What is selenium and what role does it play in the body?
Food provides us with a variety of substances we need to maintain life. These substances are essential nutrients and are classified as macronutrients (water, protein, fats, and carbohydrates) and micronutrients (vitamins and minerals).
Selenium is one of the micronutrient minerals that we need in small ‘trace’ amounts to make various important proteins, called selenoproteins, some of which help rid the body of dangerous free radicals that can damage our DNA.
In humans, selenium is essential, particularly for the effective functioning of the immune system and in controlling oxidative processes linked to cancer development.
Indeed, a growing body of evidence suggests that suboptimal intakes of selenium contribute to the development of several cancers and other major chronic diseases1.
Can you please outline your recent research on levels of blood selenium and risk of liver cancer? What were your main findings?
This new study within the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort shows an inverse correlation between selenium status and cancer risk.
The highest levels of blood selenium or of selenoprotein P (SePP), the protein that distributes selenium from the liver around the body, are associated with a decreased risk of developing liver cancer (particularly hepatocellular carcinoma, HCC), even when all other major liver cancer risk factors are taken into account2.
The study also shows that selenium level is not associated with the development of gall bladder or biliary tract tumors.
EPIC is composed of more than half a million participants across 23 centers in 10 Western European countries.
It is headed by the International Agency for Research on Cancer (IARC) in Lyon, France and its goal is to examine the relationship between diet, nutrients (including selenium), environment, lifestyle, genetics and cancer.
We identified 261 who were diagnosed with some form of hepatobiliary cancer during the follow-up of these individuals in the years since enrolment into the study and blood sample collection (comprised of 121 liver cancers and 140 gall bladder and biliary tract cancers).
These “cases” were matched with 261 cancer-free “controls” with similar age, sex, nationality, and other factors thought to influence cancer risk.
We measured selenium and SePP levels, a protein critical for selenium’s protective antioxidant functions, in the stored blood samples.
These measures indicate to us what an individual’s ‘selenium status’ is, i.e., how much selenium they have circulating in the body for important cell maintenance processes that require selenium and selenoproteins.
We then investigated whether the selenium status of cases was different from that of control subjects.
It is important to emphasize that the blood samples were taken at enrolment from all participants, this means that these samples were taken years before the cases developed their cancers, while the controls did not develop any cancers in the follow-up time period.
This is why it is termed a ‘prospective’ case-control study and this design critically helps to minimize reverse causality biases, i.e., where we get different measurements of substances like selenium in patients because they are already sick at the time of blood sampling , and this may have nothing to do with the reasons why they became ill.
The research findings tentatively suggest that where selenium is lower than the optimal level, such as for most people in Europe, increasing selenium intake may help to prevent liver cancer in addition to moderating or avoiding alcohol consumption, maintaining a healthy body weight, and stopping smoking.
The results also suggest that there is a larger risk for developing HCC among overweight and obese participants with low amounts of SePP, further underlining the importance of maintaining a healthy body weight.
However, these findings are based on a single study with a modest number of liver cancers, and thus our results need to be validated by further studies before any public health recommendations can be made.
How did you take into account other major liver cancer risk factors?
We were able to do this as we have an extensive amount of information on these factors within the EPIC cohort, which is one of the strengths of our work. When individuals were recruited into this study back in the 1990s extensive information was recorded on their diet and lifestyles including their smoking and drinking histories.
They also supplied blood samples and at this point everyone had not been diagnosed with any cancer. We continually follow up these volunteers to ascertain what cancers or other major chronic diseases they may develop.
As we have a total number of participants of around half a million it means that for each person who develops, for example, a liver cancer we can match that person (now classified as a ‘case’) very closely to another person (called a ‘control’).
We do this for details such as age, sex, recruitment center, time of blood collection, fasting status and among women, menopausal status.
Importantly, we also have measured many biologically relevant biomarkers in the blood samples of these same cases and controls.
