• Home
• About Cosic
  • European Network
• Background On Coffee
  • History of Coffee
  • Coffee Production
• Background on Caffeine
• Caffeine and Behaviour
  • Brief Summary
  • Alertness
  • Anxiety
  • Mental Performance
  • Sleep
  • Sports and Physical Performance
  • Habituation, Withdrawl, Tolerance, Reinforcement and Dependence
• Coffee and Health
  • Alzheimer's Disease
  • Asthma
  • Antioxidants
  • Bone Health
  • Cancer
  • Diabetes
  • Gastrointestinal Function
  • Gallstone Disease
  • Heart Disease
  • Coffee & Kidney Function/Fluid Balance
  • Liver Cirrhosis
  • Parkinson’s Disease
  • Pregnancy
  • Unsubstantiated Coffee and Health Links
  • Summary
• Decaffeinated Coffee - How it is produced
• Questions & Answers
• Glossary of Terms
• Contact Us
• Cosic Intranet
• PEC Intranet
• ISIC Intranet

Search

Find ANY word Find ALL words  Find EXACT phrase 

Site Search by PicoSearch

COFFEE AND CANCER

GENERAL

There are no intervention trials of effects of coffee consumption on risk of cancer at any site and consequently there is no cause effect evidence. By contrast, there are numerous case control and cohort studies of associations between coffee consumption and risk of cancer at various sites particularly the bladder, breast, colon, ovary, pancreas and kidney. Such studies of associations cannot prove cause effect relationships and are subject to confounding by other risk factors and to bias. In addition, coffee consumption might be a marker for some other aspect of lifestyle such as smoking which is a true cause of cancer.

Associations between coffee consumption and cancer risk have been reviewed at regular intervals. In 1997 the World Cancer Research Fund in association with the American Institute for Cancer Research concluded that “Most evidence on coffee suggests that coffee drinking has no relationship with cancer risk” (1). The authors of a 2000 scientific review wrote that “This updated and comprehensive overview of coffee and cancer epidemiology provides further reassuring information on the absence of any appreciable association between coffee intake and most common cancers, including cancer of the genital tract, digestive tract and of the breast” (4).

Accordingly there is no scientific reason for believing that moderate consumption of coffee increases the risk of developing cancer at any site.        

BLADDER AND LOWER URINARY TRACT CANCER

A 1991 review of associations between bladder cancer and coffee consumption identified 26 studies and analysed 22 of them (3). Sixteen studies demonstrated a higher risk of bladder cancer in coffee consumers. In 7 of these 16 studies the association was significant and in 3 there was evidence of a dose response relationship. There were no associations in the other 6 studies. When non-smokers were considered separately in 7 of these studies, the association weakened but persisted, suggesting that confounding by smoking is not the only explanation for the association. It was concluded that there was a weak positive association between risk of bladder cancer and coffee consumption but the possibility that this was due to bias or confounding could not be excluded.

A 2000 review of all types of study published since 1991 identified an additional 3 cohort studies and 12 case control studies (4). The authors stated in their abstract “Thus, a strong association between coffee drinking and bladder cancer can be excluded, although it is still unclear whether the weak association is causal or non-specific and due to some bias or confounding”. A 2001 review and meta-analysis identified 34 case control studies and 3 cohort studies (5). In agreement with previous studies, it was found that coffee consumption increased the risk of urinary tract cancer by approximately 20%.

Two meta-analyses of case control studies have been published. In 1993 thirty-five case control studies published between 1971 and 1992 were identified and 7 core studies selected for meta-analysis according to strict methodological criteria (6). The authors concluded that “the best available data do not suggest a clinically important association between the regular use of coffee and development of cancer of the lower urinary tract in men or women”. In 2000 a pooled analysis of 10 European case control studies attempted to eliminate confounding by cigarette smoking by considering non-smokers only (7). It was found that the risk of bladder cancer in coffee drinkers was no greater than in non-coffee drinkers unless consumption was ten cups or more per day. This is considerably greater than the average consumption in the United Kingdom of between 3 and 4 cups per day.

