Alcohol and cancer: a position statement from Cancer Council Australia

Alcohol use is widespread in Australia and has had a dominant role in defining Australian culture for more than 200 years.1,2 However, it is also an important cause of illness, injury and death, whether resulting from short-term episodes of intoxication or from long-term, chronic use.3

Addressing the health and social damage resulting from risky drinking is one of the three key priority areas identified by the Australian National Preventative Health Taskforce.4 Levels of harm from alcohol use are increasing, and a range of policy measures have been proposed to address the current drinking “culture” in Australia.4

In this article, we provide a brief overview of the evidence concerning alcohol use and cancer, and outline the current Cancer Council Australia (CCA) recommendations on alcohol consumption. The consensus process for developing this position statement is described in Box 1, and a summary of the key evidence-based points is provided in Box 2.

It has been known for more than 20 years that long-term chronic use of alcohol can cause cancer. In 1988, the International Agency for Research on Cancer stated that the “occurrence of malignant tumours of the oral cavity, pharynx, larynx, oesophagus and liver is causally related to the consumption of alcoholic beverages” and classified alcoholic beverages as Group 1 carcinogens — known to cause cancer in humans.5 Ethanol, the chemical present in all alcoholic beverages that induces the altered physical and mental responses experienced with alcohol use, has also been listed as a Group 1 carcinogen.6

The most recent comprehensive review of the scientific evidence by the World Cancer Research Fund (WCRF) and the American Institute for Cancer Research (AICR) concluded that there is convincing evidence that alcohol is a cause of cancer of the mouth, pharynx, larynx, oesophagus, bowel (in men) and breast (in women), and probable evidence that alcohol increases the risk of bowel cancer (in women) and liver cancer.7 Convincing and probable are the two highest levels of evidence set by the WCRF and AICR, which identify a causal relationship between a particular aspect of food, nutrition, physical activity or body composition, and cancer.7 Scientific research is continuing to identify other cancers that could be associated with alcohol use. For example, there is some evidence that heavy alcohol consumption may be associated with a higher risk of prostate cancer.8,9

There is a dose–response relationship between alcohol and cancer risk for men and women, with studies showing that the risk of cancer increases with increasing consumption of alcohol on a regular basis.7,10-12

There are a number of biological mechanisms that may explain alcohol’s contribution to cancer development. Ethanol may cause cancer through the formation of acetaldehyde, its most toxic metabolite. Acetaldehyde has mutagenic and carcinogenic properties, and bonds with DNA to increase the risk of DNA mutations and impaired cell replication.13,14 Ethanol may also cause direct tissue damage by irritating the epithelium and increasing the absorption of carcinogens through its effects as a solvent.7 In addition, alcohol can increase the level of hormones such as oestrogen, thereby increasing breast cancer risk,7 and increase the risk of liver cancer by causing cirrhosis of the liver, increased oxidative stress, altered methylation and reduced levels of retinoic acid.13 Lifestyle factors such as smoking, poor oral hygiene, and certain nutrient deficiencies (folate, vitamin B6, methyl donors) or excesses (vitamin A/β-carotene) owing to poor diet or self-medication may also increase the risk for alcohol-associated tumours.13

For some cancers, the combined effects of drinking alcohol and smoking tobacco greatly exceed the risk from either factor alone. Smoking and alcohol together have a synergistic effect on upper gastrointestinal and aerodigestive tract cancer risk.15 Compared with non-smoking non-drinkers, the approximate relative risks for developing mouth and throat cancers are up to seven times greater for people who smoke tobacco, up to six times greater for those who drink alcohol, but more than 35 times greater for those who are regular heavy users of both substances (consuming more than four alcoholic drinks and smoking 40 or more cigarettes daily).16 The synergistic effect of alcohol and smoking has been estimated to be responsible for more than 75% of cancers of the upper aerodigestive tract in developed countries.16

The relationship between alcohol consumption and body weight and fat is complex and appears to vary with sex and drinking pattern.17 From a nutritional viewpoint, alcoholic drinks represent “empty kilojoules” — that is, alcoholic drinks are high in kilojoules but low in nutritional value, especially when added to sugary mixer drinks. Alcohol itself has a comparatively high energy content (29 kJ/g) compared with other macronutrients.18

