What email address or phone number would you like to use to sign in to Docs.com?
If you already have an account that you use with Office or other Microsoft services, enter it here.
Or sign in with:
Signing in allows you to download and like content, and it provides the authors analytical data about your interactions with their content.
Embed code for: Cancer and Nutrition (EPIC) cohort study
Select a size
Coffee and tea intake and risk of brain tumors in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort study1–3 Dominique S Michaud, Valentina Gallo, Brigitte Schlehofer, Anne Tjønneland, Anja Olsen, Kim Overvad, Christina C Dahm, Birgit Teucher, Annekatrin Lukanova, Heiner Boeing, Madlen Schu ¨tze, Antonia Trichopoulou, Pagona Lagiou, Andreas Kyrozis, Carlotta Sacerdote, Vittorio Krogh, Giovanna Masala, Rosario Tumino, Amalia Mattiello, H Bas Bueno-de-Mesquita, Martine M Ros, Petra HM Peeters, Carla H van Gils, Guri Skeie, Dagrun Engeset, Christine L Parr, Eva Ardanaz, Maria-Dolores Chirlaque, Miren Dorronsoro, Maria Jose ´ Sa ´nchez, Marcial Argu ¨elles, Paula Jakszyn, Lena M Nilsson, Beatrice S Melin, Jonas Manjer, Elisabet Wirfa ¨lt, Kay-Tee Khaw, Nick Wareham, Naomi E Allen, Timothy J Key, Isabelle Romieu, Paolo Vineis, and Elio Riboli ABSTRACT Background: In a recent US cohort study, total coffee and tea consumption was inversely associated with risk of glioma, and ex- perimental studies showed that caffeine can slow the invasive growth of glioblastoma. Objective: The objective was to examine the relation between cof- fee and tea intake and the risk of glioma and meningioma in a large European cohort study, the European Prospective Investigation into Cancer and Nutrition (EPIC). Design: Data on coffee and tea intake were collected from men and women recruited into the EPIC cohort study. Over an average of 8.5 y of follow-up, 343 cases of glioma and 245 cases of meningi- oma were newly diagnosed in 9 countries. We used Cox propor- tional hazards models to examine the relation between coffee and tea and brain tumors. Results: We observed no associations between coffee, tea, or com- bined coffee and tea consumption and risk of either type of brain tumor when using quantiles based on country-speciﬁc distributions of intake. However, a signiﬁcant inverse association was observed for glioma risk among those consuming100 mL coffee and tea per day compared with those consuming ,100 mL/d (hazard ratio: 0.66; 95% CI: 0.44, 0.97; P = 0.03). The association was slightly stronger in men (hazard ratio: 0.59; 95% CI: 0.34, 1.01) than in women (hazard ratio: 0.74; 95% CI: 0.42, 1.31), although neither was statistically signiﬁcant. Conclusions: In this large cohort study, we observed an inverse association between total coffee and tea consumption and risk of glioma that was consistent with the ﬁndings of a recent study. These ﬁndings, if further replicated in other studies, may provide new avenues of research on gliomas. Am J Clin Nutr 2010;92: 1145–50. INTRODUCTION The etiology of brain tumors is poorly understood, and known risk factors, namely ionizing radiation and genetic predisposition, affect only a small proportion of the total population, which provides little opportunity for prevention. In Europe, the age- adjusted incidence rates for brain and nervous system tumors range between 4 and 6 per 100,000 person-years in women and between 6 and 8 per 100,000 person-years in men (1). Reported increases in incidence rates of brain tumors have been mostly attributed to improvements in diagnosis and classiﬁcation (2). A recent study in the United States with 335 incident cases of glioma from 3 large prospective cohorts reported a strong inverse association between total coffee and tea intake and risk of glioma (relative risk: 0.60; 95% CI: 0.41, 0.87 for 5 vs 0–1 cups/d; P for trend = 0.04) (3). An inverse association was also noted for caffeine intake, which was stronger in men than in women. Only 6 other epidemiologic studies, 1 cohort and 5 case-control studies, have measured the association between coffee, tea, or caffeinated beverages and glioma risk, and the results have been inconsistent (4–9). Caffeine has been inversely associated 1 From the Department of Epidemiology and Public Health, Imperial Col- lege, London, United Kingdom (DSM, VG, PV, and ER); Brown University, Providence, RI (DSM); the Research Group Environmental Epidemiology (BS) and Division of Cancer Epidemiology (BT), German Cancer Research Center, Heidelberg, Germany; the Department of Epidemiology, School of Public Health, Aarhus University, Aarhus, Denmark (KO and CCD); the Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark (CCD); the Institute of Cancer Epidemiology, Danish Cancer So- ciety, Copenhagen, Denmark (ATand AO); the Department of Epidemiology, German Institute of Human Nutrition Potsdam-Rehbruecke, Nuthetal, Ger- many (HB and MS); the WHO Collaborating Center for Food and Nutrition Policies, Department of Hygiene, Epidemiology and Medical Statistics, Uni- versity of Athens Medical School, Athens, Greece (AT and PL); the Hellenic Health Foundation, Athens, Greece (AT, AK, and PL); First Department of Neurology, Eginition Hospital, University of Athens Medical School, Athens, Greece (AK); the Department of Biomedical Science and Human Oncology, CPO-Piemonte, Torino, Italy (CS); the Nutritional Epidemiology Unit, Department of Preventive and Predictive Medicine, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy (VK); the Molecular and Nutri- tional Epidemiology Unit, ISPO–Cancer Research and Prevention Institute, Florence, Italy (GM); the Cancer Registry and Histopathology Unit, “Civile- M.P.Arezzo” Hospital, ASP 7 Ragusa, Italy (RT); the Department of Clinical and Experimental Medicine, Federico II University, Naples, Italy (AM); the National Institute for Public Health and the Environment, Bilthoven, Netherlands (HBB-d-M and MMR); the Department of Epidemiology, Bio- statistics and HTA, Radboud University Nijmegen Medical Centre, Nijme- gen, Netherlands (MMR); Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Netherlands (PHMP and CHvG); Am J Clin Nutr 2010;92:1145–50. Printed in USA. 2010 American Society for Nutrition 1145 by guest on May 11, 2013ajcn.nutrition.orgDownloaded from with many brain-related diseases, including Parkinson (10) and Alzheimer (11) diseases. In addition, coffee consumption has been reported to be inversely associated with liver cancer (12) and has been shown to reduce subclinical inﬂammation in a clinical trial (13). Given the limited evidence suggesting that coffee and tea intake may reduce the risk of glioma, more studies are needed to address this hypothesis. Furthermore, the relation between coffee and tea intake and the risk of meningioma has not been examined to date. In this study, we examined the relation between coffee and tea intake and risk of glioma and meningioma in the pro- spective cohort study European Prospective Investigation into Cancer and Nutrition study (EPIC). This is the ﬁrst study to examine coffee and tea intake and brain tumors in a European population. SUBJECTS AND METHODS EPIC is an ongoing prospective cohort study conducted in 10 European countries that was designed to investigate causes of cancer. The cohort consists of 521,448 male and female subjects, mostly aged 25–70 y, whowere recruited between 1991 and 2000, usually from the general population residing within deﬁned geographic areas (ie, towns or provinces), with some exceptions: women who were members of a health insurance scheme for state school employees (France), women attending breast cancer screening (Utrecht, the Netherlands and Naples, Italy), blood donors (some centers in Italy and Spain), and predominantly vegetarians (Oxford “health conscious” cohort). Eligible subjects were invited to participate in the study by mail or by personal contact. Those who decided to participate signed an informed consent form and were mailed a diet and lifestyle questionnaire, except for those at all of the centers in Spain, Greece, and Ragusa, where interviewer-administered questionnaires were used. In most countries, study subjects were invited to a center for blood collection and anthropometric measurements and to deliver the completed diet and lifestyle questionnaires. The questionnaires included questions on diet, smoking, alcohol drinking, education, occupation, history of previous illness and disorders or surgical operations, physical activity, and other lifestyle factors. Loss to follow-up (deﬁned as unknown vital status at the last follow-up time) was ,6% across centers. The study was approved by the Internal Review Board of the International Agency for Cancer Research on Cancer and the local ethics committees in the participating countries. Cancer ascertainment Cancer diagnoses are based on population cancer registries in Denmark, Italy, the Netherlands, Norway, Spain, Sweden, and the United Kingdom. For France, Germany, and Greece, active follow-up was made by using a combination of different methods, including health insurance records and cancer and mortality registries. However, because of missing histology on cancer cases reported in the