Article Text

Download PDFPDF

Original article
Occupational solvent exposure and risk of meningioma: results from the INTEROCC multicentre case–control study
  1. Dave McLean1,
  2. Sarah Fleming2,
  3. Michelle C Turner3,4,5,
  4. Laurel Kincl6,
  5. Lesley Richardson7,
  6. Geza Benke8,
  7. Brigitte Schlehofer9,
  8. Klaus Schlaefer9,
  9. Marie-Elise Parent10,
  10. Martine Hours11,
  11. Daniel Krewski12,
  12. Martie van Tongeren13,
  13. Siegal Sadetzki14,
  14. Jack Siemiatycki7,
  15. Elisabeth Cardis3,4,5
  1. 1Centre for Public Health Research, Massey University, Wellington, New Zealand
  2. 2University of Leeds, Leeds, UK
  3. 3Centre for Research in Environmental Epidemiology (CREAL), Barcelona, Spain
  4. 4Universitat Pompeu Fabra (UPF), Barcelona, Spain
  5. 5CIBER Epidemiología y salud Pública (CIBERESP), Barcelona, Spain
  6. 6Oregon State University, Corvallis, USA
  7. 7University of Montreal Hospital Research Centre (CRCHUM), Montreal, Canada
  8. 8Monash University, Melbourne, Australia
  9. 9Unit of Environmental Epidemiology, DKFZ, Heidelberg, Germany
  10. 10INRS-Institut Armand-Frappier, Montreal, Canada
  11. 11IFSTTAR, Lyon, France
  12. 12University of Ottawa, Ottawa, Canada
  13. 13Institute of Occupational Medicine, Edinburgh, UK
  14. 14Gertner Institute, Chaim Sheba Medical Centre & Tel Aviv university, Tel Aviv, Israel
  1. Correspondence to Dr David McLean, Centre for Public Health Research, Massey University, 102 Adelaide Road, Wellington 6021, New Zealand; d.j.mclean{at}massey.ac.nz

Abstract

Objective To examine associations between occupational exposure to selected organic solvents and meningioma.

Methodology A multicentre case–control study conducted in seven countries, including 1906 cases and 5565 controls. Occupational exposure to selected classes of organic solvents (aliphatic and alicyclic hydrocarbons, aromatic hydrocarbons, chlorinated hydrocarbons and ‘other’ organic solvents) or seven specific solvents (benzene, toluene, trichloroethylene, perchloroethylene, 1,1,1-trichloroethylene, methylene chloride and gasoline) was assessed using lifetime occupational histories and a modified version of the FINJEM job-exposure matrix (INTEROCC-JEM). Study participants were classified as ‘exposed’ when they had worked in an occupation for at least 1 year, with a 5-year lag, in which the estimated prevalence of exposure was 25% or greater in the INTEROCC-JEM. Associations between meningioma and each of the solvent exposures were estimated using conditional logistic regression, adjusting for potential confounders.

Results A total of 6.5% of study participants were ever exposed to ‘any’ solvent, with a somewhat greater proportion of controls (7%) ever exposed compared with cases (5%), but only one case was ever exposed to any chlorinated hydrocarbon (1,1,1-trichloroethane). No association was observed between any of the organic solvents and meningioma, in either men or women, and no dose–response relationships were observed in internal analyses using either exposure duration or cumulative exposure.

Discussion We found no evidence that occupational exposure to these organic solvents is associated with meningioma.

View Full Text

Statistics from Altmetric.com

What this paper adds

  • Meningiomas are the most frequently diagnosed primary brain tumours, although the aetiology is largely unknown.

  • The only established risk factor is ionising radiation, although associations with exposure to extremely low frequency electromagnetic fields, herbicides, lead and organic solvents have been observed.

  • The INTEROCC multicentre case–control study of 1906 cases and 5565 controls investigated associations between meningioma and a range of occupational exposures, and in these analyses we examined associations with occupational exposure to organic solvents.

