Fish oil and prostate cancer: Go beyond the headlines

I suppose it may help diffuse any relevant tensions to get the punch line out of the way: I have a prostate, and in the context of my mostly-plant diet, will continue to eat fish and take a daily omega-3 supplement (krill oil in my case). We will, as well, continue using various omega-3 supplements in varying doses for specific therapeutic purposes in my clinic.

And now, let’s turn to the news. Using Google this morning, I found over 8 million Web pages addressing the combination of fish oil and prostate cancer — just shy of 600,000 blogs and over 9,000 news items. Not all of these relate directly to the recent study suggesting a link between fish oil intake and prostate cancer, but many do.

So it’s pretty clear I was far from alone in my eagerness to pounce on this story and opine when such headlines as “Fish oils may raise prostate cancer risks, study confirms,” embellished with images of fish oil capsules, began populating print media and cyberspace alike. I have benefited from the more insightful reactions of others, mixed up though they may be with a venting of passions.

But I think it’s just as well that my wife and I had a couple of vacation days out of town in a place with negligible Internet access while this tale went through the mad, early gyrations of a fish out of water. It would be easier to get the implications of this study wrong than right, and in either direction.

The study does not prove that fish oil intake causes prostate cancer. But nor is the study dismissible rubbish, as the large crowd of fish oil proponents gathered around it with harpoons in hand would like to contend.

The study in question is observational — or at least, the relevant portion of it is — meaning no intervention was undertaken. Men were not given fish oil supplements, nor fish dinners.

Rather, blood was obtained from men at baseline as they enrolled into the study. Over time, more than 800 men developed prostate cancer. The researchers matched these men on the basis of age and race to others in the study who had not developed prostate cancer, and compared those baseline blood samples for various things, notably the levels of different kinds of phospholipid fatty acids, a metabolized form of dietary fats correlated with their intake. The authors found, and they state several times in the paper, that this was at odds with their expectations, that higher omega-3 levels were seen in the men with prostate cancer than in those without.

Because the findings were counter-intuitive, and counter to the researchers’ expectations, they compared them statistically to the results of other studies that had examined the same association in a meta-analysis. This part of the paper was too vaguely described for my purposes, but indicated that an association between higher omega-3 levels and increased prostate cancer has been observed with some consistency.

Others have waded deep into the kelp already and gone after this paper’s school of statistical details: differences in the strength of association between prostate cancer and DHA as opposed to EPA, for instance. HuffPost blogger Johnny Bowden does some of this himself, and cites others who have as well. I leave them to it, and you to the risk of drowning in the details accessible with Google or Bing should you choose to ply these waters further.

I went another way altogether, seeking the view from altitude. We are all aware of populations with especially high intake of omega-3 fat; what does epidemiology tell us about their prostate cancer risk?

The big picture is mostly, if not quite entirely, reassuring. A traditional Japanese diet is considerably higher in omega-3 fat than the prevailing Western diet, yet the Japanese have a low rate of prostate cancer. So far, so good. However, studies have shown an association between higher fish intake and higher prostate cancer incidence among Japanese men.

Scandinavian men also consume lots of omega-3 compared to others, and have relatively high rates of prostate cancer. In this population, the association between fish intake and prostate cancer is murky. Salted and smoked fish intake correlates with higher risk, but fish oil intake with lower risk. A study of men in California suggests that some kinds of fish, and even some cooking methods, might increase prostate cancer risk, while other fish and cooking methods might lower it.

And finally, there’s the Inuit, who consume omega-3 marine oils at legendary levels. Prostate cancer is rare among them, even when it is sought at autopsy, and omega-3s have been invoked as a potential explanation why.

Where does this leave us? If we are honest about it, whatever our prior inclinations, the answer is: in muddy water, and a bit confused.

The new study inducing all the fuss shows an association between higher levels of omega-3 in the blood at baseline, and the development of prostate cancer during the study follow-up. However, those who went on to develop prostate cancer also had higher PSA levels at baseline, were more educated, and were more likely to have first degree relatives with prostate cancer. If any of these caused these men to increase their intake of omega-3s as a protective measure, the study could be evidence of causality opposite the claimed direction: Perhaps risk for prostate cancer, real or perceived, increased intake of omega-3s, rather than the other way around.

When something as fundamental as the direction of a causal pathway is uncertain, so, too, are any conclusions generated by a study. An observational study such as this should only be used to generate hypotheses, and raise good questions — not answer them definitively.

Does this study raise a legitimate question? While we fans of omega-3s might prefer a rush to dismissal and derision, the reality is yes, it does. As the authors note, this is not the first study to show an association between higher omega-3 levels and prostate cancer. And in population studies around the world, higher intake of fish and higher rates of prostate cancer have run together.

But in populations, high omega-3 levels could mean many things. Perhaps it means intake of contaminants from fish. The omega-3s might be protective, but maybe there is net harm even so if contaminants in fish are carcinogenic. Our native, Stone Age intake of omega-3 fat was thought to be much higher than prevails today — but our Paleolithic forebears were reliably spared exposure to mercury, and PCBs. We, alas, tend to get our fish with just such stowaways along for the ride.

Perhaps Japanese men who eat the most fish eat the least vegetables — and are harmed by what they are omitting from their diets, instead of, or in addition to, what they are including.

But even with all of the relevant provisos collated, there is the possibility that omega-3 intake — from fish and/or supplements (the new study did not look at sources, so cannot be used to make specific assertions about supplements versus fish) does increase prostate cancer risk. Is this plausible?