French cohort, France was excluded from the analysis. We included all ﬁrst primary incident cases diagnosed through the end of follow-up (from January 2003 to November 2006 depending on center) using the International Classiﬁcation of Diseases–Oncology [ICD-O] second edition (14) for glioma (9380-9460, 9505) and meningioma (9530-9537). A total of 343 glioma (165 men, 178 women) and 245 meningioma cases (54 men, 191 women) were available for these analyses. Exposure assessment Diet was measured over the 12 mo before recruitment by using food-frequency questionnaires speciﬁcally designed for each participating country; more details on the questionnaires can be found elsewhere (15). An interview-based dietary history method combining a questionnaire with a 7-d menu book was used in Malmo ¨, Sweden. Participants reported consumption of coffee, tea, and herbal tea; questionnaires in some centers inquired about caffeinated and decaffeinated coffee consumption. The ques- tionnaire for Norway did not include questions on tea intake. Participants recorded the number of cups of coffee or tea per month, week, or day; the exact structure of the questions varied the Department of Community Medicine, University of Tromsø, Tromsø, Norway (GS and DE); the Institute of Basic Medical Sciences, Department of Biostatistics, University of Oslo, Oslo, Norway (CLP); the Public Health Institute of Navarra, Pamplona, Spain (EA); CIBER Epidemiologı ´a y Salud Pu ´blica, Spain (EA, M-DC, and MJS); the Department of Epidemiology, Regional Health Authority, Murcia, Spain (M-DC); Andalusian School of Public Health, Granada, Spain (MJS); the Public Health and Participation Directorate, Health and Health Care Services Council, Asturias, Spain (MA); the Public Health Division of Gipuzkoa, Basque Regional Health Depart- ment and CIBER Epidemiologı ´a y Salud Pu ´blica, Spain (MD); the Unit of Nutrition, Environment and Cancer, Catalan Institute of Oncology, Barcelo- na, Spain (PJ); the Department of Public Health and Clinical Medicine, Nutritional Research, Umea ˚ University, Umea ˚, Sweden (LMN); the Depart- ment of Radiation Sciences, Oncology, Umea ˚ University Hospital, Umea ˚, Sweden (BM); the Department of Plastic Surgery, Ska ˚ne University Hospital Malmo ¨ (JM) and the Department of Clinical Sciences in Malmo ¨/Nutrition Epidemiology (EW), Lund University, Malmo ¨, Sweden; the Cancer Epide- miology Unit, University of Oxford, Nufﬁeld Department of Clinical Med- icine, Oxford, United Kingdom (NEA and TJK); the Department of Public Health and Primary Care University of Cambridge, Cambridge, United King- dom (K-TK); the MRC Epidemiology Unit, Cambridge, United Kingdom (NW); and the International Agency for Research on Cancer, Lyon, France (IR). 2 The coordination of EPIC is ﬁnancially supported by the European Commission (DG-SANCO) and the International Agency for Research on Cancer. The national cohorts are supported by the Danish Cancer Society (Denmark); Ligue contre le Cancer, Socie ´te ´ 3M, Mutuelle Ge ´ne ´rale de l’Ed- ucation Nationale, Institut National de la Sante ´ et de la Recherche Medicale (France); Deutsche Krebshilfe, Deutsches Krebsforschungszentrum and Fed- eral Ministry of Education and Research (Germany); Ministry of Health and Social Solidarity, Stavros Niarchos Foundation and Hellenic Health Foun- dation (Greece); Italian Association for Research on Cancer (AIRC) and the National Research Council (Italy); the Dutch Ministry of Public Health, Welfare and Sports (VWS); the Netherlands Cancer Registry (NKR); LK Research Funds, Dutch Prevention Funds, Dutch ZON (Zorg Onderzoek Nederland); World Cancer Research Fund (the Netherlands); Statistics Neth- erlands; Norwegian Cancer Society (Norway); Health Research Fund (FIS), Regional Governments of Andalucı ´a, Asturias, Basque Country, Murcia, and Navarra, ISCIII RETIC (RD06/0020) (Spain); the Swedish Cancer Society, Swedish Scientiﬁc Council and Regional Government of Ska ˚ne and Va ¨st- erbotten (Sweden); Cancer Research UK; the Medical Research Council; Stroke Association; the British Heart Foundation; Department of Health, Food Standards Agency; and Wellcome Trust (United Kingdom). 3 Address correspondence to DS Michaud, Brown University, Box G-S121-2, Providence, RI 02912. E-mail: firstname.lastname@example.org. Received May 26, 2010. Accepted for publication August 6, 2010. First published online September 15, 2010; doi: 10.3945/ajcn.2010.29876. 