The Cholesterol Times, Issue #007 -- Cancer: It's Not the Meat, But How You Cook

[Guest guest]

Sun, 16 Oct 2005 03:33:06 UT

" Christopher Masterjohn "

<The_Cholesterol_Times

The Cholesterol Times, Issue #007 -- Cancer: It's Not the

Meat, But How You Cook It

 

 

cholesterol-and-health.com

 

Research Watch

 

Cancer: It's Not the Meat - It's How You Cook It

 

Pancreatic cancer is the fourth most common cause of cancer deaths in

the US, and is the most fatal cancer: 5-year survival rates are less

than five percent. Studies associating dietary factors with pancreatic

cancer have been inconsistent. Some have found associations with

animal products, while others have not. Most studies finding such an

association have wrongly attributed it to fat, saturated fat, or

cholesterol without justification, which is contradicted by a new

study published by Nothlings and coworkers, on October 5, 2005, in the

Journal of the National Cancer Institute.

 

The study analyzed data from the Multiethnic Cohort Study, collected

prospectively (meaning the dietary data was collected before disease

occurrence) over seven years, from more than 215,000 men and women

aged 45-75 years old, 190,545 of whom contracted pancreatic cancer.

 

While most studies claiming to indict meats and other animal products

in disease pay little attention to the quality of the meat, this study

provided analysis specific to processed meats, and discussed the

possibility of the confounding effects of high-temperature cooking and

grilling.

 

The authors wrote in the abstract that " the strongest association was

with processed meat, " noting a 68% increased risk in the uppermost

quintile of consumption compared to the lowest quintile, and noted

that the same comparison for pork and total red meat (beef, pork, and

lamb) were about 50%, while no associations were found for other

animal products, fat, or saturated fat.

 

Beef consumption itself had a 21% increased risk comparing the highest

and lowest quintiles, but the fourth (second highest) quintile had

only a two percent increased risk, only a quarter of the eight percent

increased risk seen in the second lowest quintile, making the

association inconsistent and unconvincing. The association with pork

was not linear, but was somewhat more convincing, being much higher in

the fourth and fifth quintiles (44% and 53%) than in the second and

third quintiles (14% and 12%), compared to the lowest quintile. " Red

meat, " which the authors considered to mean beef, pork, and lamb, had

a 45% increased risk when comparing the fifth quintile to the first,

yet the second-highest quintile, the fourth, had only a 3% increased

risk over the first! The third quintile had a 27% increased risk, and

the second quintile had a 6% increased risk. Again, the association

was very inconsistent and unconvincing.

 

The authors noted that " processed meats " had the highest correlation

out of all meats: 68% increased risk in the highest quintile compared

to the lowest. Again, consistency was lacking. The third (middle)

quintile actually had the highest association at a whopping 80%, while

the second (59%) and fourth (47%) quintiles both had lower

associations than the middle or highest quintile. However, the

association with processed meat is still the most convincing out of

all associations, because the associations were relatively high in all

quintiles compared to the lowest, whereas the others showed weak and

negligible correlations interspersed between strong ones.

 

Unfortunately, where this study failed to deliver was in providing

data on non-processed meats. If a substantial portion of " red meats "

were processed, then they would carry an association that might not

belong to unprocessed red meats.

 

Although the study did not consider cooking temperature or cooking

method, they rightly noted that cooking meats at high temperatures

causes the formation of heterocyclic amines, and charcoal broiling and

grilling cause the formation of polycyclic aromatic hydrocarbons, both

of which have been found to be carcinogenic in animals. This study

cited three other studies in its discussion finding that meats -

including red meats -- that were not grilled or barbecued were not

associated with pancreatic cancer, while barbecued and grilled red

meats were. It is relieving to see researchers paying attention to

these nuances and noting that the associations they have found between

meats and cancer may well be due to carcinogens formed as a result of

cooking and preserving methods rather than the meats themselves. I

would add an additional consideration: prolonged cooking of meats, or

the cooking of meats at high temperatures may not only cause the

formation of injurious compounds, bu t also the destruction of

protective compounds.

 

The study found no statistically significant correlations for poultry,

fish, dairy products, or eggs. Fat, saturated fat, and cholesterol

were not associated with pancreatic cancer. Thankfully, the authors

emphasized these negative findings by noting them in the abstract.