Although cohort studies have more robust designs than case control studies they have never been separately analysed. In the Californian Seventh Day Adventist Study, 52 cases of bladder cancer were identified in a study population of 34,198 but there was no significant association between coffee consumption and risk of disease (8). In a study of 7,995 Japanese American men living in Hawaii, 96 cases of bladder cancer were diagnosed and although coffee consumption was associated with an increase in bladder cancer risk this was not significant (9). In a study of almost 43,000 Norwegian men and women, 53 cases of bladder cancer were identified, but no significant associations between a coffee consumption greater than or equal to 7 cups per day and disease risk emerged either in men or women (10). The most recently published study of a subcohort of the Netherlands Cohort Study identified 569 bladder cancer cases in a study population of 3,123 men and women (11). After adjustment for all confounders, a non-significant association between bladder cancer risk and coffee consumption was observed in men but a significant inverse association in women. Hence there is no evidence from cohort studies that coffee consumption increases the risk of bladder cancer. 

In 2007, the World Cancer Research Fund Report brought further clarity to the debate surrounding coffee drinking and bladder cancer. In this report they state that 'The judgements of the previous report on coffee were practically the same as in this report, except that the previous report judged that drinking more than 5 cups per day was a possible cause of bladder cancer. The evidence now indicates that coffee is unlikely to have a substantial effect on risk of this cancer'.

BREAST CANCER 

A 1991 review of the literature identified 7 case control studies none of which showed any association between coffee consumption and risk of breast cancer (3). By 2000, a further three cohort studies and four case control studies had been published (4). The three cohort studies and three of the four case control studies were also unable to show any association between the risk of breast cancer and coffee consumption. By contrast, the fourth case control study from Finland demonstrated an inverse association between coffee consumption and risk of breast cancer in postmenopausal women (12). The authors concluded that “there is no appreciable relation between coffee and cancer of the breast”.

At least seven cohort studies of associations between coffee consumption and breast cancer risk have been published. Studies of 23,912 male and female Californian Seventh-day Adventists (13), 2,891 Norwegian women (14), 14,593 Norwegian women (15), 89,494 female nurses from the USA (16), 18,586 postmenopausal women living in New York State (17), 34,388 postmenopausal women living in Iowa (18) and 59,036 Swedish women (19) found no significant associations between risk of breast cancer and coffee consumption.

Male breast cancer also exists although it is a much rarer disease. A recent population-based case control study from Canada which compared 81 cases with 1905 controls found an inverse and statistically significant association between coffee consumption and risk of male breast cancer both before and after correction for confounders (20). The possibility that coffee consumption protects against male breast cancer awaits confirmation by other studies.    

There is no evidence whatever for an association between female breast cancer and coffee consumption.

COLORECTAL CANCER

Case control studies and cohort studies give different messages about associations between coffee consumption and risk of colorectal cancer. Out of the 12 “informative” case control studies identified in a 1991 review, 11 showed an inverse association between coffee consumption and the risk of colorectal cancer (3). In 5 of these case control studies the association was statistically significant and a significant dose response relationship was demonstrated in one of them. By contrast, none of the four cohort studies reviewed showed any evidence of an inverse association between coffee consumption and risk of colorectal cancer. 

A 2000 review identified 2 new cohort studies and 11 new case control studies published since 1991 (4). In their abstract the authors wrote that “Overall evidence on the coffee-colorectal cancer relation suggests an inverse association, since most case control studies found odds ratios below unity, particularly for colon cancer. The pattern of risk is less clear for cohort studies”. A population based case control study published since the 2000 review was published also reported an inverse association between coffee consumption and risk of colon cancer with evidence of a significant dose response relationship (21).

A similar conclusion was arrived at in a meta-analysis published in 1998 of the 12 case control studies and 5 cohort studies which met the selection criteria of the study (22). When the case control studies were analysed separately, then coffee consumption lowered the risk of colorectal cancer by 28%. However, when the 5 cohort studies were analysed separately, then coffee consumption lowered the risk of colorectal cancer by only 3%. When both types of study were combined the reduction in risk was 24% due to the larger number of case control studies. The lower risk of colorectal cancer in heavy coffee drinkers was observed in studies from Asia, Northern and Southern Europe and North America. The author suggested that ongoing cohort studies might help resolve the discrepancy between the results of case control studies and cohort studies.

