Does drinking alcohol increase the risk for pancreatic cancer? While researchers have long suspected it does, the evidence has remained inconsistent – until now, with a global study of more than 2m people boosting the case that a link exists. It comes as experts have reiterated the associated carcinogenic effects of alcohol, among other risks.
Medscape reports that the study, which pooled data from 30 prospective cohorts, found that daily alcohol intake was associated with a “modest” increased risk for pancreatic cancer in both women and men, regardless of smoking status.
However, the extent of the risk depended somewhat on how the researchers modelled alcohol intake. One model, which mapped continuous increases in alcohol consumption, suggested there is no safe dose of alcohol – any amount can increase the risk for pancreatic cancer, though only by 3% for every additional 10g of alcohol per day, or about two-thirds of a standard drink.
The other model, which compared risk by alcohol volume categories, found that the risk does not become significant until a certain alcohol threshold – about two to three drinks per day for men and one to two for women.
“Still, overall, our findings provide new evidence that pancreatic cancer may be another cancer type associated with alcohol consumption, a connection that has been underestimated until now,” said the study’s senior author Pietro Ferrari, PhD, head of the Nutrition and Metabolism Branch at International Agency for Research on Cancer.
Co-author Jeanine Genkinger, PhD, MHS, had a stronger take on the findings. “I think this shows that alcohol use is a robust risk factor for pancreatic cancer,” said Genkinger, associate Professor of Epidemiology, Columbia University Mailman School of Public Health, New York City, noting that even more moderate drinking levels – no more than one drink for women and two for men – might be enough to boost pancreatic cancer risk.
How much of a risk?
The latest data, published in PLOS Medicine, comes at a time of increased attention to the alcohol-cancer link.
Earlier this year, then-US Surgeon General Vivek Murthy, MD, issued an advisory calling for cancer warnings to be added to alcohol labels. Major cancer organisations have determined alcohol to be an established risk factor for seven cancer types – those of the oral cavity, larynx, pharynx, esophagus, liver, breast and colon/rectum.
Despite the strong suspicion that drinking alcohol also contributes to pancreatic cancer risk, this aggressive cancer has not yet made the official list.
The major reason is that the evidence surrounding an alcohol-pancreatic cancer link has been deemed “inconsistent”, “suggestive” and “inconclusive” by expert panels. Studies have been hampered by difficulties separating alcohol use from smoking, a known risk factor for pancreatic cancer, as well as varying findings by alcohol type and geographic location.
In addition, certain studies highlighting a link have indicated that any association between alcohol and pancreatic cancer is driven only by more extreme drinking habits – more than four drinks a day, and sometimes as many as nine drinks.
The latest analysis, Genkinger said, helped clarify uncertainty surrounding the alcohol-pancreatic cancer link, which was especially important for “a disease where we don’t have that many modifiable risk factors”.
The findings are based on cohorts spanning four continents, all part of the Pooling Project of Prospective Studies of Diet and Cancer. Just under 2.5m cancer-free participants were recruited between 1980 and 2013 (median age, 57 years), of whom 70% were alcohol drinkers, 47% were never-smokers, and 64% were alcohol drinkers and never smokers.
Most study participants were from North America (60%), followed by Europe or Australia (32%) and Asia (8%).
Alcohol intake was modelled in two ways: continuously for every 10 g/d increase and by volume threshold, using 0.1 to < 5 g/d as the reference for non-drinkers.
For context, in the US, the amount of alcohol in a standard drink is defined as 14g of pure alcohol – equivalent to a 340ml can of beer, a 150ml glass of wine, or a 44ml shot glass of distilled spirits.
Over a median of 16 years, the researchers observed 10 067 incidents of pancreatic cancers. In the continuous model, the risk for pancreatic cancer rose by 3% for every additional 10 g of alcohol consumed per day (hazard ratio [HR], 1.03; 95% CI, 1.02-1.04).
This association remained consistent and significant among women and men (HR, 1.03 for both), current smokers (HR, 1.03), former smokers (HR, 1.02), and never-smokers (HR, 1.03), and across cohorts from Australia, Europe, and North America (HR, 1.03 for all), though not Asia (HR, 1.00).
The research team also found evidence that the type of alcohol mattered: Alcohol from beer and liquor/spirits was associated with a significantly increased risk for pancreatic cancer (HR, 1.02 and 1.04, respectively) but alcohol from wine was not (HR, 1.00).
This finding is in line with some previous studies suggesting that wine may have a different relationship with cancer risk compared with other alcoholic beverages.
