December 10, 2006
High Protein Diet Ups Cancer Risk?

Should we eat less protein to increase our odds of living long lives?

Dec. 7, 2006 -- A great deal of research connects nutrition with cancer risk. Overweight people are at higher risk of developing post-menopausal breast cancer, endometrial cancer, colon cancer, kidney cancer and a certain type of esophageal cancer. Now preliminary findings from researchers at Washington University School of Medicine in St. Louis suggest that eating less protein may help protect against certain cancers that are not directly associated with obesity.

The research, published in the December issue of the American Journal of Clinical Nutrition, shows that lean people on a long-term, low-protein, low-calorie diet or participating in regular endurance exercise training have lower levels of plasma growth factors and certain hormones linked to cancer risk.

These researchers think that just because people on low calorie low protein diets have lower blood levels of insulin-like growth factor 1 (IGF-1) that this is proof that protein is a culprit for raising IGF-1.

"However, people on a low-protein, low-calorie diet had considerably lower levels of a particular plasma growth factor called IGF-1 than equally lean endurance runners," says the study's first author Luigi Fontana, M.D., Ph.D., assistant professor of medicine at Washington University and an investigator at the Istituto Superiore di Sanità in Rome, Italy. "That suggests to us that a diet lower in protein may have a greater protective effect against cancer than endurance exercise, independently of body fat mass."

But has Fontana looked hard at the body of research on calorie restriction? Are distance runners really a good gold standard to compare to? We already know that calorie restriction will boost longevity of lab mice as compared to mice who eat more and get more exercise.

Note that Fontana's group that had the lowest IGF-1 levels also ate a raw food vegetarian diet. Okay, that's lower in glycemic index, plus comes with lots of beneficial compounds in fruits and vegetables. Seems to be he changed too many variables at once between groups.

The study involved three groups of people. The first ate a low-protein, low-calorie, raw food vegetarian diet and was made up of 21 lean men and women. Another group consisted of 21 lean subjects who did regular endurance running, averaging about 48 miles per week. The runners ate a standard Western diet, consuming more calories and protein than group one. The third group included 21 sedentary people who also consumed a standard Western diet, higher in sugars, processed refined grains and animal products. The subjects were matched for age, sex and other demographic factors, and no one smoked or had diabetes, cardiovascular disease, cancer, lung disease or other chronic illness.

Protein intake was, not surprisingly, lowest in the low-protein group. They averaged a daily intake of 0.73 grams of protein per kilogram of body weight. Endurance runners ate 1.6 grams and sedentary people on the Western diet, 1.23 grams. The recommended daily allowance for protein intake is 0.8 grams. That's about three ounces of protein per day for a 220-pound man.

"It's interesting to us that both the runners and especially the sedentary people consumed about 50 percent more protein than recommended," says Fontana. "We know that if we consume 50 percent more calories than recommended, we will become obese. But there is not a lot of research on whether chronic over-consumption of protein also has harmful effects."

The conclusions these researchers draw about about protein and cancer risk are based on a known association between insulin-like growth factor 1 (IGF-1) blood plasma levels and risk of breast cancer, prostate cancer, and colon cancer.

Fontana and colleagues found significantly lower blood levels of plasma insulin-like growth factor 1 (IGF-1) in the low-protein diet group than in either the equally lean runners or the sedentary people eating a standard Western diet. Past research has linked pre-menopausal breast cancer, prostate cancer and certain types of colon cancer to high levels of IGF-1, a powerful growth factor that promotes cell proliferation. Data from animal studies also suggest that lower IGF-1 levels are associated with maximal lifespan.

What I'd like to see: A larger assortment of diets compared for effects upon blood IGF-1 levels. I do not believe they've proven their main claim against protein:

"Our findings show that in normal weight people IGF-1 levels are related to protein intake, independent of body weight and fat mass," Fontana says. "I believe our findings suggest that protein intake may be very important in regulating cancer risk."

See below for why I doubt the strength of their claim. They might be right. I'd really like to know whether they are right. But the study strikes me as having a major shortcoming.

He calls the study a hypothesis-generating paper that suggests connections between dietary protein and epidemiological studies that show associations between IGF-1 levels and the risk of cancer. But he says more research is needed to clarify what that connection is.

I see a big obvious shortcoming of this study in this paragraph. Do you see it too?

The researchers also found that the group of endurance runners in the study consumed the highest number of calories, averaging more than 2,600 per day. Those on a standard Western diet consumed just over 2,300 calories daily, while those in the low-calorie, low-protein group ate just under 2,000 calories a day. Members of the latter group also tended to weigh less than sedentary people but slightly more than the endurance runners. The average body mass index (BMI) in the low-protein, low-calorie group was 21.3. BMI averaged 21.1 among the runners and 26.5 among those who were sedentary. BMI is a measurement of weight divided by height squared. People with a BMI greater than 25 are considered overweight.

Problem: The people on the standard diet ate more calories than those on the low protein diet. So how much of the lower blood IGF-1 is due to lower calories rather than lower protein? We already know that calorie restriction causes all sorts of blood markers to shift in directions favorable to good health. Cholesterol and triglycerides go down. Markers for insulin sensitivity improve. So I would expect better IGF-1 just from the lower calorie intake.

Note that their lower calorie lower protein study participants had mower body mass indexes. Was the lower IGF-1 just due to that? Have any studies been done comparing IGF-1 levels as a function of BMI?

