Review

Continuing Medical EducationNature Clinical Practice Neurology (2006) 2, 494-503
doi:10.1038/ncpneuro0289  
Received 23 January 2006 | Accepted 7 July 2006

Epidemiology and molecular pathology of glioma

Judith A Schwartzbaum*, James L Fisher, Kenneth D Aldape and Margaret Wrensch  About the authors

Correspondence *Division of Epidemiology and Biometrics, School of Public Health, The Ohio State University, Columbus, OH 432–10, USA

Email jschwartzbaum@sph.osu.edu

Summary

Gliomas account for almost 80% of primary malignant brain tumors, and they result in more years of life lost than do any other tumors. Glioblastoma, the most common type of glioma, is associated with very poor survival, so glioma epidemiology has focused on identifying factors that can be modified to prevent this disease. Only two relatively rare factors have so far been conclusively shown to affect glioma risk—exposure to high doses of ionizing radiation, and inherited mutations of highly penetrant genes associated with rare syndromes. In addition, preliminary evidence points to a lower glioma risk among people with allergic conditions and high levels of serum IgE. Recent research has focused on identifying germline polymorphisms associated with risk of glioma, and using molecular markers to classify glial tumors into more-homogenous groups. Because gene products probably interact with environmental factors or developmental signals to produce gliomas, large studies are needed to analyze associations between polymorphisms and glioma. Cohort studies of immune factors and glioma risk are being undertaken to validate the results of case–control studies. Studies of polymorphisms of genetic pathways with strong prior hypotheses are being planned, and whole-genome scans are being proposed to study high-risk families and case–control series. The Brain Tumor Epidemiology Consortium has been formed to co-ordinate these studies.

Review criteria

PubMed was searched using Entrez for articles published up to 1 January 2006, including electronic early-release publications. Search terms included "glioma case–control", "glioma genes", "glioma polymorphisms", "glioma biomarker", "glioma survival", "glioma epidemiology" and "glioma pathogenesis", as well as these terms preceded by "brain tumor". We screened for pertinent articles by reading the resulting abstracts. The resulting articles were obtained and references were checked for additional material when appropriate.

Keywords:

brain tumors, epidemiology, glioma, molecular pathology

Top

Introduction

The term 'glioma' encompasses all tumors that are thought to be of glial cell origin. These include astrocytic tumors (World Health Organization classification astrocytoma grades I, II [astrocytoma], III [anaplastic astrocytoma], and IV [glioblastoma or GM]), oligodendrogliomas, ependymomas, and mixed gliomas.1, 2 In this article, we will use the terminology for histologic subtypes shown in the Central Brain Tumor Registry of the United States (CBTRUS) annual report (Statistical report: Primary Brain Tumors in the United States, 1998–2002);1 this report also provides a detailed table of International Classification of Diseases for Oncology (ICD-O) codes corresponding to each subtype. Many descriptive statistics are not specific to glioma, but rather include glioma among primary malignant brain and CNS tumors or—more specifically—tumors of neuroepithelial tissue. Because gliomas account for approximately 77% of primary malignant brain tumors, however, they strongly influence these statistics, and information about glioma can be inferred from them.

Approximately 13,000 deaths and 18,000 new cases of primary malignant brain and CNS tumors occur annually in the US.1 During the years 1998 to 2002, the average annual age-adjusted incidence rate of primary malignant brain and CNS tumors among adults aged 20 years and over in the US was 9.0 per 100,000 person-years1, with rates ranging from 7.3 per 100,000 person-years in Virginia to 10.5 per 100,000 person-years in Maine and Idaho. Table 1 shows the individual rates for different glioma histologic subtypes, their median ages at diagnosis, and rates for men and women. GM, the highest grade tumor (grade IV), is associated with the highest median age at diagnosis. The different characteristic age distributions of glioma subtypes are available at the CBTRUS website.1 All gliomas are more common in men than in women, although these gender differences are quite small in the case of oligodendroglioma. Because of the differing age distributions of different glioma subtypes, and because overall rates are driven by the most common glioma type, GM, the general statements made above are not representative of the demographic characteristics of gliomas in individuals below the age of 20 years.

