UK United Kingdom

Are fish oil supplements putting you at risk of prostate cancer?

A report published in the Journal of the National Cancer Institute late last week shows a potential link between omega-3 fatty acids and the risk of developing prostate cancer. But it may be premature…

News of a potential adverse link with prostate cancer comes at a time when the public is heeding the message that the omega-3s derived from marine sources are beneficial. Sam Catch/Flickr

A report published in the Journal of the National Cancer Institute late last week shows a potential link between omega-3 fatty acids and the risk of developing prostate cancer.

But it may be premature for people to reduce their omega-3 intake until we have a better understanding of what lies behind this apparent association.

Omega-3 fatty acids are a class of polyunsaturated fats found in marine and plant oils that are considered essential for health. Plants contain alpha-linolenic acid (ALA), the short-chain omega-3 precursor that is elongated to form the more beneficial long-chain acids (EPA, DPA and DHA) found in marine sources.

This conversion is limited in humans; less than 1% of ALA is converted through to DHA. Hence the Nutrient Reference Values for Australia and New Zealand and the National Heart Foundation recommend consumption of pre-formed long chain omega-3 obtainable from fish, fish oil and krill oil, in addition to consumption of ALA from plant sources such as canola, chia, flaxseed, soy and various nuts.

News of potential adverse association with prostate cancer comes at a time when the public have been heeding the message that the omega-3s derived from marine sources are beneficial for cardiovascular health, counteracting inflammatory conditions, such as rheumatoid arthritis, and possibly improving mood and cognitive performance.

Identifying risks

A potential link between prostate cancer and consumption of ALA was reported more than a decade ago but was subsequently dismissed due to lack of supportive evidence. And with the realisation that ALA contributes little to the health benefits ascribed to omega-3, attention has focused more recently on the effects of the long-chain omega-3s.

We can now assess the “omega-3 status” of an individual from the levels of these fatty acids accumulating in certain fractions of the blood. And a strength of the current report is that it examined associations between blood levels of long-chain omega-3 rather than estimating intakes from dietary records, which is notoriously imprecise.

The report indicated that DHA, the omega-3 considered most important for conferring cardiovascular benefits, has the strongest association with high-grade prostate cancer – the authors included their latest findings in a meta-analysis in the same report, which indicates a 48% increase in relative risk of high-grade prostate cancer.

Most cancer risk factors are associated with unhealthy diets and lifestyles, so, on one level, this conflicts with current nutrition advice promoting increased omega-3 consumption to reduce the risk of chronic disease. It also challenges conventional thinking on the benefits versus risks of omega-3 supplementation.

The right dose

Up until now, long-chain omega-3s have been considered very safe. There was a minor concern that fish oil supplementation may prolong bleeding, although the risk is no greater than that posed by a low dose of aspirin. Both can help reduce the risk of potentially lethal blood clots.

The levels of omega-3 that the prostate cancer study authors refer to can be achieved through consumption of fish alone. Brandon Fick

Australia has adopted the generally-recognised-as-safe limit of three grams of omega-3 fatty acids per day (equivalent to around ten standard fish oil capsules) as recommended intake.

The level recommended for reducing the risk of chronic disease, particularly for maintaining heart health, is only 0.5 grams per day (one-sixth of the upper limit) but, on average, Australians consume only 0.2 grams of omega-3 per day.

That means, ideally, they should be consuming an extra 0.3 grams per day, equivalent to the omega-3 content of a typical fish oil capsule or a weekly serve of oily fish.

Too much of a good thing?

There are cases where, at excessive intakes, beneficial nutrient supplements, such as antioxidants, may have detrimental effects.

But that doesn’t explain the present finding. The levels of omega-3 that the prostate cancer study authors refer to are not extreme – they can be achieved through consumption of fish alone, without popping capsules.

The range of omega-3 levels from low risk to high risk as described in the study are within the range attainable by variations in the amount of omega-3 consumed in typical Australian diets.

While this observation raises concern, there’s a lot more we need to understand before we can properly balance the risk-versus-benefit equation for omega-3 consumption.

