select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to genetic study looking at alcohol and cancer

A study published in the International Journal of Cancer looks at alcohol metabolism genes and risks of site-specific cancers in Chinese adults.


Prof Daniel Nebert, Professor Emeritus in the Department of Environmental Medicine and Center for Environmental Genetics, University of Cincinnati College of Medicine, said:

“This study has a very large sample size (>150,000) and solid statistical approach — including sensitivity analyses, and adjustments for population stratification. Also, the study benefits from data collected from a single large prospective cohort (the China Kadoorie Biobank). Moreover, the study focused on only two single-nucleotide variants (SNVs), each with a strong a priori probability to be relevant.

“Many of the reported P-values in this study are very significant (<0.0001), which is stringent enough to confirm the associations; this advantage mainly benefited from the large sample size of the study. Because of this, the major purpose of this article is not to answer whether or not these SNVs are associated with the cancers (confirmed), but rather to estimate the effect-sizes (i.e., hazard ratios) for different types of cancers (i.e., quantifying the effect). Their results confirm an association between the causal effects of alcohol consumption on some cancers, with a possible interaction with the ALDH2 and ADH1B genotypes.

“These two SNVs each have what is called a “small effect-size” on the phenotype (which is cancer risk caused by alcohol consumption). By “small effect-size,” we are talking about 1/1,000th or 1/10,000th. In other words, in a cohort study of only 50,000 or 100,000 subjects, the effects of these SNVs would most likely not be detectable.”


Dr Michael Jones, Senior Staff Scientist in Genetics and Epidemiology at The Institute of Cancer Research, London, said:

“Alcohol is classified as a carcinogen by The International Agency for Research on Cancer (IARC/WHO).  These findings from the China Kadoorie Biobank Study support the evidence that alcohol is carcinogenic to humans.

“Interestingly, the study found that people who drank regularly despite having a genetic intolerance to alcohol were at greater risk of upper aero-digestive tract cancers – such head and neck cancer and oesophageal cancer.  This is possibly because they metabolised alcohol differently.

“It is important to note that the study was conducted in China, where the types of alcoholic drink and the pattern of drinking may be different to consumption elsewhere in the world.  In other words, results could differ across populations.”


Dr Stephen Burgess, statistician at the MRC Biostatistics Unit, University of Cambridge said:

“While our genetic code cannot easily be altered, it can inform us about the impact of factors that can be altered.  Two genetic variants that are common in individuals of Chinese ancestry affect how alcohol is processed, and lead to the East Asian “alcohol flushing response”, whereby some East Asians go red in the face after drinking alcohol.  This is due to inefficiency in breaking down acetaldehyde, a toxic chemical produced by the body as an intermediate step when digesting alcohol.  Acetaldehyde is a noted carcinogen (cancer-causing chemical).

“As there is inherent randomness in how genetic variation is passed on from generation to generation, genetic variants can be treated similarly to randomization in a clinical trial – they divide the population into groups (those with the variant, and those without it) which differ with respect to consequences of having the genetic variant, but not with respect to other unrelated factors.  This means that any difference in disease outcomes between these groups can be reliably attributed to the consequences of having the genetic variant.  In this case, that consequence is drinking more or less alcohol, and the disease outcomes are certain cancers – in particular, oesophageal cancer and head and neck cancer.  Associations of these cancer-related genetic variants with those specific cancer outcomes is an indication that alcohol is a causal risk factor for those cancer outcomes – drinking more alcohol increases one’s risk of oesophageal cancer and head and neck cancer.  Similar associations were not observed for liver, stomach, or lung cancers.

“The statistical methods used in this study are appropriate.  The strength of the method relies on the assumption that the genetic variants are distributed in the population randomly with respect to all possible competing risk factors (“confounders”) that may serve as alternative explanations for an association with cancer risk.

“One limitation of this investigation is that the harmful effect of alcohol on cancer appears strongest in people who cannot break down alcohol efficiently.  Inability to break down alcohol efficiently is a common trait in East Asian populations, but it is less common in European ancestry populations.  Further research is needed to determine whether a similar harmful effect of alcohol on cancer holds in European ancestry populations.”


Prof Paul Pharoah, Professor of Cancer Epidemiology, University of Cambridge, said:

“There is strong evidence from observational (non-randomised) studies that alcohol consumption causes an increased risk of head and neck, oesophagus, liver, bowel, and female breast cancer.  However, observational studies can be affected by confounding and so causality cannot be proven.  The aim of this study was to provide further evidence these associations are causal using an approach known as Mendelian randomisation.  This approach uses inherited genetic variation known to be associated with a risk factor of interest – in this case alcohol intake – and evaluates the association of the genetic variant(s) with the disease of interest.  Any observed gene-disease association is less prone to confounding bias than with observational studies and provides stronger evidence for a causal relationship between the risk factor and the disease.

“This study is a well-conducted study that has been carefully analysed.  However, the results do not add much to what we already know.  The association reported in this study between genetic variants associated with alcohol consumption and oesophageal cancer and head and neck cancer have been reported previously (and strongly suggest the association between alcohol consumption and these cancers is causal).  While no association was found for other cancers the power of the study to detect an association is quite limited because of the small number of cancers occurring during follow-up.  Thus, the negative (null) findings cannot be considered definitive.

“In brief, this study provides further evidence for a causal association between alcohol consumption and oesophageal cancer and head and neck cancer.  Other evidence suggests that alcohol consumption causes an increased risk of liver, bowel, and female breast cancer.  Limiting alcohol consumption is one of many ways to live a healthier lifestyle that helps reduce cancer risk.”


Prof Darren Griffin, Professor of Genetics, University of Kent, said:

“This is an 11 year long population study with excellent data backed up by solid analysis.

“Essentially, it reports that certain genetic variants predispose a higher risk of head and neck, oesophageal, and lung cancers (the so called “upper aero-digestive tract” area).

“The study design inherently isolates the fact that it is alcohol consumption (not other lifestyle factors) that seem to cause the cancer.

“Such studies have far reaching implications for lifestyle choices, however there are range of other ways in which alcohol can be damaging to the health.  A favourable genetic variant isn’t necessarily a passport for excessive and frequent binging.”



‘Alcohol metabolism genes and risks of site-specific cancers in Chinese adults: an 11-year prospective study’ by Pek Kei Im et al. was published in the International Journal of Cancer at 09:01 UK time on Thursday 20 January 2022.

DOI: 10.1002/ijc.33917



Declared interests

Prof Daniel Nebert: “I declare no interests, financial or otherwise.”

Dr Stephen Burgess: “Conflict of interest: I was senior author of parallel but independent work in 2020 that showed no strong associations between genetic predictors of alcohol consumption and cancer outcomes in European ancestry populations (subject to some degree of statistical uncertainty).  There was some evidence of associations with oesophageal cancer and with head and neck cancer, but not enough evidence to make definite conclusions, and little consistent evidence for other cancer outcomes.  This work is cited in the published manuscript (reference 41).  I have no other relevant conflict of interest to declare.”

Prof Paul Pharoah: “I have no conflicts of interest.”

None others received.

in this section

filter RoundUps by year

search by tag