September 1998 — U.S. Edition 

Shopping for Soy
“Welcome to the exciting — and healthy — world of soy!” says the blurb for GeniSoy soy powders and bars in the online store run by Reach4Life Enterprises (

   “Researchers...have found that genistein, a flavonoid extract from soy, inhibits the growth of breast cancer cells grown in laboratory dishes,” it continues. “Additional studies have shown that isolated soy protein has special health benefits which can help prevent osteoporosis, heart disease and aid in menopause.”

   Welcome to the overblown — and under-substantiated — world of soy claims.

   Some products, like Solaray’s Genistein pills, point to “a link between higher intake of genistein and an increase in overall health and wellness.” Others get technical, noting that genistein is an “important inhibitor of capillary proliferation” or that it “can bind to hormone receptor sites and reduce the uptake of certain tissue-specific hormones.”

   The bottom line is that these claims come from the companies’ marketing departments. No one outside the firm checks to make sure they’re honest. And you can’t assume that genistein or other isoflavones are safe just because soy is a popular food in Asia.

   It’s a long way from tofu and tempeh (cooked and fermented soybean cake) to soy protein bars and shakes, and further still to isoflavone pills. According to the latest estimates, the Japanese average seven to ten grams of soy protein a day (though they might have eaten more in the past).1 That’s about 30 to 50 mg of isoflavones, says Loma Linda University soy expert Mark Messina.

   But people can get far higher doses from purified or synthetic genistein, which “opens up the possibility of toxic side effects,” says the University of Alabama’s Stephen Barnes. Though none are known, he adds, “there’s a toxic range for any compound.”

   If you want to boost your soy intake, here’s what you need to know:

Look for isoflavones. So far there’s no evidence that straight isoflavones — minus soy protein — can lower cholesterol. But you want some isoflavones, and you can’t assume that all foods that contain soy have them. Soy sauce, Ensure, and Sustacal have little or none. Ditto for many soy foods that are made from soy protein concentrate, which can be processed using an alcohol wash that destroys isoflavones. Unfortunately, unless a company lists isoflavone content on its label (and few do), there’s no way to tell from looking at the ingredient list.

Take a powder. Though brands vary, soy protein powders — which you mix into juice, milk, or whatever — are the most efficient way to get soy protein plus isoflavones. Soy protein shakes typically supply fewer isoflavones, because they have added sugar. Some powders and shakes are fortified with extra vitamins and minerals; others aren’t (check the label). Bonus: Most labels also list isoflavone levels.

Hit the bar. If you’d rather eat than drink your soy, companies like GeniSoy make it easy with vitamin-fortified not-quite-as-sweet-as-candy bars that come either bare or coated with chocolate or yogurt. You pay for the convenience with extra calories — about 200 versus roughly 100 for most powders. They come from the extra sugar and, in the coated flavors, extra fat. What’s more, palm kernel oil in some of the coatings can use up more than ten percent of a day’s worth of heart-damaging saturated fat. Check the label.

Don’t forget tofu. It may no longer be trendy, but tofu (soybean curd) tastes terrific in stir-fries, marinated and grilled, or right out of the package (if it’s a cooked, flavored brand like White Wave Baked Tofu). Tofu is a good source of protein that has no cholesterol and less saturated fat than meat or poultry.

1: Journal of Nutrition 128: 209, 1998.

Baby Soy
Fact: Infants fed soy formula have higher levels of estrogen-like isoflavones in their blood than infants fed cow’s milk formula or breast milk.

   Fact: Babies have been fed soy formulas for the past 30 or 40 years with no apparent harm. Is that because no one’s looked or because soy is really harmless?

   “No one has conducted an epidemiological study to look at possible adverse outcomes due to soy exposure,” says Daniel Sheehan, a birth defects expert at the National Center for Toxicological Research (NCTR) in Jefferson, Arkansas. “If you don’t look, you don’t find.” (Sheehan’s views are not those of the NCTR.)

