点击显示 收起
Nutrition Matters Inc
439 Calhoun Street
Port Townsend, WA 98368
E-mail: markm{at}olympus.net
Dear Sir:
The article by Keinan-Boker et al (1) in a recent issue of the Journal failed to find a relation between isoflavone intake and breast cancer risk in a large cohort of Dutch women. The observed lack of effect should not be surprising. In my opinion, there is little reason to expect typical Western isoflavone intakes to affect health. Although a few Western epidemiologic studies have noted associations between the intake of soy, isoflavones, or both and a variety of health outcomes (2-5), in all probability, these associations are spurious in nature. Furthermore, accurately quantifying isoflavone intake when exposure is so low is extremely difficult because soy protein is used as a functional ingredient (eg, to promote water retention or as a bleaching agent) in many Western foods. Isoflavone intake from these sources may be equal to or greater than that from foods actually identified as soy foods.
In the West, soy foods are consumed primarily by vegetarians and health-conscious individuals; thus, among Westerners, those who consume high amounts of soy are likely to lead very different lifestyles than do nonconsumers. Controlling for so many potentially confounding variables is difficult. In contrast, in Asia, the distinction between those who do and those who do not consume soy, especially among older persons, is likely to be much less pronounced because soy is a traditional staple.
In the study by Keinan-Boker et al (1), median isoflavone intake was just 0.67 mg/d; even among women in the fourth quartile, median isoflavone intake was only 0.77 mg/d. Such low intakes produce negligible serum isoflavone concentrations. As noted by the authors, the isoflavone intake of Dutch women is considerably less than the 50 mg/d typically consumed by Japanese women.
Few dose-response trials have been conducted in humans, so it is difficult to estimate with confidence the amount of isoflavones needed to exert physiologic effects in vivo. However, most clinical trials have used 40–90 mg isoflavones/d. Amounts within this range have produced several reported effects; for example, in postmenopausal women, 40 mg/d enhanced systemic arterial compliance (6), 50 mg/d reduced oxidative DNA damage (7), and 56 mg/d reduced LDL-cholesterol oxidation (8).
Generally, studies examining the effects of isoflavones on bone mineral density have used doses toward the higher end of the range. Two such studies found that moderate (40 and 56 mg/d) doses of isoflavones were ineffective, whereas higher (80 and 90 mg/d) doses were efficacious (9, 10). In contrast, however, a recently published 1-y study found that 26 mg isoflavones/d nonsignificantly decreased spinal bone loss and significantly increased markers of bone formation in postmenopausal women (11).
The amount of isoflavones needed to exert beneficial effects may vary according to the disease in question; thus, one cannot necessarily assume that the level of isoflavone exposure needed for skeletal and coronary benefits is the same as that needed for the prevention of breast cancer. Furthermore, it is possible that the amount of isoflavones needed for health benefits when consumed over the course of a lifetime is lower than that needed to produce benefits in short-term clinical trials. Some evidence suggests this is the case for bone health. Several Asian epidemiologic studies found soy-isoflavone intake to be associated with higher bone mineral density (12-15); intakes in these studies were lower than the doses used in the clinical trials.
Although the data overall are mixed, several Asian epidemiologic studies found the intake of one or more soy foods to be inversely related to breast cancer risk (16-18). However, few epidemiologic studies reported total soy intake. Thus, it is difficult to gain insight about optimal isoflavone intakes from these studies. In a recently published Japanese study that did show that total isoflavone intake (based on the intake of miso, tofu, and natto) was associated with a reduced risk of postmenopausal breast cancer, mean daily genistein intakes (which represent about one-half of total isoflavone intake) for the first through fourth quartiles were 6.9, 13.0, 20.0, and 25.3 mg, respectively (16).
Intriguingly, a Chinese epidemiologic study found soy intake to be inversely associated with breast cancer risk but failed to observe a monotonic dose-response relation (19). The reduction in risk that occurred between the first and second intake deciles was similar to that between the first and tenth intake deciles. The isoflavone intake cutoffs for the first and second deciles can be estimated to be 11 and 16 mg/d, respectively. These findings allow for the possibility that relatively low isoflavone intakes are protective against breast cancer. But even the first decile isoflavone intake cutoff in this study far exceeds Western isoflavone intake.
Finally, a particularly intriguing but still speculative hypothesis regarding breast cancer is that early isoflavone exposure is protective against later development of this disease (18). In the epidemiologic study that provides the most support for this hypothesis, however, isoflavone intake among women who were in the fourth intake quartile for teenage soy consumption can be estimated to be >40 mg/d.
In conclusion, extreme caution is needed when making pronouncements about the possible health effects—desirable or undesirable—of soy intake on the basis of Western epidemiologic studies involving non-Asians.
REFERENCES