Yes, it is. Unexpected, but biologically plausible just the same. Omega-3s are anti-inflammatory, which means that in general they subdue immune system responses. This is likely protective against chronic inflammation, allergy, and autoimmune disease. But while cancer, like other chronic diseases, can be propagated by inflammation, it also involves a rogue cell or cell colony escaping the immune system’s constant police work. An attenuation of that immune system vigilance in the form of an anti-inflammatory effect could, conceivably, help give cancer the upper hand. Immune system function is all about balance. Too much inflammation is bad, certainly. Too little, however, might mean increased risk for infection, and potentially, cancer.

That consideration is important. Omega-3 dosing, whether from food or supplement is not a “some is good, more must be better” proposition. There is an optimum we may not yet fully know, although we have cause to think it’s more than most of us in modern culture consume. Relevant evidence indicates that our native ratio of omega-3 to omega-6 fat was much higher than prevails in modern diets.

Higher intake of omega-3 has at times been associated with compelling potential benefit. One meta-analysis links higher omega-3 consumption to lower risk of diabetes. Another, just out, and representing almost a million women, shows an inverse association between omega-3 intake and breast cancer risk: more fish oil, less breast cancer. If omega-3 fat truly did raise prostate cancer risk but lower breast cancer risk, I guess we men dining with our wives would need to get used to saying: I certainly WON’T have what she’s having!

Is it plausible that higher omega-3 intake increases prostate cancer risk, but decreases breast cancer risk? I suppose the subtleties of carcinogenesis might allow for it, but I find it very far-fetched; if it doesn’t stretch the envelope of credibility to the tearing point, it sure comes close. One of these associations is likely to be wrong.

The new study shows an association between higher blood levels of omega-3 and prostate cancer incidence, nothing more. Despite the inclination to burn them in effigy, the authors themselves claimed nothing more than that. Specific assertions about fish oil supplements are products of media distortion, a problem that routinely bedevils the delivery of medical news — and an issue my next column will show to be a life-and-death concern for us all.

For now, I remain convinced of likely net benefit from the willful inclusion of omega-3s in our diets, from fish and/or plant sources, and from judicious use of supplements. Perhaps excessive intake of omega-3 fat does increase prostate cancer risk, perhaps contaminants in some fish exert that effect, or perhaps the association will prove spurious with more methodologically robust research. We need such research to know for sure.

In the interim, the implications of the new study are prone to distortion in both directions by proponents and opponents alike. Whatever the ideal place for fish oil in our diets, the exaggerations of fish tales make rather dubious food for thought. My advice is to consume them all accordingly.

David L. Katz is the founding director, Yale-Griffin Prevention Research Center.

Comments are moderated before they are published. Please read the comment policy.

  • Dorothygreen

    I also read one of the 9,000 articles on Omega 3 increasing risk for prostate cancer and thought it might be a fluke.

    Here Dr Katz suggests the possibility there is a valid point that too much Omega 3 may indeed cause an increase risk for prostate cancer in some folks.

    Well, I just finished reading, again, about statins – good news in the article, outrage in the comments regarding side effects. Little was said about their strong anti-inflammatory properties. So a light bulb went off and I searched – do statins increase or decrease the risk of cancer?

    Viola! I only looked at one article, there may be more.

    Do statins prevent or promote cancer? April 2008

    “In their commentary, Drs. Takahashi and Nishibori1 discuss putative antitumour effects of statins. However, prospective data suggest that statins actually increase cancer in certain segments of the population. Additionally, new findings regarding the immunomodulatory effects of statins may explain the mechanism by which that increase occurs2.

    OK, so it is not only too much Omega 3 but also too much statin that may increase risk of cancer.

    I don’t have a prostate. But like all folks I have a lot of other body parts that might be at risk for cancer. I stopped fish oil supplements. I don’t like taking food away from the big fish. And, fish oil supplements are concentrated Omega 3 fats just like Omega 6 oils are concentrated pro-inflammatory fats. It is easy to over consume both. In addition one is missing all the other nutrients in the whole foods that contain these essential fats.

    I eat wild salmon in small doses 3 to 4 times a week – about 3/4 to a lb a week and saute on low heat in a bit of coconut oil and grass fed butter, lemon, a touch of salt and good amount of pepper. Chop the salad ingredients while that is cooking – this is a fast food meal. Anything else takes time. Where I live wild salmon is on sale at end of run. I get lots of it, freeze it in small doses.

  • rs986

    Just wondering: is it perhaps possible that men who consume fish oil are also more likely to get PSAs? Because as far as I can tell, the greatest risk to getting diagnosed with prostate cancer is getting your PSA drawn.

    • jere14

      A very good question. A preponderance of prostate cancer seems to be non-progressive and non-lethal while the lethal form has poor treatment outcomes. Based on this I no longer get my PSA checked. It just incites worry and the chance to make a wrong decision based on fear and outside pressure.

  • jere14

    I think that there is too much use and reliance on nutritional supplements. Very often claims are based on anecdotal associations. Before cohort studies were run, fish oil was thought to reduce coronary heart disease based on the high intake of fish in Greece and the low incidence of heart disease. It turned out later than cohort studies did not see a reduction in deaths from heart disease. As a matter of fact the Alpha Tocopherol Beta Carotene study done in Finland (I think) showed that fish oil supplementation increased death from heart disease (risk ratio = 1.3).

    Cohort studies demonstrate association but not causation. Many researchers will end up with the results that they want. Falsified research is a big problem. Meta analyses almost always finds studies with positive associations as well as studies with no association. In such a scenario I believe that it is best to give more reliance on studies that do not challenge the null hypothesis, especially when some kind of an association is the desired result.

Most Popular