   Infants who are fed soy formula get the equivalent of six to 11 times the dose of isoflavones that’s enough to change menstrual patterns in women who are fed soy.1,2

   What harm could isoflavones conceivably cause? Menstrual or ovarian disorders, endometriosis, and low sperm counts are just a few possibilities. But that’s no reason to panic.

   “There absolutely must be a biological effect of the high exposure to isoflavones in babies fed soy formula,” says Kenneth Setchell of the Children’s Hospital Medical Center in Cincinnati.

   “The critical question is whether the effect is adverse or beneficial.” Animal studies suggest that soy formula could lead to a lower risk of breast cancer, he points out.

   Until the government or producers fund studies to examine the issue — or the industry simply removes isoflavones from soy formula — parents are left in a quandary. Studies are under way to determine whether cow’s milk formula can raise the risk of juvenile diabetes.

   The bottom line is that no one has answers yet. The only thing everyone agrees on: Breast milk is best.

1: Lancet 350: 23, 1997. 2: Amer. J. Clin. Nutr. 60: 333, 1994.

The Bottom Line
Roughly 25 grams a day of soy protein (with its naturally occurring isoflavones) is enough to lower LDL (“bad”) cholesterol by about ten percent in people who start out with an LDL above 160.

It’s too early to say whether soy or its isoflavones can reduce the risk of breast or prostate cancer, osteoporosis, or hot flashes.

To play it safe, women who have had breast cancer shouldn’t load up on soy protein or take isoflavone pills.

Soy, What’ New?
(Acrobat 39k)

   How good is the evidence? It’s clear that soy can lower blood cholesterol. Beyond that, nothing is clear.

   “There are alot of reasons to eat soy and the evidence is encouraging, but it’s still preliminary,” says soy expert Mark Messina of Loma Linda University in California.

   The soy craze could even do damage. Some researchers worry that high doses of isoflavones may harm women who already have breast cancer.

   “It’s alarming because supplement makers are rushing ahead of the game,” says Stephen Barnes of the University of Alabama at Birmingham. “We don’t know what effects high doses have in humans.”

   It all started with China and Japan. Compared to Americans, Asians have fewer heart attacks, less breast and prostate cancer, fewer hip fractures, and the women report fewer hot flashes and other symptoms of menopause.

   “For a while it looked like, ‘oh my gosh,’ soy is the answer to everything,” says researcher Margo Woods of the Tufts University School of Medicine in Boston. “But the more research that’s been done, the more things got complicated.”

   Least complicated is soy’s ability to cut blood cholesterol.

^ Top

Heart Disease
After more than five decades and dozens of studies, researchers have figured out that soy does lower cholesterol, but only if you feed the right soy foods to the right people.1

   “Some researchers have thought soy protein alone was enough, but you need isoflavones together with the protein,” says John Crouse III, professor of internal medicine at Bowman Gray School of Medicine in Winston-Salem, North Carolina.

   Crouse put 156 people with high cholesterol on a standard cholesterol-lowering diet.2 Then he gave them a daily beverage containing either 25 grams of soy protein or 25 grams of casein (a milk protein) as a placebo. But the soy beverages weren’t identical. One naturally contained 62 mg of isoflavones per serving. The others were altered to contain 37, 27, or 3 mg per serving.

   “In people who started out with LDL (‘bad’) cholesterol over 160, LDL fell by ten percent in those who got 62 mg of isoflavones and by eight percent in those who got 37 mg of isoflavones a day,” says Crouse. LDL didn’t fall significantly in those who got 3 mg or 27 mg.

   And for people who started out with an LDL between 140 and 160, cholesterol didn’t drop significantly, even if they got 62 mg of isoflavones.

   (Would pure isoflavones — without soy protein — lower LDL? In two studies, they didn’t., But so far, no one has tested isoflavones alone on people with high LDL.)

   “Would I recommend soy protein as an exclusive treatment for people who need to lower LDL by more than ten percent? No,” says Crouse.

