Soyonara? Tough times for the "miracle bean".

Is soy a "Fab bean or has-been"? wondered the Los Angeles Times in May. By July, the Orlando Sentinel didn't have to ask. "Bean's health benefits are exaggerated," it warned.

It's been a pretty tough year for the little legume.

* In August 2005, an expert U.S. government panel found unclear or insufficient evidence that soy can prevent heart disease, relieve menopausal symptoms, or prevent osteoporosis. That led the National Institutes of Health to suspend funding for new soy studies.

* In the fall of 2005, the soy industry withdrew its petition asking the FDA to allow labels to claim that soy protein helps prevent cancer.

* In February, the American Heart Association concluded that soy doesn't cut bad cholesterol as much as experts had thought.

Is soy losing its cachet as a health food hero?

Heart disease, breast cancer, prostate cancer, hot flashes, osteoporosis, brain aging. Researchers have been hunting for evidence that soy--or its estrogen-like isoflavones--can curb or prevent each of them. In most cases, the evidence is still iffy at best. Here's what we know so far.

HEART DISEASE

"Diets low in saturated fat and cholesterol that include 25 grams of soy protein a day may reduce the risk of heart disease."

That's the health claim, approved by the Food and Drug Administration in 1999, that put soy on the map ... and on supermarket shelves. Sales of soy foods have doubled since then.

"But the early research this claim was based on hasn't been confirmed by later research," says Alice Lichtenstein. She's director of the Cardiovascular Nutrition Laboratory at the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University in Boston and a member of the American Heart Association's Nutrition Committee.

"The original recommendation and the hype about soy really started with a 1995 meta-analysis published in the New England Journal of Medicine," recalls Lichtenstein.

That study, commissioned by the Dupont-owned soy-processing company Solae, pooled the results of 38 clinical trials. Its conclusion: people who ate an average of 47 grams of soy protein a day cut their LDL ("bad") cholesterol by an average of 13 percent. (1)

But soy had no effect on those with total cholesterol levels below 256, which includes most people. "The bulk of the studies in the meta-analysis were done in Italy, focused on people with very high cholesterol levels, and used a soy protein preparation that doesn't appear to be available in the U.S.," Lichtenstein points out.

Today, the picture looks very different.

"The majority of the studies conducted since then have been very well-controlled, have looked at a relatively wide range of individuals, and have used a reasonable range of soy protein products," says Lichtenstein. And they show that LDL cholesterol drops on average by only about 3 percent in soy eaters.

"That's a minimal effect at best and certainly not the very positive observation seen earlier," she adds. What's more, studies testing soy isoflavones found no impact on cholesterol. (Isoflavones are estrogen-like phytochemicals found in soybeans.)

That's why the American Heart Association jumped off the soy bandwagon in February. "The direct cardiovascular health benefit of soy protein or isoflavone supplements is minimal at best," its Nutrition Committee concluded. (2)

"Soy protein can still have a beneficial role in the diet if it's used judiciously to replace animal foods like full-fat milk or hamburgers," says Lichtenstein. "But the independent effect of soy protein on the risk of heart disease just isn't what it was once thought to be."

Is the FDA reconsidering its approval of the soy-may-reduce-heart-disease-risk health claim that appears on dozens of soy foods? "FDA generally doesn't comment on in-house re-evaluation of any particular health claim," says agency spokesperson Arthur Whitmore.

BREAST CANCER

Women in Asia, where soy is a staple, have lower rates of breast cancer than women in the United States or Europe. Do soy foods help protect women from breast cancer?

"The short answer is that we don't know," says Bruce Trock of the Brady Urological Institute and Kimmel Cancer Center at Johns Hopkins in Baltimore.

Earlier this year, Trock and his colleagues analyzed the 18 epidemiological studies on soy and breast cancer that were published from 1978 through 2004. (3) Most interviewed breast cancer patients to see if they reported having eaten more or less soy than similar cancer-free women.

"When you consider all the studies together, women who consumed larger amounts of soy foods had about a 15 percent lower risk of breast cancer than women who consumed smaller amounts," says Trock.

But he isn't enthusiastic about the results.

"The problem is that the studies are very inconsistent. The amounts of soy the women were eating varied tremendously from study to study, and so did how well the studies were carried out."

