ||||||
| VCAM1 | rs2209627 | AA | 222 (67.7) | 54 (53.5) | 1.00 | |
| AG | 92 (28.1) | 41 (40.6) | 2.60 | (1.53 – 4.42) | ||
| GG | 14 (4.3) | 6 (5.9) | 2.13 | (0.66 – 6.80) | ||
| Ptrend | 0.001 | |||||
| FCER1G | ||||||
| NDUFS2 | rs4656993 | GG | 116 (35.3) | 21 (20.8) | 1.00 | |
| AG | 155 (47.1) | 50 (49.5) | 1.98 | (1.08 – 3.61) | ||
| AA | 58 (17.6) | 30 (29.7) | 2.92 | (1.45 – 5.87) | ||
| Ptrend | 0.002 | |||||
| FCER1G | rs12094497 | GG | 270 (82.1) | 93 (92.1) | 1.00 | |
| AG | 52 (15.8) | 8 (7.9) | 0.34 | (0.17 – 0.87) | ||
| AA | 7 (2.1) | 0 (0.0) | 0.00 | 0 | | ||
| Ptrend | 0.004 | |||||
| rs11587213 | AA | 224 (67.9) | 80 (80.0) | 1.00 | ||
| AG | 93 (28.2) | 19 (19.0) | 0.53 | (0.29 – 0.96) | ||
| GG | 13 (3.9) | 1 (1.0) | 0.20 | (0.02 – 1.72) | ||
| Ptrend | 0.008 | |||||
| CD14 | ||||||
| PRO1580 | rs3822356 | AA | 208 (63.4) | 49 (48.5) | 1.00 | |
| 1.09 | ||||||
| AG | (33.2) | 47 (46.5) | 2.24 | (1.34 – 3.76) | ||
| GG | 11 (3.4) | 5 (5.0) | 2.15 | (0.65 – 7.14) | ||
| Ptrend | 0.003 | |||||
| TNFRSF18 | ||||||
| TNFRSF18 | rs9729550 | AA | 180 (54.9) | 68 (67.3) | 1.00 | |
| AC | 121 (36.9) | 29 (28.7) | 0.56 | (0.33 – 0.95) | ||
| CC | 27 (8.2) | 4 (4.0) | 0.28 | (0.09 – 0.89) | ||
| Ptrend | 0.003 | |||||
| RAC2 | ||||||
| RAC2 | rs2213430 | CC | 103 (31.2) | 38 (37.6) | 1.00 | |
| CT | 165 (50.0) | 49 (48.5) | 0.56 | (0.32 – 0.96) | ||
| TT | 62 (18.8) | 14 (13.9) | 0.35 | (0.16 – 0.75) | ||
| Ptrend | 0.004 | |||||
| XDH | ||||||
| XDH | rs207444 | GG | 291 (88.5) | 97 (96.0) | 1.00 | |
| AG | 37 (11.3) | 4 (4.0) | 0.26 | (0.09 – 0.79) | ||
| AA | 1 (0.3) | 0 (0.0) | 0.00 | 0 | | ||
| Ptrend | 0.005 | |||||
| C1D | ||||||
| C1D | rs10203061 | AA | 235 (71.7) | 56 (56.0) | 1.00 | |
| AG | 85 (25.9) | 40 (40.0) | 1.98 | (1.18 – 3.33) | ||
| GG | 8 (2.4) | 4 (4.0) | 2.81 | (0.70 – 11.2) | ||
| Ptrend | 0.006 | |||||
| CCR6 | ||||||
| CCR6 | rs9459883 | GG | 265 (80.3) | 91 (90.1) | 1.00 | |
| CG | 63 (19.1) | 10 (9.9) | 0.42 | (0.20 – 0.89) | ||
| CC | 2 (0.6) | 0 (0.0) | 0.00 | 0 | | ||
| Ptrend | 0.008 | |||||
| rs3798315 | CC | 251 (76.1) | 87 (86.1) | 1.00 | ||
| CT | 72 (21.8) | 13 (12.9) | 0.43 | (0.21 – 0.87) | ||
| TT | 7 (2.1) | 1 (1.0) | 0.31 | (0.03 – 2.81) | ||
| Ptrend | 0.008 | |||||
| TLR1/TLR10/TLR6 | ||||||
| TLR10 | rs11466657 | AA | 306 (93.9) | 97 (98.0) | 1.00 | |
| AG | 20 (6.1) | 2(2.0) | 0.19 | (0.04 – 0.84) | ||
| GG | -- | -- | ||||
| Ptrend | 0.008 | |||||
| NOS1 | ||||||
| NOS1 | rs10850803 | AA | 273 (83.2) | 73 (72.3) | 1.00 | |
| AG | 54 (16.5) | 24 (23.8) | 1.70 | (0.92 – 3.13) | ||
| GG | 1 (0.3) | 4 (4.0) | 13.50 | (1.16 – 157) | ||
| Ptrend | 0.009 | |||||
| DEFA5 | ||||||
| DEFA5 | rs10503360 | TT | 95 (28.8) | 13 (12.9) | 1.00 | |
| GT | 160 (48.5) | 56 (55.5) | 2.10 | (1.05 – 4.19) | ||
| GG | 75 (22.7) | 32 (31.7) | 2.74 | (1.28 – 5.83) | ||
| Ptrend | 0.01 | |||||
Odds ratios adjusted for sex, age, study hospital, and distance of residence from hospital.