Several of these are useful in estimating other risk factors for liver cancer development such as, among others, Hepatitis B and C positivity and markers of liver injury [alanine aminotransferase (ALT), aspartate aminotransferase (AST), g-glutamyltransferase (GGT), liver-specific alkaline phosphatase (AP), albumin, and total bilirubin].
Do you think this correlation is likely to be causal in nature?
Yes I do when I weigh up the growing body of experimental and observational evidence, including our own, suggesting that insufficient intakes of selenium contribute to the development of several cancers1.
Many of the previous studies in this area were conducted in laboratory animals and cell culture systems, and most of the human studies have been small and were retrospective, meaning they couldn’t really determine if poor selenium status preceded the onset of cancer.
The results from our prospective study thus add another important missing piece to the evidence that suboptimal selenium status may be associated with an appreciably increased risk of HCC development, together with other lines of evidence, including mouse models that have the SePP gene removed.
However, I stress that studies such as ours are observational in nature - we cannot infer causality from correlation and these data should not be used by themselves to suggest that improving selenium status necessarily reduces cancer risk.
We also discuss in our article a possible scenario where low selenium intake contributes more directly to general liver disease (due to negative effects on the adequate working of hepatocyte cells in the liver) that may progress to cancers in some people.
The importance of adequate selenium status for the health of hepatocyte cells may also explain the strong correlation with risk of HCC.
What further research is needed to validate your findings?
Further research is needed into the modifiable causes of these cancers so as to formulate effective prevention strategies. It is important to stress that our results are based on a single study with a modest total number of liver cancers, and thus our results need to be validated by further studies.
We discuss in our article that SePP may be an early and sensitive indicator of hepatocyte-related liver health, and this is a very interesting area to follow-up on.
If more strong evidence comes from other studies then we will need randomized controlled dietary intervention trials in populations where selenium is suboptimal (e.g. Western Europe) to fully assess whether increasing selenium intake may reduce the risk of liver cancer, especially for those at high risk (e.g. hepatitis B and C virus positive, or presence of liver damage / cirrhosis) for these cancers and also taking account of genetic background (genetic variants in selenoproteins affecting their adequate functioning).
As previous studies raised concern about antioxidant supplements causing adverse effects, future trials will need to be very carefully conducted and intensely monitored.
I believe the main reason for this was the inappropriate design of some of these trials, for example the SELECT study (Selenium and Vitamin E Cancer Prevention Trial)3 in north America was inadvisable and unnecessary, as all participants had adequate selenium status to begin with (we discuss these issues in our previous paper on selenium status and colorectal cancer risk4).
Thus it is important to develop and measure biomarkers that can reliably reflect nutrient status (such as SePP for selenium status) in participants prior to these trials, so as to avoid the risk of giving extra nutrients to individuals who have enough of them already.
The type of selenium provided in supplementation trials is also important to consider as we see different biological responses to the major chemical forms of dietary selenium such as selenomethionine, the major source of selenium in the human diet, and selenium selenite.
This may be particularly relevant to liver disease and liver cancer as there is emerging lines of evidence that they differentially contribute to the amount of biologically usable selenium for hepatocyte metabolism.
We discuss this, including information from other studies related to this issue, in our article2. To my mind, this is an intriguing area of future investigation.
These trials are very costly, take a lot of people and a long time to follow to cancer onset and it is difficult to know how best to go from studies such as animal models and human observation studies to full trials.
I think we need more confirmation prospective studies first, then pilot trials to show effects on secondary biomarkers, e.g., to show reductions in reactive oxidant species and other biomarkers of liver disease.
Additionally, there is evidence that some selenoproteins could possibly have dual effects that may promote cancer, especially when a cancer is progressing but has been undetected. Thus, there is a lot more to understand before we move on to full trials of robust design.
Unfortunately I personally might not be able to progress my work in this area much further as I have only a short-term part-time contract in my current institution (thanks to my current grant funding from the Health Research Board of Ireland). So I am currently looking for suitable new opportunities elsewhere to take this research forward!
However, in this new current funding I am leading a large study also in the EPIC cohort to examine the association of selenium status with breast cancer risk. This will be the largest observational study ever conducted in this area, involving approximately 2,500 women with breast cancer and the same number of control women without cancer.
