Prof Patrick Holford of Teesside University excites remarkable admiration; so much so that Holford Watch is regularly reproved for merely questioning the Holford interpretation of science and nutrition.
We’ve previously referred to the oddities of the Holford way with simple numbers whereby, e.g., 4 days is the equivalent of a month and a striking inability to interpret simple messages accurately.
Looking through Holford’s latest newsletter to his 100%health members (sadly, not available online), there is an embarrassment of advice that is grounded in a quirky interpetation of the referenced research.
Take heart disease. Studies with seal oil have proven better for reducing cardiovascular disease risk than fish oil (not that I’m recommending you eat seals!). [ref 1] There’s good reason to believe it’s the DPA that makes the difference. While EPA and DHA both thin the blood – this happens by stopping little ‘platelets’ in the blood from clumping or sticking together – recent research shows that DPA does it better. [ref 2] Also, when omega 3 intake is low, arterial lesions develop faster, this is what you want to stop because this can ultimately block an artery. While EPA inhibits this process, DPA is approximately six times more effective. [ref 3]
DPA may also be important for preventing or slowing down cancer. For tumours to grow they have to develop a blood supply. This is called angiogenesis. DPA has been shown to inhibit this process and slow down growth of breast, colon and prostate cancer cells. [ref 5]
Well, yes and no. The general reader would need good luck with finding some of these references as they are not available on Entrez Pubmed which gives access to the indexed medical journals. Even when the reader is fortunate enough to track down some of these references it seems as if several of them report studies with rabbits, cows, rats and mice rather than humans and are typically laboratory studies in vitro rather than in vivo and living creatures.
Ref 1 was to two papers. Mann et al. The effectiveness of DPA rich seal oil compared with fish oil in lowering platelet activation in healthy human objects. International Society for the Study of Fatty Acids and Lipids (ISSAFI), July 2006 (I think that this is CS 19.2. Also, Meyer et al, ISSFAL, July 2006 (P0 16). Those are the references that I could find; you will notice that the correct title for the first paper refers to “human subjects” rather than “objects”. It is worth noting that the Mann reference is to the abstract of a conference paper and the Meyer to a conference poster. If the links do not work (they are a little flaky) then you may need to go to ISSAFI, navigate to the archives, to the 2006 conference, Monday 24 July and then the Poster Session A for the Meyer poster (and the later Sparkes, mentioned below) or navigate to Thursday 27 for the Mann paper.
For the Mann paper, there were 30 healthy participants (age and gender unknown) and these were randomised to 3 groups of 10 to test either a placebo, seal oil or fish oil. The study period was 14 days and the blood analyses were performed on samples drawn at the start and close of the study. The authors concluded:
This study further supports the suggestion that DPA may be more beneficial than EPA or DHA in lowering CVD risk factors related to platelet activation and blood lipids.
So, we don’t know anything about the age, lifestyle, nationality or gender of the participants although all of these may be relevant to fatty-acid dietary intake, uptake and metabolism. We don’t know that the changes in blood lipids etc. would be maintained if supplementation were continued or if it would revert to baseline or otherwise alter. There were 10 people in the group that took seal oil capsules. The researchers looked at some parameters that are associated with cardiovascular disease risk and specified ones associated with platelet activation or aggregation and the blood lipid profile. They reported:
Platelet activation (as measured by p-selectin expression) was decreased after consumption of seal oil only, however platelet aggregation demonstrated no changes post supplementation in any of the three groups. [Emphasis added.]
This is particularly interesting in the light of Holford’s above quoted segue into the assertion that DPA is better than EPA or DHA at preventing platelet aggregation: Holford provides another reference that I shall discuss below. However, for completeness here, it is worth emphasising that neither of these studies has been published in a peer-review journal and there is insufficient detail available about the methodology, materials and other matters to form a proper opinion as to the merit of either; as such, Holford is crediting these studies with a status that they might not as yet merit by failing to qualify his interpretation of their reported findings. It seems rather premature to be discussing these studies as if they are adequate to support recommendations for fatty acid supplementation to reduce CVD risk for the general population.
Ref 2 S Akiba et al. Involvement of lipooxygenase pathway in docosapentaenoic acid-induced inhibition of platelet aggregation. Biological and Pharmaceutical Bulletin (2000), vol 23 (11), pp 1293-1297.
The full paper provides a reasonable overview of some of the work that has been done with seal oil which suggests that DPA has antithrombotic properties (inhibits the clumping of blood platelets) although the biological mechanism is not as yet fully described. It does seem as if there are interesting results from very small scale studies although it is worth remembering that there is little peer-reviewed work in this area and not much has been carried out in humans.