 

Nothlings, et al., " Meat and Fat Intake as Risk Factors for Pancreatic

Cancer: The Multiethnic Cohort Study, " Journal of the National Cancer

Institute, Vol. 97 No. 19 (2005) 1458-1465.

 

Alzheimer's Study Takes Cheap Shot At Cholesterol

 

An October study published in the Archives of Neurology found strong

correlations between body mass index and cholesterol levels, and the

risk of Alzheimer's Disease (AD).

 

Although the unique angle of the study was that it measured body mass

index, the authors considered their measurements of cholesterol levels

to be important enough to note in the abstract and discussion. The

authors noted that high cholesterol levels (above 251 mg/dL) about

doubled the risk of AD, and that in combination with high blood

pressure and obesity, they yielded a six-fold increase in the risk of AD.

 

This study constitutes nothing more than a cheap shot against

cholesterol. If the authors considered their findings on cholesterol

to be worth devoting half of the results section of the abstract to,

then it deserved a more careful analysis. Although total cholesterol

levels above 251 mg/dL showed an association with Alzheimer's, no

attempt was made to analyze a dose-response effect between cholesterol

levels and AD (that is, to demonstrate a consistent effect of higher

cholesterol levels yielding higher rates of Alzheimer's). More

importantly, no adjustment of cholesterol levels for the

apolipoprotein E epsilon-4 allele (ApoE4) or familial

hypercholesterolemia was made.

 

In multivariate analysis (that is, adjusting for possible confounding

factors), a body mass index (BMI) over 30 carried a 144% increased

risk of AD when adjusted for age, sex, education, and follow-up time;

a BMI over 30 carried a 110% increased risk of AD when adjusted for

blood pressure, cholesterol level, and smoking; leastly, BMI carried a

substantially reduced 88% association with AD when adjusted for ApoE4,

diabetes, and myocardial infarction. Fat intake and physical activity

had no effect on the association between AD and BMI.

 

These figures indicate that the third model, which adjusted for ApoE4,

diabetes, and myocardial infarction, contained the most important

confounding factors, accounting for about half of the association with

obesity.

 

Despite these strong effects of confounding factors on the association

with obesity, the authors made no such attempt to adjust the

association with total cholesterol levels with multivariate analysis.

 

As explained in Part One and Part Two of Cholesterol-And-Health.com's

Alzheimer's articles, a contributing factor to Alzheimer's disease may

be an inability to incorporate cholesterol into cell membranes,

possibly due to an inefficiency of the ApoE4 allele. ApoE4 is

associated with high cholesterol levels, possibly for this very

reason: free cholesterol levels outside of cells are higher when the

cholesterol is not properly incorporated into cells and into their

membranes. Some researchers have hypothesized that the ApoE4 allele is

only harmful when combined with a high-carbohydrate diet,

demonstrating that the ApoE4 allele has largely been weeded out in

ancestries with a long history of agriculture.

 

High cholesterol levels would also include those with familial

hypercholesterolemia, which is a genetic disease that involves the

lack of a properly functioning LDL receptor, not a disorder merely of

high cholesterol levels. Associations with cholesterol levels that do

not account for this confounding factor have no validity.

 

The authors of any study asserting an association between cholesterol

and Alzheimer's disease owes their readers a thoughtful analysis of

that association's significance, including adjustment for familial

hypercholesterolemia, the ApoE4 allele, carbohydrate intake, and other

factors hypothesized to interact with the ApoE4 allele.

 

Kivipelto, et al., " Obesity and vascular risk factors at midlife and

the risk of dementia and Alzheimer disease, " Archives of Neurology 62

(2005) 1556-1560.

 

High-Cholesterol, High-Fat Diet Reduces Inflammation in Baboons -- But

You Wouldn't Know It From the Abstract

 

An October 2005 study published in the American Journal of Clinical

Nutrition measured the effect of a high-cholesterol, high-fat (HCHF)

diet on inflammation in baboons. The authors noted the inconsistency

with which cholesterol levels are associated with heart disease in

humans, and sought to examine biomolecular changes in the baboons that

are difficult to study in humans. They used both blood samples and

biopsies of the femoral artery.

 

The study was not controlled with an untreated sample of baboons.

Instead, one group of baboons was fed a " normal " diet for a period of

time, and then switched to an HCHF diet. Blood samples were taken at

baseline (the beginning of the study period), three weeks, and 7

weeks, while biopsies were taken from the femoral artery at baseline

and seven weeks, and put through different tests.