Table 1 Median age at diagnosis and age-adjusted average annual (1998–2002) incidence rates of adult glioma stratified by sex. Reported in CBTRUS 2005–2006 Statistical Report: Primary Brain Tumors in the United States, 1998–2002.1
Table 1 - Median age at diagnosis and age-adjusted average annual (1998|[ndash]|2002) incidence rates of adult glioma stratified by sex. Reported in CBTRUS 2005|[ndash]|2006 Statistical Report: Primary Brain Tumors in the United States, 1998|[ndash]|2002.
Full tableFigures & Tables indexDownload PowerPoint slide (165K)

Examination of brain tumor incidence data from CBTRUS for the 10-year period from 1985 to 1994 revealed a slight but statistically significant average annual percentage increase in incidence (0.9%).1 It is likely, however, that most, if not all, of this increase is attributable to improvements in diagnostic imaging (CT and MRI), increased availability of medical care and neurosurgeons, changing approaches to the medical treatment of older patients, and changes in the classifications of specific histologies of brain tumors from benign to malignant.3, 4, 5, 6, 7

There is around a fourfold difference in the incidence of primary malignant brain tumors between countries with a high incidence (e.g. Australia, Canada, Denmark, Finland, New Zealand and the US) and regions with a low incidence (e.g. Rizal in the Philippines and Mumbai in India).7, 8 Differences in diagnostic practices and completeness of reporting make all geographic comparisons difficult.7 In addition, higher incidence rates appear in countries—and perhaps in states within the US—with greater access to health care and better medical care.7, 8 Some geographic variation is not easily attributable to this phenomenon, however, as malignant brain tumors are twice as common in Northern Europe as they are in Japan. Therefore, environmental factors might account for some of the observed geographic differences.

The prognosis for patients with glioma is often very poor (only approx2% of patients aged 65 years or older, and only 30% of those under the age of 45 years at GM diagnosis, survive for 2 years or more),1 and treatments to cure GM have yet to be devised. Consequently, our present strategy is to identify genetic, behavioral, environmental and developmental contributors to glioma risk through epidemiological studies, with the ultimate goal of reducing the disease burden. Survival time for patients with a malignant brain tumor is related to age at diagnosis and histologic type of their tumor. Two-year relative survival probabilities according to age at diagnosis and histologic tumor type are shown in Figure 1.1 For each age-group, relative survival probability is lowest for patients with GM, and within each histologic type, survival probability is lowest for those in older age-groups.

Figure 1 Glioma 2-year relative survival probabilities by age at diagnosis and histologic subtype, based on the follow-up of individuals diagnosed between 1973 and 2002.
Figure 1 : Glioma 2-year relative survival probabilities by age at diagnosis and histologic subtype, based on the follow-up of individuals diagnosed between 1973 and 2002. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com

Source: Surveillance, Epidemiology and End Results (SEER), compiled by the Central Brain Tumor Registry of the United States (CBTRUS) and reported in tabular form in: CBTRUS (2005) Statistical Report: Primary Brain Tumors in the United States, 1998–20021 Abbreviation: NOS, not otherwise specified.

Full figure and legend (35K)Figures & Tables indexDownload PowerPoint slide (123K)

Most of the recent epidemiological research on primary malignant brain tumors has focused on glioma, and we therefore restrict the present review to this tumor. Furthermore, recent research has primarily examined genetic or molecular factors, so our discussion focuses on these topics. We refer interested readers to several current reviews of general brain tumor epidemiology and pathogenesis.9, 10, 11 The epidemiology of meningioma, the most common primary non-malignant brain tumor, has also been comprehensively reviewed by Claus and colleagues.12 The purposes of the present article are to present recent molecular epidemiological findings and suggest promising paths for future research.

Top

Environmental and genetic risk factors for glioma

The only firmly established exogenous environmental cause of glioma is exposure to therapeutic or high-dose radiation, although high-dose chemotherapy for treatment of cancers at sites other than the brain has also been linked to this condition.13, 14 Genetic factors determine the degree of risk from these exposures; Relling et al.15 showed that among children treated with cranial irradiation and intensive antimetabolite therapy for acute lymphocytic leukemia, those with germline polymorphisms leading to low or absent thiopurine methyltransferase activity were significantly more likely than those without such polymorphisms to subsequently develop brain cancer. Similar studies need to be conducted among adults.

Abundant animal and other data support the biological plausibility of neurocarcinogenicity of endogenous and exogenous chemicals (e.g. N-nitroso compounds, reactive oxygen and nitrogen species, several industrially used chemicals, and polycyclic aromatic hydrocarbons). Although people might be exposed to many of these chemicals through essential cellular metabolism, diet, occupation and personal habits, studies of glioma epidemiology in humans have generated inconsistent or null findings for these risk factors.7, 14, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 Possible explanations for the failure to find consistent and statistically significant findings for chemical risk factors include the following: small study sample sizes; false-positive results (related to both small sample sizes and lack of precise research hypotheses);26 invalid or imprecise exposure measures (resulting from use of proxy respondents when brain tumor patients are unavailable, or from patient or control errors in exposure history recall); inherited or developmental variation in metabolic and repair pathways; unaccounted-for protective exposures or conditions (such as allergies); differential diffusion of chemicals across the blood–brain barrier; differentially expressed metabolic and repair pathways in the brain; and disease heterogeneity.14, 27, 28, 29 Progress has been made in understanding environmental contributors to other cancers by joint consideration of inherited variation in detoxification, metabolism or repair, and histological or molecular tumor subtypes. A classic example includes a demonstration that specific tumor protein p53 (TP53) mutations in hepatocellular carcinoma occur exclusively in carriers of specific variants in the genes that encode epoxide hydrolase and glutathione S-transferase M1.30 Nonetheless, it is worth noting that some of the biases noted above, as well as publication bias of positive results, could lead to overestimation of risks, so it is possible that the failure to find strong and consistent environmental risk factors for brain tumors (except with high-dose radiation) might be attributable to the absence of true strong environmental associations. This possibility is supported by the much lower magnitude of geographic variation than that found for cancers with well-established and strong environmental risk factors such as smoking with lung cancer or sunlight with melanoma. Given the paucity of studies that have examined environmental risk factors for well-defined glioma subtypes, however, it is premature to conclude that environmental risks do not exist.