1146 MICHAUD ET AL by guest on May 11, 2013ajcn.nutrition.orgDownloaded from somewhat by country and questionnaire. Total consumption (in mL/d) was calculated for each center. Data on tobacco use were obtained at baseline by means of questions on smoking status (current, past, or never smoker), type of tobacco (cigarettes, cigars, or pipe), number of cigarettes currently smoked, and the age when participants started and, if applicable, quit smoking. Height and weight were measured in all EPIC centers except for France, Norway, and Oxford (United Kingdom), for which self-reported height and weight were as- sessed via questionnaire (15). Education and occupation and other existing medical conditions were also recorded. Statistical analyses For this analysis, we excluded all prevalent cancers cases at baseline and individuals with incomplete follow-up (n = 27,082), those with missing dietary data (n = 6157), France because of missing histology (n = 69,427), and poorly completed ques- tionnaires based on the ratio for energy intake compared with energy expenditure (excluding the top and bottom 1%; n = 8295). We further excluded brain cancer cases with missing histology or cases not classiﬁed as gliomas or meningiomas (n = 178). The ﬁnal data set for the analysis included 410,309 par- ticipants with a mean follow-up time of 8.5 y. Person-time was calculated from the date of recruitment until the date of incident brain cancer diagnosis, death, date of last contact, or end of follow-up period, whichever came ﬁrst. The hazard ratios (HRs) and their corresponding 95% CIs for brain tumors were esti- mated by using the Cox proportional hazards regression model, with age as the primary time variable. All models were stratiﬁed by EPIC-participating center, to account for center effects re- lated to different recruitment and follow-up procedures, by sex, and by age at recruitment in 1-y categories and to reduce sen- sitivity to any violations of the proportional hazards assumption. Control for age in the models versus stratiﬁcation resulted in the same associations for coffee and tea intake. For the main anal- yses, quantiles for the different beverages were created based on the distribution of intake within each country, given that coffee and tea vary substantially by country (especially volume and concentration of coffee). Our decision to present combined coffee and tea intake was based on the hypothesis that caffeine may play a role in risk of glioma (as described in the discussion) and because of the lack of an estimated “caffeine” variable in this data set; on average, coffee and tea intake combined are estimated to contribute 90% of all caffeine intake (3). We also conducted analyses in which categories were based on absolute volumes consumed rather than country distributions. There are no established risk factors for brain cancer in the general pop- ulation, but we adjusted for BMI, smoking, and education in our ﬁnal models because these variables are common cancer risk factors. However, the inclusion of these variables did not noticeably affect the coffee or tea estimates. Furthermore, among women, we examined whether menopausal status, hor- mone replacement therapy, and oral contraceptive (OC) use were confounding factors. Tests for trend across categories were calculated by assigning the median value to each quartile of intake and entering this variable as a continuous term in the Cox regression models. All statistical analyses were carried out with SAS 9.1 (SAS Institute Inc, Cary, NC). RESULTS Daily coffee and tea intake varied substantially across Euro- pean countries participating in the EPIC cohort. The highest mean coffee consumption was reported in Denmark (798 mL/d) and the lowest in Italy (98 mL/d). For tea, the highest mean consumption was observed in the United Kingdom (532 mL/d) and the lowest in Spain (6.2 mL/d) (Table 1). Participants consuming higher volumes of coffee and tea tended to be slightly older, more educated, current smokers, and have a lower BMI (Table 2). Given that the range of intakes varied substantially by country, we used the distribution of intake for each country to create quantiles of coffee, tea, and combined coffee and tea (Table 3). In these analyses, we observed no associations for coffee or tea in relation to either glioma or meningioma risk, and the associa- tions were similar for men and women (data not shown). These results were similar when mutually adjusted for tea and coffee consumption in the same model (data not shown) or when controlled for hormone replacement therapy or OC use. All models with tea or coffee and tea combined excluded Norway because Norway did not have data on tea consumption. TABLE 1 Distribution of