Critique 274 – Multi-ancestry study of the genetics of problematic alcohol use in over 1 million individuals


Zhou H; Kember RL; Deak JD; Xu H; Toikumo S; Yuan K; Lind PA; et al (48 authors)


Nature Medicine (2023) https://doi: 10.1038/s41591-023-02653-5

Author’s Abstract

Background Problematic alcohol use (PAU), a trait that combines alcohol use disorder and alcohol-related problems assessed with a questionnaire, is a leading cause of death and morbidity worldwide.

Methods Here we conducted a large cross-ancestry meta-analysis of PAU in 1,079,947 individuals (European, N = 903,147; African, N = 122,571; Latin American, N = 38,962; East Asian, N = 13,551; and South Asian, N = 1,716 ancestries).

Results We observed a high degree of cross-ancestral similarity in the genetic architecture of PAU and identified 110 independent risk variants in within- and cross-ancestry analyses. Cross-ancestry fine mapping improved the identification of likely causal variants. Prioritizing genes through gene expression and chromatin interaction in brain tissues identified multiple genes associated with PAU. We identified existing medications for potential pharmacological studies by a computational drug repurposing analysis. Cross-ancestry polygenic risk scores showed better performance of association in independent samples than single-ancestry polygenic risk scores. Genetic correlations between PAU and other traits were observed in multiple ancestries, with other substance use traits having the highest correlations.

Conclusions This study advances our knowledge of the genetic aetiology of PAU, and these findings may bring possible clinical applicability of genetics insights – together with neuroscience, biology and data science – closer.

Forum comments

Problematic alcohol use (PAU) contributes to approximately 6% of annual morbidity and mortality in the USA, for example, and similarly worldwide (Stahre et al. 2014). It is a chronic psychiatric disorder characterised by harmful consumption patterns leading to negative emotional, physical and social ramifications. Because of the interaction of genetics and environment, a person cannot be born with an alcohol use disorder (NIAAA, 2008). Although people can have genes that predispose them to developing an alcohol use disorder, genetics only account for approximately half of a person’s overall risk (NIAAA, 2008). The rest of these predispositions come from the social and environmental factors that a person encounters throughout their childhood and life (Edenber and Foroud, 2013; NIAAA, 2008).

This impressive genetic study published in the prestigious journal Nature Medicine concerned an extended follow-up study on the correlations between genes and their variants and the outcome Problematic Alcohol Use (PAU) (Zhou et al., 2023). PAU consists of alcohol use disorder and alcohol-related problems assessed with a questionnaire (AUDIT-P). Interesting parts in this study are the extension into cross-ancestry analysis and the study of targets for drugs and medication based on these genetic analyses.

Multi-ancestry genetic analysis (cross-ancestry analysis) seems to be a new and complex way to identify new genes that may contribute to the aetiology of a specific disease. Whereas the first study appeared in 2014 on the genes underlying the susceptibility to glaucoma, 46 studies were published in 2023 alone on various disease outcomes. By combining data of populations of various descent common pathways or common predispositions for a disease may be identified and specific and differentiating pathways and predispositions may be discovered. This new approach may lead into new insights and treatment avenues for multifactorial diseases.

The scope of this study was extensive. The authors not only used various populations in the world representing various ancestries, to analyse the genes (GWAS) associated with PAU, they also included, amongst other genetic and statistical methods, transcriptomics data (TWAS), phenotypic data (pheWAS), included brain chromatin interaction profiles and investigated drug repurposing.

One important aspect is how much of the problems observed may be explained by this new approach. The authors state that approximately 50% of PAU may be explained by genetic factors. The new approach increased the percentage of genetic factors identified contributing to PAU from 5% before up to 12% now. This means that the total percental explanation has increased from 2.5% (5% from 50%) to 6% (12% from 50%). It is progress, but these numbers also indicate that there is still a long way to go.

There are several inherent problems with this type of genetic association studies. Of course, the multi-ancestry approach provides insight in the joint alleles involved in the disease and therefore the generalizability of the results. However, results may be confounded by cultural and environmental factors that vary across ancestries affecting the expression of the various genetic traits.

Also, it may be challenging to have an equal and equivalent representation of each of the ancestries involved in the analysis. In this specific case the European and African cohorts were best represented, whereas the European cohort comprised the vast majority (83.6%) of the total population under study. Cross-ancestry replication of the European variants (85 variants) was good for the African cohort (76 variants), which corresponds to approximately 90%. Of these 76 variants, 64 (about 84%) had the same direction of effect.

The most significant allele identified was the ADH1B*rs1229984 a variation in the gene code for one of the alcohol dehydrogenase enzymes. This allele has been investigated in various other large cohorts (Millwood et al., 2023) and has not only been associated with all-cause mortality and risk for specific cancers (Zhang et al., 2023), but also for alcohol use disorder (AUD) (Kilcoyne et al., 2014). This allele is a so-called missense variant, so one nucleotide difference in the DNA that leads to a change in the coded protein, the enzyme alcohol dehydrogenase. These variants are associated with an increased catalytic activity, so a faster breakdown of alcohol and, therefore, a potential protective effect on long-term consequences of alcohol use (Polimanti & Gelernter, 2018). On the other hand, a higher ADH activity may lead into higher alcohol consumption levels, higher oxidative stress levels and change the rewarding effect of alcohol so that more alcohol is needed to get the rewarding effect. The current allele identification indicates that this specific allele is really involved in PAU across various ancestries and may also be interpreted as a positive control for the methodology applied.

