Friday, June 17, 2011

Mice and breast cancer

Laura and Elizabeth (thanks for full text) both forwarded links to this study.

Low carb diets, in the correct mouse model, delay onset and slow progression of breast cancer.

It says the sorts of things you would expect it to say and, if you really feel this highly artificial mouse model has relevance to the sorts of breast cancer humans might develop, it certainly suggests that a low carbohydrate diet might have some benefits. But the paper itself is awful.

Two giggles did come out of it. First was the use of high percentage of casein as the sole protein source. Now somewhere, sometime, there was a vegan nut who screamed that casein was carcinogenic. China? China Syndrome? China Study? T. Colon Campbell???? Shrug. These has-been vegans get everywhere.

The second is the extreme fat phobia of the authors. I know these people have to make a crust and funding is not what it used to be and fat bashing is always helpful but there eventually comes a point when people really believe that fat causes cancer. Even highly saturated milk fat.

You can just imagine that cows evolved casein and palmitic acid to kill their calves. Or humans are not mammals in the same way as cows are, human breasts having evolved to sell newspapers rather than to feed offspring. Human babies should be fed sucrose water with a little soya bean oil added of course. It's a strange world.

I have reached the point where I no longer give any credence to high fat diet studies where 30% of the calories are coming from sucrose or the pellets are stained red to signify Crisco. Not so the current authors.

Ultimately, while sucrose and trans fats are excellent substances to study when looking at the effects of pushing the profitability of the food industry to its absolute maximum, they have nothing to do with a high fat diet based on Food.

Reading through the full text there are so many failures of perception and basic biochemistry that it might be worth a post in the end, but here's a typical blooper. Not only do we have Gourmand rats, we also have mice who need false teeth!

First we have to have another black box warning

Untested ad hoc hypotheses can make you look very stupid.


Here we go:

"Although mice on 8% CHO diet had slower growing tumors, they lost weight, weighing, on average, 20% less than mice on 5058 diet (Fig. 1D). This was consistent with the mice eating less than the 5058 group (data not shown), likely because the 8% CHO pellets were significantly harder to chew."

Executive summary: We're idiots.

Extended translation:

Diet 5058 is standard breeding colony crap-in-a-bag with 55% of calories from starch. It appears to be mildly obesogenic compared to 8% of calories from starch... That MUST be because the lower carb diet is too hard to chew. We couldn't be arsed to have a control group offered a harder diet with 55% carbs because we're idiots, as are our scrutineers.

GCBC anyone?

Oh, and another giggle: 5058 is described, COMPLETELY incorrectly, as a "Western Diet". It's a starch based, sucrose free, 20% fat, mostly PUFA, trans free diet, remarkably similar to Barnard's idiotic vegan diet for the progression of diabetes in humans. It's standard mouse chow.

Where do funding bodies find these people to throw their money at?

Eeeh, yer has ter larph.



Anonymous said...

As a norwegian I somtimes find your ironic take om crappy research hard to follow. Not so today. :)

Thanks for the giggle.

Peter said...

it's frustrating that we have to rely on this level of work to get even pointers about what may help on a practical basis... Ah well.


blogblog said...
This comment has been removed by the author.
blogblog said...

In cancer patients the body frequently tries to cure itself via cachexia. There is no diet more ketogenic than starvation.

The response of the treating oncologist when faced with cachexia is to immediately recommend a mushy high carbohydrate diet to "keep up your strength."

Result: A slow and painful death.

Unknown said...

i agree that labdiet 5058 is a crappy control, but do you like the use of say for example Research Diets D12492 vs D12540B? They also make up the caloric increase of fat (lard) with maltodextrin and sucrose

Unknown said...

... and corn starch. Do you think thats an appropriate mix?

Peter said...

Hi Dave, I have absolutely no problem with mouse chow as the control. I particularly like the fact it is more fattening than the 8% carb diet. My objection is that it is described as a Western Diet. This is simply incorrect and should have been shown no mercy by the scrutineers, who were clearly asleep. The 8% carb diet would have utterly knocked spots off of a true Western diet in terms of obesity control, as it would in humans... Even 5058 can do this when matched against D12492.


Peter said...

Jeff, ridiculous is an excellent description. That good old TD.88137, 30% of calories from sucrose. That's a lot of fat to put into an hepatocyte... Makes Homer Simpson look like a health freak (I exaggerate, a little).


John said...

What was the macro breakdown for the low carb group?

The authors remind me of coworkers and casual friends who somehow think, because I don't eat grains, I eat "all protein," as if high fat is unfathomable...If I had a nickel for every time I was offered a "high protein snack."

Peter said...

Table is here. I'd not eat any of the diets personally...


Stan Bleszynski said...

I am not sure if that vegan Cornell professor who screamed that casein is carcinogenic, was really a nut. He wrote some bestselling books, got paid, has had a lifelong secure employment, probably never had to work for a living like us, and is highly regarded in the circles and groups of the same kind, unlike us. 8-:)

pyker said...