  • Despite the large size of the INTEROCC study, the prevalence of occupational exposure to solvents was low, with a total of 6.5% of participants ever exposed to any solvent, 5.7% to aromatic hydrocarbons, 2.5% to aliphatic and alicyclic hydrocarbons, and only one case to chlorinated hydrocarbons.

  • We found no evidence of an increased risk of meningioma associated with occupational exposure to organic solvents. Further analyses are required to examine the effect of multiple exposures.

Introduction

Meningiomas are the most frequently diagnosed primary brain tumours, accounting for approximately 20% of all intracranial tumours in males and 38% in females in the USA.1 ,2 The incidence of meningioma increases markedly with age, and the disease occurs at a ratio of approximately 2:1 in females compared with males. In contrast to other types of cancer, meningioma does not appear to be influenced strongly by lifestyle factors such as cigarette smoking, diet and alcohol.3 ,4

Much of the aetiological research on brain tumours conducted to date has investigated primary malignant brain tumours (which are predominantly gliomas), and relatively few have investigated specific risk factors for meningioma. Increased risks of the various brain tumour types have been observed in a number of studies to be associated with specific occupations or occupational exposures, most notably farming and pesticides,5–8 organic solvents5 ,9 ,10 and metals,11–13 although the evidence remains equivocal for most of these exposures.14

There have been relatively limited findings to date on environmental risk factors for meningioma. The only established risk factors for meningioma are high-dose ionising radiation, with significantly elevated risk observed in atomic bomb survivors,15 and following lower-dose radiation therapy given to Israeli children for scalp ringworm;16 ,17 however, these account for only a small proportion of cases. Several studies have also shown elevated risk associated with exposure to full-mouth dental X-rays, although it is thought that this may be associated with the higher doses that were given in the past.18–20

Other environmental exposures for which associations with meningioma have been observed include extremely low frequency electromagnetic fields,21 herbicides in women but not in men22 and lead (most apparent in individuals with the ALAD G177C polymorphism).23–25 A number of occupations with potential exposure to organic solvents have also been implicated as risk factors for meningioma. The highly lipophilic nature of organic solvents means that they are readily absorbed through the respiratory system and skin, and readily cross the blood–brain barrier, where the resulting irritation of cell membranes and tissues causes neurotoxicity through mechanisms including interaction with neuronal receptors, inhibition of excitatory receptors/channels and/or upregulation of inhibitory receptors/channels.26 In a Chinese case–control study, a non-significantly elevated risk was observed in individuals with occupational exposure to petrol or diesel oil, or to benzene,23 while in another Swedish study a significantly elevated risk was observed in those estimated to have ‘possible’ exposure to solvents, although no risk was apparent in those with ‘probable’ solvent exposure.24 Significantly elevated risks have also been observed in a US case–control study in potentially solvent-exposed occupations, including automotive body and related repairers, automotive body painters and designers and decorators, with a suggestion of a dose–response relationship; however, the number of cases was small and the CIs wide.25 Other studies, however, have found no association with solvent exposure in women or with exposure to chlorinated solvents.13 ,27

In recognition of the paucity of information on the aetiology of meningioma, the Brain Tumour Epidemiology Consortium recently released a consensus document recommending further epidemiological research into risk factors for meningioma, with improved exposure assessment, and using large-scale case–control studies on an international level.3 In the year 2000, an international group mounted a study of glioma, meningioma, acoustic neurinoma and malignant parotid gland tumours in order to evaluate possible carcinogenic effects of mobile phone use (the INTERPHONE study). The study design and epidemiological methods, and a description of the INTERPHONE study population, are described in detail elsewhere.28 An important additional component of the INTERPHONE study was the inclusion by seven of the participating countries (Australia, Canada, France, Germany, Israel, New Zealand and the UK) of an additional section in the study questionnaire obtaining a complete occupational history of all jobs held for more than 6 months to investigate associations with occupational exposure to selected chemical agents.