How do soil levels of selenium vary around the world?
Selenium is found in small amounts in foods like shellfish, salmon, Brazil nuts, meat, eggs, grains, and onions. However, selenium levels in foods depend largely on the levels of selenium in the soil where the food is grown and animals graze.
Soil levels tend to be low in many regions in Europe, contributing to lower body levels of selenium in those populations compared with people living in regions with higher soil selenium concentrations, such as North America.
In agreement with this pattern of worldwide selenium distribution, data from intervention trials and studies such as ours that were able to test selenium levels before cancer onset (prospective studies) suggest that any protective effect of higher selenium intake against cancer development is probably more apparent in populations with low selenium availability, such as many across Europe, and not in areas with higher selenium intake levels as for North America.
Have the geographical patterns in incidence mortality rates of liver cancer been studied to date?
Yes - liver cancer is a major problem in the developing world. The geographic variation in liver cancer incidence rates mirrors the prevalence of two established marked risk factors for these cancers: viral hepatitis B and C and aflatoxin exposure5.
In 2012, worldwide, there are estimated to have been 782 000 new cases of liver cancer. It is the second most common cause of death from cancer worldwide, estimated to have been responsible for nearly 746 000 deaths in 2012 (9.1% of all cancer-related deaths that year)6.
The prognosis for liver cancer is very poor (with an overall ratio of mortality to incidence of 0.95), so the geographical patterns in incidence and mortality are very similar6. Liver cancers are often diagnosed at late stages and have limited treatment options.
Why do you think the incidence of liver cancers in developed countries is increasing?
It is indeed markedly increasing in traditionally lower-risk industrialized countries. In Europe, for example, liver cancer is now the seventh most common cause of death from cancer6. I believe it is primarily driven by the dramatic changes in Western lifestyle and dietary habits over recent decades.
Poor diets, excess alcohol use, smoking, and sedentary lifestyles commonly lead to obesity, insulin resistance, metabolic and hormonal changes, low vitamin D levels, and eventually to the development of type 2 diabetes and/or non-alcoholic fatty liver disease. These are risk factors for liver cancer development.
How much further research is needed before public health recommendations can be made with regards to selenium intake and liver cancer?
A typical answer from a researcher - a lot! And it is really true. As I already mentioned, our results alone only provide an indication that improving selenium status may reduce liver cancer risk.
There is a long way to go and our results need to be validated by further studies, such as the ones I touched on earlier, before I would feel comfortable in endorsing public health recommendations in regards to levels of selenium intake to help prevent liver cancer development.
Although based on the evidence of what level of selenium we need for the best working order of key selenoproteins such as SePP in our body, I would state that we require a selenium status (that is the actual amount of selenium circulating in our bodies) of at least 124 nanograms per liter (as ascertained in plasma; this issue is discussed in our recent paper on colorectal cancer and selenium status4).
An optimum dietary selenium level for cancer prevention may vary according to life-stage, sex, general state of health, organ site, and genetic background. These are all areas where more research is needed.
Is selenium intake likely to affect the risk for any other cancers?
Yes. Selenium intake may well affect the risk of several cancers1. There is good evidence for this for breast cancer, colorectal cancer and some types of prostate cancer. In the EPIC cohort we recently published a similar study showing that a higher selenium status was associated with a reduced colorectal cancer risk4.
A follow-up study we are currently preparing for submission will show that this risk is modified by interaction between selenium status and genetic variants we inherit in genes coding for selenoproteins, which require selenium for their manufacture in the body.
Where can readers find more information?
Please Note: The research is a joint project involving the International Agency for Research on Cancer (IARC), RCSI, Charité Medical School Berlin, and collaborators in the European Prospective Investigation into Cancer and Nutrition (EPIC). The study was jointly funded by the French National Cancer Institute (Institut National du Cancer; INCa) and the Health Research Board of Ireland (HRB).
About Dr David Hughes
Dr David Hughes is an honorary research lecturer at the Royal College of Surgeons in Ireland (RCSI) in Dublin, Ireland and a PI in the Centre for Systems Medicine within the RCSI.