Of the 100 variants identified, 80 were previously reported to be associated with PAU and of these, 53 variants were in protein-coding genes, of which 9 were missense variants. So, possibly roughly 10% of all variants identified potentially lead into functional protein changes.

From the identified cross-ancestry variants, candidate drug targets were searched for. Among them OPRM1 and GABRA4 were found, both targets for naltrexone and acamprosate respectively, again confirming what is currently used in medical practice. An additional nine other genes were identified as druggable. Trichostatin-a was put forward as a potentially re-purposable drug. The drug has been around for several years and tested in cells and animals and has growth arresting effects (Taddei et al., 2005).

Other re-purposable drugs include spironolactone and chlomethiazole. Spironolactone reduced alcohol use in both rats and humans in a recent study (Farokhnia et al., 2022) according to the authors. The study however, showed clear effects in rodents, but did not include human data obtained in clinical trials. The data were generated in a pharmacoepidemiologic cohort study showing a significant association between spironolactone treatment and reduction in self-reported alcohol consumption. The effect of spirolactone on self-reported alcohol consumption, however, was small. The authors are aware of the potential pitfalls in drug development. Many of those exist, specifically in multifactorial diseases like PAU, and the authors state “however, the pathway from genetic variants to the function of encoded protein to a biologically important therapeutic target is complicated and intricate, requiring more work in many modalities.”

The study uses PAU as an outcome parameter. PAU is a combination of alcohol use disorder (AUD) and AUDIT-P which differ phenotypically and genetically from typical alcohol consumption. The general idea is that only problematic alcohol use and alcohol dependence are included in this term. Interestingly, others have tried to characterize the genetic influences shared across various substance use disorders and found that all substance use disorders had one common so-called Addiction-Risk-Factor. The ‘Addiction-Risk-Factor’ included aspects like risk-taking, preoccupation and negative affect. This factor showed that there was a common genetic liability for all addictions tested and were largely independent of the substance of use (Hatoum et al., 2022). It may be interesting to see what common genetic variants exist in the various substance use disorders. Rather than cross-ancestry analysis that may be confounded by cultural and environmental factors, cross-substance-use-disorder analysis may better elucidate the genetic predisposition for problematic substance use including problematic alcohol use.

Specific Comments from Forum Members

Reviewer Ellison noted: “I agree that this is an important paper.  If people do not just say ‘It is all in the genes, so there is nothing we can do,’ the data presented will be very helpful in development of potential ways for preventing and managing alcohol misuse.  Genetic analyses such as this one will especially help direct pharmaceutical endeavours to find drugs that may be useful in these disorders.

Still, we must remember that approximately one half of factors that influence the risk of alcohol misuse relate to cultural or environmental effects, and we must not give up on public health approaches (e.g., advocating the consumption of alcoholic beverages with food and other aspects of the pattern of drinking) for the prevention of these disorders.”

Reviewer Harding comments that he “questions the premise set out at the beginning that ‘excessive alcohol use and alcohol use disorder are leading causes of death and morbidity worldwide, 2.2% of female deaths and 6.8% of male deaths.’ I don’t doubt that excessive alcohol use is a serious problem, but is it really a leading cause of death to the extent that it is responsible for 6.8% of all male deaths worldwide?  According to UK Government data published by the Office of National Statistics, in 2020 there were 8,974 deaths due to alcohol specific causes, representing 1.47% of all deaths.  It is hard to believe that other western countries are much different, or in European populations that comprise the majority of the cohort studied in the Zhou et al. (2023) paper.  If the figures quoted above are correct, other countries must have much higher levels of deaths due to alcohol consumption to arrive at the worldwide figures quoted.  Excessive alcohol use does not feature in the top 10 causes of death 2000-2019 globally according to WHO.”

Reviewer Waterhouse also considered that “this is a very helpful dataset, but I am also concerned about over reliance on the genetic information it reveals.  In particular, I can imagine a situation where if an individual has any one of the identified genetic variants, they will be presumed to be on the road to PAU if they touch alcohol.  I hope that future clinical advice will be more nuanced than that.”

Concluding comments

As suggested by Reviewer Goldfinger, “this is an ambitious study and identifies genetic variables that may be associated with problematic alcohol use, certainly a societal plague with serious consequences. Its potential utility for therapeutic intervention is an idealistic goal.

I am struck by the reference to The Global Burden of Disease Study 2016, which would suggest systemic alcoholism in the USA, with 29.5 million persons with an alcohol abuse disorder, that is, about 9% of the American adult population (328M according the 2020 US Census Bureau). This study has been repeatedly quoted, by a very many news agencies, as a justification to challenge any alcohol consumption whatsoever.

