Is soy safe to eat?

From our September 2000 issue

Many articles, including several in this newsletter, have touted the health benefits of soy which are largely related to plant estrogens (phytoestrogens) called isoflavones. In addition, the Food and Drug Administration recently approved a health claim for the cholesterol-lowering effects of soy protein. But is there a limit to how much soy is good for us?

Some critics of soyfoods note that any foods or compounds with such potent biological properties even healthful ones are bound to have some unwanted side effects. But defenders of soy cite the long history of consumption in Asian countries without evidence of problems, as well as the lack of problems in Western vegetarian populations who eat soy.

The following discussion looks at three complex areas that have been the focus of these kinds of debates breast cancer, thyroid function, and cognitive function.

Soy and Breast Cancer

Although there has been much interest in the possibility that soy could reduce cancer risk, recently, questions have been raised about its effect in women who have estrogen-positive (ER+) breast tumors. These are tumors that are stimulated by estrogen and therefore might be stimulated by the weak estrogenic activity of the soybean isoflavones. Some studies actually suggest this is the case. Based on the results of a two-week study, British researchers found that soy exerted weak estrogenic effects on breast tissue in premenopausal women. A 10-month study involving both pre- and postmenopausal women produced similar findings. Both studies used about two servings of soyfoods per day.

However, these were preliminary findings, and other research shows different results. For example, soy consumption has been shown to favorably alter the metabolism of estrogen in a way that makes it less likely to stimulate tumor growth. And in animals, soy consumption in general inhibits tumor development often by as much as 50 percent. In fact, in one animal study miso (a fermented soy product) in combination with tamoxifen (a drug used to treat breast cancer) inhibited the development of tumors to a greater extent than tamoxifen alone.

The action of genistein, the primary isoflavone in soybeans, adds to the confusion of the picture. At concentrations similar to those that occur in the blood after soyfoods are consumed, it slightly stimulates the growth of ER+ cancer cells. However, at higher concentrations, genistein inhibits both ER+ cancer cells as well as cells that don't respond to estrogen. Interpreting these findings is difficult because some data suggest that genistein is actually more potent in animals and humans than in laboratory tests. Also, soy may contain other anticarcinogens in addition to isoflavones. All of this suggests that it is not quite so simple to determine the effects of soy on breast cancer risk in women with ER+ tumors.

When we look at human populations in Asian countries where soy intake is higher than in the United States and compare cancer rates to soy intake, there is no association between soy intake and cancer risk.

Since this issue is largely unresolved, women who have ER+ cancer should probably not begin to consume soy therapeutically that is, in large amounts specifically meant to prevent recurrence even though there are many mechanisms by which soy may inhibit breast cancer development. However, women with ER+ breast cancer need not avoid modest amounts of soy for example, three servings per week. However, they should be sure to discuss these dietary decisions with their physician.

Soyfoods and Thyroid Function

A number of foods, including soybeans, contain compounds called goitrogens that interfere with thyroid function (see the April 2000 issue of VNHL). In extreme cases they may cause an enlarged thyroid gland, or goiter, a symptom of inadequate iodine status. However, in animals, increasing iodine intake prevents this goitrogenic effect of soy. Some research also indicates that heating soy a part of the process for making all commercial soyfoods seems to eliminate the goitrogenic effects somewhat.

In the 1950s, several cases of goiter were identified in infants fed formulas based on soy flours that were not fortified with iodine. This type of formula is no longer used; today's infant formulas are based on soy protein isolates and are fortified with iodine. In animal studies, soy protein isolate has been shown to be less goitrogenic than soy flour. No cases of goiter have been seen in babies reared on current formulas. One area of concern involves infants who have congenital hypothyroidism, a syndrome of thyroid deficiency. These babies need synthetic thyroid hormones, and research shows that those consuming soy formula need about 25% more of the hormone. However, bear in mind that fiber supplementation has also been shown to increase the need for thyroid medication in thyroid patients. It may be that soy, like fiber, causes less thyroid hormone to be absorbed, rather than exerting some adverse effect directly on the thyroid.

It isn't clear which factors in soybeans are goitrogens. Some research suggests that it is the isoflavones, but other research contradicts this. In healthy postmenopausal women given 90 mg of isoflavones a day there were no differences in thyroid hormones compared to women who received a placebo.¹⁴ In a study at the University of Minnesota, consuming isoflavone-rich soy products had no effect on thyroid hormones in either pre- or post-menopausal women. Likewise, consuming isoflavone-rich soy protein over a four-week period had no effect on thyroid hormones in men.

Most of the evidence indicates that there is little risk to thyroid function in healthy people who consume soy. However, it is possible that those with marginal intakes of iodine could be at risk. Some research suggests that vegetarians, particularly vegans, have low iodine intake compared to the rest of the population. That may place them at risk if they consume generous amounts of soyfoods or other foods that contain goitrogens. The answer is definitely not to give up soy, however, but rather, to increase iodine intake, which is easily done by using a small amount of iodized salt.

Soy and Cognitive Function

Human research on soy and cognitive function is very limited, but one recent study did find a negative effect of soy. In the Honolulu Heart Program, an ongoing study of the health of Japanese-American men living in Hawaii, those who consumed the most tofu (two to four times per week) during middle age showed the most signs of mental deterioration in later years. Although this study controlled for other factors, it is possible that tofu consumption was linked with something else that was responsible for the cognitive decline. However, there is a plausible biological explanation for this finding if soy isoflavones act as an anti-estrogen on brain tissue. But this is speculative at this point.