The information collected from these seven countries has been further analysed for the Occupational Exposure and Brain Cancer (INTEROCC) study using a job-exposure matrix to assess occupational exposure in order to evaluate the effects of exposure to selected chemicals in the workplace on the risk of brain cancer. For this component of the INTEROCC study in which we evaluate risk factors for meningioma, a total of 1906 cases and 5565 population-based controls were included, making this by far the largest case–control study of meningioma after the INTERPHONE study. In this paper, we present results for occupational exposure to a range of organic solvents.

Methods

The case–control study

The INTEROCC study is a multicentre population-based case–control study conducted between 2000 and 2004 in 10 study centres from seven countries, namely, Australia, Canada, France, Germany, Israel, New Zealand and the UK. The sampling frame for this study depended on the local situation in each country;28 however, in most centres, the study population consisted of all residents in the study regions aged 30–59, with eligible cases being all residents of those regions diagnosed (either histologically or in 25% of cases based on unequivocal diagnostic imaging) during the study period with a confirmed first primary meningioma. The age range of cases varied in some centres, with cases up to the age of 69 years included in Germany, cases aged 18–69 in the UK and in Israel cases aged 18+ years were recruited to allow for greater case ascertainment. Again using methods suitable for each local situation, controls were randomly selected from the source population in each centre, either individually matched or frequency matched on year of birth (within 5-year categories), sex and study region. Wherever possible study subjects were interviewed face to face by trained interviewers using a computer-assisted personal interview questionnaire, although a small number of proxy interviews were conducted where the study subject had died or was too ill to be interviewed. The study questionnaire covered demographic factors, including socioeconomic and lifestyle factors, and a full occupational history.

Occupational exposure assessment

The occupational history information obtained included all jobs held for at least 6 months, the job title, the specific tasks performed in that job, the company name, a description of the activities performed by the company and the start and end year of each job. An occupational hygienist from each of the seven participating countries coded local jobs according to the international occupation and industry classifications, that is, the International Standard Classification of Occupations 1968 edition (ISCO68) and 1988 edition (ISCO88) for occupations,29 ,30 and the International Standard Industrial Classification 1971 edition (ISIC71) for industry.31 To maximise the quality and comparability of coding between countries, a common set of coding guidelines was established, a group training exercise was held and a web-based forum was established to discuss difficult cases. Before and after comparison trials were conducted to determine the reliability of coding between different countries, and pairwise agreements between coders for ISCO68 and ISCO88 showed a moderate improvement of at least 10% between trials.32

Based on a review of the literature in relation to possible occupational risk factors for brain tumours, a total of 29 chemical agents, including selected organic solvents, combustion products, metals, dusts, formaldehyde, oil mist and sulfur dioxide, were selected for investigation in the INTEROCC study. The retrospective assessment of exposure to the selected chemicals for all study participants was made by linking the occupational history information on job and industry codes to a job-exposure matrix (INTEROCC-JEM). The INTEROCC-JEM was based on the Finnish job-exposure matrix (FINJEM),33 modified by using exposure estimates developed for a population-based case–control study of lung cancer in Montréal.34 Assignments of exposure were peer reviewed by an international panel of occupational hygienists to ensure that INTEROCC-JEM estimates better reflected the prevailing exposure patterns in the seven participating countries.35

For this component of the INTEROCC study, occupational exposure to the following organic solvents was assessed: aliphatic and alicyclic hydrocarbons (ALHC), aromatic hydrocarbons (ARHC), toluene (TOLU), benzene (BENZ, chlorinated hydrocarbon solvents (CHC), perchloroethylene (PER), methylene chloride (MCH), 1,1,1-trichloroethane (TCE), and trichloroethylene (TRI), gasoline (GASO), and other organic solvents (OSOL). The INTEROCC-JEM assigns both p (probability of exposure as judged by the proportion of workers in that occupation who were considered to be exposed) and I (intensity of exposure). In determining the exposure status of individual study participants, it was decided to assign exposure only to those occupations where the estimated probability of exposure in that job category was greater than or equal to 25% (p≥25%). To ensure that the main analyses were based on a clean non-exposed group, we also excluded from the reference group any individuals who had held occupations in which the estimated prevalence of exposure p was greater than or equal to 5% but less than 25%. A minimum period of exposure of 1 year was applied, and to allow for latency a lag period of 5 years was also applied.

