This 'wonder food' may be more hype than help
by Janet Walzer

You'll find enough soy products in the supermarket to merit a separate aisle: Soy burgers, soy hotdogs, soy cheese, to name a few. These products have proliferated due to the government's labeling of soy as a food that reduces the risk of heart disease. But in fact, the scientific jury is still out as to soy's healing powers. Although including soy in your diet makes sense—it is full of nutrients and low in saturated fat—as Tufts researchers have found, evidence for the therapeutic benefits of soy is debatable.

The wonderful world of plants
Soy is full of phytochemicals, which are substances found in plants that are not nutrients but might have health properties nonetheless. There are several classes of phytochemicals, and flavonoids, which are found in soy, make up one class. In fact, all plants produce flavonoids, and certain plant-based products rich in this compound, such as green tea and red wine, are receiving considerable scientific attention for their potential health benefits.

"It's a brave new world with phytochemicals," comments Johanna T. Dwyer, director of the Frances Stern Nutrition Center and senior scientist at the Jean Mayer USDA Human Research Center on Aging (HNRCA) at Tufts, who is working on several projects involving flavonoids. "There are thousands of flavonoids, [with] two dozen probably the most critical," explains Dwyer, who along with nutrition students Julie Peterson, N02, and Kenneth Chui, N03, are collaborating with the USDA's Beltsville Agricultural Research Center on the development of a food composition database. Peterson, who has been working on this project for five years for her thesis, is combing through the literature on flavonoids, analyzing the data and then entering the information in what should become an invaluable resource on these substances, which might have cancer-fighting characteristics. Dwyer points out that the government only collects information on about 120 substances, so the promise of phytochemicals, such as flavonoids, is exciting.

Isoflavones, which are a class of flavonoids and are abundant in soy, have been the object of much scientific scrutiny, particularly in the last decade as more researchers consider their relationship to human estrogen. Some isoflavones are "phytoestrogens" or plant estrogens, and their possible effect is one piece of the soy puzzle: A phytoestrogen can both mimic human estrogen or block it. This dual role becomes particularly significant in medical conditions unique to women.

Good for the heart?
In 1999, the Food and Drug Administration (FDA) gave the go-ahead for products with a minimum of 6.25 grams of soy protein to receive the following label: "25 grams of soy protein a day as part of a diet low in saturated fat and cholesterol may reduce the risk of heart disease." This amount is equal to four servings of soy protein daily. But before considering a daily lunch of Tofu Pups™, it is worth reviewing the research behind this assertion.

The FDA based its health claim primarily on a meta-analysis that appeared in The New England Journal of Medicine in 1995. The study authors statistically analyzed 38 studies and concluded that eating approximately 50 grams of soy protein a day reduced total cholesterol, low-density lipoproteins (LDL, known as the "bad cholesterol") and triglycerides. But what has been found since the 1995 study?

"The data have not been strong in that direction since then," remarks Alice H. Lichtenstein, senior scientist and director of the Cardiovascular Nutrition Research Program at the HNRCA. Lichtenstein is referring to several studies that have shown that soy has no effect on health as well as to the research she and her colleagues conducted at the HNRCA from 1998 to 2001 on the effects of soy protein and soy isoflavones on blood lipid levels.

The study participants consisted of 24 postmenopausal women and 18 men, each of whom rotated through four different diets; this rotation, Lichtenstein points out, is statistically more powerful than having four groups of 42 people because each person serves as his or her own control. A reduced triglyceride effect was found, but there was no impact on cholesterol with soy isoflavones, and the effect with soy protein was negligible. When asked if these results were expected, Lichtenstein replies, "We were not surprised by [our] findings because when we re-looked at the meta-analysis, the studies showed that the most impact came from one laboratory and were not supported by other studies."

Because the FDA has attached a public health claim to the use of soy, Lichtenstein, the Stanley N. Gershoff Professor at the Gerald J. and Dorothy R. Friedman School of Nutrition Science and Policy, says a thorough assessment demonstrating its accuracy is essential. She allows that it is possible that the processing of the soybeans to make the final product might have something to do with the discrepancies among studies, perhaps by altering the phytochemical content. To tease out these issues, Lichtenstein is now investigating the effect of soy-derived foods, including soy burgers and tofu, on cholesterol.

"This is a really good example of what initially looks like an easy answer or quick fix," reflects Lichtenstein. "Just take some soy and no need to worry about cholesterol levels." However, she continues, "Just because there doesn't seem to be a strong independent factor for reduced cholesterol levels doesn't mean that substituting soy burgers for hamburgers won't yield some type of [health] effect." In other words, replacing saturated fat with unsaturated fat can have a positive effect on lipid levels. Still, Lichtenstein believes that soy protein has taken on a mythical role in the American diet. "If you have high cholesterol, you can't rely on soy products to reduce those levels."

Isoflavone vs. placebo
Soy researchers have also focused on menopausal symptoms due to the isoflavones' capacity to act like the natural hormone estrogen. It has been thought that ingesting soy either through food or a supplement might provide a reprieve from one of the most annoying menopausal symptoms, hot flashes.

Margo N. Woods, associate professor of family medicine and community health at Tufts School of Medicine, and colleagues conducted a study to determine the impact of a soy supplement on hot flashes. The study consisted of 85 menopausal women who had five or more hot flashes a day. The participants were given two bars of 45 milligrams of isoflavones (and some soy protein) daily for three months. They took a break for one month, and then the same women ate two placebo bars for another three months. Neither the participants nor the researchers knew which bar they were given, making it a double-blind study.

Woods found that there were no differences in the number or intensity of hot flashes when the soy and placebo time periods were compared. Surprised by these outcomes, Woods and colleagues asked these questions: Did we give enough soy? Did we give enough isoflavones? Were levels biologically significant? The answer was yes in all cases.

This lack of effect is echoed by Barry R. Goldin, professor of family medicine and community health at Tufts, regarding a premenstrual (PMS) study he and colleagues recently completed. In a double-blind study, 74 women with mild to moderate symptoms of PMS were either given a placebo pill or an isoflavone pill for three months. Each woman kept a diary of her symptoms. The results showed that 40 percent of the participants improved with both treatments. "It is tough to say that isoflavones have no effect with such a huge placebo effect," concludes Goldin. "You would need a very strong effect to see a difference, and it is not clear if isoflavones can do it."

Not totally benign
More worrisome than not knowing if soy is more effective than placebo is determining whether soy, in fact, can cause harm. There has been evidence indicating that high levels of isoflavones might increase the risk of breast cancer in women who are high risk for the disease. Yet studies of Asian women suggest that increased soy consumption could be protective for breast cancer, and actually, this population of women has a better prognosis when diagnosed with breast cancer. Many differences exist between the lives of American and Asian women, however, with factors such as environment confounding the issue. Woods' study did show that women who ate the soy supplement had lower levels of estrogen in their blood compared with their levels after eating the placebo bar. These lower levels would then reduce estrogen exposure, thereby decreasing the risk of breast cancer.

To demystify the relationship between soy consumption and breast cancer, more research is necessary, particularly randomized clinical trials in the United States, which both Woods and Goldin point out, have never been conducted in this country. What is not confusing is the impact of soy supplements on women taking tamoxifen for the treatment of breast cancer. It has been shown that this level of isoflavones directly competes with this drug therapy, and therefore, soy supplements should not be taken by women being treated for breast cancer.

For the rest of the population, soy in moderate amounts can be part of a well-balanced diet. Says Woods, "It is clear that soy is a good alternative to other proteins...but it is not a medication. It is a food."