^ Top

Beyond Cholesterol
The heart-and-soy story doesn’t stop there. The isoflavones in soy are phytoestrogens — weak estrogen-like substances made by plants. And they may protect the heart like the estrogens that are naturally produced by premenopausal women or the estrogen pills taken by many postmenopausal women.

   “Women have a high frequency of anginal pain that’s not associated with blockage of the coronary arteries,” explains Bowman Gray researcher Tom Clarkson.

   “Estrogen is necessary to enable coronary arteries to dilate when they get a signal,” he adds. Without estrogen, they constrict, and that causes the pain. But when he feeds soy with isoflavones to female monkeys that have clogged arteries, “the arteries are able to dilate when they need to.”

   And when researchers in Australia gave 80 mg of isoflavones a day — without soy — to 21 women, they found a 26 percent improvement in the elasticity of the women’s arteries.3

   What’s more, the arteries of monkeys that are fed a diet high in saturated fat get less clogged if they’re also fed soy protein. That’s because estrogen helps keep the number of LDL particles that go from the blood into the artery wall close to the number that come out, says Clarkson.

   “As long as the LDL particles move out of the lining of the arteries as often as they enter, there’s no atherosclerosis,” he explains. “Estradiol [the major form of estrogen that occurs naturally in women], conjugated estrogens from animals [the estrogen in pills], and soy phytoestrogens are all capable of these cardiovascular benefits.”

^ Top

Breast Cancer
Can soy prevent breast cancer? “It’s the most interesting question, because everyone wants it to,” says Loma Linda’s Mark Messina, who consults for the soy industry. “But it’s also the most speculative and the least likely to show anything definitive in the immediate future.”

   Genistein, a major isoflavone in soy, is a weak estrogen. If estradiol promotes cancer, soy could protect the breast by interfering with estradiol — that is, by acting as an anti-estrogen.

   “If phytoestrogens compete with your natural estrogen to bind to the body’s estrogen receptors, then they would decrease the receptors’ ability to pick up estradiol,” says Woods. “At least that’s what we think happens in premenopausal women.”

   But the story isn’t that simple. Phytoestrogens may also act as estrogens, she adds.

   “In postmenopausal women, where you have little of your own estradiol, phytoestrogens could have a small estrogen-like effect because they do bind to the estrogen receptors and translate some of the estrogen message to the cells,” explains Woods.

   Genistein’s impact on cancer cells isn’t simple either.

   “In test tubes, genistein either stops cancer cells from growing or makes them grow,” says Barnes. “It depends on the kind of cells. We don’t know how to interpret it.” Ditto for studies in animals.

   Barnes’ colleague, Coral Lamartiniere, has given genistein to rats soon after birth or shortly before puberty. In either case, the animals got fewer tumors when treated with carcinogens later in life.7 “They had lifelong protection from the cancer,” he says.

   But in two other studies, genistein alone had a weaker impact or no impact on the incidence of tumors in adult animals.8,9 And in some studies that fed soy protein to adult rats, tumor rates were lower; in others they weren’t. “In adult animals, the results on soy and breast cancer are mixed,” says Messina.

   The same holds for human studies.

   “The population studies on soy and breast cancer are very limited and not consistent,” says Regina Ziegler, a nutritional epidemiologist at the National Cancer Institute in Bethesda, Maryland.

   For example, studies in China found no difference in soy intake between breast cancer patients and their healthy counterparts.10 In Singapore, investigators found a link between a higher soy intake and a lower risk of breast cancer in premenopausal, but not postmenopausal, women.11

   And Ziegler and colleagues found that Asian-American women with breast cancer ate less tofu as adults than those without cancer.12 But the tofu may not explain the lower risk. “Soy could just be an indicator of something else that alters risk, like lower weight, more physical activity, fewer calories, less meat, or more vegetables,” she explains.

   Others agree. “There’s reason to be hopeful, but we won’t know until we have studies that are better designed and evaluated,” says Lamartiniere.