Ah, fantastic. I'm definitely going on the "hard to chew" diet. First I'll whip up a batch of kelp jerky...

karl said...

I wonder if some of the basic research in other fields has been effected by the standard diet they give these mice (herbivores) - Seems a diet more tied to what they evolved to eat might be more appropriate -(a paleo-mouse diet?)

A read of the VARIOUS Standard diets at led me to doubt that different studies are actually comparing apples to apples - diet details could skew results of many studies out side of lipids - prevent reproduction of results etc..

blogblog said...

you would be hard pressed to find a worse mammalian model for human diseases than rats and mice. The only reason rats and mice are used is because they are small, short-lived and people generally tolerate them being experimented upon.

Rats and mice are primarily seed eaters. Their natural diet is normally extremely low in fat (<5%) and very high in fibre.

Rats and mice are also extreme athletes. They can effortlessly run the rodent equivalent of a world record marathon every day. Unlike humans they need a 1-2 hours a day of vigorous amount of exercise just to stay healthy.

Stipetic said...

Also, until mice breasts evolve to sell newspapers, we shouldn't seriously consider them as apropos models.

Stephan Guyenet said...

Hi Peter,

If we're discussing glucoagon, why not discuss amylin, which is a more potent satiety peptide (that has actually been shown to cause fat loss and is currently in clinical weight loss trials) released by the pancreas in response to carbohydrate ingestion? I think you forgot to mention that one.

The larger point is that insulin cannot be viewed in isolation. You can't say "carbs increase insulin, which decreases lipolysis, therefore fat accumulates". The body mounts a complex response to any meal that includes dozens of peptide and nerve signals, one of which is insulin. These are designed to steward the constructive use of nutrients. But it makes the system very complex, and one cannot extrapolate anything from observing insulin's anti-lipolytic action in isolation.

The only way to determine if carbs cause fat gain is by testing it directly. It has been tested, many times, in lean, overweight and obese people, and they don't. If carb-induced insulin suppressed lipolysis and that were relevant for fat balance, you would see that in metabolic ward studies where volunteers eat isocaloric diets of different macronutrient composition. Fat would be trapped in fat cells in the carb group, their energy expenditure would decrease and they would accumulate fat. However, all available evidence (and there is a lot of it) suggests that isocaloric diets have the same effect on body fatness, regardless of carb:fat ratio, regardless of whether they are designed for weight loss or weight maintenance, and regardless of whether they are conducted in lean, overweight or obese people.

I have seen no evidence whatsoever to support the claim that insulin is a physiologically relevant regulator of fat storage under normal conditions. That hypothesis has been around for a long time, but obesity and metabolism researchers considered it rejected by the mid 1980s if not earlier:

I also have seen no evidence to support your statement that "Once you are insulin resistant carbohydrates become spontaneously fattening".

Low-carb diets are not the miracle cure for obesity that you might expect them to be if carb-insulin were a dominant factor. If you look at long-term outcomes, they cause on average 2-6 kg of weight loss at one year, similar to higher-carb diets such as the Mediterranean diet:

LC is no cure for the obesity epidemic. Some individuals have very good success with LC, while others gain weight, but there is no indication that has anything to do with insulin.

I know it is convenient to reject the evidence from genetics offhand, but researchers find this evidence to be informative, particularly because it is consistent with the rest of the literature on how body fatness is regulated.

"Discarding insulin as a factor in obesity because there are circumstances in which starch does nor invariably elevate insulin is a serious case of throwing the baby out with the bath water." There was never any baby here Peter. The carb-insulin-obesity hypothesis never had any serious traction except in the lay press. If it were that simple, obesity would be a thing of the past and we'd be singing kumbaya by now.

Asim said...

Hey all,

First time poster, recent reader.. Love your blog Peter, sometimes it's over my head, by your a sane voice among few, like Dr. Ayers, in this world of blogging about nutrion. Just wanted to make a comment on the below:

"In cancer patients the body frequently tries to cure itself via cachexia. There is no diet more ketogenic than starvation."

I just wanted to pointed out that cachexia is not starvation, and feeding calories to a person in advanced stages of cancer, when cachexia sets in, does little to the victim in maintaing weight. Cachexia is not a method in which the body tries to cure itself, it's a negative result of the body trying to fight the cancer, without not actually knowing what it's fighting.

Cachexia is a state of chronic inflammation at it's worst, i.e. the body is wasting away as a result of the damage being inflicted by the response of the immune system'.

This does not mean a 'ketogenic' diet wouldn't work, but the mechanism is clearly related to it's reduction in inflammation in general. Any measures, in theory, to reduce inflammation would impact positively for the victim the cancer progression, but halting or slowing progression is not to be confused with cure.

For a cure to happen from the body itself, the cancer has to be recognized by the immune system and targetted by it.

Peter said...

Hi Asim,

Point taken