In the absence of trials that can demonstrate a causal relationship, we need to have a clearer understanding of possible underlying mechanisms. It’s known, for example, that omega-3 oils can influence certain male reproductive mechanisms, such as sperm production and motility, but there is currently no hypothesis to explain how they might influence prostate cancer risk.

At the same time, evidence is mounting for protective effects of omega-3 consumption in breast and other cancers

Until now, the risk factors for fish oil supplementation have been considered to be minimal. Even if a causative relationship between omega-3 consumption and prostate cancer is confirmed, consumers will need to weigh up their cancer risk against the known cardiovascular benefits before giving up on omega-3 fatty acids.

Sign in to Favourite

Join the conversation

35 Comments sorted by

  1. George Michaelson


    If a weekly serve of oily fish is enough to meet the RDI (effectively 1/7th of a single oily fish based meal in this definition) then any culture which has more than this amount of oily fish in the diet should have demonstrably higher rates of prostate cancer, noting other effects might mitigate.

    (I am not an epidemiologist so I can well believe its not this simple)

    So on that basis, do we actually see significantly, or even raised, if not extreme, higher rates of prostate cancer in iceland, norway, finland, sweden, or even the netherlands?

    1. Colin MacGillivray

      Architect, retired, Sarawak

      In reply to George Michaelson

      Yes George, got me worried for a moment. Fish here in Sarawak is cheap- small ones at AUD5 a kilo or less, so I eat it every day. Really boosts the HDL.
      Surely there's a difference between food and pills, which pill producers are probably not keen to investigate?
      Haven't tried krill yet Rosemary! (following comment)

    2. Paul Rogers


      In reply to George Michaelson

      Good question, George.

      I found at least one recent study that looked at fish consumption and prostate cancer risk in a high fish consuming country (Iceland).

      "Consumption of Fish Products across the Lifespan and Prostate Cancer Risk"

      Conclusions: "Salted or smoked fish may increase risk of advanced prostate cancer, whereas fish oil consumption may be protective against progression of prostate cancer in elderly men. In a setting with very high fish consumption, no association was found between overall fish consumption in early or midlife and prostate cancer risk."

      More elaboration on this is required it seems. Like other supplement studies, it may turn out that the whole product is superior to the supplement..

    3. Chris O'Neill

      Victim of Tony Abbotts Great Big New Tax

      In reply to George Michaelson

      I know Sweden has one of the worlds highest prostate cancer death rates. There is a website showing world maps of country death rates for many causes of death. Carelessly I forgot to bookmark it.

    4. Rosemary Stanton

      Nutritionist & Visiting Fellow at University of New South Wales

      In reply to George Michaelson

      Australia does not have an RDI for long chain omega 3 fatty acids. Where there is insufficient evidence to set an RDI, we have an AI (adequate intake) based on the average daily intake of apparently healthy people and assumed to be adequate. This is 90mg/day for women and 160mg/day for men.

      The NRVs also include a table with suggested dietary tragets to reduce chronic disease risk. These are based on the 90th centile of current population intake. For long chain omega 3 fatty acids, these are 430mg/day for women and 610mg/day for men (approx Peter Howe's 500mg/day). Note however, that the NRVs specifically recommend these levels be "attained by replacing energy-dense, low nutrient foods and drinks with long chain omega 3 rich foods such as tuna, salmon and mackerel, lean beef or low energy-density long chain omega 3 enriched foods". There is no recommendation for supplements.

    5. Sue Ieraci

      Public hospital clinician

      In reply to Chris O'Neill

      WHO lists Sweden as equal 4th (with Australia) for longevity (overall population lifespan of 82). That could be part of the issue. Remember, again, that everyone has to die of something. People who die of prostate cancer have not died earlier of something else.

    6. Geoffrey Harold Sherrington


      In reply to Sue Ieraci

      Hi Sue,
      As a post-70 year person, I've been barraged with articles and recommendations about how to handle possible early signs of prostate cancer. IIRC, one of the more popular pop-sci journals had an edition about the subject some 15 years ago, or whenever the PSA test became widespread after approval about 1995. I felt then that unless a new development was engineered, the medico was faced with a "Justice of Solomon" type of predicament. Has there been a supplementary test that assists the PSA test early in the sequence and indicates a preference for intervention rather than inaction (assuming age has been factored in)?