DISCUSSION
In our exploratory study of risk of adult meningioma with common tagging SNPs in 148 innate immune genes and their surrounding regions, we identified seven genetic regions of particular interest within four innate immune pathways: TNFSRF18, FCER1G and VCAM1 (integrins/cell surface receptors), NFKB1 and CD14 (pattern recognition and antimicrobials), C1D (complement) and CCR6 (chemokines).
The gene regions for NFKB1 and FCER1G were particularly intriguing given the relationship between T-cell regulation and chronic inflammation (NFKB1), and IgE and allergic reactions (FCER1G). Three statistically significant SNPs were observed in each of these regions at the p<0.01 level. NFKB1 encodes the subunit p50/p105 for a pleiotropic transcription regulator activated by a variety of intra- and extra-cellular stimuli. The transcription inhibitor is part of the DNA-binding subunit of the NFKB protein complex involved with T cell regulation and chronic inflammation (15). Epidemiologic studies have observed associations between NFKB1 polymorphisms and risk of some cancers, including glioma, non-Hodgkin lymphoma and Hodgkin’s lymphoma (5, 16–17), and inflammatory diseases (18), but not for other cancers such as breast, colorectal, or renal cell cancers (19–20). The NFKB pathway has also been associated with inflammatory conditions in the brain, such as Alzheimer’s disease (7–8). Less is known about FCER1G and its relevance to the etiology of meningiomas. The FCER1G is a subunit of the high-affinity IgE receptor mediating allergic reactions. Functional polymorphisms in FCER1A, which also codes for a subunit of the IgE receptor, were strongly associated with serum IgE levels in a genome-wide association study (21).
Several additional SNPs of interest lay within gene regions coding for integrin/cell surface receptors, including TNFSRF18, FCER1G and VCAM1. This pathway is of particular interest given that changes in integrin pattern expression have been observed in a variety of meningiomas (9–10) and may influence their invasive biological behavior.
The few existing studies of common genetic variation and risk of meningioma have reported statistically significant associations in genes in the DNA repair, apoptosis/cell cycle, IGF, folate metabolism, p53, and oxidative response pathways (22). To our knowledge, this is the first study of meningioma to explore a large number of genetic variants in the innate immunity pathway. Comparing our results for innate immunity genes and risk of meningioma with a previous examination of the same SNPs for glioma risk (5), we found that among the top ten hits, NFKB1 was significant in both studies (p<0.01). However, gene region-based hits (p<0.05) differed between glioma (ALOX5, SELP and SOD) and meningioma (TNFSRF18, FCER1G, VCAM1, NFKB1, CD14, C1D, and CCR6).
Although our results provide interesting clues, they are subject to some caveats. While the hospital-based design of this study allowed accrual of incident meningiomas, and we imposed strict quality control criteria for blood collection and genotyping of DNA samples, the sample size for this study remains small. The SNPs in this study were chosen as tagging markers for the genetic region and not based on known function, thus the observed associations could be due to linkage disequilibrium with the true unobserved causal SNPs. Replication of these findings with increased coverage of the identified genes of interest and larger sample size (for example, in multicenter studies of meningioma) is required to rule out the possibility of chance findings.
