Is soy a cancer risk?

December 1, 2010 12:10 PM

Is there a definitive medical consensus on the risk of cancer from eating soybeans?

I've been eating more soy lately (say two or three times a week as part of a meal and once or twice a week as a snack). A friend noticed and flipped out, saying I'm increasing my risk of cancer.

Googling reveals answers all over the map. So is there definitive and current answer to the question of whether soy increases the risk of cancer? Does that answer vary, in any way, based on ones age, sex, or race?

posted by nomadicink to health & fitness (24 answers total)

No, there is no definitive answer, but any answer there is is a lot closer to no than to yes.

And yes, of course the answer varies depending on at least age, sex, genetic background, exposure, and probably about 20 other things.

There is no reason for you to stop or slow your eating of soy. Many many people eat a lot more of it than you, and there is -- at the very best -- scant, not-very-reliable, sketchy evidence that it maybe possibly slightly increases cancer risk, and honestly, I would call that a favorable assessment.
posted by brainmouse at 12:23 PM on December 1, 2010 [1 favorite]

Here's a well written recent review, if you have access to an academic library.

The cancer research is not conclusive. Some studies have shown decreased risk of breast cancer; others have shown increased risk. There's likely no need to worry if you keep your consumption moderate and do not have any additional major risk factors for breast cancer (family history, for example).

There's more concern about potential effects during development, so there might be more cause for caution regarding the consumption of soy products by pregnant women.

There have also been some studies showing benefits: reduced risk of heart disease and osteoporosis.
posted by mr_roboto at 12:25 PM on December 1, 2010 [1 favorite]

Your friend is dramatically oversimplifying things. As your consultation with Dr. Google indicates, the results are all over the map. A similar PubMed search backs this up. The main reason is that "cancer" is in fact many, many different diseases, so even if there were definitive evidence regarding soy's negative impact on one kind of cancer, it might be mitigated by a positive impact on another. For example, this recent meta-analysis finds a 21% reduction in colorectal cancer risk in women but not men. This one finds that consumption of some soy products might reduce prostate cancer risk. This one finds that consumption of soy may reduce breast cancer risk in Asian but not Western women.

In sum, consumption of soy at the levels you are doing may increase your risk for some cancers, and it may decrease your risk for others. But the evidence is not strong enough to unequivocally conclude one way or another.
posted by googly at 12:28 PM on December 1, 2010

Googling reveals answers all over the map.

I don't see any answers that say "Soy as a small part of a varied diet increases risk of cancer" that come from reliable sources.

The data that link soy consumption to increased risk of prostate cancer among men in Africa include, among the studied group, subjects with very limited diets.

This seems like a very detailed and objective overview.
posted by Sidhedevil at 12:28 PM on December 1, 2010

People have been eating soy for thousands of years, and there is plenty of evidence to suggest that it is healthy to do so, especially if it lowers your overall meat intake. However, soy protein isolate has only been around since the 1960s, and forms the basis for most processed soy foods like boca burgers. I personally like to stick to Michael Pollen's rule of eating only food my grandma would recognize, which in this case would be traditional foods like miso, tofu, tempeh, soymilk and edamame.

A 2006 AHA study offered some good evidence that there are differences between the two. They conclude that soy may offer some health benefits (though not as much as previously thought), but soy protein isolate does not.
posted by susanvance at 12:30 PM on December 1, 2010 [4 favorites]

The Japanese eat a lot of soy, in soy sauce and as tofu. (Also natto.)

If soy consumption resulted in a non-negligible increase in cancer rates, doctors there would have noticed it a long time ago.
posted by Chocolate Pickle at 12:33 PM on December 1, 2010

If you were genuinely worried, you could focus more on fermented soy products like tempeh, miso, and natto. No one's ever had anything bad to say about those, and many sources say they're much better for you than unfermented soy.
posted by Go Banana at 12:39 PM on December 1, 2010

If you were genuinely worried, you could focus more on fermented soy products like tempeh, miso, and natto. No one's ever had anything bad to say about those, and many sources say they're much better for you than unfermented soy.

Hold on, hold on, hold on.

The issue here is dietary phytoestrogens, and while it's true that fermented products have lower levels of glucoside phytoestrogens, they have higher levels of the aglycones daidzein, genistein, and glycetein (source). All of these molecules are potential endocrine disrupters.
posted by mr_roboto at 12:48 PM on December 1, 2010

People have been eating soy for thousands of years, and there is plenty of evidence to suggest that it is healthy to do so, especially if it lowers your overall meat intake.

Just want to point out that this is totally wrong and backwards.

If you're a man, eating soy raises the risk of estrogen/testosterone problems. Avoid it at all costs. Eat meat and get healthy and strong.
posted by unixrat at 12:49 PM on December 1, 2010 [2 favorites]

Huh. I hadn't heard anything about soy and cancer. I have heard about consuming 'too much' soy and perhaps the effect of the phytoestrogens in it having some kind of effect (positive OR negative, depending on your particular sex and case), but that's it. I have also heard that non-processed soy is better for you (as susan says above).
posted by bitterkitten at 12:49 PM on December 1, 2010

Oops, I'm too slow.
posted by bitterkitten at 12:50 PM on December 1, 2010

I think the issue with eating soy products these days is that there can be higher levels of hormones, which can sometimes cause cancer.

FWIW, we have soy products every day. I'm more worried about the drywall plaster that's used to solidify most tofu you find in the store, not the risk of cancer.
posted by KokuRyu at 1:06 PM on December 1, 2010

If you're a man, eating soy raises the risk of estrogen/testosterone problems. Avoid it at all costs. Eat meat and get healthy and strong.