Genetic syndromes, familial aggregation and linkage

Evidence indicating genetic susceptibility to glioma has come from studies of genetic syndromes, familial aggregation, linkage and mutagen sensitivity. Although the few genetic syndromes (caused by inherited rare mutations) associated with increased risk of brain tumors account for a small proportion of cases,9, 31 they provide an important starting point for identifying candidate genes and pathways that could be involved in gliomagenesis (Table 2). The inherited syndromes neurofibromatosis 1 and 2, tuberous sclerosis, retinoblastoma, Li–Fraumeni syndrome, and Turcot's syndrome and multiple hamartoma have been associated with increased risks for gliomas and other types of primary brain tumor. The roles of more-common variants in many of these genes (and related pathways) in sporadic glioma are unknown.9

Table 2 Inherited mutations in members of families at increased risk of glioma.
Table 2 - Inherited mutations in members of families at increased risk of glioma.
Full tableFigures & Tables indexDownload PowerPoint slide (105K)

Demonstration of familial aggregation does not prove a genetic etiology, because families share common environments, but is often among the first indicators that genetic susceptibility might play a part in the pathogenesis of a complex disease. Relative risks of glioma among family members of patients with glioma have ranged from one to ten; in large, well-conducted studies, familial glioma risks are about two times greater than glioma risks among people without a family history of the disease. This risk is similar in magnitude to familial association of breast and other cancers for which susceptibility genes have been identified.31, 32, 33 Because the baseline risk for brain tumors is substantially less than for breast cancer, however, the absolute risk of brain tumors to family members of a brain tumor patient is also considerably less than the risk of breast cancer to family members of a breast cancer patient. The pattern of brain tumor occurrence in families has been attributed to environmental causes in one study,34 and to multifactorial, polygenic or autosomal recessive inheritance in others.32, 35, 36 Paunu et al.37 recently published evidence of statistically significant linkage to 15q23–q26.3 for 15 Finnish families with multiple cases of glioma (after stringent control for multiple testing, the conservative P-value was 0.03).

Because only a small proportion of gliomas are likely to be due to the effects of high-dose radiation or inherited rare mutations, researchers have turned their attention to polymorphisms in genes that might influence susceptibility to glioma in concert with environmental exposures.

Immunologic risk factors and germline polymorphisms

Among the most consistent findings of the past decade are statistically significant inverse associations between adult glioma and histories of allergies or chicken pox, IgG antibodies to varicella-zoster virus (VZV), and high levels of serum IgE.38, 39, 40, 41, 42, 43, 44, 45, 46, 47 Schwartzbaum and colleagues41 used germline polymorphisms in genes associated with asthma and allergies as biomarkers for the presence of asthma and other allergic conditions. They argued that although self-report of allergic conditions provides a better measure of allergic conditions than do allergy-related germline polymorphisms, self-report might be influenced by the presence of glioma, whereas germline polymorphisms are not. That is, glioma patients or their proxies might be less likely to remember allergic conditions than healthy controls, or the glioma might suppress immunologic responses; however, neither recall nor immunosuppression by the tumors would influence a person's germline polymorphisms.

By comparing distributions of asthma and allergy-related polymorphisms in patients with GM and controls,41 these authors were able to address lingering doubts that the negative association between allergy history and GM might result from recall bias. They recently reported that single-nucleotide polymorphisms (SNPs; see Box 1) in genes related to asthma are also related to GM; in addition—confirming the inverse association between asthma and GM—they found that genotypes that increase asthma risk are associated with decreased GM risk.41 They examined five SNPs in three genes, interleukin 4 receptor alpha (IL-4RA), IL-13 and ADAM33; the IL-4RA SNPs T478C TC, CC and A551G AG, AA were significantly positively associated with GM—odds ratios were 1.64 (95% CI 1.05–2.55) and 1.61 (95% CI 1.05–2.47), respectively—whereas the IL-13 SNP C1112T CT, TT was inversely associated with GM (odds ratio 0.56 [95% CI 0.33–0.96]).

Box 1 Single-nucleotide polymorphisms.