Statistical analysis

Conditional logistic regression models were used to calculate ORs and 95% CIs for risk of meningioma associated with each solvent or solvent class. Analyses were based on strata defined by the matching variables: sex, age (5-year groups) and country-region, and adjusted for education. Three exposure indices were used: (1) ever/never exposed; (2) lifetime cumulative exposure categorised in quartiles of the distribution among exposed controls and (3) total duration of exposure in categories decided a priori of 1–4 years, 5–14 years and >15 years. All subjects never exposed to ‘any solvent’ were used as the reference category. Sensitivity analyses were conducted by excluding proxy respondents, subjects over 69 years and subjects with neurofibromatosis and tuberous sclerosis, and by including those with uncertain exposure (ie, duration of exposure<1 year) in the unexposed group. We also repeated the analyses using different cut-off points for defining exposure (ie, p≥5%, p≥50%) and using different lag periods (ie, 1, 5 and 10 years).

Results

Table 1 describes selected characteristics of the study population of the 1906 meningioma cases and 5565 controls that were included in the analyses. The majority of cases (73.4%) were women, the mean age at diagnosis was 55 years and the majority of cases (60.1%) had at most a primary or secondary education. A total of 4.6% of the case interviews were conducted with a proxy respondent. Overall participation rates were 81% for cases and 50% for controls.

Table 1

Description of the study population

The lifetime occupational exposure prevalence, mean cumulative exposure and mean duration of exposure to organic solvents are shown in table 2. Where p≥25%, a total of 6.5% of all participants were ever exposed to any solvent, with a greater proportion of controls (7%) compared with cases (5%). Where p≥5% was used as a cut-off point, a total of 14.3% of participants were ever exposed to any solvent, while where a cut-off point of p≥50% was used the total ever exposed was 4.7%. The specific solvents with the greatest number of participants ever exposed at p≥25% were ARHC 5.7%, ALHC 2.5%, TOLU 2.3% and OSOLs 1.6%. Of the CHCs, no cases were ever exposed (at p≥25%) to MCH, PER or TRI, and only one case was exposed to TCE. The highest mean cumulative exposures were for ARHC (10 464.20 ppm) and TOLU (10 446.80 ppm), while the longest mean durations of exposure were to OSOL (12.4 years), ARHC (10.9 years), ALHC (10.4 years) and TOLU (10.3 years). Controls tended to have greater mean cumulative exposure to all solvents compared with cases and also tended to have higher mean durations of exposure to all solvents compared with cases, with the exception of BENZ. As can be seen in table 3, with the exception of those ever exposed to ARHC (where less than 40% were ever exposed to any other agent), joint exposure to different solvents or classes of solvents was common.

Table 2

Prevalence, mean cumulative exposure and mean duration of solvent exposure in cases and controls*

Table 3

Joint exposure to solvents in the INTEROCC study*

As can be seen in table 4, no association was observed between exposure to any solvent and meningioma in the overall study population (OR=1.05, 95% CI 0.82 to 1.36), or in women (OR=1.02, 95% CI 0.75 to 1.39) or men (OR=1.16, 95% CI 0.73 to 1.84) separately. For those specific solvents with at least five exposed cases, the ORs ranged from 0.65 to 1.33 overall, and none were statistically significant. In internal analyses, there were also no significant trends of increasing risk with either increasing cumulative exposure or duration of exposure, either overall or when stratified by gender. Sensitivity analyses excluding proxy respondents, including those with uncertain exposure (less than 1 year in duration) in the reference group, excluding unresponsive respondents, excluding subjects greater than 69 years of age or excluding subjects with neurofibromatosis or tuberous sclerosis, generally revealed little change in ORs (results not shown). Other sensitivity analyses using different cut-off points to define exposure and using different lag times also resulted in little change to OR estimates and showed no significant trends according to exposure duration or cumulative exposure. Similarly analyses including additional potentially confounding variables (socioeconomic status, marital status, smoking, use of a proxy, a medical history of atopic conditions or age at first exposure) in the model also generally resulted in little change in the ORs. There was also no significant interaction according to sex, age, age at first exposure or smoking status for ever exposure (data not shown).