 

Although the authors didn't mention it in the abstract, discussion, or

anywhere in the text of the study, the high-cholesterol, high-fat diet

contained only 0.9% of calories as omega-3 fatty acids -- which are

precursors to anti-inflammatory hormones -- compared to 3.4% on the

" normal " diet -- nearly 4 times lower a concentration. Similarly, the

HCHF diet contained 3.3 times fewer omega-6 fatty acids, which are

precursors to both inflammatory and anti-inflammatory hormones. This

was shown in Table 1, but the significance of the failure to control

for these inflammatory and anti-inflammatory factors in the diet in a

study measuring inflammation was not discussed anywhere in the text.

 

The authors referred to the changes seen on the HCHF diet as " arterial

endothelial dysfunction. " Endothelial cells line the insides of

arteries and are the first line of defense against injurious factors

within the arteries. In the abstract, they wrote: " We found that the

HCHF diet induced a high inflammatory status, as indicated by

increased concentrations of interleukin 6 (IL-6), tumor necrosis

factor alpha (TNF-alpha), and monocyte chemoatractant protein-1 (MCP-1). "

 

What the authors failed to note in the abstract was that this effect

was temporary, and the long-term effect of the high-fat,

high-cholesterol diet was to dramatically decrease these markers of

inflammation. While TNF-alpha doubled at three weeks, its

concentration was eight times lower at seven weeks compared to

baseline. Likewise, while IL-6 concentration was nearly tripled at

three weeks, it was seven times lower at seven weeks than at baseline.

MCP-1 remained 60% higher than baseline at seven weeks, and another

chemical called interleukin-8 was also much higher at seven weeks than

baseline, while oxidized cholesterol levels began declining between

three and seven weeks, antioxidant status remained unchanged on the

diet, and other endothelial dysfunction markers remained unchanged.

 

The high-fat, high-cholesterol diet -- which contained 6 times the fat

and 318 times the cholesterol concentrations of the " normal " diet --

produced a doubling of total cholesterol levels, while the ratio of

total to HDL cholesterol remained unchanged. No changes were found in

C-reactive protein, which is regarded as a better predictor of heart

disease than cholesterol levels. No atherosclerotic lesions were

observed, nor were any fatty streaks observed in the arteries.

 

In the cultured cells -- one set taken at baseline and the other taken

at seven weeks of a high-fat, high-cholesterol diet -- output of MCP-1

and several insoluble inflammatory markers were about doubled on the

HFHC diet, while no difference was found in the concentrations of

cytokines or other soluble inflammatory markers.

 

It is difficult to know exactly what this change would mean within the

living baboon, since cells normally interact with many other factors

secreted by neighboring cells to which they are attached, or from

far-away cells in the body. When the researchers tested how these

cells reacted to stimulators of inflammation, what they found was a

dramatic decrease in inflammatory response.

 

The authors hypothesized that feeding an HCHF diet might precondition

cells to react with a greater degree of inflammation to substances

that normally induce an inflammatory response. Yet they found the

opposite. Cells from HCHF-fed baboons had an 8-fold lower VCAM-1 (a

chemical that helps inflammatory cells infiltrate blood vessels) and

6-fold lower mE-selectin response to TNF-alpha, and a 9-fold decreased

VCAM-1 response to lipopolysacccharide. The HCHF diet completely

abolished the inflammatory response to both native and oxidized LDL.

 

The authors concluded: " In summary, our results clearly show that the

HCHF diet resulted not only in increased serum cholesterol

concentrations but also in inflammation and endothelial dysfunction. "

Yet, as one can see, the results are anything but clear. The authors

disingenuously cited increases in inflammatory markers in the abstract

when those same inflammatory markers were actually dramatically

decreased by the high-fat, high-cholesterol diet when it was allowed

to be fed over seven weeks. And even seven weeks is a very short time.

Were the few other inflammatory markers that didn't eventually fall

after seven weeks there to stay, or was it only a matter of a few more

weeks before they also would adjust?

 

What does this study tell us about how high-cholesterol, high-fat

diets affect humans? Nothing. The baboons themselves are not only not

human, but, as noted in the text, " the baboons selected for this study

were previously classified as high responders because their serum

LDL-cholesterol concentrations increased after an HCHF diet. " In other

words, the baboons were selectively bred to respond to such a diet

with increased cholesterol concentration, when we know that

high-cholesterol feeding usually causes very little change in

cholesterol levels in most humans.