^ Top

Prostate Cancer
“Autopsy studies show that Japanese men have the same amount of non-invasive prostate cancer cells as American men,” explains Barbara Winters of the American Health Foundation in New York. “But the cells don’t progress to prostate cancer as they do here, so the men are likely to die of something else first.”

   It’s probably not genes that protect Japanese men. Their prostate cancer rates rise if they move to the U.S. And — perhaps because the Japanese diet is becoming more like ours — prostate cancer rates are up in Japan.

   Researchers know that isoflavones slow the growth of prostate cells in test tubes and in some animal studies.13,14,15 But so far, no studies have been done on people.

   Enter PSA, or prostate specific antigen. With a simple blood test, PSA levels predict who will get prostate cancer and whether it will return in men who have had surgery to remove a cancerous prostate gland. That makes it practical to design trials on soy and the prostate.

   Working with Sloan-Kettering Memorial Hospital in New York, the American Health Foundation has a pilot study under way to see if men who have had their prostates removed can follow a low-fat soy-supplemented diet. The men will consume a low-fat or ordinary diet with or without a soy beverage that’s fortified with 200 mcg of selenium and 800 IU of vitamin E. The beverage contains 40 grams of soy and 40 mg of isoflavones.

   “Some of the men have elevated cholesterol and it’s well established that soy may lower it,” says Winters. “So it’s not a bad thing for total health.”

   Although most animal studies have tested genistein, Winters’ study uses soy powder, not pills. “We don’t know about the safety of isoflavone pills,” she says. “It’s safer to start with a food source because that’s what the epidemiology is based on.”

^ Top

Millions of postmenopausal women in the U.S. take estrogen to stem the bone loss that leads to osteoporosis. Could they eat soy foods or take phytoestrogen supplements instead?

   Comparing U.S. women to Asian women doesn’t answer the question.

   Japanese women have a lower rate of hip fractures, but their pelvic structure may be more resistant to breaks. “And their risk of spinal fracture may be higher than ours,” says Messina.

   Other evidence is more encouraging. In rats that lose bone because their ovaries are removed, soy protein prevents the loss.16 And ipriflavone, a synthetic drug derived from isoflavones, increases bone mass in women with osteoporosis or low bone mass.17 But studies on soy and human bones are still preliminary.

   At the University of Illinois, researchers compared 40 grams a day of milk protein to 40 grams of soy protein that contained either 56 or 90 mg of isoflavones.18 (The women who took soy were given enough calcium to match the amount in the milk protein.)

   “In the postmenopausal women who got soy protein with 90 mg of isoflavones every day for six months, we found higher bone mineral density in the lumbar spine,” says Susan Potter, now at Protein Technologies International, a soy producer in St. Louis. “Other bones weren’t affected, but that’s not surprising, because the spine responds more quickly.”

   She also saw no improvement in bone density among the women who got only 56 mg of isoflavones. “The results are promising, but the study needs to be done for one to three years,” says Potter.

^ Top

Menopausal Symptoms
Estrogen pills cut the frequency of hot flashes by 60 percent in women going through menopause. It’s too early to say whether soy can do the same.

   “Most studies say that soy has a slight effect, but the placebo effect is high,” says Tufts’ Margo Woods.

   For example, in a small Australian study, both the women who got soy flour and the women who got wheat flour (as a placebo) reported fewer hot flashes.19 And in a pilot study at Bowman Gray School of Medicine, women who got 34 mg of isoflavones a day in soy powder had less severe (though no fewer) hot flashes than women who got a placebo.20

   Perhaps the best studies are still in progress. In one, Woods is giving 100 women who reported at least five hot flashes a day 45 mg of isoflavones in two soy bars every day for three months and two (lookalike but soy-free) placebo bars for another three months.

   What can women do in the meantime? “When I get calls from women with hot flashes, I suggest that they try soy foods — not the isoflavone pills — and see if they help,” says Woods. Her only caution is for women who have had breast cancer.