Table 4

Risk estimates for meningioma associated with exposure to selected organic solvents

Discussion

This multicentre case–control study of 1906 cases and 5565 controls provides no evidence of any association between occupational exposure to organic solvents and meningioma in the overall study population or in women or men separately. There were also no significant trends of increasing risk with either increasing cumulative exposure or duration of exposure. However, interpretation of these negative findings should take into account the strengths and limitations of the study.

While the INTEROCC study is the largest case–control study conducted to date investigating associations between occupational exposure to organic solvents and meningioma, the relatively low-exposure prevalence in the study population limited its power to identify significant associations and in particular after stratification by gender. Potential sources of selection and recall bias in the design and conduct of the INTERPHONE study on which this study was based have been addressed in detail elsewhere;28 ,36 however, these related primarily to the somewhat more problematic issue of recall of cell phone use by study participants and it is unlikely that the same biases would apply to occupational chemical exposure with the methods used in this study to assign exposure. The vast majority of interviews in this study were conducted face to face, with only 4.6% of the case interviews being conducted with a proxy respondent. There were differential response rates between cases and controls (estimated as 81% of cases and 50% of controls); however, there is little evidence of systematic response bias apart from gender distribution and, to a lesser extent, country-region, between cases and controls, and these differences were controlled for in the statistical analyses.

The retrospective assessment of occupational exposure is the major challenge in all community-based case–control studies, particularly in multicentre studies where there may be either true differences in exposure prevalence and intensity between countries or differences in the quality of the exposure estimation made locally.37 ,38 The different methods available for estimating occupational exposures retrospectively, that is, analysis by job or industry title, self-reports of exposure, job-exposure matrices and case-by-case expert evaluation,39 each have advantages and disadvantages and involve some trade-offs between cost and validity.35 With over 35 000 jobs having been held by INTEROCC study participants and the relatively limited information available in the occupational history, the only feasible option for this study was the use of a job-exposure matrix.

Notwithstanding the limitations associated with the use of a job-exposure matrix approach to estimate past exposure, we believe that the process we followed to modify the FINJEM for application to the INTEROCC study resulted in improved consistency and better estimates of exposure across the seven participating countries. The assessment of exposure based on study participants’ reported occupational histories (which most people can recall with accuracy) and the assignment of exposure through linkage with a job-exposure matrix substantially reduce the risk of recall and reporting bias; however, assigning a single exposure measure to all subjects when it is known that there is typically wide variability in exposure levels between workers in a given job title does introduce error. This error, however, is non-differential with respect to case/control status and is, therefore, likely to lead to the attenuation of true risk estimates.

A number of previous studies have suggested associations between occupational exposure to organic solvents and meningioma. However, this evidence has come from two relatively small studies using exposure assessment based on responses to specific questions about chemical exposures,23 or analyses using expert assessments to group jobs determined to have similar tasks and exposures.25 Another large Swedish study involved linkage of cancer registry data with census data from 1960 and 1970 to determine occupation and used a Swedish job-exposure matrix to estimate exposure.24 While this study found some evidence of an association between occupations categorised as having ‘possible’ exposure to solvents and meningioma, no association was observed for occupations categorised as having ‘probable’ exposure to solvents. Despite the large sample size in the INTEROCC study, the prevalence of exposure to CHCs was extremely low, with only one case ever exposed to a CHC. We were unable, therefore, to assess risk associated with exposure to MCH, PER, TCE or TRI; however, the exposure prevalence was adequate for the other solvents. The recent US study using expert assessment to estimate exposure found no evidence of an association between occupational exposure to CHC solvents and meningioma.27

In conclusion, this study provides no evidence of an association between occupational exposure to selected organic solvents and meningioma.

