By Dr. Neville Wilson. ( 30 July 2013 )
The recent widespread media reports, implicating dietary omega-3 as a risk factor for prostate cancer in males, is likely to generate public concerns about the purported health and safety benefits of dietary oily fish or fish oil supplements.
Understandably, health conscious consumers of omega-3 supplements will be confused by these recent sensationalist media reports, and may be prompted to question, or even abandon, their long held beliefs and practices regarding the protective nature of supplemental fish oils.
A disturbing link between omega-3 polyunsaturated fats (N-3 PUFAS) and increased prostate cancer risk was proposed by researchers in a review of data from a previously conducted trial ( SELECT ), designed to evaluate the protective role of vitamin E and selenium in cancer risk.
The study was commenced in July 2001 and was discontinued in September 2008, on the grounds that no protective effects for selenium and vitamin E were demonstrated for the development of prostate cancer.
(The study in question is a review of data from the previous SELECT study, known as a case-cohort design nested within the Select study ).
Despite the failure of this study to achieve its endpoint goal, and the acknowledgement of it’s authors that “neither of these findings proves an increased risk from the supplements and may be due to chance”, the SELECT trial was promoted as proof that antioxidant nutrients do not have proven benefits of cancer protection. (1)
The concluding remarks by the lead author of the study may be viewed as a hidden bias against the use of anti-oxidant supplements, given his advice that dietary supplements be used with caution.
According to the senior author, Dr. Alan Kristal, “we’ve shown once again that the use of nutritional supplements may be harmful”.
To their credit, the trial authors concede that a limitation of the study was the reporting of fatty acids as weight proportions, “because an increase in the percentage of one type of fatty acid requires a decrease in others”.
THE DATA :
In Table 2 of the study report the distribution of the several plasma phospholipid fatty acids among the SELECT participants is not given as a specific measurement, but rather as a percentage of the total distribution of fatty acids.
Since the percentage of EPA and DHA represents a very small percentage of the total distribution of fatty acids, we are left without answers as to which fats comprised the major proportion of the total distribution, and what their potential impact might have been on the end result.
A curious observation was the correlation between high trans-fat levels and an increased level of protection against the risk of high-grade cancer, a finding which is not supported by the wider body of scientific research.
Data from this study was collected and analysed to further investigate whether high consumption of omega-3 fatty acids could contribute to prostate cancer risk, and the authors reported that “these findings contradict the expectation that high consumption of long – chain omega-3 fatty acids and low consumption of omega-6 fatty acids would reduce the risk of prostate cancer”
The authors report that these findings are not novel, and have been reported in 2 previous studies, with the current findings replicating previous outcomes, but concede that “it is unclear why high levels of long chain omega-3 PUFA would increase cancer risk, and further study will be needed to understand the mechanisms underlying the findings reported here”
The authors have thus conceded that the study outcome is not evidence of a cause and effect relationship between omega -3 and high grade prostate cancer, and that the correlation observed needs to be further assessed in a study specifically designed for that purpose.
The sensationalist reporting by the media did not include these cautionary remarks by the study authors, and they are thus guilty of misleading the public through such false representations of the facts.
HOW RELIABLE ARE THESE CONCLUSIONS ?
The hypothesis of risk, as speculated by the authors of the SELECT study, which gave rise to the media reports, is not supported by hard evidence from within the study, and the researchers concede only a correlation, but not a cause and effect relationship, between omega 3 consumption and prostate cancer.
They also omit valuable information about the possible use of prescription drugs (like statins) by the participants, anti-oxidant supplement intake, the type, frequency, dosages and duration of dietary fat intake, level of physical activity, and the presence or absence of stress, each of these being important factors which inevitably impact on risks to health.
They also provide no details as to whether or not oily fish or fish oil supplements was taken on a regular basis over the duration of the study.
It is not known whether any of the participants were taking statin drugs, widely prescribed for lowering cholesterol levels, or any other pharmacological agents.
Recent findings on the health effects of omega-3 fatty acids and statins, and their interactions, suggest that statins may inhibit the protective benefits of dietary omega-3 fatty acids. ( 2)
OTHER STUDIES :
The outcomes from several reputable studies contradict the speculative conclusions by the SELECT authors, showing “fish oil consumption may be protective against progression of prostate cancer in elderly males” (3) and these findings are supported by population based studies involving Japanese, Swedish and Eskimo males, who have a low incidence of prostate cancer and who consume liberal portions of oily fish regularly.