 

A single-nucleotide polymorphism (SNP) is a type of genetic sequence variant involving a single base pair change that is found in at least 1% of a population. As these variants might affect the structure or function of the gene product, they might account for individual differences in disease risk in combination with other variants or exposures. SNPs can be arbitrarily named, although if the function or protein product of the gene on which they are found is known, they are sometimes named for this function or product (e.g. an SNP on the gene that codes for the interleukin 4 receptor is named IL4RA T478C). The number associated with this SNP indicates its location on the gene and the letters T and C show the two bases that can occur at that location on the gene. Sometimes, the SNP is labeled by the amino acid that is coded for by the affected codon, rather than the base pair. Other conventions have also been used in naming polymorphisms. We use the official gene symbols given by the National Center for Biotechnology Information (NCBI) obtained through Entrez.69

As germline polymorphisms were used as biomarkers for susceptibility to asthma and allergies, the results cannot be attributed to either recall bias or effects of GM on the immune system, so these findings would appear to validate the association between self-reported allergic conditions and GM. It is possible that IL-13 or its shared receptor with IL-4, IL-4RA, play independent roles in allergic conditions and GM. Alternatively, some aspect of allergic conditions themselves might reduce GM or glioma risk.

Tang and colleagues48 showed that GM is positively associated with the human leukocyte antigen (HLA) genotype B*13 and the HLA haplotype B*07-Cw*07 (P = 0.01 and P <0.001 respectively), and is inversely associated with the genotype Cw*01 (P = 0.05). Interestingly, if confirmed, these results could partially explain the increased GM incidence in white individuals, because B*07 and B*07-Cw*07 are much more common in this population than in other ethnic groups.

A variety of other viral (simian virus 40, JC virus, BK virus, other papovaviruses, adenoviruses, retroviruses, the herpes viruses cytomegalovirus and human herpesvirus 6, and influenza) and parasitic infections (Toxoplasma gondii) have been investigated in relation to gliomagenesis in experimental animals and limited epidemiological studies, but definitive results regarding possible roles of these infectious agents in human gliomas have remained elusive.7, 8

On the basis of these as yet inconclusive relationships between infections and immunologic factors and glioma, we have hypothesized that it might be the specific nature of the immune system's response to antigens, and not exposure to the antigen per se, that is responsible for the inverse associations with glioma,43, 44 because there is near-universal exposure among the general population to both influenza virus and antigens that provoke allergies (e.g. pollen, food).

Glioma and DNA repair gene polymorphisms

Inherited variation in components of DNA repair pathways represents another category of genes that have been extensively studied with respect to cancer, because of the importance of DNA repair in maintaining genomic integrity.49, 50, 51 In 2002, Goode et al.52 reviewed 30 studies reporting on DNA repair polymorphisms with respect to adult glioma and cancer of the bladder, breast, lung, skin, prostate, head and neck, stomach, and esophagus, and the number of studies published has more than quadrupled in the past 3 years. Glioma or specific glioma subtypes have been significantly associated with variants in ERCC1, ERCC2, the nearby gene GLTSCR1 (glioma tumor suppressor candidate of unknown function), PRKDC (also known as XRCC7), and MGMT,50, 52, 53, 54, 55, 56 but too few studies have been done to assess consistency.

The complexity of DNA repair is becoming increasingly apparent, with 130 known genes involved in base excision repair, direct reversal of damage, mismatch repair, nucleotide excision repair, homologous recombination, non-homologous end-joining, sanitization of nucleotide pools, activity of DNA polymerases, editing and processing of nucleases, and postreplicative repair. Genes associated with sensitivity to DNA damaging agents are also involved in DNA repair.51 Studies focusing on constellations of DNA repair variants involved in these pathways might help to elucidate their role in gliomagenesis and progression. If different pathways make necessary contributions to various stages of tumor development (e.g. mutation, angiogenesis), interactions among these pathways will have to be evaluated. For example, if germline variants that reduce brain inflammation and angiogenesis are present, a glioma might not develop, even in the presence of environmental exposure and germline variants associated with high rates of DNA repair errors. Simultaneous consideration of immune and DNA repair pathways would allow proper evaluation of both sets of polymorphisms.

Glioma and polymorphisms in cell cycle regulation

Dysregulation of the cell cycle (control of proliferation and apoptosis) is a hallmark feature of most gliomas,9 and MDM2 is a key molecule in maintaining the fidelity of this process. In one study,57 the G variant of SNP309 in the MDM2 promoter led to higher expression of MDM2 with concomitant reduced expression of TP53, and was found to be significantly associated with earlier age of tumor development and multiple tumor sites in subjects with Li–Fraumeni syndrome, of which brain tumors are one component. Because of smaller sample sizes, however, false-positive findings are even more likely to occur among subgroups (e.g. young people) than among the study sample as a whole. Such statistically significant subgroup analyses from a single study are difficult to interpret and require replication.