coffee and tea intake in the data set for this analysis [EPIC (European Prospective Investigation into Cancer and Nutrition)], by country Participants Coffee intake Tea intake Mean Median 10th–90th Percentile Mean Median 10th–90th Percentile mL/d mL/d Denmark 55,005 798 900 86–1600 302 85.7 0–900 Germany 49,478 428 392 48–864 129 21.4 0–398 Greece 25,615 181 140 11–354 9 0.5 0–11 Italy 44,507 98 90 6–180 39 5 0–150 Netherlands 36,288 552 500 125–1000 261 238 0–594 Norway1 35,217 416 360 60–780 — — — Spain 40,003 118 93 0–275 6.2 0 0–0 Sweden 48,679 468 400 107–800 76 0 0–250 United Kingdom 75,517 378 339 4–856 532 475 2–1140 1 No questions on tea in the dietary questionnaire. COFFEE, TEA, AND RISK OF BRAIN TUMORS 1147 by guest on May 11, 2013ajcn.nutrition.orgDownloaded from We also examined the association with coffee and tea intake by using absolute cutoffs measured in volume (mL/d) (data not shown). No associations were observed for coffee and tea intake and risk of meningioma or glioma. However, for glioma, a sug- gestion of a threshold effect was observed at ’100 mL/d. A comparison of those consuming 100 mL coffee and tea/d with those consuming ,100 mL/d showed a statistically signiﬁcant inverse association (HR: 0.66; 95% CI: 0.44, 0.97; P = 0.03). This inverse ﬁnding (hazard ratios: 0.21–0.81) was consistent across 6 of 7 countries, although individually the associations were not statistically signiﬁcant (Figure 1); the Netherlands was excluded because there were no cases in the lowest cate- gory. The association was slightly stronger in men (HR: 0.59; 95% CI: 0.34, 1.01) than in women (HR: 0.74; 95% CI: 0.42, 1.31). Further control for menopausal status or OC use in women did not inﬂuence the estimates. Control for soft drinks (which TABLE 2 Distribution of baseline characteristics by coffee and tea intake in the EPIC (European Prospective Investigation into Cancer and Nutrition) cohort Coffee and tea intake combined (mL/d) Characteristic ,100 100–599 600 P value1 n 52,423 167,735 190,151 Age at recruitment (y) 50.1 6 10.22 50.2 6 10.1 51.9 6 10.5 ,0.001 Male (%) 32.7 33.4 36.6 ,0.001 Smoking status (%) ,0.001 Never 59.4 48.8 42.9 Former 21.7 26.6 30.7 Current 18.9 24.6 26.4 BMI (kg/m2) 26.7 6 4.6 26.0 6 4.3 25.4 6 4.0 ,0.001 Education (%) ,0.001 None/primary 54.5 40.2 28.3 Technical/professional 13.3 23.1 32.0 Secondary 16.6 16.9 14.3 University 15.6 19.8 25.4 1 Derived by chi-square and one-factor ANOVA or Kruskal-Wallis test. 2 Mean 6 SD (all such values). TABLE 3 Hazard ratios (HRs) and 95% CIs for coffee and tea intake and risk of glioma and meningioma in the EPIC (European Prospective Investigation into Cancer and Nutrition) cohort1 Person-years Glioma Meningioma Cases HR2 (95% CI) Cases HR2 (95% CI) Coffee Quintile 1 682,062 57 1.0 (referent) 49 1.0 (referent) 2 737,184 83 1.21 (0.86, 1.71) 64 1.03 (0.71, 1.51) 3 706,526 77 1.15 (0.81, 1.63) 47 0.91 (0.60, 1.37) 4 644,138 62 1.27 (0.87, 1.84) 43 0.94 (0.61, 1.44) 5 723,334 64 0.98 (0.67, 1.41) 42 0.71 (0.46, 1.08) P for trend 0.68 0.23 Tea3 Quartile 1 688,486 70 1.0 (referent) 41 1.0 (referent) 2 1,219,802 125 0.76 (0.52, 1.10) 74 0.95 (0.58, 1.55) 3 665,444 54 0.80 (0.55, 1.17) 35 0.85 (0.52, 1.37) 4 706,738 82 1.05 (0.75, 1.48) 52 1.02 (0.66, 1.60) P for trend 0.59 0.83 Total coffee and tea3 Quintile 1 644,221 67 1.0 (referent) 49 1.0 (referent) 2 672,211 73 1.03 (0.73, 1.44) 39 0.72 (0.47, 1.10) 3 655,259 62 0.89 (0.63, 1.27) 33 0.61 (0.39, 0.95) 4 641,130 58 0.85 (0.60, 1.22) 39 0.73 (0.48, 1.12) 5 667,649 71 1.02 (0.72, 1.44) 42 0.71 (0.46, 1.10) P for trend 0.85 0.38 1 Quintile (or quartile for tea) cutoffs are based on the distribution of intake in each country. 2 HRs are stratiﬁed by age, country, and sex and adjusted for smoking status, BMI, and education. 3 Norway is excluded from these analyses (no data on tea). 1148 MICHAUD ET AL by guest on May 11, 2013ajcn.nutrition.orgDownloaded from may contain caffeine) did not modify the associations. No as- sociation was found for meningioma risk when the same cutoffs for coffee and tea intake combined (HR: 0.84; 95% CI: 0.50, 1.43) or when a higher cutoff of 200 mL/d (HR: 0.80; 95% CI: 0.50, 1.30) was used, given the suggested threshold in Table 3. Most of the coffee consumed in this population was caffeinated because the consumption of decaffeinated coffee was very low (mean: 32 compared with 217 mL caffeinated coffee/d; based on data from Northern Italy, United Kingdom, the Netherlands, Greece, and Germany). We did not have a total caffeine variable, but we were able to examine caffeinated coffee consumption