There is strong evidence that alcohol causes seven types of cancer and probably others, a review has concluded.
A study of existing research found strong evidence of a direct, harmful effect of drinking, even though scientists are unsure of the exact biological reasons why alcohol causes cancer.
Writing in the journal Addiction, Jennie Connor, from the University of Otago in New Zealand, said alcohol is estimated to have caused about half a million deaths from cancer in 2012 alone - 5.8% of cancer deaths worldwide.
The highest risks are from heavy drinking, but even people who drink at low levels are at risk.
Her review linked alcohol to cancer of the mouth and throat, larynx, oesophagus, liver, colon, bowel and breast.
She said: "There is strong evidence that alcohol causes cancer at seven sites, and probably others.
"Confirmation of specific biological mechanisms by which alcohol increases the incidence of each type of cancer is not required to infer that alcohol is a cause."
She said that based on current evidence, there is no safe level of drinking with respect to cancer though the risks are reduced for some cancers when people stop drinking.
She added that the supposed health benefits of drinking - such as red wine being good for the heart - were "seen increasingly as disingenuous or irrelevant in comparison to the increase in risk of a range of cancers".
In January, the UK's chief medical officers said no level of regular drinking is without risks to health.
Publishing a raft of recommendations, they said men should consume no more than 14 units of alcohol per week, down from the previous 21 units, bringing them into line with the recommendation for women.
Modelling for the study showed that, compared with non-drinkers, women who regularly drink two units a day have a 16% increased risk of developing breast cancer and dying from it.
Those who regularly consume five units a day have a 40% increased risk.
For every 1,000 women who do not drink, 109 will develop breast cancer. This rises to 126 women for those who drink 14 units or less per week, and 153 women for those who drink 14 to 35 units a week.
Scientists are still researching how alcohol can lead to cancer. One theory is that alcohol damages DNA.
Susannah Brown, science programme manager for the World Cancer Research Fund (WCRF), said of the new work: "Many people believe that alcohol consumption is only linked to liver cancer, but this review confirms the findings of our Continuous Update Project that alcohol is strongly linked to an increased risk of a number of different cancers, including two of the most common - bowel and breast cancer.
"Among other evidence, we see the risk increasing as the amount of alcohol consumed increases, and we agree that there is solid evidence to conclude that alcohol consumption directly causes cancer.
"For cancer prevention, we have long recommended that people should not drink alcohol at all, but we understand that this can be easier said than done."
The WCRF has previously said that drinking three alcoholic drinks or more per day increases the risk of stomach cancer
It also found strong evidence for a link with other cancers, including mouth and throat, liver, bowel and breast.
Elaine Hindal, chief executive officer from the industry-funded alcohol education charity, Drinkaware, said: "We know that around 3.5 million middle-aged men are drinking more than the low risk guidance of 14 units or 6 pints of 4% beer per week. It is why we have launched our 'Have a little less, feel a lot better' campaign to target those who could be storing up serious health problems for the future."