  2. Rosemary Stanton

    Nutritionist & Visiting Fellow at University of New South Wales

    Foods contain a range of omega 3 fatty acids. To talk about 'omega 3' is equivalent to talking about 'vitamin'.

    And please can we leave the krill for the aquatic creatures who really need them.

    1. Paul Weldon

      Research Fellow

      In reply to Rosemary Stanton

      My Nephrologist specialist recommends 3-4g of fish oil a day for chronic kidney disease (krill or other), as well as lowering salt and protein intake. The medical community are increasingly alert to the potential benefits of EPA/DHA and the general consensus (up until I read this article at least) was that even if it didn't benefit, there was no evidence of negative risk.
      It doesn't help that many websites that claim to offer nutritional advice for people with issues such as CKD leave you feeling you shouldn't eat anything!

  3. Chris O'Neill

    Victim of Tony Abbotts Great Big New Tax

    Of course, the only way to eliminate most of the risk of dying from prostate cancer is to get your prostate removed before the age at which prostate cancer starts to develop (45-50). This could be called the Angelina Jolie technique (applied to a different type of cancer in her case).

    Some people have the idea that removing the prostate after cancer develops is helpful but it's only helpful in the sense that it removes currently healthy prostate cells that could one day mutate into a fatal cancer. If you already have prostate cancer stem cells then they are already spread around your body, even for the earliest of cancers. Something like 20% of old Australian men are getting prostatectomies but the prostate cancer death rate has only been reduced by 20%.

  4. Gary Cassidy

    Following hip surgery I've put up with aching hips for a few years, recently I decided to try 3g dose EPA/DHA daily (previously I'd taken ~600mg with no noticeable effect), and my outcome has been very positive (better than anti-inflammatory (15mg mobic), as good or better than paracetamol).
    Just as I was starting to sing praises for fish oil I read about this. Damn it!

    1. Rosemary Stanton

      Nutritionist & Visiting Fellow at University of New South Wales

      In reply to Gary Cassidy


      Omega 3 fatty acids do have an anti-inflammatory effect, although in well conducted trials, benefits have mainly been noted for rheumatoid arthritis (not the more common osteoarthritis).

      Also other early studies (mainly on fish eating populations) for cardiovascular disease have not been backed - see
      Wei MY, Jacobson TA. Effects of eicosapentaenoic acid versus docosahexaenoic acid on serum lipids: a systematic review and meta-analysis. Curr Atheroscler Rep. 2011;13(6):474-83.
      The ORIGIN…

      Read more
  5. Geoffrey Harold Sherrington


    There is a table of extras - some might have changed by now - available from Australian Health Funds as reported in 'The Australian Skeptic'.
    It is difficult to throw out the junk and retain customary therapies like dental, optical. The fund rules can make it hard.
    If the medical funds of Australia (apart from the Doctors' Fund) recognise this assortment of weird therapies, then I'm not surprised that fish oil supplements are up for discussion here.

    Read more
  6. Sue Ieraci

    Public hospital clinician

    ''But it may be premature for people to reduce their omega-3 intake until we have a better understanding of what lies behind this apparent association.''

    PLEASE PLEASE PLEASE let's not all start saying ''omega-3 causes prostate cancer'' on the basis of a single report.

    Clinical practice should not change on the basis of one publication - it requires a solid body of evidence.

    Let's also remember that one major reason for the prominence of prostate cancer in our society - and, indeed, many cancers, is that we live so long. According to the WHO (2011 figures) Australia is ranked at equal top 4, with an overall life expectancy of 82 years. In contrast, Sierra Leone, ranked 193rd, has an overall lifespan of 47 years.

    This is not an argument for ignoring any type of cancer but we have to remember that, eventually, everyone dies of something. When we look at the incidence of any type of disease, it has to be seen in age context.

  7. Andrew Sinclair

    Professor of Nutrition Science

    I think there has been far too much noise about this study.

    The Brasky et al study was NOT a study about fish oil.

    It was study that was designed to test the effects of vitamin E and/or selenium on prostate cancer rates, the SELECT trial.