 

The study also failed to control for inflammatory and

anti-inflammatory hormone precursors in the diet. Since the HCHF diet

had several times less omega-3 fatty acids, precursors to

anti-inflammatory hormones, there is no way to know what portion of

the results were due to a reduced quantity of this factor.

Nevertheless, most measures of inflammation were either unchanged or

even declined, some dramatically.

 

The ultimate lesson of this study is how an abstract can be written to

tell a very different story from the one told in the tables of data

within the text. Just below to read the abstract for

yourself -- and then consider whether you would have the same

impression of the study if the abstract had been the only piece of

information about it that you had read.

 

Shi, et al, " Arterial endothelial dysfunction in baboons fed a

high-cholesterol, high-fat diet, " American Journal of Clinical

Nutrition, 82 (2005) 751-759.

 

Inhibition of Inflammatory Factor Decreases Atherosclerosis in Mice

With No Effect on Cholesterol or Triglycerides

 

Yet another piece of research verifies that atheroclerosis can be

inhibited by substances that have no effect on cholesterol levels,

strongly suggesting that the beneficial effects seen in the use of

cholesterol-lowering statin drugs are coincidental to, and not because

of, their ability to lower cholesterol.

 

Cholesterol-And-Health.com's review of Dr. Duane Graveline's Statin

Drugs Side Effects and the Misguided War on Cholesterol discusses how

statins inhibit nuclear factor-kappa B, a molecule that activates the

inflammatory process. Statins also inhibit cholesterol synthesis

independently.

 

A new study published in the Journal of Physiology and Pharmacology

shows that pyrrolidine dithiocarbamate (PDTC), a substance that, like

statins, inhibits nuclear factor-kappa B, but, unlike statins, does

not lower cholesterol levels, substantially reduces atherosclerosis in

mice. Like statins, PDTC also has antioxidant effects. The experiment

used mice that were genetically altered to not produce apolipoprotein

E (apoE) or the LDL receptor (LDLR). These mice develop severe

dyslipidemia and athersclerosis.

 

Blood vessels of mice treated with PDTC had only 62% as much area

affected by atherosclerosis when examined longitudinally and only 51%

as much area affected by atherosclerosis when measured

cross-sectionally. PDTC treatment had no effect on cholesterol or

triglyceride levels.

 

This study is a small part of a large body of evidence that refutes

the lipid hypothesis - that is, the idea that the formation of

atherosclerotic lesions and the development of arterial obstruction

and heart disease is a function of the concentration of cholesterol in

the blood. In Issue #005 of this newsletter, I reported on a

meta-analysis that found omega-3 fatty acids to be much more effective

than statins at reducing deaths from heart disease, despite having no

effect on cholesterol levels. Conversely, fibrates, niacin and " diet "

were effective at reducing cholesterol levels but had no effect on

deaths from heart disease, and fibrates increased death from cancer.

The same issue reported on a study finding that resveratrol, a

component of red wine, reduced atheroclerotic obstruction of rabbit

arteries without affecting cholesterol levels. In The Cholesterol

Myths Dr. Uffe Ravnskov pointed out that pathogenic changes in

arterial plaques tend to occur after middle-age peaks in blood

cholesterol levels have begun to decline, and Issue #006 of this

newsletter reported on a new hypothesis put forth by Wilbert Gamble,

in which he pointed out that arteries of mammary glands, which have

very high concentrations of cholesterol and triglycerides, do not

develop atherosclerosis. Anthony Colpo and Shane Ellison have made

numerous other very important points refuting the lipid hypothesis.

 

Hopefully this study and others like it, illuminating mechanisms by

which the non-cholesterol-lowering effects of statin drugs can reduce

heart disease risk, will eventually help waken us from the lull of

junk science that has brought us the theoretical wasteland that is the

lipid hypothesis.

 

Jawien, et al., " Inhibition of nuclear factor-kappa B attenuates

atherolsclerosis in apoE/LDLR - double knockout mice, " Journal of

Physiology and Pharmacology, Vol. 56 No. 3 (2005) 483-489.