A Swedish study of 6272 males, over a period of 30 years, showed reduced rates of prostate cancer by oily fish consumption. (4)
Japanese males consuming omega-3 fatty acids reduced their risks for prostate cancer (5)
New Zealand males with high levels of DHA ( a long chain omega-3 fatty acid) reduced their prostate cancer risk by 38 %. (6)
Researchers at the University of California looked for a link between omega-3 intake and prostate cancer in 2009. In a case controlled study of 466 males diagnosed with aggressive prostate cancer they found that an increasing intake of long chain n-3 (omega – 3 ) fatty acid was strongly associated with decreased risk of aggressive prostate cancer. (7)
The absence of hard data from the SELECT study about the duration of fish oil consumption, and the source for the fish, or supplemental products, (if consumed) is one of several weaknesses which characterises this study, rendering it less than an authoritative guide to healthy dietary practice involving supplemental or dietary oily fish consumption.
We are not told what the nature and source was of any ingested omega-3 fats by the participants, prior to their initial blood sampling, and although we do know that they were requested not to use nutritional supplements during the period of the trial, there is no evidence that some may have done upon discovery of raised PSA levels.
The source of dietary oily fish, or fish oil supplements, may impact significantly on the level of health hazard attributed to trial participants, given the high levels of environmental contaminants (PCBs, mercury) that may be present in certain areas of farmed salmon, or inadequately purified omega-3 supplemental products. (8)
Exposure to persistent organic pollutants results in mitochondrial dysfunction in both animal and humans ( 9 ) and may inhibit the protective effects of n-3 fatty acids by an alteration of mitochondrial function.
The study in question did not compare the outcome of fish oil or fish oil supplement by consumption by one group of men, compared to that for another similar group of men who took a placebo instead of supplements.
A randomized placebo controlled study would be required to properly test the hypothesis of risk for prostate cancer.
According to the reported study method, serum was collected from participants in the previous Prostate Cancer Prevention Trial and, in this study, tested for phospholipid long chain omega- 3 fatty acids, together with several other fatty acids, and the distribution (percentages of omega-3 of the total fatty acids) was correlated with prostate cancer incidence, looking at 4 groups of men, graded for (a) no cancer, (b) total cancer (c) low grade cancer and (d) high grade cancer.
OMEGA 3 INDEX:
It is important to note that in this study omega-3 percentages were calculated as a proportion of total phospholipids in the plasma, taken at the commencement of the study, and not repeated at its conclusion.
Omega-3 fatty acids exert their protective functions at the level of cellular membranes, and the study does not reflect cellular membrane omega-3 levels.
The measurement of omega-3 in red cell blood membranes can be obtained by use of a special test known as the HS-OMEGA 3 test, formulated by Drs William Harris and C Von Schacky. (10) A low Index ( <4 ) is associated with an increased risk for morbidity and mortality, while a high Index ( >4 ) is associated with improved survival rates, the target range for a healthy HS-OMEGA-3 is 8-11%.
According to Dr. William Harris, the lowest quartile of omega-3 in the SELECT participants would correlate to an HS-OMEGA 3 Index < 3.16 % and the highest quartile to an HS-OMEGA 3 Index of > 4.77%, suggesting that men in the highest quartile of the study ( Table 3 of the study ) still had low levels of membrane omega – 3, and they were the ones who were at greater risk for high – grade prostate cancer.
Since blood samples to evaluate the levels of EPA and DHA (omega-3 fatty acids) were not repeated during the duration of the study it is not possible to evaluate a pattern of dietary oily fish or supplemental intake, or dosages consumed.
In the final analysis, only DHA proportions (and not EPA) correlated with high grade prostate cancer, and interestingly, ( and surprisingly ) higher proportions of trans fatty acids, known for their harmful effects, correlated with a lower incidence of prostate cancer.
(The study might well have served to show that an increase of dietary trans-fats correlates with a decreased risk for prostate cancer, but no scientific evidence exists to support such a proposal !)
The wide range of fatty acids isolated in this study included EPA, DHA, trans fats 18:1, trans-fats 18:2, trans-fats 16, ALA, LA (18.2 omega 6), arachidonic acid, and mystery fats, which comprised the largest proportion of the total phospholipid count.
Any one of these unknown fatty acids could have been implicated in the prostate cancers observed.
