Facebook Question: Fish Oil & Prostate Cancer

We received the question below following our recent series discussing fish oil and prostate cancer.

Question: 

Thank you for addressing this sensationalized issue. But I am still confused. The study only revealed a correlation with DHA and not with EPA. Yet the amount DHA in the bloodstream can be elevated just by following a low-fat diet - without eating any fish or taking any fish oil supplements at all. Reference here (Raatz, SK et al.  Total fat intake modifies plasma fatty acid composition in humans.  Journal of Nutrition 2001; 131:231-234).

Our Response: 

Thank you for pointing out the reference by Susan Raatz and colleagues, and for raising the question of its implications for the study by Brasky et al.  You are raising a rather technical question, and we will have to give a somewhat technical answer.  First of all, we would like to point out that Brasky et al did not find a correlation of high grade prostate cancer only with DHA.  The correlation that they explored the most in their statistics was the sum of EPA, DHA and DPA omega-3 fatty acids.  Nevertheless, among the three fatty acids, DHA was, indeed, the one that showed the most difference between the low-grade and high-grade prostate cancer. 

Raatz and colleagues performed a study in which volunteers were fed either a high fat (45%) or a low fat (20%) diet for a month, and then switched to the opposite diet, after a month of their usual diet, to wash out the effect of the first experimental diet.  While the diets differed in the amount of fat they included, they had similar ratios of saturated fat, monounsaturated fat and polyunsaturated fat, with roughly similar proportions of omega-3 and omega-6 fats.  The way they did this was to put a lot of sugar in the form of fruit juice and table sugar into the low fat diet to dilute the fat – not something we recommend!!

Subjects’ blood was analyzed for fatty acids after both the high fat and the low fat diets.  The plasma phospholipids were analyzed, just as in Brasky’s study.  You are correct that Raatz and colleagues found that DHA could be increased by the low fat diet, without any change in fatty acid proportions or supplementation with fish or fish oil. The authors speculate that this has to do with competition between omega-3 and omega-6 fatty acids for enzymes that turn plant-derived omega-3s from sources like soybean oil into EPA and DHA.  We’d note that Brasky completely overlooks this source of EPA and DHA in his discussion of this article, and ascribes all EPA and DHA to fish or fish oil.  That’s actually not an accurate characterization, since we do derive some of our EPA and DHA from a fatty acid called alpha-linolenic acid, though our bodies don’t do such a great job of converting it.  Just one of several deficits in the Brasky article!

But let’s look at exactly what the Raatz and Brasky studies found.  Could it actually be that a low-fat diet is implicated in high-grade prostate cancer?  Here is a table that shows the amounts of DHA and EPA in both the Raatz study of low fat and high fat diets, and the Brasky study of low grade and high grade prostate cancers.  We are also showing the amounts of the major omega-6 acid in the blood, linoleic acid, abbreviated LA.  This omega-6 fatty acid is a step in the body’s synthesis of inflammatory prostaglandins. 

Plasma phospholipid %

	Raatz low fat	Raatz high fat	Brasky low grade	Brasky high grade
DHA	4.53*	3.54*	3.01*	3.09*
EPA	0.71*	0.51*	0.66	0.65
LA	21.55*	28.55*	18.91	19.09

The numbers that have asterisks beside them are significantly different from the corresponding numbers in their articles.  So,  DHA, EPA and LA differ significantly between the low fat and high fat diets in Raatz’s study, and DHA differs significantly between the low grade and high grade prostate cancer patients in Brasky’s study.  The numbers that are underlined are statistically similar.  So, EPA and LA are similar between the low grade and high grade prostate cancer patients.

Comparing the levels of the fatty acids between the studies gives some interesting insights.  First, it confirms something Brasky pointed out, that the men in his studies had somewhat low levels of omega-3 fats, especially DHA.  But let’s look at the levels of LA in the two studies.  In Brasky’s study, the levels of LA are the same in the low grade and high grade prostate cancer patients. They are both at more or less the same level as the low fat diet in Raatz’s study, perhaps a little lower.  However, in Raatz’s study, the high fat diet resulted in significantly and noticeably higher levels of linoleic acid.  This is not seen in Brasky’s study, where the LA numbers are statistically similar.  If a low fat diet was causing the elevated DHA levels in Brasky’s study, as you suggested based on the results of Raatz’s study, wouldn’t it also cause a lower level of linoleic acid, just as happened in Raatz’s study?  But the levels are the same, so we can’t  see how to ascribe the higher DHA content in the high grade prostate cancer patients to a low fat diet. We also note that in Raatz’s study, the low fat group had increased levels of EPA and the high fat group had less EPA.  The levels of EPA were virtually the same in the low grade and high grade cancer patients.

Ultimately, it is impossible to tell what was really going on in Brasky’s study, because of the lack of any dietary data.  So we’ll have to leave the discussion somewhat unresolved.  In the meantime, you may be interested in reading the blog article we posted this past Friday about the recent study by Lenore Arab, which implicates the consumption of high amounts of red