As a society, we have exposure to many morbid conditions. Persons living in cold climates wear sweaters and jackets, those with familial hyperlipidemia take statins or other life-saving lipid-lowering therapies, and so on. As suggested by others, those with genetic predisposition to alcoholism (very frequently identified during a family history at a GP office visit without the benefit of genetic mapping) can be addressed with counselling in moderation or abstinence as indicated. Novel pharmacologic therapies is certainly of interest.”


Edenber, H.J., Foroud, T. (2013) Genetics and alcoholism. Nature Reviews Gastroenterology & Hepatology, 10(8), 487-494.

Farokhnia, M., Rentsch, C. T., Chuong, V., McGinn, M. A., Elvig, S. K., Douglass, E. A., Gonzalez, L. A., Sanfilippo, J. E., Marchette, R. C. N., Tunstall, B. J., Fiellin, D. A., Koob, G. F., Justice, A. C., Leggio, L., & Vendruscolo, L. F. (2022). Spironolactone as a potential new pharmacotherapy for alcohol use disorder:  convergent evidence from rodent and human studies. Molecular Psychiatry, 27(11), 4642–4652.

Hatoum, A. S., Johnson, E. C., Colbert, S. M. C., Polimanti, R., Zhou, H., Walters, R. K., Gelernter, J., Edenberg, H. J., Bogdan, R., & Agrawal, A. (2022). The addiction risk factor: A unitary genetic vulnerability characterizes  substance use disorders and their associations with common correlates. Neuropsychopharmacology : Official Publication of the American College of  Neuropsychopharmacology, 47(10), 1739–1745.

Kilcoyne, B., Shmulewitz, D., Meyers, J. L., Aharonovich, E., Greenstein, E., Frisch, A., Weizman, A., Spivak, B., Edenberg, H. J., Gelernter, J., & Hasin, D. S. (2014). Alcohol consumption mediates the relationship between ADH1B and DSM-IV alcohol  use disorder and criteria. Journal of Studies on Alcohol and Drugs, 75(4), 635–642.

Millwood, I. Y., Im, P. K., Bennett, D., Hariri, P., Yang, L., Du, H., Kartsonaki, C., Lin, K., Yu, C., Chen, Y., Sun, D., Zhang, N., Avery, D., Schmidt, D., Pei, P., Chen, J., Clarke, R., Lv, J., Peto, R., … Chen, Z. (2023). Alcohol intake and cause-specific mortality: conventional and genetic evidence in a prospective cohort study of 512 000 adults in China. The Lancet. Public Health, 8(12), e956–e967.

National Institute on Alcohol Abuse and Alcoholism. (2008). Genetics of alcohol use disorder.

Polimanti, R., & Gelernter, J. (2018). ADH1B: From alcoholism, natural selection, and cancer to the human phenome. American Journal of Medical Genetics. Part B, Neuropsychiatric Genetics : The  Official Publication of the International Society of Psychiatric Genetics, 177(2), 113–125.

Stahre, M., Roeber, J., Kanny, D., Brewer, R. D., & Zhang, X. (2014). Peer reviewed: contribution of excessive alcohol consumption to deaths and years of potential life lost in the United States. Preventing Chronic Disease, 11.

Taddei, A., Roche, D., Bickmore, W. A., & Almouzni, G. (2005). The effects of histone deacetylase inhibitors on heterochromatin: implications for anticancer therapy? EMBO Reports, 6(6), 520–524.

Zhang, B., Peng, Y.-H., Luo, Y., Hong, C.-Q., Lin, Y.-W., Zhang, Y.-L., Xu, Y.-W., Su, X.-F., & Wu, F.-C. (2023). Relationship between esophageal squamous cell carcinoma risk and alcohol-related ALDH2 and ADH1B polymorphisms: Evidence from a meta-analysis and Mendelian randomization analysis. Cancer Medicine, 12(20), 20437–20449.

Zhou, H., Kember, R. L., Deak, J. D., Xu, H., Toikumo, S., Yuan, K., Lind, P. A., Farajzadeh, L., Wang, L., Hatoum, A. S., Johnson, J., Lee, H., Mallard, T. T., Xu, J., Johnston, K. J. A., Johnson, E. C., Nielsen, T. T., Galimberti, M., Dao, C., … Gelernter, J. (2023). Multi-ancestry study of the genetics of problematic alcohol use in over 1 million individuals. Nature Medicine, 29(12), 3184–3192.

Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:

Henk Hendriks, PhD, Netherlands

Creina Stockley, PhD, MBA, Independent consultant and Adjunct Senior Lecturer in the School of Agriculture, Food and Wine at the University of Adelaide, Australia

R Curtis Ellison, MD, Section of Preventive Medicine/Epidemiology, Boston University School of Medicine, Boston, MA, USA

Richard Harding, PhD, Formerly Head of Consumer Choice, Food Standards and Special Projects Division, Food Standards Agency, UK

Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark

Mladen Boban, MD, PhD, Professor and Head of the Department of Pharmacology, University of Split School of Medicine, Croatia

Andrew Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis.

Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, University of Arizona School of Medicine, Tucson, AZ, USA

Ulrich Keil, MD, PhD, Professor Emeritus, Institute of Epidemiology & Social Medicine, University of Muenster, Germany