Top

Molecular pathology of glioma

Commonly altered chromosomal regions in glioma

Classical cytogenetic and array-based comparative genomic hybridization studies of gliomas have identified copy number changes (deletions, amplifications, gains) in several regions; deletions and loss of heterozygosity in tumors might point to genes involved in tumor suppression, whereas amplifications and gains might point to genes involved in tumor initiation or progression (e.g. oncogenes). The chromosomal alterations that are most regularly observed in glioma are summarized in Table 3.9, 58 Although several well-known tumor suppressor genes and oncogenes are located in the regions affected by these alterations, many genes in these regions have yet to be examined for their specific relationship with gliomagenesis.


Dysregulated pathways in astrocytic gliomas

Classical tumor molecular and cytogenetic studies, as well as the newer array-based assays of comparative genomic hybridization and RNA expression, indicate substantial genetic and gene-expression heterogeneity within and between histologic grades of astrocytic tumors and between different histologic types of gliomas.9, 10, 59, 60, 61, 62, 63 (M Wrensch et al., unpublished data) The degree to which the mature tumor is independent of 'causal' environmental or genetic pathways is unknown, however, and whether variability of dysregulated pathways among tumors reflects different causes has yet to be determined.

Dysregulation of genetic pathways can occur through a variety of genetic and epigenetic mechanisms, including gene mutation, amplification, deletion, methylation or demethylation, whole or partial chromosome gains or losses, and transcriptional interference.2, 9, 54, 64 There are several inter-related pathways that are well established as being dysregulated in GM and other gliomas. Recurring themes are that these pathways are involved in cell cycle signaling and control (proliferation and apoptosis), cytoskeleton regulation, transcription, growth signaling, cell adhesion, cell migration and cytokine response.9

Molecular pathways to glioblastoma

Progress has been made with respect to elucidating the genetic pathways that lead to GM. It is now believed that GMs arise by one of at least two pathways that can be defined in clinical terms: one pathway results from tumor progression from lower grade astrocytomas ('secondary' GM), whereas the second pathway does not involve a clinically evident precursor ('primary' or 'de novo' GM). Interestingly, examination of two molecular aberrations, TP53 mutation and EGFR amplification, has revealed correlations with the type of GM defined on a clinical basis.61, 65 Specifically, tumors with TP53 mutations are more likely to be secondary GM, arising from lower grade precursors, whereas de novo GM is more likely to harbor EGFR amplification. This distinction is not absolute and has recently been called into question,66 and it does not take into account the substantial proportion of GM tumors that harbor neither TP53 mutation nor EGFR amplification. It also does not explain the elevated risk of hospitalization for epilepsy five or more years before first-time high-grade glioma diagnosis that was observed in a large population-based case–control study.67 The two-pathway hypothesis does, however, raise the possibility that distinct subtypes of GM, although similar histologically, might display clinical differences, specifically in response to therapeutic agents. Molecular subtyping is currently not routine, but is likely to be of use in the future as an adjunct to histology in the classification of these tumors.

Population studies of tumor markers in astrocytic gliomas

Only two studies have so far presented genetic and molecular tumor marker data for relatively large numbers of population-based glioma cases.14, 54 Although each study assessed five or six tumor markers, only two—EGFR amplification and TP53 mutation—were measured in both studies. Interestingly, the percentage of GM tumors with EGFR amplification was identical in the two studies (36% of 371 de novo GM cases from Zürich and 36% of 386 GM cases from San Francisco Bay Area). The percentage of tumors with TP53 mutation varied from 28% of 386 GM tumors from Zürich to only 15% of 409 GM tumors from San Francisco; percentages in clinical series have ranged from 20% to 30%.68 Despite some inconsistencies, these findings support the hypothesis, from smaller clinical series, that astrocytic tumors might arise through different pathways and might reflect the action of different causal mechanisms.

Top

Conclusions and future prospects

One focus of molecular genetic studies should be the interaction between the known environmental risk factor, exposure to therapeutic doses of ionizing radiation, and DNA repair polymorphisms. If the DNA repair pathway that is relevant to ionizing radiation exposure and glioma risk could be identified, it might be used to investigate links between additional environmental factors and glioma risk.

One of the most promising areas for future glioma research is the role of immune factors in tumor development. All observations from case–control studies based on molecular biomarkers (e.g. viral antibodies, IgE levels) that could possibly be altered by the presence of the tumor need to be confirmed in cohort studies where serum has been collected well in advance of tumor development. Studies examining the role of genetic pathways in glioma development need to be driven by strong a priori hypotheses to avoid false-positive results. Therefore, future studies must be planned by multidisciplinary teams collaborating to identify potential genetic and environmental risk factors. The Brain Tumor Epidemiology Consortium (BTEC) brings together glioma research teams from around the world to plan and conduct these studies. The purpose of this international organization of epidemiologists and other scientists, which meets twice a year, is to understand brain tumor etiology and survival, share resources, and plan and conduct studies. We have established working groups on DNA repair polymorphisms, meningioma, oligodendroglioma, childhood cancer, and familial linkage of glioma, and we plan to form an additional working group focusing on immunological factors and glioma.