Acknowledgments

We would like to thank Timo Kauppinen of the Finnish Institute of Occupational Health for allowing us to use and modify the FINJEM as well as sharing his valuable expertise on the exposure assessment committee in the company of Nils Plato of the Karolinska Institute in Stockholm. Jordi Figuerola and Rodrigo Villegas of CREAL have been the backbone of the team in cleaning and preparing the data for analysis. We would also like to thank Avital Jarus-Hakak (Israel), Louise Nadon (Canada), Florence Samkange-Zeeb (Germany), Anne Sleeuwenhoek (UK) and Hélène Tardy, who coded occupations and assisted in the data clean-up. We are grateful to Drs Bruce Armstrong (Australia), Mary McBride (Canada), Maria Blettner (Germany), Joachim Schüz (Germany), Alistair Woodward (New Zealand) and Patricia McKinney (UK) for the use of the occupational data from their INTERPHONE study centres for the INTEROCC project.

References

View Abstract

Footnotes

  • Contributors DM was involved in exposure assessment, analysis strategy and manuscript writing. SF was involved in data collection and analysis strategy. MCT was involved in data management and analysis. LK was involved in study coordination, data management and coordination of exposure assessment. LR was the coordinator involved in study design, exposure assessment, data management and statistical analysis. GB was involved in exposure assessment and analysis strategy. MvT was involved in exposure assessment and analysis strategy. EC was the overall coordinator of INTEROCC study, supervision of database management, exposure assessment strategy, data analysis and interpretation of results. All other coauthors were responsible for original data collection and/or decisions taken on exposure assessment, analysis strategy and review of the manuscript. All authors read and reviewed the final manuscript.

  • Funding This work was funded by the National Institutes for Health (NIH) Grant No. 1R01CA124759-01. Coding of the French occupational data was in part funded by AFSSET (Convention N° ST-2005-004). The INTERPHONE study was supported by funding from the European Fifth Framework Program, ‘Quality of Life and Management of Living Resources’ (contract 100 QLK4-CT-1999901563) and the International Union against Cancer (UICC). The UICC received funds for this purpose from the Mobile Manufacturers’ Forum and GSM Association. In Australia, funding was received from the Australian National Health and Medical Research 5 Council (EME Grant 219129) with funds originally derived from mobile phone service license fees; a University of Sydney Medical Foundation Program; the Cancer Council NSW and The Cancer Council Victoria. In Canada, funding was received from the Canadian Institutes of Health Research (project MOP-42525); the Canada Research Chair programme; the Guzzo-CRS Chair in Environment and Cancer; the Fonds de la recherche en sante du Quebec; the Canadian Institutes of Health Research (CIHR), the latter including partial support from the Canadian Wireless Telecommunications Association; the NSERC/SSHRC/McLaughlin Chair in Population Health Risk Assessment at the University of Ottawa. In France, funding was received by l'Association pour la Recherche sur le Cancer (ARC) (Contrat N85142) and three network operators (Orange, SFR, Bouygues Telecom). In Germany, funding was received from the German Mobile Phone Research Program (Deutsches Mobilfunkforschungsprogramm) of the German Federal Ministry for the Environment, Nuclear 45 Safety, and Nature Protection; the Ministry for the Environment and Traffic of the state of Baden-Wurttemberg; the Ministry for the Environment of the state of North Rhine-Westphalia; the MAIFOR Program (Mainzer Forschungsforderungsprogramm) of the University of Mainz. In New Zealand, funding was provided by the Health Research Council, Hawkes Bay Medical Research Foundation, the Wellington Medical Research Foundation, the Waikato Medical Research Foundation and the Cancer Society of New Zealand. Additional funding for the UK study was received from the Mobile Telecommunications, Health and Research (MTHR) program, funding from the Health and Safety Executive, the Department of Health, the UK Network Operators (O2, Orange, T-Mobile, Vodafone, ‘3’) and the Scottish Executive.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Various ethics committees in each country and study centre.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.