Cohort studies have a stronger experimental design than case control studies. In a study of nearly 43,000 Norwegian men and women where 130 cases of colon cancer and 79 cases of rectal cancer were diagnosed, there were no significant associations between coffee consumption and disease risk (10). When 27,111 male Finnish smokers were studied, 106 cases of colon cancer and 79 cases of rectal cancer were identified but no associations between coffee consumption and disease risk were reported (23). In the most recently published study, 460 new cases of colorectal cancer were diagnosed in a cohort of 61,463 Swedish women but no association between risk of colorectal cancer and coffee consumption was shown (24).

The disagreement between the results of case control studies and cohort studies remains to be resolved. On the one hand, the results of prospective cohort studies showing no protective effect of coffee against colorectal cancer are more reliable than the results of retrospective case control studies showing a protective effect. On the other hand, there is a remarkable degree of consistency in the results of case control studies. It is hard to imagine how a methodological artefact could account for this consistency.

OVARIAN CANCER

Twelve case control studies of associations between consumption of coffee and risk of ovarian cancer have been published (25-36). Nine of these studies were unable to show any consistent or significant associations (25-33). However, two of these studies indicated a consistent and significant association between coffee consumption and risk of ovarian cancer was demonstrated (34, 35). A population-based case control study from the USA of 549 cases of ovarian cancer and 516 controls showed no significant association between coffee or caffeine consumption and risk of ovarian cancer (36).

However, when pre-menopausal and postmenopausal women were analysed separately, the consumption of coffee or caffeine was significantly associated with risk of ovarian cancer in pre-menopausal but not in postmenopausal women.

Two cohort studies have also examined associations between coffee consumption and ovarian cancer risk. A study of 23,912 Californian Seventh-day Adventists found no association between coffee consumption and fatal ovarian cancer (13). A study of 21,238 Norwegian women found 93 cases of ovarian cancer but no significant associations with coffee consumption (10).     

It can be concluded that there is no consistent evidence for a link between coffee or caffeine consumption and the risk of developing ovarian cancer.

PANCREATIC CANCER

A case control study from the USA published in 1981 suggesting a two to three fold increase in risk of pancreatic cancer associated with drinking three or more cups of coffee per day stimulated an enormous amount of research in this area (37). In 1987 a pooled analysis of nine epidemiological studies gave a relative risk of 1.3 for moderate coffee drinkers and 1.6 for heavy coffee drinkers when the original study was included in the analysis but only 1.2 and 1.4 when it was excluded (38). By 1990 when thirty epidemiological studies were reviewed it was concluded that the evidence did not support the hypothesis that coffee consumption increases the risk of pancreatic cancer (39). In 1991 twenty-six case control studies were reviewed and it was concluded that “the data are suggestive of a weak relationship between high levels of coffee consumption and the occurrence of pancreatic cancer, but the possibility that this is due to bias or confounding is tenable” (3). In addition none of the six cohort studies reviewed reported a significant association between pancreatic cancer risk and coffee consumption. By 2000 an additional twelve case control studies and four cohort studies had been published (4). Nine of these case control studies reported no association, one reported an inverse association and two reported a positive association. None of the four cohort studies reported any significant associations. It was concluded that “a strong association between coffee and pancreatic cancer can now be excluded; however, the presence of some moderate and inconsistent association may deserve further investigations”.

As has recently been pointed out, there is a particular reason for distrusting the results of case control studies on pancreatic cancer as poor survival leads to reduced participation rates by cases in interviews and consequently more interviews with surrogates (40). A clearer picture of the relationship between coffee consumption and pancreatic cancer is likely to emerge from a consideration of cohort studies only. 

At least thirteen cohort studies of associations between coffee consumption and pancreatic cancer risk have been published and the vast majority of these found no significant associations. Thus studies of 23,912 Californian Seventh-day Adventists (13), 50,000 USA college alumni (41), 7,355 Japanese men living in Hawaii (42), nearly 43,000 Norwegian men and women (10), 34,000 Californian Seventh-day Adventists (43), 122,894 men and women living in California (44), 17,633 American men (45), 13,979 elderly Americans (46), 175,000 Americans (47), 47,794 American men and 88,799 American women (40) and 12,204 Swedish women and 9,608 Swedish men (48) were all unable to show significant associations between coffee consumption and risk of pancreatic cancer. By contrast, a study of 265,118 Japanese men (49) and 33,976 postmenopausal women from Iowa (50) reported a significant positive association between pancreatic cancer risk and coffee consumption. Nevertheless the weight of cohort evidence remains firmly against the concept that coffee drinking increases the risk of developing pancreatic cancer.