The study examined the impact of n-3 fatty acids on platelets derived from rabbits blood which were then exposed to various products to induce aggregation. For some of the provocations, the authors reported that there was no significant difference between the performance of EPA, DHA and DPA although DPA showed greater potency for inhibiting collagen- and arachidonic acid-induced aggregation.
The authors suggested that it was plausible to examine whether these fatty acids were inhibiting platelet aggregation via some putative interference with cyclooxygenase-dependent arachidonic acid metabolism and whether the greater potency of DPA was confirmed in this study involving a test substance where it had previously outperformed EPA and DHA. As might have been expected, all of the n-3 fatty acids directly inhibited cyclooxygenase but DPA returned the greatest inhibition.
Follow-up testing of the effect of n-3 fatty acids on the lipoxygenase pathway indicates that DPA may be more potent in vitro than EPA and DHA.
However, the authors are careful to state that:
the inhibitory effect of DPA was more potent than that of EPA or DHA in response to collagen and arachidonic acid…In [other]-stimulated aggregation, however, there was no difference.
It is an interesting study of the impact of n-3 fatty acids on rabbits blood platelets: it may be a useful pointer towards the impact of specific n-3 fatty acids on several inflammatory mechanisms. However, it seems to be a substantial leap from that finding to the claim that:
[s]tudies with seal oil have proven better for reducing cardiovascular disease risk than fish oil.
The next reference was supposed to point towards substantiation that DPA is more effective than EPA at inhibiting arterial lesions in conditions of low omega 3 intake (as above).
Ref 3 M Tsuji et al. Prostaglandins Leukotrienes & Essential Fatty Acids (2003), vol 68, pp 337-342. Also T Kanayosu-Toyoda et al., Prostaglandins Leukotrienes & Essential Fatty Acids (1996), vol 54, pg 319. (I’ve provided the links to save a difficult search both papers, particularly as the latter should have read pp 319-325).
Tsuji et al report that they:
examined the effect of docosapentaenoic acid (DPA, 22:5, n-3), an elongated metabolite of EPA, on tube-forming activity in bovine aortic endothelial cells (BAE cells) incubated between type I collagen gels. [Emphasis added.]
It’s not entirely clear that a laboratory study of the effect of DPA on the inner cells of coronary blood vessels in cows is entirely ready for the primetime publicity that it has in this newsletter.
Similarly, the Kanayosu-Toyoda et al. paper offers some interesting results but it is, again, an in vitro study involving mechanisms of wound repair in cells from bovine blood vessels. You might wish to avert your eyes for the quibble that this paper looked at the different potency of the influence of DPA that was added directly to the sample, or DPA as a metabolite of a dose of EPA that was added to a sample. The authors conclude:
these data suggest that the stimulative effect of EPA on [endothelial cell] migration occurs via DPA, and that DPA may act as a powerful anti-atherogenic factor.
Neither of the papers for this reference directly validate the statements that:
when omega 3 intake is low, arterial lesions develop faster, this is what you want to stop because this can ultimately block an artery. While EPA inhibits this process, DPA is approximately six times more effective.
In particular, neither of these studies directly relates to humans.
Holford makes a particularly interesting claim that:
DPA may also be important for preventing or slowing down cancer. For tumours to grow they have to develop a blood supply. This is called angiogenesis. DPA has been shown to inhibit this process and slow down growth of breast, colon and prostate cancer cells. [ref 4]
Ref 4 W E Hardman. n-3 Fatty Acids and Cancer Therapy. Journal of Nutrition (2004), vol 134, pp 3427S-3420S. Also, DP Rose & JM Connolly. Regulation of tumor angiogenesis by dietary fatty acids and eicosanoids. Nutrition & Cancer (2000) vl 37 (2), pp 119-27.
Hardman’s conference paper provides an overview of studies that report that supplementing the diet of tumour-bearing mice or rats with oils containing (n-3) (omega-3) or with purified (n-3) fatty acids and reports that the oils slows the growth of various types of cancers, including lung, colon, mammary, and prostate. Hardman does not explicitly address the role of DPA rather than Omega-3 fatty acids in this review. In contrast to Holford’s interpretation, Hardman limits her conclusion:
preclinical studies indicate that (n-3) fatty acids should be beneficial for cancer treatment. Mechanistic studies indicate feasible mechanisms for the influence of (n-3) fatty acids on tumor growth, survival, and response to chemotherapy. A limited number of clinical studies indicate that (n-3) fatty acids may be beneficial when consumed before chemotherapy. It seems important to commence human trials using an (n-3) fatty acid as a supplement to standard chemotherapy.
