Yesterday’s blog post looked at the study by Thomas Brasky
and colleagues, which suggested that men who had high-grade prostate cancer had
slightly higher blood levels of the omega-3 molecules from fish, EPA and
DHA. Based on this result, the authors
cautioned men about using fish oil or eating fish. I reviewed the quite substantive criticisms
of the article, but also mentioned that looking into the background of this
study uncovered some other potentially useful insights about fish and prostate
cancer. My staff and I also examined the
literature in this area and came across some surprising findings from other
studies.
One of the major issues with Brasky’s study is the lack of
dietary data in the study, since it only examined levels of omega-3s in the
blood. In 2010, a meta-analysis was
published by Symanski and colleagues that analyzed the relationship between
fish consumption and prostate cancer. In
a meta-analysis, researchers gather all the human studies on a specific
topic. The studies can be either
clinical trials or epidemiological studies.
The data from all the studies are grouped together statistically and analyzed
to answer the question at hand. This
gives a much more comprehensive and unbiased view of a subject than simply
gathering whatever studies you happen to know about, or collecting only the
studies that support your own position.
In the case of the fish and prostate cancer meta-analysis, researchers
found that, based on multiple epidemiological studies of different designs, the
amount of fish that men consumed did not predict whether they were going to get
prostate cancer. However, they also
found that higher fish consumption was linked to lower mortality rates for
prostate cancer, suggesting that fish consumption might be beneficial for men
who already had prostate cancer. They
could not analyze the relationship between fish and incidence of high-grade or
advanced prostate cancer, because there were too few studies to draw
conclusions.
But along with this study indicating that fish does not
increase prostate cancer risk is another meta-analysis, by Chua and colleagues,
in which studies similar to that of Brasky and colleagues were collected and
analyzed together. Yes, there have been
other studies like that of Brasky – they just have not caught the attention of
the press the way this one has! The 12
studies analyzed in this meta-analysis explored the relationship of EPA and DHA
in blood plasma or serum or in red blood cells to prostate cancer occurrence.
This gives us a broader picture than Brasky’s study, which only analyzed one
possibly questionable blood component.
What this study found was that blood levels of EPA and DHA were not
related to overall prostate cancer risk (although, interestingly, a minor fish
oil constituent called DPA, was related to lower prostate cancer risk). This is similar to the results of Szymanski’s
meta-analysis on fish consumption. However, Chua found that levels of EPA and
DHA may be linked to slightly higher rates of high-grade prostate cancer. This result is a little questionable, though,
because they had to exclude one study (the Physician’s Health Study) from the
analysis because it was quite different from the other studies and did not
account for the prevalence of risk factors in its study population. So, this study shows us that there is, again,
no link of fish constituents to overall prostate cancer risk, but that there is
a possible link to high-grade prostate cancer.
A possible link to high-grade prostate cancer was also found
in another study on fish consumption, published in 2012 by Joshi and
colleagues. This study found that
advanced prostate cancer was more common in men who ate large amounts of white
fish (flounder, halibut, cod, sole) that was cooked using high-temperature
cooking methods (grilling, pan-frying, oven-broiling or barbequeing) until it
was well-done. Consumption of white fish
cooked at low temperatures or cooked until just done was not related to
prostate cancer. Nor was consumption of
tuna or dark fish (salmon, mackerel and others). While we think of salmon as being the main
source of omega-3s, the fact is that flounder, cod, sole and halibut are also
rich in omega-3s. And many people eat
more of these fish than of salmon, often in the form of fish sticks, fish fries
or commercial fried fish sandwiches (one popular brand of fish sandwich uses pollock,
another high omega-3 fish). And, all of
these are examples of high-temperature cooking, which, can be extremely adverse
clinically, with the potential for compromising health consequences, and are
quite relevant to the findings of these studies.
This study was followed up by a laboratory study led by Dr
Chelsea Catsburg, which explored a possible mechanism for the link to prostate
cancer. These researchers noted that
cooking white fish at high temperatures generates carcinogenic molecules called
heterocyclic amines (HAs) and polycyclic aromatic hydrocarbons (PAHs). The exact same phenomenon is seen when red
meat is cooked at high temperatures, and is behind the caution against eating
grilled, charred meat, also thought to be carcinogenic. HAs and PAHs generate free radicals in the
body, which can initiate cancer-causing mutations. Catsburg’s team suspected that variation in
genes involved in the detoxification of carcinogens might be involved in the
link between overcooked white fish and prostate cancer. They analyzed the genes involved in
detoxifying carcinogens in localized prostate cancer patients and advanced
prostate cancer patients. They found
that men with a particular form of one gene (PTSG2 765 G/C) who ate lots of
white fish did, in fact, have a higher risk of having advanced prostate cancer,
especially if they ate well-done white fish.
Men with other forms of the gene did not have the risk for prostate
cancer even if they ate well-done white fish.
Catsburg’s study tells us that it may not be the EPA and DHA
in fish that is the potential culprit in causing prostate cancer. Instead, it may be the HAs and the PAHs
lurking in all the fried white fish consumed by American men. The EPA and DHA may just “be along for the
ride!” This issue highlights why the
lack of dietary data in Brasky’s study is significant. Were those men with high-grade prostate
cancer eating too many fish sticks and fish sandwiches? We just don’t know, since this study has no
dietary correlations. Yet this missing
information is extremely relevant. In
addition, since all of the men in Brasky’s study apparently had rather low
omega-3 levels, one suspects that they may have been eating suboptimal diets,
perhaps including the regular consumption of fried fish and fish sandwiches.
Now let’s take a look at an interesting characteristic of
both EPA and DHA. These molecules are
prone to getting oxidized – in other words, rancid. Rancid oils from fish and other sources contain
damaging free radicals. In fact, one
concern with fish oil supplements is that some of them may be partially
oxidized; in other words, they may contain damaging free radicals. If this was the case with the fish oils in
the Brasky study, for example, it makes sense that this was a contributing
factor to the study’s findings. So far
then, we have 2 mechanisms by which fish or fish oil may be linked to prostate
cancer, though neither implicates the fish itself! One involves the method of cooking – we know
that cooking at high temperatures causes the production of carcinogenic chemicals
– and the other involves potential rancid fish oil constituents. In addition, there is a THIRD mechanism that
has been suggested as possibly being linked to prostate cancer: mercury pollution. Exposure to mercury has been linked to
chronic inflammation of the prostate in rats.
Inflammation increases the risk for prostate cancer.
In tomorrow’s blog, I will detail what I am advising our
patients in light of this research, and what we should all be concerned
about.