    The reason why people think it is about fish oil is that after that trial, the investigators measured the levels of fatty acid levels in blood that can serve as an index of how much fish or fish oil the subjects might have been taking. The most striking…

    Read more
    1. Gary Cassidy

      In reply to Andrew Sinclair


      RE "the no cancer group had mean levels of 4.48%, the low grade cancer group had mean values of 4.66 % and the high grade cancer group had levels of 4.71%. By way of comparison, vegetarians who consume no fish have blood long chain omega 3 levels of 4.1% (Li et al, Amer J Clin Nutr 1999) and subjects fed a fish-rich diet for 2 weeks have blood levels of 21.5% (Mann et al Lipids 1997)."
      Thank you for putting this into context.

      Makes me wonder how this paper could pass the review process with conclusions framed in such a manner.

    2. Sue Ieraci

      Public hospital clinician

      In reply to Gary Cassidy

      Gary - this is what I keep trying to explain - peer review for publication does not mean that the paper is somehow valid or correct - it just means it is judged as suitable for publication in that particular journal.

      Real peer review occurs when an informed community reads the paper, analyses the methodology, and evaluates it alongside existing evidence, and the science or techonology underlying that professional practice.

      That is actually the purpose of publication - so that the work can be evaluated and judged by people like Andrew Sinclair or Rosemary Stanton - who can put it in context with existing knowledge.

      We are overplaying the significance of ''peer review'', and misunderstanding the purpose of publication.

    3. Gary Cassidy

      In reply to Sue Ieraci

      Hi Sue,
      RE: "We are overplaying the significance of ''peer review''"
      Do you think that it should be changed to editorial review? "Peer" indicates a suitably qualified researcher operating within a similar field of study. I have no problem with interesting or controversial results being published and subsequently scrutinised by other researchers - this is important to science. However, I do have a problem with unjustified or misleading conclusions passing through the peer review process. Particularly when it is likely that the conclusions are going to hit the headlines.
      A quick google search would indicate that neither group had high blood concentrations of LCω-3PUFA relative to normal populations.

    4. Geoffrey Harold Sherrington


      In reply to Gary Cassidy

      Puts a new meaning on lube & oil change, eh?
      What does it mean that X% of omega 3 in blood levels is measured?
      Is it that (a) the omega 3 is on its way to excretion, without having done anything of note;
      or (b) that the omega 3 has done work in finding the niche that needs it most, but there is an excess that reports as X% in the blood;
      or (c) that the blood it about to take the omega 3 to places of benefit, so that the starting levels of X% stated are pre-work and mean nothing?

      Surely one needs to measure the active chemical in the location where its mechanisms affect the body. To broaden the argument, think of a range of anti-oxidants and their magical powers.

    5. Sue Ieraci

      Public hospital clinician

      In reply to Gary Cassidy

      Gary - the term ''peer review'' prior to publication is correct in the sense that most journals choose reviewers who work, or have expertise, in the same broad area as the subject of the paper. This does go beyond editorial review, as the peer reviewers are asked to critique the methodology and validity of the results, not just the language and style.

      What has happened, though, is this: research has moved from the specialised journals, which generally were only read by an informed audience, to…

      Read more
    6. Gary Cassidy

      In reply to Sue Ieraci

      HI Sue,

      I don't disagree with what you're saying, however the reality is that individual journal papers are reported on in the media and elsewhere. Often the more controversial ones relating to lifestyle, health and disease get most attention. In addition to this the research institutes often release press releases at the time of publication. In this case the release is titled
      "Study confirms link between high blood levels of omega-3 fatty acids and increased risk of aggressive prostate cancer…

      Read more
    7. Sue Ieraci

      Public hospital clinician

      In reply to Gary Cassidy

      ''In my opinion journal editors and reviewers should take into account the reality of media reporting and play an active role in ensuring that conclusions are an appropriate (and responsible) representation of the study results.''

      I agree, but the reality is that journals vary widely in their purpose, ownership, structure and governance systems. It doesn't seem possible to regulate the quality of journal editing - it is more about ''buyer beware''. That makes it all the more important for journalists to evaluate their sources before going to print.