But Genkinger pointed out that this finding could “reflect the ways in which people tend to drink different types of alcohol”. Wine, she noted, is often part of a meal, and people who favour wine may be less likely to binge drink than those who typically choose other types of alcohol.
This study, however, did not survey participants about specific drinking patterns, including binge-drinking. In the threshold model, however, the increased risk only became significant once alcohol intake reached a certain level.
For women, drinking one to two standard drinks per day raised their risk for pancreatic cancer by 12% compared with little to no drinking. For men, the threshold was a little higher: consuming two to four drinks a day was associated with a 15% increase in risk, whereas drinking more than that was tied to a 36% greater risk.
Overall, this research contributes to the growing body of evidence that pancreatic cancer should be added to the official alcohol-cancer risk list, according to Alison Klein, PhD, MHS, professor of oncology, pathology, and epidemiology, at Johns Hopkins School of Medicine, Baltimore, who was not involved in the research.
Having the Conversation
The recent Surgeon General’s advisory encouraged clinicians to inform their patients that drinking is a cancer risk factor, something unknown to most Americans, according to survey findings.
“I think this study is a good reminder to talk to our patients about their alcohol use,” said Edward Thomas Lewis III, MD, an addiction psychiatrist and clinical assistant professor at the Medical University of South Carolina, Charleston.
Providers can take opportunities for those discussions during routine care, such as when prescribing a medication that can interact with alcohol or when a patient’s health condition, such as high blood pressure or heart disease, can be exacerbated by drinking.
What’s challenging, Lewis said, was advising patients on what level of drinking is “Okay”, given that even lower levels of alcohol consumption – around one to two drinks per day – may carry some risk. “There is no zero-risk alcohol use,” Lewis said. But, he added, people do not necessarily have to abstain to see benefits, either.
“So it may be that a patient…is able to reduce their alcohol consumption by two or three standard drinks over a week. That still has a positive net effect,” he said.
Upping the risks
The US Surgeon General’s recommendations – to update the warning label on alcohol to include cancer, counselling patients about alcohol as a cancer risk factor, and promoting alcohol intervention strategies in clinical settings – have also been lauded by Yale School of Medicine experts, who have reiterated, in JAMA, the associated carcinogenic effects of alcohol, among other risks.
Kenneth Morford, Jeanette Tetrault and Patrick O’Connor write:
Alcohol-related cancers include cancers of the mouth, pharynx, larynx, oesophagus, breast, colorectum, and liver – in 2020, some 740 000 cancer cases globally (4.1%) were attributable to alcohol consumption, mostly in men (76.7%).
In the US, alcohol was identified as the third-leading modifiable cancer risk factor after cigarette smoking and excess weight, contributing to 5.4% of cases and 4.1% of deaths among adults 30 or older in 2019.
The primary carcinogenic effects of alcohol (ethanol) are attributed to its metabolite acetaldehyde, which can disrupt DNA synthesis, repair, and methylation.
Carriers of the ALDH2* variation, most prevalent among East Asian people, metabolise acetaldehyde slower, causing greater carcinogenic effects at lower amounts of alcohol. Additionally, alcohol can produce inflammation and oxidative stress.
Heavy alcohol use (five or more drinks/d), typically over at least five to 10 years, can cause cirrhosis, a risk factor for hepatocellular carcinoma, while alcohol also increases oestrogen levels, increasing breast cancer risk, and acts as a solvent for other carcinogens, such as those in tobacco, increasing the risk of oral and gastrointestinal cancer.
Other proposed carcinogenic mechanisms of alcohol include disruption of one-carbon and folate metabolism essential for DNA methylation and synthesis; altered retinoid metabolism affecting cell growth, differentiation, and apoptosis; reduced immune function; and dysbiosis of the oral and intestinal microbiome, increasing acetaldehyde levels and inflammation.
Dose-response relationship
Cancer risk increases with higher amounts of alcohol consumption. Although some studies categorise drinking as light, moderate, and heavy, these quantities are not standardised, and definitions vary across studies.
Additionally, the definition of a standard drink and alcohol consumption levels differ across countries. Because evidence is based on observational studies, risk of bias affects the evaluation of these associations, particularly at lower consumption levels.
Potential confounding factors include cigarette smoking, body mass index, physical activity, diet, and education. Additionally, abstainer bias, which includes those who have stopped drinking or never drank due to health conditions, further complicates analyses when these individuals are included in the comparison group.