It can be concluded that while some case control studies, particularly the earlier ones, suggest that coffee drinking is associated with an increase in the risk of pancreatic cancer, the vast majority of cohort studies do not support the idea. Since there are no intervention trials published on the effect of coffee drinking on the risk of pancreatic cancer then there is no evidence for a cause effect relationship.

RENAL CANCER

As noted on a 1991 review (3), out of four case control studies, three showed that coffee consumption was associated with a slight increase in risk of transitional cell cancers of the renal pelvis and ureter but none of the associations was significant. In addition, six case control studies and one cohort study failed to provide consistent evidence of an association between coffee drinking and adenocarcinoma of the kidney.

Subsequent studies have also failed to provide any consistent evidence for an association between coffee consumption and renal cancer. No significant associations were shown between coffee consumption and renal cancer in case control studies of 203 cases from the USA (51), 240 cases from Italy (52), 196 cases from France (53) and 518 cases from Canada (54). A cohort study from Norway was also unable to demonstrate any significant associations between renal cancer and coffee consumption (10). Finally, a case control study carried out in Australia, Denmark, Sweden and the United States which analysed 1,185 cases of renal cancer was also unable to show any consistent or significant associations nor any evidence for a dose response relationship (55). A 2000 review of these data concluded that “epidemiological data on the relation between coffee consumption and kidney cancer risk are reassuring” (4).

OCHRATOXIN A

Ochratoxin A is one of a family of mycotoxins produced by  the mould Penicillium verrucosum and by several species of Aspergillus including A. ochraceus, A. carbonarius and A. niger. Chemically it consists of a chlorinated isocoumarin moiety linked through a carboxyl group to L-phenylalanine via an amide bond. The major food contaminated by ochratoxin A is cereals but much lower levels of contamination may be found in grape juice and red wine, coffee, cocoa, nuts, spices and dried fruits, and other agricultural products subjected to conditions that result in mould growth. Ochratoxin A is an accepted nephrotoxin  and in animals is a carcinogen and is also teratogenic and immunotoxic. The Joint FAO/WHO Expert Committee on Food Additives (JECFA) has set a provisional tolerable weekly intake (PTWI) of 100 ng/kg body weight although the Canadian authorities have set a lower tolerable daily intake (TDI) of 1.5 to 5.7 ng/kg body weight and the European Commission a TDI of not more than 5 ng/kg body weight (56). 

Ochratoxin A production by Penicillium verrucosum is characteristic of cereals. By contrast, a recent study of 408 Brazilian coffee samples identified Aspergillus ochraceus as the major source of ochratoxin A followed by Aspergillus carbonarius (57). Aspergillus niger was an unimportant source. It was also found that there was little infection of coffee cherries while on the tree but infection occurred postharvest, the most likely sources being soil, equipment and drying yard surfaces. Improvements in good agricultural practices have reduced ochratoxin A contamination of coffee.

Processing also lowers the ochratoxin A content of coffee. Roasting dramatically lowers the ochratoxin A content of coffee by 50-90% (58), 30-90% (59) or 81% (60). It has also been reported that decaffeination lowers the ochratoxin A content of coffee by 92% (61).

It has recently been estimated that mean total intakes of ochratoxin A are 45 ng/kg body weight per week assuming a body weight of 60 kg (62). Cereals and wine contribute about 25 and 10 ng/kg body weight per week respectively whereas grape juice and coffee each contribute only 2-3 ng/kg body weight per week. Other foods such as dried fruits, beer, tea, milk, cocoa, poultry and pulses contributed less than 1 ng/kg body weight per week. These estimates underline the conclusion of the Ministry of Agriculture, Fisheries and Food that coffee is not a major source of ochratoxin A in the normal diet (63).