Acknowledgments
We thank Henry Chen, Michael Stagner, Bob Wheeler, Jane Wang, and Leslie Carroll of Information Management Systems for help with statistical programming and biospecimen coordination. This research was supported by intramural funds from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services, and has been funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under contract N01-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
Abbreviations
- NCI
National Cancer Institute
- OR
Odds Ratio
- 95% CI
95 % Confidence Interval
- SNP
Single Nucleotide Polymorphism
- AA
Amino Acid
- nt
Nucleotide
References
- 1.Central Brain Tumor Registry of the United States (CBTRUS) 2008. Statistical Report: Primary Brain Tumors in the United States, 2000–2004. [Google Scholar]
- 2.Schoemaker MJ, Swerdlow AJ, Hepworth SJ, van Tongeren M, Muir KR, McKinney PA. History of allergic disease and risk of meningioma. Am J Epidemiol. 2007;165:477–85. doi: 10.1093/aje/kwk048. [DOI] [PubMed] [Google Scholar]
- 3.Linos E, Raine T, Alonso A, Michaud D. Atopy and risk of brain tumors: a meta-analysis. J Natl Cancer Inst. 2007;99:1544–50. doi: 10.1093/jnci/djm170. [DOI] [PubMed] [Google Scholar]
- 4.Litman GW, Cannon JP, Dishaw LJ. Reconstructing immune phylogeny: new perspectives. Nat Rev Immunol. 2005;5:866–79. doi: 10.1038/nri1712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Rajaraman P, Brenner AV, Butler MA, et al. Common variation in genes related to innate immunity and risk of adult glioma. Cancer Epidemiol Biomarkers Prev. 2009;18:1651–8. doi: 10.1158/1055-9965.EPI-08-1041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mrass P, Weninger W. Immune cell migration as a means to control immune privilege: lessons from the CNS and tumors. Immunol Rev. 2006;213:195–212. doi: 10.1111/j.1600-065X.2006.00433.x. [DOI] [PubMed] [Google Scholar]
- 7.Fontalba A, Gutierrez O, Llorca J, et al. Deficiency of CARD8 is associated with increased Alzheimer’s disease risk in women. Dement Geriatr Cogn Disord. 2008;26:247–50. doi: 10.1159/000160956. [DOI] [PubMed] [Google Scholar]
- 8.Moynagh PN. The interleukin-1 signalling pathway in astrocytes: a key contributor to inflammation in the brain. J Anat. 2005;207:265–9. doi: 10.1111/j.1469-7580.2005.00445.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Beschet I, Brunon J, Scoazec JY, Mosnier JF. Expression of beta1 and beta4 integrins in normal arachnoid membrane and meningiomas. Cancer. 1999;86:2649–58. [PubMed] [Google Scholar]
- 10.Bello L, Zhang J, Nikas DC, et al. Alpha(v)beta3 and alpha(v)beta5 integrin expression in meningiomas. Neurosurgery. 2000;47:1185–95. doi: 10.1097/00006123-200011000-00035. [DOI] [PubMed] [Google Scholar]
- 11.Inskip PD, Tarone RE, Hatch EE, et al. Cellular-telephone use and brain tumors. N Engl J Med. 2001;344:79–86. doi: 10.1056/NEJM200101113440201. [DOI] [PubMed] [Google Scholar]
- 12.The International HapMap Project. Nature. 2003;426:789–96. doi: 10.1038/nature02168. [DOI] [PubMed] [Google Scholar]
- 13.http://tagzilla.nci.nih.gov/. [cited 2008 February 15]; Available from:
- 14.Dudbridge F, Koeleman BP. Rank truncated product of P-values, with application to genomewide association scans. Genet Epidemiol. 2003;25:360–6. doi: 10.1002/gepi.10264. [DOI] [PubMed] [Google Scholar]
- 15.Chang M, Lee AJ, Fitzpatrick L, Zhang M, Sun SC. NF-kappa B1 p105 regulates T cell homeostasis and prevents chronic inflammation. J Immunol. 2009;182:3131–8. doi: 10.4049/jimmunol.0803637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chang ET, Birmann BM, Kasperzyk JL, et al. Polymorphic variation in NFKB1 and other aspirin-related genes and risk of Hodgkin lymphoma. Cancer Epidemiol Biomarkers Prev. 2009;18:976–86. doi: 10.1158/1055-9965.EPI-08-1130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wang SS, Purdue MP, Cerhan JR, et al. Common gene variants in the tumor necrosis factor (TNF) and TNF receptor superfamilies and NF-kB transcription factors and non-Hodgkin lymphoma risk. PLoS ONE. 2009;4:e5360. doi: 10.1371/journal.pone.0005360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sun XF, Zhang H. NFKB and NFKBI polymorphisms in relation to susceptibility of tumour and other diseases. Histol Histopathol. 2007;22:1387–98. doi: 10.14670/HH-22.1387. [DOI] [PubMed] [Google Scholar]
- 19.Curran JE, Weinstein SR, Griffiths LR. Polymorphic variants of NFKB1 and its inhibitory protein NFKBIA, and their involvement in sporadic breast cancer. Cancer Lett. 2002;188:103–7. doi: 10.1016/s0304-3835(02)00460-3. [DOI] [PubMed] [Google Scholar]
- 20.Riemann K, Becker L, Struwe H, et al. No association of the NFKB1 insertion/deletion promoter polymorphism with survival in colorectal and renal cell carcinoma as well as disease progression in B-cell chronic lymphocytic leukemia. Pharmacogenet Genomics. 2006;16:783–8. doi: 10.1097/01.fpc.0000230414.74726.f6. [DOI] [PubMed] [Google Scholar]
- 21.Weidinger S, Gieger C, Rodriguez E, et al. Genome-wide scan on total serum IgE levels identifies FCER1A as novel susceptibility locus. PLoS Genet. 2008;4:e1000166. doi: 10.1371/journal.pgen.1000166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gu J, Liu Y, Kyritsis AP, Bondy ML. Molecular epidemiology of primary brain tumors. Neurotherapeutics. 2009;6:427–35. doi: 10.1016/j.nurt.2009.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]