Are any readers aware of studies of people on the high protein Atkins diet that looked at blood IGF-1 levels?

What is needed: Comparison of IGF-1 levels of people on different ratios of fats, carbohydrates, and protein on a normal calorie diet. Then, repeat the same experiment on people who are on calorie restriction diets. I certainly expect the people on lower calorie diets to have lower IGF-1. But will there be differences in IGF-1 based on the relative contributions of protein, fat, and carbos as calorie sources?

While I'm asking for experiments: I'd like to see comparisons of IGF-1 for diets where the carbos come from different sources such as low glycemic index versus high glycemic index foods and high fructose versus high glucose foods.

I'd also like to see the effect of BMI on IGF-1. Will high BMI people have high IGF-1 even if, say, they go on a low protein diet. I'd expect they would. Can any sort of diet that does not bring off weight lower IGF-1? Does high dose resveratrol lower IGF-1?

Can anyone point out studies in the research literature that control for factors that Fontana's team apparently didn't separately control for?

Update: The abstract provides more details:

Plasma concentrations of insulin, free sex hormones, leptin, and C-reactive protein were lower and sex hormone–binding globulin was higher in the low-protein, low-calorie diet and runner groups than in the sedentary Western diet group (all P < 0.05). Plasma insulin-like growth factor I (IGF-I) and the concentration ratio of IGF-I to IGF binding protein 3 were lower in the low-protein, low-calorie diet group (139 ± 37 ng/mL and 0.033 ± 0.01, respectively) than in the runner (177 ± 37 ng/mL and 0.044 ± 0.01, respectively) and sedentary Western (201 ± 42 ng/mL and 0.046 ± 0.01, respectively) diet groups (P < 0.005).

But, again, how much of the result was due to A) lower protein, B) lower calories, or C) lower BMI? The latter two will both lower IGF-1 and markers for inflammation such as C-reactive protein. See the comments for pointers to other research that suggests, yes, protein restriction can lower unfavorable indicators in blood such as reactive oxygen species (ROS). So maybe a lower protein diet will help.

Share |      Randall Parker, 2006 December 10 09:12 AM  Aging Diet Cancer Studies

crush41 said at December 10, 2006 12:42 PM:

Seven miles a day and only 2,600 calories? Lean indeed!

Randall Parker said at December 10, 2006 1:25 PM:


Yes, CR lowers IGF-1. The Wikipedia entry mentions that BMI influences it without spelling out that high BMI probably raises it.

But did I miss anything else at those links about other dietary factors that change IGF-1?

Your second link mentions that cutting calories while maintaining protein reduces loss of muscle mass. That might increase longevity by increasing heart muscle strength. But does it also increase IGF-1?

Your second link also includes an excerpted paragraph that does suggest that protein restriction might bring benefits: (my bold emphases added)

Many previous investigations have consistently reported that caloric restriction (40%), which increases maximum longevity, decreases mitochondrial reactive species (ROS) generation and oxidative damage to mitochondrial DNA (mtDNA) in laboratory rodents. These decreases take place in rat liver after only seven weeks of caloric restriction. Moreover, it has been found that seven weeks of 40% protein restriction, independently of caloric restriction, also decrease these two parameters, whereas they are not changed after seven weeks of 40% lipid restriction. This is interesting since it is known that protein restriction can extend longevity in rodents, whereas lipid restriction does not have such effect. However, before concluding that the ameliorating effects of caloric restriction on mitochondrial oxidative stress are due to restriction in protein intake, studies on the third energetic component of the diet, carbohydrates, are needed.

So maybe protein restriction provides some benefits. But I'd really like to see a controlled study that did protein restriction without calorie restriction and also had a control group and two other groups that did lipid and carbo restriction. If the study measured IGF-1 and ROS levels in the blood we'd learn a great deal.

boxfullofspam said at December 11, 2006 12:24 AM:

If you restrict protein while maintaining constant calories, what do you replace it with?
Either fat or carbohydrates. Since carbohyrates stimulate IGF-1 production (through insulin release) to a greater degree than protein, you'll actually get a net increase in IGF-1 levels unless you employ a high-fat diet. I suspect that is not what the researchers had in mind.
This sort of thing has been extensively discussed in bodybuilding circles, in the context of manipulating diet to INCREASE IGF-1 levels, and they long since came to the conclusion that you need plenty of carbs to maintain high IGF-1 levels, and consequently you can't "gain mass" on a (very)high-protein low-carb diet. Once protein gets over 30% of intake it isn't worth ading more, because it comes at the expense of insulin-producing carbs.
In other words an Atkins-style diet lowers IGF-1 levels because it replaces carbohydrates with protein and fat, and bodybuilders will avoid low-carb diets during the bulking up phase for this reason.

The implication from this line of research is that the lowest possible IGF-1 levels for a given calorie level will be obtained on a 90% fat diet, but good luck trying to fund that research. Even on rats.

Buzz Lightbeer said at December 13, 2006 9:19 AM:

A hypothesis I'd like to see tested is that people who eat more protein have a higher cancer risk partly because of the 'crap' in conventionally (e.g. feedlot) raised meat, incl. the lopsided Omega-6 vs. Omega-3 FA ratio, residual growth-promoting hormones, etc.

Nick said at December 8, 2010 7:00 AM:

This is really good info. Very interesting. I also found which has some great stuff on diets and weight loss.

Nick x

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©