in a separate analysis; we observed no associations for this variable (data not shown). No associations were observed for herbal tea intake and risk of glioma or meningioma when examined in the centers that had this information (data not shown). DISCUSSION In this large EPIC cohort, daily coffee and tea consumption of 100 mL was associated with a lower risk of glioma. The slightly stronger inverse association in men was consistent with a recent prospective cohort analysis in the United States (3). No associations were noted for coffee and tea and risk of meningioma. In a recent publication, 3 prospective cohorts from the United States were combined to examine coffee, tea, and caffeine intakes in relation to glioma risk (3). An inverse association was noted for combined coffee and tea consumption: 5 compared with 0–1 cups coffee and tea/d (HR: 0.60; 95% CI: 0.41, 0.87). A strong inverse association for caffeine intake was observed in men in a comparison of the top with the bottom quintile of total caffeine intake (HR: 0.46; 95% CI: 0.26, 0.81). Only 5 case-control studies have examined coffee or tea intake in relation to brain tumors, and the results have been inconsistent (5–9), although 2 of these studies reported inverse associations when estimating total caffeine consumption from beverages (7, 8). Coffee and tea both contain caffeine and many other com- pounds, some of which have antioxidant properties. In fact, coffee has a greater total antioxidant capacity (ie, cumulative capacity of food components to scavenge free radicals) than any given fruit or vegetable (16). Given that we did not observe an association between coffee and tea consumption and meningioma risk, it is possible that the effect of coffee, if causal, is acting late in the process of carcinogenesis by preventing tumor growth. A recent study showed that caffeine can slow the invasive growth of glioblastoma invarious in vitro assays by inhibiting inositol 1,4,5 triphosphate receptor subtype 3–mediated calcium release (17). Another potential mechanism that might be implicated in- volves the DNA repair protein O6-methylguanine-DNA methyl- transferase (MGMT). The coffee components kahweol and cafestol have been reported to increase MGMT activity in rat liver (18). Similarly, the tea component polyphenol (2)- epigallocatechin-3-gallate reactivates methylation-silenced genes, including MGMT, in cancer cell lines (19). The higher activation state of MGMT is thought to have a protective effect against the development of several cancer types, including colon cancer (20) and glioma (21). Furthermore, the MGMT genetic poly- morphism has been associated with glioma risk (22). Coffee and tea consumption patterns vary substantially by country. The type of brewing method used for coffee can affect the concentration of coffee substantially. For example, espresso coffee, created by forcing hot pressurized water through ground coffee, is signiﬁcantly more concentrated than is coffee that is gravity-brewed (eg, ﬁltered coffee). For example, 1 ﬂuid oz (29.6 mL) espresso contains 64 mg caffeine, whereas 8 oz (236.6 mL) gravity-brewed coffee contains 95 mg caffeine (23). Given these differences, the total volume consumed on a daily basis may not accurately reﬂect the actual consumption of caf- feine or the concentration of other compounds contained in tea and coffee. Thus, differences in brewing methods by country will automatically result in some measurement error when it comes to estimating caffeine or coffee and tea compounds; this may ex- plain why the results from the EPIC study were different from those reported in the US cohorts, in whom a lower risk of glioma was apparent at a higher intake (3). Nonetheless, given the large variability in coffee concentrations due to different brewing methods and inevitable measurement error, the consistency in the inverse associations found in these 2 studies is noteworthy and is unlikely to have been due to chance alone. This was the ﬁrst study to examine the relation between coffee and tea consumption and risk of meningioma and the second prospective study to examine these beverages in relation to glioma. By using a prospective design, we avoided many typical methodologic issues that arise in retrospective studies (eg, se- lection and recall bias) and others that often arise when dealing with brain tumors (eg, the need for use of proxy data for deceased cases and cognitive impairment in cases). In this study, we were able to examine coffee and tea consumption across different countries in Europe with a large range of intakes of both