Just chiming in here, and there is exactly zero scientific evidence to support this theory.

A lot of talk about this centres around a UNC Chapel Hill Study that dealt with massive doses of isolated isoflavones with little to no effect. We are talking the equivalent of 20-30 cups of soy milk per day.

The study concluded that there was no reason for men to avoid soy products.
posted by WinnipegDragon at 1:06 PM on December 1, 2010 [10 favorites]

[comment removed - question is not about eating meat, take that derail to metatalk or email the non-anon OP please, thank you.]
posted by jessamyn at 2:00 PM on December 1, 2010

I can't find the story, but a couple of years ago I read a NYT article about using soy as an alternative to hormone replacement therapy. That article stated that subjects had to eat upwards of 75 grams of soy protein a day to have any sort of detectable effect on their hormones, no mention of cancer increases.
posted by lunalaguna at 2:05 PM on December 1, 2010

There is no credible medical evidence of any negative effects from soy.

Most anti-soy things out there are on the intellectual level of the NewsMax article forwarded by my imbecilic ex-boss, which claimed soy milk was turning kids "gay." (They actually put it in quotes, as if it were some exotic medical condition.)
posted by drjimmy11 at 2:25 PM on December 1, 2010

As other's have said, there is no medical consensus that eating soy in the amounts you mention will increase your risk of cancer. Here is one article with some citations about health effects of soy. This article is also good and mentions the testosterone issue.
posted by davar at 2:37 PM on December 1, 2010

I had breast cancer that was estrogen-receptor positive and I was advised by my doctors to not eat soy or soy products because of the phytoestrogens present. However, they said the science was inconclusive and that they couldn't see that a small amount, such as the tofu in hot and sour soup, would do any harm. They just advised me not to start drinking soy milk or having lots of soy products.

But I think there's a long road between "you've had a type of cancer that likes estrogen so avoid any extra estrogens if you can" and "OMG! Soy causes cancer!"
posted by angiep at 3:03 PM on December 1, 2010

Thanks ya'll!
posted by nomadicink at 3:34 PM on December 1, 2010

I have a history of benign cysts, but like angiep, I've had doctors advise me to avoid soy because the estrogens *may* promote cyst growth and the type of cysts I have put me in a higher risk group for someday developing estrogen-receptor positive cancer.

There is a lot of anti-soy research summarized here that I think goes beyond the "imbecilic" anti-soy point of view that drjimmy11 referred to.

My conclusion is that foods like tofu, that have been eaten by tons of people forEVER, are probably pretty harmless. But newer frankenfoods, like processed soy veggie burgers, probably are best treated like any other processed food.
posted by chez shoes at 3:55 PM on December 1, 2010

you might already be doing this, but it doesn't hurt to aim for organic and/or non genetically modified soy products when you can. In the US, 93 percent of soybean crops are modified (see chart).
posted by changeling at 3:57 PM on December 1, 2010

There is a lot of anti-soy research summarized here that I think goes beyond the "imbecilic" anti-soy point of view that drjimmy11 referred to.

You are right: they only cite anti-soy research. That should be somewhat telling. I prefer the more balanced approaches that I and other people linked (seriously - they think that a 1953 study about feeding birds a certain kind of soy meal says anything at all about humans eating tofu?). There is a very big anti-soy agenda, which you should keep in mind when reading about soy. See also this article in The Guardian, where the author critizes the scientific methods of the Weston A. Price foundation.

Their stated arguments don't have much to do with research about soy. "MSG is often added to soy foods" is not an arguments against soy at all, "Soy foods contain high levels of aluminium" is as a general statement not supported by research. "Fragile proteins are denatured during high temperature processing to make soy protein isolate and textured vegetable protein." is again, not an argument against soy, etc. etc.
posted by davar at 12:56 AM on December 2, 2010

The cause(s) of cancer(s) are really complicated; it's incredibly difficult to definitively prove that any substance causes cancer. So, okay, forget proving direct causation. Strong correlations are pretty compelling, right? But research exploring correlations between exposure to a substance with incidence of cancer is complicated, too, for a bunch of reasons, such as:

* Foods are not themselves uncomplicated compounds, and no particular foodstuff-x is really consistent, either. Recreated chemical composition of foodstuff does not tend to work in an organism the way foodstuff does.

* Self-reported dietary studies can study consumption of actual foodstuff, but they require participants to be accurate in reporting, and also, again, foodstuff-x is not consistent.

* The effect of that foodstuff on a person can be influenced by lots of factors -- inherited genetic factors, overall health, other specific medical conditions, and past or present exposure to other substances -- that make identifying the "key" to the effect a very noisy process.

The answer to "does x definitely cause cancer" is pretty much always "it depends," and it's safe to say that you should look critically at any source that tells you that the answer is simple. Same goes for prevention of cancer, by the way, except moreso, since it's an attempt to prove a negative.

However, all of the cancer epidemiologists I know are happy to point out that in the face of all of this uncertainty, anyone can lower their risk of cancer in general by eating a balanced and varied diet, getting some exercise, and not smoking.
posted by desuetude at 6:28 PM on December 2, 2010

(Oh, and please don't take the above to mean that those studies exploring correlation aren't incredibly valuable to all sorts of research. It's just that their intended value tends to get garbled in simplistic snappy mainstream news stories.)
posted by desuetude at 2:47 PM on December 3, 2010