Key points

  • Only rare familial syndromes and exposure to high therapeutic doses of ionizing radiation are known causes of glioma

  • Asthma and other allergic conditions decrease glioma risk, and this protective association has been confirmed for glioblastoma by objective evidence from asthma-related germline polymorphisms

  • The general absence of consistent findings of associations between DNA repair and cell cycle regulation polymorphisms and glioma risk might be attributable to unexamined interactions between these genes and immune regulatory genes or with as yet unknown environmental factors or both

Top
Acknowledgments

This work was supported by grants R03CA10337, RO1CA52689 and P50CA097257 from the National Institutes of Health.

Top

References

  1. Central Brain Tumor Registry of the United States [http://www.CBTRUS.org]
  2. Kleihues P and Cavenee WK (1997) Tumors of the central nervous system: pathology and genetics. Lyon, France: International Agency for Research on Cancer
  3. Davis FG et al. (1997) The rationale for standardized registration and reporting of brain and central nervous system tumors in population-based cancer registries. Neuroepidemiology 16: 308–316 | PubMed | ChemPort |
  4. Helseth A (1995) The incidence of primary central nervous system neoplasms before and after computerized tomography availability. J Neurosurg 83: 999–1003 | PubMed | ChemPort |
  5. Legler JM et al. (1999) Cancer surveillance series [corrected]: brain and other central nervous system cancers: recent trends in incidence and mortality. J Natl Cancer Inst 91: 1382–1390 | Article | PubMed | ChemPort |
  6. Smith MA et al. (1998) Trends in reported incidence of primary malignant brain tumors in children in the United States. J Natl Cancer Inst 90: 1269–1277 | Article | PubMed | ChemPort |
  7. Wrensch M et al. (2002) Epidemiology of primary brain tumors: current concepts and review of the literature. Neuro-oncol 4: 278–299 | Article | PubMed | ISI |
  8. Inskip PD et al. (1995) Etiology of brain tumors in adults. Epidemiol Rev 17: 382–414 | PubMed | ISI | ChemPort |
  9. Ichimura K et al. (2004) Molecular pathogenesis of astrocytic tumours. J Neurooncol 70: 137–160 | Article | PubMed |
  10. Ohgaki H et al. (2004) Genetic pathways to glioblastoma: a population-based study. Cancer Res 64: 6892–6899 | Article | PubMed | ISI | ChemPort |
  11. Preston-Martin S and Mack WJ (1996) Neoplasms of the nervous system. In Cancer Epidemiology and Prevention, 1231–1281 (Eds Schottenfeld D and Fraumeni JF) New York: Oxford University Press
  12. Claus EB et al. (2005) Epidemiology of intracranial meningioma. Neurosurgery 57: 1088–1095 | Article | PubMed |
  13. Edick MJ et al. (2005) Lymphoid gene expression as a predictor of risk of secondary brain tumors. Genes Chromosomes Cancer 42: 107–116 | Article | PubMed | ISI | ChemPort |
  14. Ohgaki H and Kleihues P (2005) Epidemiology and etiology of gliomas. Acta Neuropathol (Berl) 109: 93–108 | Article | PubMed |
  15. Relling MV et al. (1999) High incidence of secondary brain tumours after radiotherapy and antimetabolites. Lancet 354: 34–39 | Article | PubMed | ISI | ChemPort |
  16. Chen H et al. (2002) Diet and risk of adult glioma in eastern Nebraska, United States. Cancer Causes Control 13: 647–655 | Article | PubMed |
  17. Cobbs CS et al. (2003) Inactivation of wild-type p53 protein function by reactive oxygen and nitrogen species in malignant glioma cells. Cancer Res 63: 8670–8673 | PubMed | ISI | ChemPort |
  18. Gurney JG et al. (2002) Null association between frequency of cured meat consumption and methylvaline and ethylvaline hemoglobin adduct levels: the N-nitroso brain cancer hypothesis. Cancer Epidemiol Biomarkers Prev 11: 421–422 | PubMed |
  19. Huncharek M et al. (2003) Dietary cured meat and the risk of adult glioma: a meta-analysis of nine observational studies. J Environ Pathol Toxicol Oncol 22: 129–137 | PubMed | ChemPort |
  20. Krishnan G et al. (2003) Occupation and adult gliomas in the San Francisco Bay Area. J Occup Environ Med 45: 639–647 | PubMed |
  21. Maekawa A and Mitsumori K (1990) Spontaneous occurrence and chemical induction of neurogenic tumors in rats—influence of host factors and specificity of chemical structure. Crit Rev Toxicol 20: 287–310 | PubMed | ChemPort |