Another problem: a large intake (dose) did not have a greater effect (response) than a smaller intake. "You're more convinced that something's really going on if you see a dose-response, and we didn't see one," says Trock. "It's very difficult to come up with an explanation for why that would be the case."

Animal studies don't help.

First, "there are really no animal studies that use doses of soy as low as what you find being consumed in Western societies," says Trock. "The animal studies that show decreases in tumor growth use very high levels of soy or soy isoflavones that you can't get by diet alone."

To complicate matters, he adds, "the animal studies that use amounts of soy that are achievable through diet but that are higher than what's typical in Asian diets actually show an increase in tumor growth."

Still, Trock is reassured by the fact that women in Asian countries who eat a lot of soy have a low risk of breast cancer.

"So if a woman wants to add soy foods to her diet, we don't think there's anything wrong with that, if she eats regular soy foods in amounts consistent with what Asian women eat."

That would be something in the neighborhood of one serving a day--a cup of soy milk, a few ounces of tofu, or a handful of roasted soy nuts--says Trock.

"But the effects may not be comparable to an Asian diet," he points out, "because Asian women have probably been eating soy throughout much of their lives."

In animals, soy decreases breast tumor growth only if they start eating it at birth or around puberty, not in adulthood. A handful of studies in women also "suggests that soy needs to be consumed at an early age for it to be protective," says Trock.

But those studies--which ask breast cancer patients to remember what they ate decades before--are far from definitive.

"The sad thing is that when you look at all the data, there's no clear answer we can give," concedes Trock. "There's some evidence that soy foods are protective, but there's also a lot of caveats and inconsistencies."

One thing is certain, he adds: "There's no evidence that high doses of isoflavones in pills, powders, or other highly processed forms prevent breast cancer." High means "probably at least 60 mg a day, though there's no precise definition."

MENOPAUSE

Millions of women know all too well the discomfort of hot flashes, night sweats, and vaginal dryness that often confirms the arrival of menopause. And now it's clear that estrogen replacement therapy has a dark side--an increased risk of coronary heart disease, stroke, breast cancer, deep vein thrombosis, and dementia. That makes soy look even more tempting.

"We were hoping that the research would show that soy isoflavones work for hot flashes," says Heidi D. Nelson of Oregon Health and Science University's Oregon Evidence-based Practice Center in Portland.

"A lot of women tell me they love their soy, that it makes them feel better, but it's hard to sort that out from a placebo effect, since hot flashes come and go on their own in most women."

Nelson and her colleagues recently pooled the results of six studies looking at soy isoflavones and hot flashes, the most easily measured symptom of menopause. (4)

"None of the trials is considered good quality," she cautions. "They usually had small numbers of subjects, and the results were inconsistent."

Women who were given daily isoflavone pills for three to six months averaged about one less hot flash a day than placebo takers. Taking the pills for a shorter period of time, four to six weeks, had no major effect.

Women reported fewer hot flashes in only three of the five trials that used SO to 70 milligrams of isoflavones a day. "The sixth study tried 150 mg and it showed no effect, which is contrary to what you would expect," says Nelson.

Her conclusion: "We don't have strong evidence about soy's effectiveness because we're lacking high quality trials that compare isoflavones with a placebo. The results of current trials don't negate the whole idea that isoflavones may work to some degree, but we need bigger and better studies before recommending soy for hot flashes."

What can women do in the meantime?

"If they find that a little bit of soy or soy isoflavones makes a difference for them, it's hard to argue with that," says Nelson. "But there's no good evidence that taking more than 50 to 70 mg a day helps."

As for vaginal dryness, "soy has no effect," says women's health researcher Mindy Kurzer of the University of Minnesota. (5)

THE AGING BRAIN

In animals, estrogen stimulates the growth of brain cells and the connections among them. So it was reasonable for researchers to hope that estrogen--and estrogen-like soy isoflavones--might keep the brain sharp.

But the two biggest tests of estrogen and phytoestrogen pills on mental performance-both conducted in women--have come up empty.

In the Women's Health Initiative Memory Study, more than 2,100 women who took estrogen and progestin for four years scored no better on a test for cognitive impairment--and were at increased risk for dementia--compared with the roughly 2,200 women who took a placebo. (6)

And after 88 women took supplements of soy protein containing 99 mg of isoflavones every day for one year, they scored no better on tests of memory, verbal fluency, or attention than 87 women who were given a placebo. (7)

But that doesn't necessarily end the story, says Pauline Maki of the Center for Cognitive Medicine at the University of Illinois at Chicago. Maki has been a co-investigator on more than a dozen trials testing hormones on the brain.