bev- erages, but we were unable to examine different brewing methods because thesedatawere not systematicallycollectedinall countries. Furthermore, as with other dietary studies, there was the potential for measurement error because food-frequency questionnaires measure average intake over a 1-y period and were only admin- istered once in the EPIC study, and we were unable to examine lifetime exposure to coffee and tea intake or different time periods. Given the few known risk factors for brain tumors, any confounding would have been due to as yet nonestablished risk factors. Overall, in this large prospective study, we reported a lower risk of glioma for men and women consuming 100 mL coffee FIGURE 1. Hazard ratios and 95% CIs for glioma associated with coffee and tea intake (.100 mL/d compared with 100 mL/d) by country in the EPIC (European Prospective Investigation into Cancer and Nutrition) study. The Netherlands was excluded from this plot because of small numbers in the low coffee and tea group. Norway was excluded because no data were collected on tea intake. COFFEE, TEA, AND RISK OF BRAIN TUMORS 1149 by guest on May 11, 2013ajcn.nutrition.orgDownloaded from and tea per day. No associations were observed for coffee or tea consumption and risk of meningioma. More studies are needed to conﬁrm these observations. The authors’ responsibilities were as follows—DSM: statistical analysis and writing of the manuscript; BS and BT: recruitment and follow-up of the Heidelberg cohort; HB and MS: recruitment and follow-up of the Potsdam cohort; AT and PL: recruitment and follow-up of the Greek cohort; CS, GM, RT, VK, and AM: recruitment and follow-up of the 5 Italian cohorts; HBB-d-M and MMR: recruitment and follow-up of the Bilthoven cohort; PHMP: recruitment and follow-up of the Utrecht cohort; MA, MJS, M-DC, EA, PJ, and MD: recruitment and follow-up of the 6 Spanish cohorts; K-TK and NW: recruitment and follow-up of the Cambridge cohort; NEA and TJK: recruitment and follow-up of the Oxford cohort; JM and EW: recruit- ment and follow-up of the Malmo cohort; GS and DE: recruitment and follow- up of the Norway cohort; BSM and LA: recruitment and follow-up of the Umea cohort; AJ and AO: recruitment and follow-up of the Copenhagen co- hort; KO and CD: recruitment and follow-up of the Aarhus cohort; and ER: coordination of the entire EPIC collaboration. All authors contributed to the interpretation of the results and preparation and approval of the ﬁnal manu- script. The sponsors were not inﬂuential in the study design, data collection or analysis, interpretation of results, or writing of the manuscript. None of the authors reported a conﬂict of interest. REFERENCES 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. GLOBOCAN 2008: cancer incidence and mortality worldwide. Lyon, France: IARC Press, 2010. 2. Arora RS, Alston RD, Eden TO, et al. Are reported increases in in- cidence of primary CNS tumours real? An analysis of longitudinal trends in England, 1979-2003. Eur J Cancer 2010;46:1607–16. 3. Holick CN, Smith SG, Giovannucci E, Michaud DS. Coffee, tea, caf- feine intake, and risk of adult glioma in three prospective cohort studies. Cancer Epidemiol Biomarkers Prev 2010;19:39–47. 4. Eﬁrd JT, Friedman GD, Sidney S, et al. The risk for malignant primary adult-onset glioma in a large, multiethnic, managed-care cohort: cigarette smoking and other lifestyle behaviors. J Neurooncol 2004;68:57–69. 5. Lee M, Wrensch M, Miike R. Dietary and tobacco risk factors for adult onset glioma in the San Francisco Bay Area (California, USA). Cancer Causes Control 1997;8:13–24. 6. Burch JD, Craib KJ, Choi BC, Miller AB, Risch HA, Howe GR. An exploratory case-control study of brain tumors in adults. J Natl Cancer Inst 1987;78:601–9. 7. Blowers L, Preston-Martin S, Mack WJ. Dietary and other lifestyle factors of women with brain gliomas in Los Angeles County (California, USA). Cancer Causes Control 1997;8:5–12. 8. Giles GG, McNeil JJ, Donnan G, et al. Dietary factors and the risk of glioma in adults: results of a case-control study in Melbourne, Australia. Int J Cancer 1994;59:357–62. 9. Hochberg F, Toniolo P, Cole P, Salcman M. Nonoccupational risk in- dicators of glioblastoma in adults. J Neurooncol 1990;8:55–60. 10. Costa J, Lunet N, Santos C, Santos J, Vaz-Carneiro A. Caffeine exposure and the risk of Parkinson’s disease: a systematic review and meta- analysis of observational studies. J Alzheimers Dis 2010;20(suppl 1): S221–38. 11. Santos C, Costa J, Santos J, Vaz-Carneiro A, Lunet N. Caffeine intake and dementia: systematic review and meta-analysis. J Alzheimers Dis 2010;20(suppl 1):S187–204. 