Other sources of potential bias include under-reporting of alcohol consumption and selection bias.
Heavy alcohol consumption is associated with the greatest risk for all seven cancer types. At lower levels, conclusions from meta-analyses and expert reports often differ, depending on cancer type.
In a 2015 meta-analysis of 572 case-control and cohort studies published between 1956 and 2012, drinking more than 50g of alcohol (approximately >3.5 drinks) per day, compared with not drinking or occasional drinking, was associated with increased risk of cancers of the mouth and pharynx, larynx, oesophagus, female breast, colorectum, and liver.
The 2018 World Cancer Research Fund (WCRF) Third Expert Report, a comprehensive synthesis that included the 2015 meta-analysis, found strong evidence supporting an increased risk of colorectal cancer associated with consuming approximately greater than or equal to 30 g (≥two drinks) per day and of liver cancer associated with consuming approximately greater than or equal to 45 g (≥three drinks) per day.
Low to moderate consumption
A 2025 report by the National Academies of Sciences, Engineering, and Medicine (NASEM) systematically reviewed 23 cohort and two case-control studies, published between 2010 and 2014, that focused specifically on alcohol consumption of less than or equal to 28 g/d (≤two drinks/d) in men or less than or equal to14 g/d (≤one drink/d) in women and cancer risk.
The NASEM report only considered studies that used “never drinkers” in the comparison group, unlike the 2015 meta-analysis, which included individuals who previously or occasionally consumed alcohol in the comparison group.
However, restricting studies to never drinkers yielded few studies and therefore limited evidence.
In the NASEM report, drinking less than or equal to 14 g/d (≤one drink/d) in women was associated with a 10% increased breast cancer risk based on a meta-analysis of four cohort studies (n ≥100 937) and a 5% increased risk for every 10 to14 g of higher daily consumption (starting at 0 g/d) based on a meta-analysis of seven cohort studies (n ≥409 592).
For cancers of the oral cavity, pharynx, larynx, and oesophagus, evidence was insufficient to support an association between alcohol consumption of less than or equal to 28 g/d (≤two drinks/d) in men or less than or equal to 14 g/d (≤one drink/d) in women.
The 2015 meta-analysis similarly reported positive associations between breast cancer risk and both light (≤12.5 g/d [approximately ≤one drink/d]) and moderate (>12.5 to ≤50 g/d [approximately >one to ≤3.5 drinks/d]) alcohol consumption, and increased colorectal cancer risk with moderate, but not light, consumption.
In contrast to the NASEM report, the 2015 meta-analysis reported that any consumption level was associated with an increased risk of oral, pharyngeal, and oesophageal cancers in a dose-response manner.
However, when cohort studies were analysed separately from case-control studies, associations were no longer statistically significant for light and moderate drinking with oral and pharyngeal cancers or light drinking with oesophageal cancer.
The 2018 WCRF concluded that convincing evidence supports increased risk of cancers of the mouth, pharynx, larynx, oesophagus, and female breast with any amount of alcohol consumption, following a dose-risk relationship with every 10-g increase in alcohol consumed per day starting at zero intake.
It also reported increased risk at any alcohol consumption level for oestrogen-receptor-positive, but not oestrogen receptor–negative breast tumours, with the strongest evidence for postmenopausal breast cancer.
Clinical implications
Clinicians should discuss with patients the cancer risks associated with alcohol use. For people who choose to drink, the Dietary Guidelines for Americans recommend not exceeding two drinks daily for men and one for women.
However, even at these levels, cancer risk may be increased for patients who smoke, carry the ALDH2*2 genetic variation (evident by a flushing response to alcohol), have a family history of alcohol-related cancer or alcohol use disorder (AUD), or have other alcohol-related diseases associated with cancer, like liver disease.
Any daily alcohol consumption is associated with certain cancers, and the risk of cancer increases with higher alcohol intake. Patient education, routine alcohol use screening, and AUD treatment, when indicated, can decrease these risks and promote overall health.
Medscape article – Alcohol and Pancreatic Cancer: New Evidence About Risk (Open access)
JAMA Network article – Alcohol and cancer risk (Open access)
See more from MedicalBrief archives:
Surgeon-General wants cancer warnings for alcohol
400 UK cancer cases daily from junk food, alcohol and sunbathing
Light to moderate alcohol may reduce risks of cancer, CVD and mortality
Alcohol consumption linked to rising cancer cases – US study
Alcohol consumption associated with nearly 5% of US cancer cases