LIVER CANCER

In recent years studies have suggested that coffee drinking may be protective against the development of hepatocellular carcinoma independently of its aetiology. Coffee has been studied extensively in relation to other conditions affecting the liver and this is reported elsewhere on this site. In 2005 Japanese researchers (64) published their findings after conducting a large-scale population-based cohort study that confirmed a statistically significant inverse association between habitual coffee drinking and hepatocellular carcinoma. A further hospital based case-control study conducted in Italy (65) with 250 cases and 500 controls reached similar conclusions. Finally, the findings of these studies were further endorsed by Shimazu et al (66) who, using pooled analysis consisting of over 60,000 people, found a significant inverse association between coffee consumption and the risk of liver cancer.
Research has continued to investigate the inverse association between coffee consumption and liver cancer, with three large studies published in 2007. One of these, an Italian case-control study (67) explored the relationship between coffee drinking and hepatocellular carcinoma risk in a population whose coffee consumption is highly variable. The results support the hypothesis of a favourable effect of coffee on hepatocellular canrcinoma risk. A meta-analysis of four cohort and five case-control studies, involving 2260 cases and 239,146 controls, suggested that an increase in coffee consumption may reduce the risk of liver cancer. The authors point out that the mechanisms involved and the substances in coffee responsible for the relation remain to be elucidated (68). A further meta-analysis (69) of 10 studies covering both European and Japanese populations provided quantitave evidence of an inverse association between coffee drinking and liver cancer.These authors also point out that the mechanisms of action ivolved remain unclear at this time.

References:

1. World Cancer Research Fund. Food, Nutrition and the Prevention of Cancer: a Global Perspective, 1997.

2. Department of Health, Committee on Medical Aspects of Food Policy. Report on Health and Social Subjects, 48, p 130. HMSO: London, 1998. 

3. IARC. IARCMonographs: Evaluation of Carcinogenic Risks to Humans, Vol. 51, Coffee, Tea, Maté, Methylxanthines and Methylglyoxal. International Agency for Research on Cancer, Lyon, 1991.

4. Tavani, A. and La Vecchia, C. European Journal of Cancer Prevention, 9, 241-256, 2000. 

5. Zeegers, M.P.A. et al. International Journal of Epidemiology, 30, 353-362, 2001.

6. Viscoli, C.M. et al. Lancet, 341, 1432-1437, 1993.

7. Sala, M. et al. Cancer Causes and Control, 11, 925-931, 2000.

8. Mills, P.K. et al. American Journal of Epidemiology, 133, 230-239, 1991.

9. Chyou, P-H. et al. Annals of Epidemiology, 3, 211-216, 1993.

10. Stensvold, I. and Jacobsen, B.K. Cancer Causes and Control, 5, 401-408, 1994.

11. Zeegers, M.P.A. et al. Cancer Causes and Control, 12, 231-238, 2001.

12. Mannisto, S. et al. Journal of Clinical Epidemiology, 52, 429-439, 1999.

13. Snowden, D.A. and Phillips, R.L. American Journal of Public Health, 74, 820-823, 1984.

14. Jacobsen, B.K. et al. Journal of the National Cancer Institute, 76, 823-831, 1986.

15. Vatten, L.J. et al. British Journal of Cancer, 62, 267-270, 1990.

16. Hunter, D.J. et al. American Journal of Epidemiology, 136, 1000-1001, 1992. 

17. Graham, S. et al. American Journal of Epidemiology, 136, 1327-1337, 1992.

18. Folsom, A.R. et al. American Journal of Epidemiology, 138, 380-383, 1993.

19. Michels, K.B. et al. Annals of Epidemiology, 12, 21-26, 2002.

20. Johnson, K.C. et al. European Journal of Cancer Prevention, 11, 253-263, 2002.

21. Woolcott, C.G. et al. European Journal of Cancer Prevention, 11, 137-145, 2002.

22. Giovannucci, E. American Journal of Epidemiology, 147, 1043-1052, 1998.

23. Hartmann, T.J. et al. Nutrition and Cancer, 31, 41-48, 1998.

24. Terry, P. et al. Gut, 49, 87-90, 2001.

25. Hartge, P. et al. International Journal of Cancer, 30, 531-532, 1982.

26. Byers, T. et al. Journal of the National Cancer Institute, 71, 681-686, 1983.

27. Miller, D.R. et al. Ovarian cancer and coffee drinking In MacMahon B. and Sugimura, T. (eds) Coffee and Health (Banbury Report 17), Cold Spring Harbor, NY, CSH press, 1984. 

