There have been several huge and expensive studies published lately which showed that plausible interventions (eg: vitamins, lowfat diet) failed to change cancer incidence in humans. One explanation is that these things do not work. Another is that people are hard to study because they tend to do all sorts of things and you cant make them stop. For example, people in the control group in the lowfat study might have started eating lowfat diets by themselves. People who were supposed to be on the lowfat diet might have eaten a bunch of fat. People might not have eaten their vitamins, or eaten them when they were not supposed to, or had their food supplemented with vitamins by their governments when they were not watching.

I was thinking that one could study cancer prevention in prisoners. But a lot of people do not stay in prison for long, and there are ethical concerns with studying prisoners. Not good.

But what about pets? They lead supervised and controlled lives, except for my neighbor's dog. Cancer is a leading cause of death for pet dogs and cats. Animal lifespans are shorter than people and so cancer events will be more concentrated in a given time period. Animal diets are more straightforward than those of people. Their owners are often more motivated to care for the pets than they are for themselves. Plus there are a lot of pets, thus distributing the costs of housing and caring for them and also introducing random elements of environment similar to those that humans encounter.

I propose that pet animals be used for large cancer prevention studies. Initial studies could focus on breeds with high cancer rates - such as goldens or boxers. These breeds have enthusiasts dedicated to them who would likely be glad to help with the study. Studies could be lowfat vs highfat diets, comparisons of vitamin mixes, and so on. Eventually, cancer prevention might be a side project regularly involving breed enthusiasts such that when one study meets accrual goals the next will open.
-- bungston, Feb 13 2009

Vitamins did not prevent cancer http://www.cnn.com/...ltivitamins.cancer/
[bungston, Feb 13 2009]

Lowfat diet did not prevent cancer http://seniorjourna...atDietWithFruit.htm
[bungston, Feb 13 2009]

This is a good idea apart from its major flaws.

First, good human trials take into account human fallibility.

Second, when it comes to dietary interventions, animals are not a great model for humans. They're not bad, but they are not as useful as when testing (say) a drug. Things like vitamin requirements (and responses) differ widely.

Third, why swap unreliable human subjects for the pets of unreliable humans? Are the pet owners going to be any more rigorous in their animal feeding routines than human volunteers are in their own?

Fourth, breeds with high cancer rates are the worst systems for testing dietary preventative measures. Almost certainly, these breeds have specific genetic lesions which make them susceptible; most humans will not have such lesions. Any findings made in such animals are likely not to be relevant in most humans. (Cancer-prone animals *are* used in research - eg Oncomouse - but usually for looking at mechanisms of cancer progression or treatment rather than prevention.)

Animal studies relating to diet are not ideal (see point 2 above). But, if you're going to get anything worthwhile, you would do it using experimental animals where you can control the numerous other variables (excercise, genetic background etc).

Human studies are complicated by the fact that humans are hard to control, but they do have the considerable advantage of being directly relevant to humans.

Both human and animal studies of cancer prevention are widespread already.

Overall, alas, this idea lacks merit.
-- MaxwellBuchanan, Feb 13 2009


Man, have you ever looked into what they put into dog food? I had to switch because I thought my dog might have an allergy to one of the red dyes that is in most of them - it seemed to work, no more scratching, but they put some crap in most of those foods. Unfortunately, the good ones are pretty expensive.
-- Zimmy, Feb 13 2009


I don't know about you but i'm not really interested in having a pet that is prone to cancer. Unless we adopt pets that closely model human cancers, and most animals don't, you just won't get any worthwhile data. Studying a model with such a low degree of statistical correlation is a waste of time.
-- WcW, Feb 13 2009



random, halfbakery