Other evidence suggests that soy isoflavones (phytoestrogens) may protect against cognitive decline since estrogen therapy may help to maintain cognitive function in postmenopausal women and it may even protect against Alzheimer's disease. Some observations lend support to this theory. Japanese men living in Hawaii have more dementia and Alzheimer's disease than those living in Japan. And, among Japanese women living in the West, those who maintain a more "Japanese" lifestyle have better cognitive function. Presumably, a more Japanese lifestyle would include consumption of soyfoods. But since there are many dietary and lifestyle factors that affective cognitive function, it's difficult to establish the importance of any one dietary component based on these observations. In addition, criteria for diagnosing Alzheimer's disease vary among countries so that it is difficult to compare rates of this disease.

The findings from Hawaii are confusing and certainly provide justification for further investigation in this area. But epidemiologic studies don't show cause and effect, and many times, individual epidemiologic studies report findings that are not borne out in other studies. In fact, a study of Japanese-American women living in Seattle did not find any relationship between tofu consumption and cognitive function decline over a two-year period. This is a short period of time to detect differences, however, and the data have not been fully analyzed, so it's not yet possible to draw conclusions from this research or appropriate to make changes in eating habits.

The limited number of animal studies in this area indicate that soy has either no effect on cognitive function or that it is protective. Until there are more clinical and epidemiological studies in this area, it's not possible to draw any conclusions about soy intake and cognitive function.

Conclusion

Arguably, even the most healthful foods contain components that produce unwanted effects when they are tested in isolation in the laboratory. Broccoli, cabbage, lentils, and grapefruit all contain many naturally-occurring pesticides that can cause mutations. But that doesn't mean you should reduce your intake of any of these foods. It takes big doses of the compounds to elicit these effects much more than people could normally consume just by eating food. And we have a long history of consuming them without problems. On the other hand, just because a food has a history of consumption, it isn't necessarily benign. Such foods could still produce unwanted effects in some situations or in some individuals. For nearly all healthy individuals soy appears to be perfectly safe when consumed in reasonable amounts (up to two to three servings per day), as part of an overall healthy diet. Women with ER+ breast cancer, however, should be more restrictive in their use of soyfoods until more is known about their effects.

An update on soy and chronic disease

The vast majority of research on soy focuses on chronic disease reduction. Here is a summary of the latest findings:

Cancer: Soybeans contain a number of compounds that have anticarcinogen activity. The low rates of breast cancer in Asian women, compared to Westerners, has fueled speculation that soy is one of the protective factors for this population. However, studies within Asia don't show much difference in soy intake between women who have breast cancer and those who don't. Controlled studies in which soy is fed to women show mixed results. In contrast, many animal studies show that adding soy to typical laboratory diets reduces mammary tumor development by 25 to 50 percent. Animal studies indicate that it may be consumption of soy early in life, before puberty, that gives later protection against breast cancer. A recent human study found that soy intake in adolescence but not in adulthood was protective against breast cancer later in life.

Although there is much less research on prostate cancer, the research does suggest a protective effect for soy. Research at Loma Linda University found that men who consumed one to two servings of soymilk per day were 70 percent less likely to develop prostate cancer. And animal studies have found that soybean isoflavones directly inhibit growth of prostate tumors. While there are still many questions about soy and cancer, the data do suggest that compounds in soybeans may be beneficial in protecting against some cancers.

Heart Disease: Dosing up on soy protein (about 25 to 50 grams per day or the amount in three to six servings of soyfoods) has been shown to reduce blood cholesterol levels in people whose levels are elevated. The effect for those with moderately high levels below 240 mg/dl is fairly modest. But soy may help to protect against heart disease in other ways. The isoflavones in soy which are plant estrogens may exert some of the same protective effects as the hormone estrogen, which directly affects blood vessels.

Osteoporosis: Estrogen promotes bone health and it is possible that soy isoflavones have a similar effect. In fact, a synthetic isoflavone called ipriflavone has been used for 15 years to reduce bone loss in women. The effects of soy on bone health isn't yet clear, since the only studies to date have been either in laboratory animals or in very small groups of women, but they are certainly promising. However, since animal protein causes increased calcium loss relative to soy protein, there may be a double advantage to replacing animal foods in the diet with soyfoods.

Soyfoods in children's diets

It's back to school and, for the first time, the US Department of Agriculture says that your children can purchase soyfoods as a meat substitute in school lunches. But are they safe for little ones to eat Soyfoods, including soy infant formula, contain biologically active compounds the soy isoflavones that have some hormone-like effects.

We can look around the world and gain some assurance that these foods are indeed safe. Asian children have been regularly consuming soy for centuries. And, any problems with soy intake would be more likely to occur very early in life.

This brings us to questions about the safety of commercial soy infant formula, which is a main source of nutrition for 20 to 25 percent of infants at some point. During the earliest periods of development, infants are extremely sensitive to environmental influences. And exposure to soy isoflavones is indeed high on a body-weight basis in infants who consume soy infant formula as indicated by their high blood levels of isoflavones. Infants who weigh only ten to twenty pounds can consume about the same amount of isoflavones that has been shown to have biological effects in adults. However, short-term studies indicate that infants fed soy infant formula grow and develop normally. One small study found no hormonal abnormalities in three-year olds who had consumed soy formula in infancy. Furthermore, there have been no reports of hormonal abnormalities in people who were fed soy formula as infants and this includes millions of people in the past 30 years.

Nevertheless, critics of soy formula say that since there have not been any long-term studies of this population, subtle effects could go unnoticed. And it's true that we won't have all the answers until well-designed research is conducted aimed specifically at looking for these problems.

For now, the evidence indicates that there isn't sufficient reason to avoid soy formulas. But neither cow's milk nor soy formulas can match the advantage of breast milk always the best choice for babies before their first birthday.