   She kept them out of the study on hot flashes because the phytoestrogens in soy might interfere with the anti-estrogen effect of tamoxifen, a drug often prescribed to women with breast cancer. “With whole foods, we don’t expect soy to raise their risk, but we have no data.”

   Soy isn’t the only step women can take. “Hot flashes are more likely if you’re under stress and less likely if you exercise,” says Woods.

^ Top

The Waiting Game
The good news: A surprising number of studies on soy are in progress.

   For example, Tom Clarkson’s team is studying 240 women who are close to or going through menopause. Each is consuming a daily soy “shake” with 58 mg, 42 mg, or essentially no isoflavones. They’re tracking bone density, cholesterol, and blood pressure and are doing mammograms, uterine biopsies, and carotid ultrasound (to measure clogging in neck arteries).

   The trial is due to end next year. “It would be foolhardy to make recommendations until we finish,” says Clarkson. “I suspect that soy is safe and effective, but neither is known.”

   In the meantime, men who have high cholesterol could drink a soy beverage to help knock it down (see “Shopping for Soy”). But should women who are at high risk of — or already diagnosed with — breast cancer load up on soy or isoflavones?

   Two studies have raised red flags. In one, 24 pre- and postmenopausal women who got a daily beverage with 38 grams of soy protein and 38 mg of genistein had breast fluid aspirated every month for six months.

   Researchers expected to find a decrease in fluid, as they find in Asian women. Instead they found an increase, and increased estradiol levels, in the premenopausal women. And in seven of the 24 women, the fluid showed an increase in cell proliferation.21 Any time cells multiply, there’s an increased risk of cancer.

   “We thought we were going to suppress the breast, but instead we stimulated it,” says Nicholas Petrakis of the University of California, San Francisco. “We don’t know if that’s a problem or not.”

   In another study, British researchers at the University of Manchester gave 60 grams of soy protein (with 45 mg of isoflavones) a day to women for two weeks before they were scheduled to have breast biopsies.22

   Again, soy stimulated breast cell proliferation. “Perhaps the effect is only short-term,” says Messina. “But we don’t know.”

   “We’re playing with a complicated biology that we don’t understand,” says Walter Willett, chair of the nutrition department at the Harvard School of Public Health.

   The bottom line, say scientists, is that we just have to sit tight. Says Tufts’ Margo Woods: “We’re going to have to wait for the answers, even if they’re 15 years down the road.”

^ Top


1: New Eng. J. Med. 333: 276-282, 1995.
2: Circulation 97: 816, 1998.
3: Arter. Thromb. Vasc. Biol. 17: 3392, 1997.
4: Journal of Nutrition 128: 728, 1998.
5: Fertility and Sterility 67: 148, 1997.
6: Arter. Thromb. Vasc. Biol. 17: 2524, 1997.
7: Carcinogenesis 17: 1451, 1996.
8: Anticancer Research 16: 3293, 1996.
9: Proc. Soc. Exp. Med. Biol. 12: A828, 1998.
10: British Journal of Cancer 71: 1353, 1995.
11: Lancet 337: 1197, 1991.
12: Cancer Epidem. Bio. Prev. 5: 901, 1996.
13: Proc. Soc. Exp. Med. Biol. 12: A658, 1998.
14: Proc. Amer. Urol. Assoc. 153: 269A, 1995.
15: The Prostate 22: 335, 1993.
16: Journal of Nutrition 126: 161, 1996.
17: Calcified Tissue Internat. 54: 377, 1994.
18: Amer. J. Clin. Nutr., in press.
19: Maturitas 21: 189, 1995.
20: Menopause, in press.
21: Cancer Epidem. Bio. Prev. 5: 785, 1996.
22: Second International Symposium on the Role of Soy in Preventing and Treating Chronic Disease: p. 35, 1996.

See Also: Phytoestrogens for Menopause

Nutrition Action Healthletter Subscribe Today! Customer Service