  22. Navas-Acien A et al. (2002) Occupation, exposure to chemicals and risk of gliomas and meningiomas in Sweden. Am J Ind Med 42: 214–227 | Article | PubMed |
  23. Peterson DL et al. (1994) Animal models for brain tumors: historical perspectives and future directions. J Neurosurg 80: 865–876 | PubMed | ChemPort |
  24. Tedeschi-Blok N et al. (2001) Dietary calcium consumption and astrocytic glioma: the San Francisco Bay Area Adult Glioma Study, 1991–1995. Nutr Cancer 39: 196–203 | PubMed | ChemPort |
  25. Zheng T et al. (2001) Occupational risk factors for brain cancer: a population-based case–control study in Iowa. J Occup Environ Med 43: 317–324 | PubMed | ChemPort |
  26. Wacholder S et al. (2004) Assessing the probability that a positive report is false: an approach for molecular epidemiology studies. J Natl Cancer Inst 96: 434–442 | Article | PubMed |
  27. Gobbel GT et al. (1998) Response of postmitotic neurons to X-irradiation: implications for the role of DNA damage in neuronal apoptosis. J Neurosci 18: 147–155 | PubMed | ISI | ChemPort |
  28. Silasi G et al. (2004) Selective brain responses to acute and chronic low-dose X-ray irradiation in males and females. Biochem Biophys Res Commun 325: 1223–1235 | Article | PubMed | ChemPort |
  29. Watts LT et al. (2005) Astrocytes protect neurons from ethanol-induced oxidative stress and apoptotic death. J Neurosci Res 80: 655–666 | Article | PubMed | ChemPort |
  30. McGlynn KA et al. (1995) Susceptibility to hepatocellular carcinoma is associated with genetic variation in the enzymatic detoxification of aflatoxin B1. Proc Natl Acad Sci USA 92: 2384–2387 | Article | PubMed | ChemPort |
  31. El-Zein R et al. (2002) Epidemiology of brain tumors. In Cancer in the Nervous System, 2nd Edition, 252–266 (Ed. Levin VA) New York: Oxford University Press
  32. Bondy M et al. (1994) Genetics of primary brain tumors: a review. J Neurooncol 18: 69–81 | Article | PubMed | ChemPort |
  33. Wrensch M et al. (1997) Familial and personal medical history of cancer and nervous system conditions among adults with glioma and controls. Am J Epidemiol 145: 581–593 | PubMed | ChemPort |
  34. Grossman SA et al. (1999) Central nervous system cancers in first-degree relatives and spouses. Cancer Invest 17: 299–308 | PubMed | ChemPort |
  35. de Andrade M et al. (2001) Segregation analysis of cancer in families of glioma patients. Genet Epidemiol 20: 258–270 | Article | PubMed | ChemPort |
  36. Malmer B et al. (2001) Genetic epidemiology of glioma. Br J Cancer 84: 429–434 | Article | PubMed | ChemPort |
  37. Paunu N et al. (2002) A novel low-penetrance locus for familial glioma at 15q23–q26.3. Cancer Res 62: 3798–3802 | PubMed | ChemPort |
  38. Brenner AV et al. (2002) History of allergies and autoimmune diseases and risk of brain tumors in adults. Int J Cancer 99: 252–259 | Article | PubMed | ChemPort |
  39. Schlehofer B et al. (1992) Medical risk factors and the development of brain tumors. Cancer 69: 2541–2547 | Article | PubMed | ChemPort |
  40. Schlehofer B et al. (1999) Role of medical history in brain tumour development: results from the international adult brain tumour study. Int J Cancer 82: 155–160 | Article | PubMed | ChemPort |
  41. Schwartzbaum J et al. (2005) Polymorphisms associated with asthma are inversely related to glioblastoma multiforme. Cancer Res 65: 6459–6465 | Article | PubMed | ChemPort |
  42. Schwartzbaum J et al. (2003) Cohort studies of association between self-reported allergic conditions, immune-related diagnoses and glioma and meningioma risk. Int J Cancer 106: 423–428 | Article | PubMed | ChemPort |
  43. Wiemels JL et al. (2004) Reduced immunoglobulin E and allergy among adults with glioma compared with controls. Cancer Res 64: 8468–8473 | Article | PubMed | ChemPort |
  44. Wiemels JL et al. (2002) History of allergies among adults with glioma and controls. Int J Cancer 98: 609–615 | Article | PubMed | ISI | ChemPort |
  45. Wrensch M et al. (2001) Prevalence of antibodies to four herpesviruses among adults with glioma and controls. Am J Epidemiol 154: 161–165 | Article | PubMed | ISI | ChemPort |
  46. Wrensch M et al. (2005) History of chickenpox and shingles and prevalence of antibodies to varicella-zoster virus and three other herpesviruses among adults with glioma and controls. Am J Epidemiol 161: 1–10 | PubMed |