"Those two studies were conducted in older women," she notes. The estrogen-progestin study recruited women 65 and older, while the soy isoflavone trial tested women 60 to 75 years old.

"It appears that the functionality of estrogen receptors changes with age," Maki explains. To have an impact, estrogen or phytoestrogens must link up to a receptor on a cell wall, like a key in a lock.

"There's some evidence that the estrogen receptors respond more positively in younger women," says Maki. "That might explain why they seem to have a different response to estrogen or phytoestrogens than older women."

Smaller trials of isoflavones in younger women suggest some benefit, she says.

When British researchers at King's College London gave 60 mg a day of soy isoflavones to 25 women aged 51 to 66 for six weeks, the women performed better on tests of mental flexibility and planning ability than women who got a placebo. (8)

Earlier, the same scientists gave 60 milligrams of isoflavones a day to 18 women aged 50 to 65. After 12 weeks, the women were better able to recall pictures, sustain attention, and plan a task than women who got a placebo. (9)

"Recent studies like these suggest that the effect of phytoestrogens is greater for those skills that rely on the frontal lobe of the brain--such as mental flexibility and planning--than it is for memory, which is located in another brain region called the hippocampus," says Maki.

That could mean that phytoestrogens might slow normal aging, not Alzheimer's or other dementias, Maki speculates.

"We all experience some declines in memory as we grow older," says Maki, "but normal aging produces a greater deterioration in the frontal lobes of the brain than in the hippocampus."

It's too early to know if the King's College results will be confirmed by larger, longer studies. But if they are, how early you start to eat soy foods may matter, at least if you're a woman.

"It's almost as though, after menopause, estrogen receptors become less sensitive," says Maki. "So the idea could be that if you initiate the use of estrogen or phytoestrogen early, the receptors are used to seeing it, don't turn off, and continue to respond."

BONES

In the Women's Health Initiative trial on 16,000 postmenopausal women, nine years of hormone replacement therapy raised the risk of heart attacks, strokes, blood clots, and breast cancer.

The only bright spot: the combination of estrogen and progestin built up bone density within three years and cut the risk of hip fractures by one-third over the course of the study. (10)

Could the plant estrogens in soy provide a kinder, gentler boost to bones?

"Only one study has looked at the relationship between soy protein intake and bone fractures," says soy expert Mark Messina of Loma Linda University in California. Messina consults with soy companies on the latest scientific research.

Among more than 24,000 postmenopausal Chinese women, those who ate the most soy foods had a 37 percent lower risk of suffering a broken bone than those who ate the least. (11)

That's encouraging, but the results might not apply to U.S. women, who consume more dairy foods--and are more prone to hip bone fractures--than Chinese women. What's more, something else about soy eaters might have cut the Chinese women's risk, Messina notes.

Researchers can look at some two dozen trials that tested soy isoflavones or soy protein (or a placebo) on bone density, bone formation, or bone breakdown.

"For every one of these studies that shows no benefit, two show something positive," says Messina. But, he adds, "the positive studies aren't necessarily consistent."

For example, one study may suggest that isoflavones increase the density of the spine, another the density of the hips.

"When you throw in that kind of inconsistency, along with the fact that the studies used different products in different types of people--and that most lasted a year or less, which is probably not long enough--it's not possible to draw definitive conclusions," says Messina.

Still, he adds, "the evidence is sufficiently encouraging for health professionals to recommend that women concerned about bone health consider including a source of isoflavones in their diet."

Clearer answers will have to await the results of three large, long-term, government-funded trials. One of them, a $4.5 million, two-year study at the Baylor College of Medicine in Houston, is giving 400 women a European product extracted from soy germ called SoyLife.

But eating SoyLife isn't like eating soy foods, says Messina. "It has a very different profile of isoflavones than the soybean itself and than the extract being tested in the two other government-funded trials." That's because SoyLife contains almost no genistein, which is one of soy's two major isoflavones.

What's more, SoyLife didn't boost bone density in a pilot study by Baylor researchers. "If the SoyLife study shows nothing, it won't be clear whether it's because isoflavones do nothing for bones or it's because you need the genistein," explains Messina.