12. Nkondjock A. Coffee consumption and the risk of cancer: an overview. Cancer Lett 2009;277:121–5. 13. Kempf K, Herder C, Erlund I, et al. Effects of coffee consumption on subclinical inﬂammation and other risk factors for type 2 diabetes: a clinical trial. Am J Clin Nutr 2010;91:950–7. 14. Percy C, Muir CS, Van Holten V. International classiﬁcation of diseases for oncology (ICD-O). 2nd ed. Geneva, Switzerland: World Health Organization, 1991. 15. Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr 2002;5(6B):1113–24. 16. Pellegrini N, Seraﬁni M, Colombi B, et al. Total antioxidant capacity of plant foods, beverages and oils consumed in Italy assessed by three different in vitro assays. J Nutr 2003;133:2812–9. 17. Kang SS, Han KS, Ku BM, et al. Caffeine-mediated inhibition of cal- cium release channel inositol 1,4,5-trisphosphate receptor subtype 3 blocks glioblastoma invasion and extends survival. Cancer Res 2010;70: 1173–83. 18. Huber WW, Scharf G, Nagel G, Prustomersky S, Schulte-Hermann R, Kaina B. Coffee and its chemopreventive components kahweol and cafestol increase the activity of O6-methylguanine-DNA methyl- transferase in rat liver–comparison with phase II xenobiotic metabolism. Mutat Res 2003;522:57–68. 19. Fang MZ, Wang Y, Ai N, et al. Tea polyphenol (-)-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines. Cancer Res 2003;63:7563–70. 20. Esteller M, Toyota M, Sanchez-Cespedes M, et al. Inactivation of the DNA repair gene O6-methylguanine-DNA methyltransferase by pro- moter hypermethylation is associated with G to A mutations in K-ras in colorectal tumorigenesis. Cancer Res 2000;60:2368–71. 21. Ohgaki H. Epidemiology of brain tumors. Methods Mol Biol 2009;472: 323–42. 22. Liu Y, Shete S, Hosking FJ, Robertson LB, Bondy ML, Houlston RS. New insights into susceptibility to glioma. Arch Neurol 2010;67: 275–8. 23. USDA. USDA National Nutrient Database for Standard Reference. Release 22. Washington, DC: US Department of Agriculture, Agricul- tural Research Service, 2009. 1150 MICHAUD ET AL by guest on May 11, 2013ajcn.nutrition.orgDownloaded from ss, 2010. 2. Arora RS, Alston RD, Eden TO, et al. Are reported increases in in- cidence of primary CNS tumours real? An analysis of longitudinal trends in England, 1979-2003. Eur J Cancer 2010;46:1607–16. 3. Holick CN, Smith SG, Giovannucci E, Michaud DS. Coffee, tea, caf- feine intake, and risk of adult glioma in three prospective cohort studies. Cancer Epidemiol Biomarkers Prev 2010;19:39–47. 4. Eﬁrd JT, Friedman GD, Sidney S, et al. The risk for malignant primary adult-onset glioma in a large, multiethnic, managed-care cohort: cigarette smoking and other lifestyle behaviors. J Neurooncol 2004;68:57–69. 5. Lee M, Wrensch M, Miike R. Dietary and tobacco risk factors for adult onset glioma in the San Francisco Bay Area (California, USA). Cancer Causes Control 1997;8:13–24. 6. Burch JD, Craib KJ, Choi BC, Miller AB, Risch HA, Howe GR. An exploratory case-control study of brain tumors in adults. J Natl Cancer Inst 1987;78:601–9. 7. Blowers L, Preston-Martin S, Mack WJ. Dietary and other lifestyle factors of women with brain gliomas in Los Angeles County (California, USA). Cancer Causes Control 1997;8:5–12. 8. Giles GG, McNeil JJ, Donnan G, et al. Dietary factors and the risk of glioma in adults: results of a case-control study in Melbourne, Australia. Int J Cancer 1994;59:357–62. 9. Hochberg F, Toniolo P, Cole P, Salcman M. Nonoccupational risk in- dicators of glioblastoma in adults. J Neurooncol 1990;8:55–60. 10. Costa J, Lunet N, Santos C, Santos J, Vaz-Carneiro A. Caffeine exposure and the risk of Parkinson’s disease: a systematic review and meta- analysis of observational studies. J Alzheimers Dis 2010;20(suppl 1): S221–38. 11. Santos C, Costa J, Santos J, Vaz-Carneiro A, Lunet N. Caffeine intake and dementia: systematic review and meta-analysis. J Alzheimers Dis 2010;20(suppl 1):S187–204. 12. Nkondjock A. Coffee consumption and the risk of cancer: an overview. Cancer Lett 2009;277:121–5. 13. Kempf K, Herder C, Erlund I, et al. Effects of coffee consumption on subclinical inﬂammation and other risk factors for type 2 diabetes: a clinical trial. Am J Clin Nutr 2010;91:950–7. 14. Percy C, Muir CS, Van Holten V. International classiﬁcation of diseases for oncology (ICD-O). 2nd ed. Geneva, Switzerland: World Health Organization, 1991. 15. Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr 2002;5(6B):1113–24. 16. Pellegrini N, Seraﬁni M, Colombi B, et al. Total antioxidant capacity of plant foods, beverages and oi