If people drink alcohol in addition to their normal dietary intake — that is, without a compensatory reduction in energy intake — they are liable to gain weight. Alcohol provides extra kilojoules, and slows fat and carbohydrate oxidation. On the other hand, if drinking replaces healthy eating patterns, it can lead to nutritional deficiencies and serious illness.17,19

Therefore, as well as being a direct cause of several cancers, alcohol might also contribute indirectly to those cancers associated with excess body fatness. There is convincing evidence that body fatness increases the risk of cancers of the oesophagus, pancreas, bowel, breast (in postmenopausal women), endometrium and kidney, and probable evidence that body fatness increases the risk of gallbladder cancer.7

Earlier research reporting that low-to-moderate levels of alcohol consumption might reduce the incidence of coronary heart disease may be flawed.20 For example, misclassification error may be a factor in studies in which the category of non-drinkers includes former drinkers who might have stopped drinking for reasons such as ill health or becoming older.21 It might reasonably be assumed that this population would be more likely to have coronary heart disease.21 Other reviews have suggested that unmeasured confounding in epidemiological studies of alcohol and heart disease is likely to be widespread, and that it is almost impossible to account for this confounding without randomised controlled trials.22-24

The putative benefits of moderate alcohol consumption on heart disease appear to be confined to middle-aged and older people.25 However, the ongoing debate over the potential impact of uncontrolled confounders on estimates of the size of the cardioprotective effect, and whether or not moderate alcohol consumption should be recommended for protection against heart disease, is difficult to resolve in the absence of randomised controlled trials. Acknowledging these issues, the World Health Organization stated in 2007 that “from both the public health and clinical viewpoints, there is no merit in promoting alcohol consumption as a preventive strategy”.20 In Australia, the National Heart Foundation explicitly advises against the consumption of red wine and other types of alcoholic drinks for the prevention or treatment of heart disease.26

Several estimates of the numbers of cases of cancer attributable to alcohol use in Australia have been calculated using different methods.4,27-29 However, these calculations predate the confirmation of alcohol use as a convincing cause of bowel cancer in men. Because the incidence of bowel cancer in Australia is high,30 calculations which exclude bowel cancer are likely to lead to a substantial underestimate of the true burden of alcohol-caused cancer in Australia.

In order to estimate cancer incidence attributable to alcohol use in Australia, a set of attributable fractions developed by the WCRF and AICR for cancers (including bowel cancer) associated with alcohol for the United Kingdom31 was applied to Australian cancer incidence data for 200528 (Box 3). Of the four preventability estimates calculated (United States, UK, Brazil and China), exposure data for Australia (39%) most closely matched the UK estimates (24%). The other countries had a substantially higher proportion of the population who did not drink alcohol (US, 63%; Brazil, 79%; China, 91%).31,32

Using this method, it is estimated that 5070 cases of cancer (or 5% of all cancers) are attributable to long-term chronic use of alcohol each year in Australia. This figure includes cancers for which there is convincing evidence that alcohol use increases the risk of disease. When cancers for which the risk is probably increased by alcohol use are included, the tally rises to 5663 (or 5.6% of all cancers).

Alcoholic drinks and ethanol are carcinogenic to humans.5,6 There is no evidence that there is a safe threshold of alcohol consumption for avoiding cancer, or that cancer risk varies between the type of alcoholic beverage consumed.7

CCA recommends that to reduce their risk of cancer, people limit their consumption of alcohol, or better still avoid alcohol altogether. For individuals who choose to drink alcohol, consumption should occur within the National Health and Medical Research Council guidelines.33 CCA’s key recommendations are outlined in Box 4.

CCA is a strong advocate for evidence-based action to reshape social attitudes concerning drinking, and to reduce the burden of morbidity and mortality caused by alcohol use. These issues are addressed in policy statements adopted by CCA, available from http://www.cancer.org.au/Healthprofessionals/PositionStatements/alcohol.htm.