  47. Wrensch M et al. (2006) Serum IgE, tumor epidermal growth factor receptor expression, and inherited polymorphisms associated with glioma survival. Cancer Res 66: 4531–4541 | Article | PubMed | ChemPort |
  48. Tang J et al. (2005) Positive and negative associations of human leukocyte antigen variants with the onset and prognosis of adult glioblastoma multiforme. Cancer Epidemiol Biomarkers Prev 14: 2040–2044 | Article | PubMed | ChemPort |
  49. Berwick M and Vineis P (2000) Markers of DNA repair and susceptibility to cancer in humans: an epidemiologic review. J Natl Cancer Inst 92: 874–897 | Article | PubMed | ChemPort |
  50. Mohrenweiser HW et al. (2003) Challenges and complexities in estimating both the functional impact and the disease risk associated with the extensive genetic variation in human DNA repair genes. Mutat Res 526: 93–125 | Article | PubMed | ISI | ChemPort |
  51. Wood RD et al. (2001) Human DNA repair genes. Science 291: 1284–1289 | Article | PubMed | ISI | ChemPort |
  52. Goode EL et al. (2002) Polymorphisms in DNA repair genes and associations with cancer risk. Cancer Epidemiol Biomarkers Prev 11: 1513–1530 | PubMed | ISI | ChemPort |
  53. Wang LE et al. (2004) Polymorphisms of DNA repair genes and risk of glioma. Cancer Res 64: 5560–5563 | Article | PubMed | ISI | ChemPort |
  54. Wiencke J et al. (2005) Molecular features of adult glioma associated with patient race/ethnicity, age, and a polymorphism in O-methylguanine-DNA-alkyltransferase. Cancer Epidemiol Biomarkers Prev 14: 1774–1783 | Article | PubMed | ChemPort |
  55. Wrensch M et al. (2005) ERCC1 and ERCC2 polymorphisms and adult glioma. Neuro-oncol 7: 495–507 | Article | PubMed | ISI | ChemPort |
  56. Yang P et al. (2005) Polymorphisms in GLTSCR1 and ERCC2 are associated with the development of oligodendrogliomas. Cancer 103: 2363–2372 | Article | PubMed | ChemPort |
  57. Bond GL et al. (2004) A single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor suppressor pathway and accelerates tumor formation in humans. Cell 119: 591–602 | Article | PubMed | ISI | ChemPort |
  58. Wrensch M et al. (2005) The molecular epidemiology of gliomas in adults. Neurosurg Focus 19: E5 | PubMed |
  59. Freije WA et al. (2004) Gene expression profiling of gliomas strongly predicts survival. Cancer Res 64: 6503–6510 | Article | PubMed | ISI | ChemPort |
  60. Kleihues P and Ohgaki H (2000) Phenotype vs genotype in the evolution of astrocytic brain tumors. Toxicol Pathol 28: 164–170. | Article | PubMed | ChemPort |
  61. Lang FF et al. (1994) Pathways leading to glioblastoma multiforme: a molecular analysis of genetic alterations in 65 astrocytic tumors. J Neurosurg 81: 427–436 | PubMed | ISI | ChemPort |
  62. Nigro JM et al. (2005) Integrated array-comparative genomic hybridization and expression array profiles identify clinically relevant molecular subtypes of glioblastoma. Cancer Res 65: 1678–1686 | Article | PubMed | ISI | ChemPort |
  63. Rasheed BK et al. (1999) Molecular pathogenesis of malignant gliomas. Curr Opin Oncol 11: 162–167 | Article | PubMed | ChemPort |
  64. Kleihues P et al. (1997) Tumors associated with p53 germline mutations: a synopsis of 91 families. Am J Pathol 150: 1–13 | PubMed | ISI | ChemPort |
  65. von Deimling A et al. (1995) Molecular pathways in the formation of gliomas. Glia 15: 328–338 | Article | PubMed | ChemPort |
  66. Okada Y et al. (2003) Selection pressures of TP53 mutation and microenvironmental location influence epidermal growth factor receptor gene amplification in human glioblastomas. Cancer Res 63: 413–416 | PubMed | ISI | ChemPort |
  67. Schwartzbaum J et al. (2005) Prior hospitalization for epilepsy, diabetes, and stroke and subsequent glioma and meningioma risk. Cancer Epidemiol Biomarkers Prev 14: 643–650 | Article | PubMed |
  68. James CD et al. (2002) Genetic and molecular basis of primary central nervous system tumors. In Cancer in the Nervous System, 239–251 (Ed. Levin VA) New York: Oxford University Press
  69. National Center for Biotechnology Information [http://www.ncbi.nlm.nih.gov/entrez]
Top
Competing interests

The authors declared no competing interests.

Top

Contact the journal about this article

Subject areas under which this article appears: Neuro-oncology