Rose and Connolly explored the observation that high-fat, n-6 fatty acid-rich diets are associated with a relatively poor prognosis in breast cancer patients and the cancerous transformation of pre-malignant lesions. They report that in an appropriate laboratory mouse model, a similar high Omega-6 diet enhanced breast cancer progression, whereas Omega-3 fatty acids exerted suppressive effects that were associated with the impaired formation of new blood vessels that might service a tumour. This paper refers to Omega-3 fatty acids and does not explicitly mention the impact of DPA. The authors of this study do not go as far as Holford but limit their conclusions to the following:
Review of the experimental data suggests that selective inhibitors of eicosanoid-synthesizing enzymes and dietary intervention with n-3 fatty acids merit clinical evaluation as adjuvant therapy and chemopreventive agents. [Emphasis added.]
Holford uses the above papers to support his claim that:
DPA has been shown to inhibit this process and slow down growth of breast, colon and prostate cancer cells.
Although (as above) Holford does qualify his optimistic claims for DPA with the qualifier that it
may also be important for preventing or slowing down cancer
neither of these papers explictly discusses DPA. Neither of these papers makes comparable assertions for cancer in humans.
The papers that are referenced by Holford may be interesting but they do not validate his claims for them or, by extension, his recommendations for supplementation and particular dosages.
I recommend the following daily levels:
If you suffer from depression or have recently had a heart attack, on the basis of what is known now, I’d also supplement an additional 500mg of EPA. But, for all we know, DPA may also prove to be a potent mood booster. Maybe that’s how the Eskimos stay cheerful during the winter, with four months of total darkness.
One last thing-whenever you take a fish oil supplement, ideally derived from salmon, it’s important to check that it has been purified to the highest standards to remove any dioxins and PCBs that sadly pollute all our oceans.
The “for all we know” might be more generally applicable to Holford’s claims and recommendations in this article than this particularly waggishness about DPA and reasons to be cheerful in dark months. Without further justification, it is a little odd that the reader is expected to accept the findings of small studies that praises seal oil in conference papers and posters, but Holford chooses to ignore a review by Cassandra Sparkes et al. in PO25 at the same conference (see above for navigation notes if the link fails) that reports that:
[f]or depression, lower doses (1g/d) may be more beneficial than higher doses (2 or 4 g/d).
So, a review recommends that 1g per day may be more beneficial than a larger dose but Holford has recommended up to 700mg of EPA alone. If this sounds like a sour point, then consider how much ferreting about in obscure places has been required to examine Holford’s specific claims in this newsletter item and to demonstrate that there is substantial reason to find those claims unsubstantiated.
There are relatively few peer-reviewed studies about seal oil available on Entrez Pubmed. It might be reasonable to suggest that some studies suggest that seal oil supplementation may provoke shift in fatty acid composition in serum toward a putative anti-inflammatory profile. However, no matter what Holford claims, it is not clinically established that DPA offer significant advantages for a long-term supplementation strategy for the general population beyond the availability of Omega-3 from the regular consumption of oily fish.
One may feel the truthiness that some references support one’s viewpoint but it is not acceptable to use that in preference to evidence-based arguments to promote those viewpoints or recommend dosages for a general population.
Update 20 Sept: James C has pointed us towards Holford’s somewhat different attitude to DPA in the April 2007 edition of New Optimum Nutrition for the Mind.
Another negative factor is alcohol. It not only blocks the conversion of fats into DHA, but dissolves fatty acids within the brain’s membranes and replaces DHA with a poor substitute docosapentaenoic acid (DPA). This may be one of the main reasons why alcohol consumption is associated with mental impairment.
Alcohol worsens your memory by dissolving fatty acids within brain cells and replacing the beneficial brain-building omega-3 DHA with a poor substitute, docosapentaenoic acid or DPA.
Holford Watch has thought about the matter and intuited that Holford may have had Hibbeln’s theories in mind but he has provided no references on this matter; The Guardian carries a reasonable introduction to the Hibbeln’s theories of fatty acid supplementation and violent, alcoholic people. At the risk of being tedious, there may be significant differences between the brain material of people who eat very different diets (e.g., Australians, Americans and Japanese people). It would also be helpful if Holford clarified in situ whether he was discussing the Omega-3 or Omega-6 varieties of DPA.
The newsletter carries no caveats or explanations for this volte-face so it must be an exercise for the reader without any Holford guidance. The quality of Holford’s research and writing is truly extraordinary; I have a twinge of empathy with the dissolving brain cells.