His current research focuses on nutritional, genetic and microbial epidemiology of cancers of the colorectum, breast, liver and pancreas. He currently leads large prospective cohort studies of the influence of selenium, zinc, and copper status on colorectal, liver and breast cancer risk (2011-current; e.g. PMIDs 27357089, 25042282).
Other recent research highlights among many other published studies include the first study to propose a link between Fusobacterium nucleatum (an oral and gut bacterium) levels and clinical outcome from colorectal cancer (PMID: 24599709), leading the analysis of the implications of the pilot strategy for iFOBT colorectal cancer screening in Ireland (PMID: 23746062), and helping to show that methylation of the TFAP2E gene predicts patient response to chemotherapy (PMID:22216841).
He received his BSc Biochemistry degree from the University of Leeds, England in 1990, his PhD in Medical Genetics from Queen’s University Belfast (QUB) in Northern Ireland in 1996, and a Postgraduate Diploma in Health Professions Education at the RCSI in 2015.
Since 2014, he has been a steering committee member for colorectal cancer projects within the EPIC study. Dr Hughes worked as a research fellow at Trinity College Dublin from 2007-2011 where he instigated nutritional epidemiology and biomarker studies in colorectal cancer, including contributing to the first strong evidence that genetic variants in selenoprotein genes increase colorectal cancer risk (PMID: 20378690).
From 2000-2006 he was employed as a scientist at the World Health Organization’s cancer research headquarters (the International Agency for Research on Cancer in Lyon, France), where he helped define the first genetic modifiers of the BRCA1 and BRCA2 genes and common susceptibility alleles for breast cancer in adequately powered studies (e.g. PMIDs 1828356, 17999359, 17293864, 17018785).
As a postdoctoral fellow, Dr Hughes instigated functional genomic projects on the model nematode, C. elegans following the sequencing of the C. elegans genome (PMID:9851916) at the Sanger Institute, Cambridge, England (1998-2000).
His thesis work at QUB (including a subsequent short-term postdoctoral fellowship at the Institute for Human Genetics in Hanover Medical School, Germany) defined the spectrum of CFTR mutations (and pioneered the technique of microsatellite haplotype directed mutation testing for rare CFTR variants) as thereafter used in Cystic Fibrosis screening in Northern Ireland (e.g. PMIDs 11288718, 8956039, 8889582, 7535745).
His first research post in 1991-1992 was as a research assistant to Prof John Hardy at St Mary’s Medical School, Imperial College London, where the group discovered that mutations in the amyloid precursor protein gene (APP) gene can cause Alzheimer’s disease, the first gene linked with this disorder (PMID:1944558).
- Méplan C. Selenium and chronic diseases: a nutritional genomics perspective. Nutrients. 2015 May 15;7(5):3621-51. Review. PMID:25988760.
- Hughes DJ, Duarte-Salles T, Hybsier S, Trichopoulou A, Stepien M, Aleksandrova K, et al. (2016). Prediagnostic selenium status and hepatobiliary cancer risk in the European Prospective Investigation into Cancer and Nutrition cohort. Am J Clin Nutr. Available from: http://dx.doi.org/10.3945/ajcn.116.131672.
- Lippman SM, Klein EA, Goodman PJ, Lucia MS, Thompson IM, Ford LG, Parnes HL, Minasian LM, Gaziano JM, Hartline JA, Parsons JK, Bearden JD, 3rd, et al. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA 2009;301(1):39-51.
- Hughes DJ, Fedirko V, Jenab M, Schomburg L, Méplan C, Freisling H, et al. Selenium status is associated with colorectal cancer risk in the European prospective investigation of cancer and nutrition cohort. Int J Cancer 2015 Mar 1;136(5):1149-61.
- Gomaa A-I. Hepatocellular carcinoma: epidemiology, risk factors and pathogenesis. World J Gastroenterol 2008;14:4300.
- Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, et al. (2013). GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer. Available from: http://globocan.iarc.fr.