The other two trials, funded by the National Institutes of Health, won't be completed until 2008 at the earliest. Both use Novasoy, a soybean extract that more closely matches the isoflavones in soybeans than SoyLife.

In any case, researchers aren't expecting to discover a miracle cure.

"The thinking is that soy isoflavones will not be a treatment for osteoporosis, since they're not as biologically potent as estrogen," says Lee Alekel, the Iowa State University researcher in charge of the two trials.

"The hope is that incorporating more soy foods into the diets of Americans will have a little bit of an effect on helping to prevent or ward off osteoporosis."

The Bottom Line

* Soy foods only lower LDL ("bad") cholesterol by an average of a few points. They can lower LDL more, though, if you eat them instead of meat, cheese, or other sources of bad (saturated or trans) fat.

* It's too early to say whether soy can reduce the risk of breast cancer, delay prostate cancer, relieve menopausal symptoms, strengthen bones, or boost brainpower.

* If you want to give soy a try, stick to no more than 50 to 70 milligrams of isoflavones a day. That's the amount in 1 or 2 (8 oz.) cups of soy milk or 6 to 9 oz. of tofu (see "On Isoflavones").

On Isoflavones

If soy can help prevent prostate cancer, hot flashes, osteoporosis, or brain aging, its estrogen-like isoflavones may be responsible. But some companies don't know (or won't disclose) the isoflavone levels in their soy foods. What's more, levels vary from brand to brand. And some of these numbers might be inflated because isoflavone numbers often include a sugar-like compound that accounts for about 40 percent of their weight.

With those limitations in mind, here's a snapshot of the isoflavone levels in some popular soy foods. To play it safe, stick to no more than about 50 to 70 milligrams of isoflavones a day.

Beverages                                             Isoflavones (mg)

Edensoy Organic Soymilk (8 oz.) (1)                                 50
Starbucks Caffe Latte, soy (grande-16 oz.)                       30-55
Silk Soymilk (8 oz.) (1)                                         20-35
8th Continent Soymilk (8 oz.) (1)                                 5-20

Soybeans, Tofu, & Tempeh

Edamame (parboiled green soybeans),                                 35
  shelled (1/2 cup)
Lightlife Tempeh (4 oz.) (1)                                        35
Soy nuts, dry roasted (1 oz.-1/4 cup)                               35
Tofu, soft or firm (3 oz.-1/3 cup) (1)                           15-30

Veggie Meat

Amy's Texas Veggie Burger (1)                                       10
Morningstar Farms Prime Grillers (1)                              5-10
Burger King BK Veggie Burger (1)                                     5
Morningstar Farms Bacon Strips (2)                         less than 5
Morningstar Farms Sausage Links (2)                        less than 5
  or Patties (1)
Morningstar Farms America's Original                                 0
  Veggie Dogs (1)

Soy Dairy

Stonyfield Farm O'Soy Yogurt (6 oz.) (1)                            30
Silk Cultured Soy Yogurt (6 oz.) (1)                             20-30
Soy Delicious Purely Decadent frozen                                 5
  dessert (1/2 cup) (1)
Tofutti frozen dessert (1/2 cup) (1)                                 5
Organic So Delicious frozen dessert (1/2 cup) (1)          less than 5
Veggie Cheese Alternative (1 slice-o.6 oz.) (1)            less than 5

Cereals

Kashi GOLEAN (1 cup)                                                25
Barbara's Bakery Organic Soy Essence (3/4 cup)                      20
Nature's Path Optimum Power (1 cup)                                 15
Kashi GOLEAN Crunch (1 cup)                                          5
Quaker Oatmeal Nutrition for Women (1 packet) (1)                    5

(1) Average or range of the entire line.

Sources: companies, U.S. Department of Agriculture.

Men and Isoflavones

Men's bodies normally produce small quantities of estrogen. How do they respond to the larger amounts of plant estrogens in soy isoflavones?

Prostate Cancer

U.S. men are about three times more likely to die of prostate cancer than Japanese men, and 16 times more likely to die of prostate cancer than Chinese men.

That disparity has led researchers to wonder whether soy foods can slow or prevent the disease.

"Japanese and Chinese studies suggest that there may be a role for soy in preventing prostate cancer," says Omer Kucuk of the Barbara Ann Karmanos Cancer Institute at Wayne State University in Detroit.