    8. rory robertson
      rory robertson is a Friend of The Conversation.

      former fattie

      In reply to Sue Ieraci

      Dr Sue argued that: "peer review for publication does not mean that the paper is somehow valid or correct"

      In fact, Dr Sue, "peer review" prior to publication - and no matter if the reviews were incompetent or ignored - has for some time been the only test the University of Sydney applies when assessing whether a paper is "somehow valid or correct".

      Here is the quote from Vice-Chancellor, Dr Michael Spence: (in blue at the bottom of the first page)

      Read more
  8. Andrew Sinclair

    Professor of Nutrition Science

    Furthermore, the authors could have referred to the ecological data from different populations which indicate that the 25-fold difference in incidence of prostate cancer is unrelated to differences in seafood consumption in stark contrast to predictions of the Brasky et al study (

  9. Alexandra McManus

    Public health practitioner

    I support Andrew's comments. The study did not focus on omega 3s. In fact, the omega 3 levels reported were only pseudo measures extrapolated from the total fatty acid levels measured in blood samples.

    There was no evidence of causation in the studies reviewed.

    The methods used to summarise the studies(not a meta-analysis as indicated) was flawed. The authors did not account for the different target groups, methods or measures used in the studies. The limitations of each were not even noted…

    Read more
    1. David Bindoff


      In reply to Alexandra McManus

      Thanks to all the professionals in this field for clarifying the context of this article. It is very helpful and reassuring to this long-term low level omega 3 supplementer!

  10. Geoffrey Harold Sherrington


    In an earlier blog I wrongly used 'omega 3' as shorthand, and apologise for the simplification.
    Prof Howe wrote "There are cases where, at excessive intakes, beneficial nutrient supplements, such as antioxidants, may have detrimental effects."
    As a chemist, I have no idea what this is meant to convey. Anti-oxidants (we chemists name them 'reductants') form a spectrum of chemicals rather in the way that pH characterises a spectrum of chemicals from acid to alkaline.
    In inorganic chemistry, a reductant…

    Read more
    1. Sue Ieraci

      Public hospital clinician

      In reply to Geoffrey Harold Sherrington

      Geoffrey - this paper largely supports what you are saying:

      ''Free radicals and antioxidants – quo vadis?''

      and another commentary here:

      Dietary fads are a fertile area for marketing and pseudo-science in our current society. The same issue occurs with the pseudo-science of ''alkalinising'' foods, when we know that the health kidney and lung and buffers do a good job of keeping the body's pH within a tight range.

  11. Geoffrey Harold Sherrington


    Peter Howe,
    By coincidence, a long wait in a medical waiting room this morning led me to finish a browse of the glamour pictures in the womens' mags and then to pick up an early 2013 copy of 'Scientific American'. Therein was a main article of some findings about antioxidants, longevity, free radicals and so on. The parts of this article referenced from pre 2007 or so were similar to those I had made in 'The Australian Skeptic' about then and in part repeated in a blog comment elsewhere here.

    Read more
  12. Lewis Rassaby

    logged in via Facebook

    Thanks, Andrew Sinclair for your analysis of the paper. It might be useful in this discussion to consider what advice to patients is being dispensed by general practitioners. I am getting the impression that in the patient population a lot of people are taking from 3-10gms daily, using the highly concentrated liquid forms of fish oil for mainly heart and joint concerns, sometimes for depression and brain function. Andrew; do you have a comment on those sort of doses?

  13. NYC Body

    Owner of NYC Body at New York City Body

    In my opinion, I felt that the researchers in this study—as well as others they reported on--had a negative bias against nutritional supplements. This wasn’t a double-blind, placebo controlled trial about omega-3s—in fact, we don’t even know if the participants in this study took omega-3s. Instead, the researchers drew a conclusion based on a.2% difference in omega-3s—one that can show association, but not causation.

    - James from

  14. Alan Miller

    logged in via email

    Too much of anything can be bad whether it is a supplement, sugar or alcohol. However taking Omega 3 supplements has been shown to provide positive health benefits. Resveratrol is another nutritional supplement packed with antioxidants that has the potential to provide numerous health benefits. Visit to find out more about resveratrol supplements.