 

Fructose Enhances Atherosclerosis in Cholesterol-Fed Rabbits

 

One hypothesis of heart disease considers oxidized lipoproteins,

rather than lipoprotein levels themselves, to contribute to heart

disease, or to be markers of other oxidative stress that cause heart

disease. One process that increases oxidative stress is called

glycation. Glycation is the addition of a sugar molecule to another

molecule. Glycation is an important process when performed by enzymes

in the body, but it is believed that random glycation occurs when

blood sugar levels are elevated, which causes random damage that can

increase other measures of oxidative stress such as lipid peroxides.

 

Since fructose is believed to be better at enhancing glycation than

glucose, a recent study published by Tokita and coworkers in the

Journal of Atherosclerosis and Thrombosis tested the effect of

fructose feeding on the development of atherosclerosis in

cholesterol-fed rabbits.

 

Unsurprisingly, the rabbit -- an herbivorous animal adapted to its

herbivorous diet -- develops atherosclerotic lesions and heart disease

when fed a 1% cholesterol diet, which is 34 times the concentration of

cholesterol that would be consumed on a diet of pure egg yolk. Yet

even in the rabbit, and even at this concentration of cholesterol that

is impossible even for an animal on an exclusively carnivorous diet to

encounter in nature, it appears that it is not the cholesterol that is

harmful, but the increased concentration of lipoproteins that results

in the blood, which carry cholesterol on the inside, but contain

phospholipids on the outside of their membranes that are vulnerable to

oxidative stress. Thus, as reported in an earlier issue, when rabbits

are fed the antioxidant resveratrol, the atherogenic effect of a 1%

cholesterol diet is radically diminished.

 

The Tokita study divided male Japnese white rabbits into three groups.

Each group was fed for eight weeks on their respective diets: one, a

control diet; another, the control diet supplemented with 1%

cholesterol; the third, the control diet supplemented with 1%

cholesterol and 10% fructose water in place of plain water.

 

The degree of atherosclerotic lesions formed on the

cholesterol-fructose diet was 58% greater than that observed in

rabbits fed cholesterol alone. Supporting the hypothesis that

oxidative stress is involved in the effect of cholesterol-feeding, the

cholesterol-fed rabbits had a significant increase in peroxide levels

in the aorta compared to the controls. Yet the

cholesterol-fructose-fed rabbits had a substantially greater level of

peroxides in the aorta than did the cholesterol-fed rabbits.

 

Unfortunately, this study didn't control for caloric intake. Adding

fructose to the water rather than replacing another source of energy

with fructose must have increased the caloric intake of the rabbits.

The value of the study would have been enhanced if a fourth group of

rabbits fed a 10% glucose water was also included.

 

What does this mean for humans? The connection between oxidized

lipoproteins in humans isn't very clear. In Issue #003 of this

newsletter, I reported on a study that found a strong correlation

between oxidized lipoproteins and coronary obstruction in persons

under the age of 60, but in persons over the age of 60 there was no

relationship. Since most heart disease occurs in older rather than

younger individuals, the significance of the finding is questionable.

On the other hand, since insulin resistance and diabetes are occurring

more often, and since some young people do experience heart disease,

possibly related to this trend, and since fructose is also capable of

inducing insulin resistance, this may shed light on how our increasing

consumption of fructose is involved in our increasing decline in many

parameters of health.

 

Fructose is the predominant sugar in fruit and fruit juice,

constitutes half of table sugar, and constitutes an equal or much

higher proportion of the sugars in high fructose corn syrup, found in

many sodas and packaged foods. Consumption of fructose has

dramatically increased over the past several decades. Unlike

cholesterol, which is consumed in milligram per serving quantities,

tends of grams of fructose can be found in single servings of foods

and drinks. It is impossible to consume a 1% cholesterol diet outside

of a laboratory setting, while it is very possible to eat a 10%

fructose diet, or a diet much higher than that in fructose, when

selecting from the typical menu of modern foods and junk foods. If the

increase in oxidative stress found in this study is due to the

fructose, it represents one more reason to moderate our fructose

consumption and avoid excessive consumption of sugar and other junk foods.

 

Tokita et al., " Fructose ingestion enhances atherosclerosis and

deposition of advanced glycated end-products in cholesterol-fed

rabbits, " Journal of Atherosclerosis and Thrombosis, Vol. 12, No. 5

(2005) 260-267.

 

How t