28. Miller, D.R. et al. International Journal of Epidemiology, 16, 13-17, 1987.

29. Trichopoulos, D. et al. International Journal of Cancer, 28, 691-693, 1981. 

30. Trichopoulos, D. et al. International Journal of Cancer, 36, 291-297, 1986.

31. Tzonou, A. et al. European Journal of Cancer and Clinical Oncology, 20, 1045-1052, 1984.

32. Cramer, D.W. et al. Obstetrics and Gynaecology, 63, 833-838, 1984.

33. Polychronopoulou, A. et al. International Journal of Cancer, 55, 402-407, 1993.

34. La Vecchia, C. et al. International Journal of Cancer, 33, 559-562, 1984.

35. Whittemore, A.S. et al. American Journal of Epidemiology, 128, 1228-1240, 1988.

36. Kuper, H. et al. International Journal of Cancer, 88, 313-318, 2000.

37. MacMahon, B. et al. New England Journal of Medicine, 304, 630-633, 1981.

38. La Vecchia, C. et al. International Journal of Cancer, 40, 309-313, 1987.

39. Gordis, L. Cancer Letters, 52, 1-12, 1990.

40. Michaud, D.S. et al. Cancer Epidemiology Biomarkers and Prevention, 10, 429-437, 2001.

41. Whittemore, A.S. et al. Journal of Chronic Disease, 36, 251-256, 1985.

42. Nomura, A. et al. Journal of the National Cancer Institute, 76, 587-590, 1986.

43. Mills, P.K. et al. Cancer, 61, 2578-2585, 1988.

44. Hiatt, R. et al. International Journal of Cancer, 41, 794-797, 1988.

45. Zheng, W. et al. Cancer Causes and Control, 4, 477-482, 1993.

46. Shibata, A. et al. International Journal of Cancer, 58, 46-49, 1994.

47. Friedman, G.D. and van den Eeden, S.K. International Journal of Epidemiology, 22, 30-37, 1993. 

48. Isaksson, B. et al. International Journal of Cancer, 98, 480-482, 2002.

49. Hirayama, T. et al. Japanese Journal of Clinical Oncology, 19, 208-215, 1989.

50. Harnack, L.J. et al. Cancer Epidemiology Biomarkers Prevention, 6, 1081-1086, 1997.

51. Maclure, M. and Willett, W. Epidemiology, 1, 430-440, 1990.

52. Talamini, R. et al. Cancer Causes and Control, 1, 125-131, 1990.

53. Benhamou, S. et al. International Journal of Cancer, 55, 32-36, 1993.

54. Kreiger, N. et al. Cancer Causes and Control, 4, 101-110, 1993.

55. Wolk, A. et al. International Journal of Cancer, 65, 67-73, 1996.

56. Walker, R. Advances in Experimental Medicine and Biology, 504, 249-255, 2002.

57. Taniwaki, M.H. et al. International Journal of Food Microbiology, 82, 173-179, 2003.

58. Micco, C. et al. Food Additives and Contaminants, 6, 333-339, 1989.

59. Wilkens, J. and Jorissen, U. In: 21^st Mycotoxin Workshop, Jena(BgVV), June 7-9, 1999.

60. Blanc, M. et al. Journal of Agricultural and Food Chemistry, 46, 673-675, 1998. 

61. Heilmann, W. et al. European Food Research Technology, 209, 297-300, 1999.

62. JECFA (Joint FAO/WHO Expert Committee on Food Additives), Safety evaluations of certain mycotoxins in food. WHO Food Additives Series 47. World Health Organisation, Geneva, 2001.

63. Ministry of Agriculture, Fisheries and Food, Surveillance of Ochratoxin A in Retail Coffee Products (Food Surveillance Information Sheet No. 73), London: Joint Food Safety and Standard Group, 1995.

64. Inoue, M. et al. Journal of the National Cancer Institute, 97, 2005.

65. Gellati, U. et al. Journal of Hepatology, 42, 2005.

66. Shimazu, T. et al. International Journal of Cancer, 116, 2005.

67. Montella, M. et al. International Journal of Cancer, Online/In Press, 2007.

68. Larsson, S & Wolk, A. Gastroenterology, 132, 2007.

69. Bravi, F. et al. Hepatology, Online/In Press, 2007.