ELEVATED DHA REPORTED:
The elevated DHA levels that were documented may have resulted from a previously ingested oily fish, but may also reflect a low fat diet, since low fat diets have been shown to raise DHA levels. (11 )
In a study by Raattz et al a low fat diet comprised of 20% fat (n-6 / n-3 ratio 11.1) was compared with a high fat diet of 45 % (n-6 / n-3 ratio 12.3 ) and resulted in a 28% higher membrane DHA level
Conventional dietary advice continues to promote low fat diets as nutritionally preferable to high fat diets, despite the evidence that low fat diets are usually higher in carbohydrate content, and more likely to precipitate insulin resistance, triglyceride increases, weight gain and attendant pathologies which are more likely than with high fat intake.
It is not known whether some of the study participants had been following a conventional low fat diet prior to, and throughout the duration of the study.
OMEGA- 3 TOXICITY ? :
Since Omega-3 fatty acids are long chain carbon bonds they are potentially unstable, and thus susceptible to oxidation, given an unfavourable pro-oxidant environment.
EPA ( eicosapentaenoic acid) is an omega-3 (n-3) polyunsaturated fatty acid with 20 carbons and 5 double bonds.
DHA ( docosahexaenoic acid ) is an omega-3 ( n-3 ) polyunsaturated fatty acid with 22 carbons and 6 double bonds.
The longer the chain, and the greater the number of double bonds, the greater the risk for oxidative damage to these fatty acid chains, and this should be taken into account if large quantities of omega-3 are consumed, particularly in the absence of protective anti-oxidants.
These long chains are formed from their parent fatty acid (ALA) by a series of desaturation (oxidation) and elongation, and while they are conditionally essential to health, they are relatively unstable and can be readily denatured if exposed to further oxidation.
Anti-oxidants are usually added to omega-3 supplements in order to protect them against oxidation.
Thus, too little, or too much, omega-3 in an unpurified or unprotected form, may be potentially harmful, and this effect has been demonstrated in experiments with mice. (12 ).
In conjunction with alcohol and sugar, excessive doses of omega-3 produced harmful outcomes for pregnant rats. ( 13 ).
High levels of omega-6, which characterise the western diet, will reduce levels of omega-3 through competition for the delta -5 and delta -6 desaturase enzymes, and tilt the omega-6 / omega-3 balance in favour of inflammatory mechanisms in the body.
An excess of dietary omega-6 will stimulate the formation of prostaglandin E2, via the cyclo-oxygenase (COX) enzyme pathway, thereby enhancing new tumour growth through angiogenesis (new vessel growth). (14).
The potential for cancer development is therefore present when the ratio of omega-6 / omega- 3 is increased, the ideal ratio being 2:1 or even 1:1.
We have no data on the nutritional status of the participants in this study, and are thus incapable of assessing the underlying reasons for the increased incidence of prostate cancer.
OXIDATION OF DHA :
It has been shown that the oxidation of the longer chain deoxyhexaenoic acid (DHA) can produce a protein called carboxyethylpyrrole (CEP ), which is not produced by the oxidation of other poly-unsaturated fats, and which may be implicated in new vessel formation (angiogenesis ) within the retina of the eye, with subsequent risk for new vessel growth, retinal damage, and blindness, or even tumour progression. (15)
The anti-oxidants glutathione peroxidase and superoxide dismutase are present in the retina, and because of the high intensity of photogenerated radicles within the eye, these anti-oxidants are replaced every 12 days.
In the absence of protective dietary anti-oxidants such as glutathione peroxidase, superoxide dismutase, selenium, astaxanthin, vitamin E – gamma tocopheral, and vitamin C, the risks for tumour progression may be increased, and we are not told whether the SELECT participants had taken anti-oxidant supplements or not, despite having been requested not to do so.
INTERACTIONS OF OMEGA-3 WITH STATINS :
In accordance with conventional dietary advice, large sections of the population are likely to observe a low fat dietary regimen, and many are also taking statin drugs, falsely believing that such practice is health protective.
The protective efficacy of omega-3 has recently been called into question after several recent randomized controlled studies, since 2005, failed to show a reduction in mortality rates from coronary heart disease for persons taking omega-3 supplements.
These recent “negative” findings of omega-3 benefit contrast sharply with earlier (before 2005) “positive” findings, where significant reductions in cardiovascular mortality were reported in those participants taking omega – 3 fatty acids.