According to an industry-funded meta-analysis of eight studies in the United States, Canada, China, and Japan, men who consumed soy foods more often or in greater amounts had a 30 percent lower risk of being diagnosed with prostate cancer than men who ate soy less frequently or in smaller amounts. (1) The soy eaters ate from about two to eight ounces of soy foods a day.

"But it's still only a hypothesis," Kucuk cautions. Among the problems with the meta-analysis: six of the eight studies relied on the ability of men with prostate cancer (or their wives) to remember what they had been eating years earlier.

Studies that give soy to men with prostate cancer also have shortcomings.

For example, when Kucuk and his coworkers gave men with prostate cancer 200 milligrams a day of soy isoflavone pills for six months, their levels of prostate-specific antigen (PSA) rose less or not at all--compared to before they started on the soy.

"But we studied only a small number of patients," he notes. And since the researchers didn't compare isoflavones to a placebo, it's hard to know whether PSA levels might have remained stable without soy. (Rising PSA levels may signal a growing cancer, falling levels a shrinking cancer.)

Kucuk's study is one of 11 that have tested soy isoflavones on PSA levels.

"Three were in healthy men and had no effect," says soy expert Mark Messina of Loma Linda University in California. "In the eight studies on cancer patients, four showed some modest changes and four didn't."

The bottom line: "All the evidence we have is preliminary," says Kucuk. "Based on that, we can't make any definitive recommendations yet about consuming soy foods or supplements to avoid prostate cancer."

The Brain

"We don't have evidence that the potential benefits of soy isoflavones apply to men," cautions Pauline Maki of the University of Illinois at Chicago. Nearly all of the participants in studies testing isoflavones on the brain have been women.

"There was one observational study six years ago showing that middle-aged men in Hawaii who ate the most tofu were also the most likely to suffer cognitive impairment and brain atrophy later in life," Maki notes. (2)

"At first, people thought that was kind of silly, but now we're not so sure because of a recent impressive study in male monkeys who ate a high-soy diet."

Researchers fed 44 adult male monkeys diets with either (isoflavone-free) dairy protein or soy protein that supplied the human equivalent of either 60 or 129 milligrams of isoflavones a day. (3)

"During the 15 months of the study, the monkeys eating the most soy isoflavones were more aggressive and less social than the other monkeys," says investigator Jay Kaplan, director of the Department of Comparative Medicine at Wake Forest University in Winston-Salem, North Carolina.

They were more likely to grab, slap, bite, and chase the other males. "And they spent more time beyond the touching distance of the other animals, so they were less social in that sense."

Female monkeys showed no changes in behavior when they ate a high-soy diet, says Kaplan. What's more, among the soy-eating monkeys, the males were more likely to be insulin resistant than the females. And while soy lowered cholesterol levels substantially in the females, it wasn't as effective in the males.

"A number of things lead you to suspect that males are doing something different with isoflavones than females," says Kaplan. "And consuming large amounts of isoflavones, particularly in supplement form, may have effects that we don't yet know about and that could as easily be adverse as beneficial."

Maki agrees. "There should be some caution about recommending large amounts of isoflavones for men."

What's large? The monkeys who were given the human equivalent of 60 milligrams of isoflavones a day showed no signs of harmful effects, says Kaplan. That amount which you'd get in around one or two cups of soy milk or six to nine ounces of tofu (see "On Isoflavones," p. 5)--is at the upper end of what soy-eating Asians get, he notes.

"We just don't have enough data in men about the safety of taking more than that on a regular basis."

(1) Int. J. Cancer 117: 667, 2005.

(2) J. Am. Coll. Nutr. 19: 242, 2000.

(3) Horm. Behav. 45: 278, 2004.

(1) N. Engl. J. Med. 333: 276, 1995.

(2) Circulation 113: 1034, 2006.

(3) J. Natl. Cancer Inst. 98: 459, 2006.

(4) JAMA 295: 2057. 2006.

(5) Fertil. Steril. 83: 137, 2005.

(6) JAMA 289: 2663, 2003.

(7) JAMA 292: 65, 2004.

(8) Menopause 12: 193. 2005.

(9) Pharmacol. Biochem. Behav. 75: 721, 2003.

(10) JAMA 290: 1729, 2003.

(11) Arch. Intern. Med 165: 1890. 2005.