The well known DART (16), GISSI- Prevenzione (17), and MEDITERRANEAN alpha-linolenic (18) demonstrated significant cardio-protective properties for omega-3 fats, confirmed by a large meta-analysis, including prospective, as well as randomized controlled studies, by Mozaffarian et al, in 1999. (19)
Several recent studies have not supported these earlier findings, suggesting that other factors may have inhibited the previously demonstrated protective properties of omega-3 fatty acids.
A reasonable explanation for this discrepancy, as suggested by Michel de Lorgeril, is the fact that statins were not in use during the earlier trials, and their use in the more recent trials may have inhibited the protective potential of omega-3 used in those trials. (20 )
In a meta-analysis of randomized, double- blind, placebo controlled trials, researchers Kwak et al found that participants taking EPA and DHA decreased their risk of cardiovascular events when they were NOT receiving statins, but increased their risk when they were taking statins. (21), suggestive of a strong interaction between omega-3 fatty acids and statin drugs.
In the GISSI-HF trial the effects of n-3 EPA/DHA was evaluated in patients with chronic heart failure, and in a combined study, the effects of a powerful statin (rosuvastatin / CRESTOR) were also evaluated, and neither the omega-3 nor the statin, were shown to be protective for patients with chronic heart failure. ( 22)
Again, a possible interaction between the statin and omega-3 fatty acid resulted in this reciprocal inhibition of the one by the other, as proposed by Michel de Lorgeril (20 )
OTHER STATIN EFFECTS:
However, the failure of rosuvastatin (CRESTOR) to provide protection against heart failure patients in the absence of omega-3 in a trial by Kjekshus J et al, (23) must be explained by another mechanism, such as Co-Enzyme Q10 depletion, for which statins are notorious.
Statins deplete Co-enzyme Q10, a vital respiratory factor essential for effective mitochondrial function in the cells of heart muscle, and are toxic for the mitochondria in a dose-dependent manner. (24)
Disturbances in mitochondrial functioning may lead to a wide range of pathological states, including infectious diseases, myocardial infarction, heart failure and cancer. (25)
Other statin effects which may inhibit the protective properties of omega-3, are raised omega-6 levels as a result of statin induced arachidonic acid increases.
Omega-3 is more effective when levels of omega-6 are low, and less effective when omega-6 levels are high.
If statins had been used by any of the participants in this study, their hidden potential for inhibiting the protective properties of omega-3 would have gone unnoticed and not reported.
The failure of this study to take account of several confounding factors, as mentioned above, the absence of data regarding the actual dietary intake of omega-3 fatty acids by the participants, and plasma measurements which do not reflect cellular membrane levels, as obtained from the HS-OMEGA 3 Index, renders the reported outcome of this study inconclusive for a causal relationship between omega -3 and prostate cancer.
While association studies are useful in that they identify an area for further research, they do not prove causality, and cannot be interpreted as such. A specifically designed interventive study with a clearly defined clinical endpoint is required to corroborate the conclusions of this study.
Dr. Neville Wilson.
The Leinster Clinic.
- The Journal of the National Cancer Institute. July 10, 2013.
- BMC Medicine 2013, 11:5 Michel de Lorgeril et al.
- PLoS 8(4): Oct 11,2012
- Lancet 2001, June 2; 357 (9270): 1746-6
- Cancer Science vol 90, issue 9, p914-921,Sept 1999
- Br J Cancer 1999, Dec; 81 (7): 1238-1242
- Clin Cancer Research 2009 April 1; 15(7): 2559-66 Fradet et al
- Urol Oncol 2012, March-April; 30(2) 216-9
- PLoS ONE 2011, be5170
- www.omegamatrix eu
- The Journal of Nutrition, Feb 1, 2001, vol 131, no 2, 231-234. Raatz S K et al
- Gastroenterology, 1995 Aug; 109 (2): 547-54
- Neurotoxicol Teratol 2010 Mar-Apr; 32(2):171-181
- J Biol Chem 2003 Oct 24;28 (43): 42027-35
- Lipid Research Lab- Robert G Salomon
- Lancet, 1989,2:757-761
- Lancet 1999, 354:447-455
- Lancet 1994, 343:1454-1459
- JAMA 2006, 296: 1885-1899
- BMC Med 2013, 11:5
- Arch Intern Med 201§2, 172: 686-694
- GISSI-HF Lancet 2008, 372; 1223-1230/1231-1239
- NEJM 2007, 357: 2248-2261
- Cell Mol Life Sc 2006, 63: 2415-2425
- NEJM 2011, 364: 829-841