Additionally, for a subset of 255 participants whose mothers were alive and living in the US, the mothers were asked about their daughter's early childhood exposures.

By comparing the highest and lowest soy intake values for soy-based foods such as tofu, miso and natto, Korde and co-workers calculated that women with the highest soy intake during childhood (ages 5 to 11) had a 58 per cent lower risk of breast cancer as adults as the women with the lowest soy intake as children.

The corresponding reductions for adolescent and adult intake were about 25 per cent, they said.

The underlying mechanism is not known, said the researchers, but they hypothesised that the oestrogenic effects of soy isoflavones cause changes in breast tissue during childhood that may decrease sensitivity to carcinogens later in life. A similar protective effect has been found in studies of overweight girls, perhaps because fat tissue also secretes oestrogens, said Korde.

"Hormonal exposures in adulthood, such as use of oestrogen and progesterone replacement therapy, are established breast cancer risk factors. However, a growing body of evidence suggests that hormonally related exposures early in life may also modify susceptibility to breast cancer," said Korde.

Regina Ziegler, senior investigator on the study, added a note of caution, however, and said that it would be premature to recommend changes in childhood diet.

"This is the first study to evaluate childhood soy intake and subsequent breast cancer risk, and this one result is not enough for a public health recommendation," she said. "The findings need to be replicated through additional research."

Indeed, the study does have several limitations, including asking women to evaluate and quantify adolescent dietary intakes, as well as asking mothers about a daughter's dietary intake during childhood, the accuracy of both are dependent on the recall ability of the interviewees.

The research does fit with a growing body of research that supports the potential cancer-protecting properties of soy.

A recent animal study published in the journal Cancer Research (Vol. 66, Issue 2) reported that high dietary intake of soy protected against breast cancer in postmenopausal monkeys.

This supports another study from the University of Ulster that focussed on the inverse link between soy and breast cancer. In this study, funded by the EU's “Quality of Life and Management of Living Resources” project, soy isoflavones were reported to inhibit breast cancer cell invasion in vitro.

NutraIngredients.com has not seen the full data presented by the National Cancer Institute scientists.

But one area in which the evidence is controversial is the fatty acid's role in reducing the risk of cancer.

The new research, by scientists at Harvard School of Public Health and Brigham and Women's Hospital, analysed data from 22,071 participants in the Physicians' Health Study (PHS), a randomized, double blind, placebo-controlled clinical trial primarily designed to investigate the effect of aspirin and beta-carotene supplements on development of cancer and cardiovascular disease.

The participants also filled out a one-time food questionnaire 12 months after starting the study in order to determine if habitual dietary fish consumption (tuna fish, dark meat fish, a general fish category, and shellfish) had a different effect on men who received aspirin for five years compared to men who weren't randomized to use aspirin.

Lead author of the study, Megan Phillips, said that almost 10 per cent of the men ate fish less than once a week, 31 per cent ate it less than two times a week, 48 per cent ate fish less than five times a week, and about 11 per cent ate it five times or more a week.

After an average of 19.4 years of follow-up, they calculated that five or more servings of fish a week was associated with a 40 per cent reduction in colorectal cancer risk, compared to men who ate fish less than once a week.

The relative risk of eating fish 2-5 times a week was 20 per cent lower, and 13 per cent lower among participants who ate fish less than twice a week.

The same protective effects were observed whether the men were in the aspirin or placebo groups.

The underlying mechanism was not studied by the researchers, but they suggested that the omega-3 fatty acids content of the fish inhibit the cyclooxygenase-2 (COX-2) enzyme that plays a role in inflammatory responses and has been linked to cancer development.

"We already know that eating fish can reduce the risk of sudden cardiac death, and this might provide another reason to add fish to your diet," said Phillips.

However, Phillips said that while the results are promising, the study does have notable limitations, including being based on the assumptions that dietary fish consumption was constant, since the relationships were calculated from only one food frequency questionnaire.

She also noted that men who consumed more fish may also have a healthier lifestyle perhaps including better cancer screening.

The researchers called for additional study, and said that that a definitive answer might require a randomised trial.

NutraIngredients.com has not seen the full data presented by the Boston-based scientists.

The potential protective benefits of omega-3 fatty acids against cancer was the subject of a review, published in January 2006 in the Journal of the American Medical Association (Vol. 295, pp. 403-415).

Researchers scrutinized 38 studies published between 1966 and October 2005 that investigated the purported link between omega-3 and different types of cancer and met certain criteria. The studies had to describe the effects of omega-3 fatty acid consumption on tumour incidence, be prospective cohort in design, and be conducted on a human population.

Despite finding 65 estimates of association across 20 different cohorts for 11 different types of cancer and six different ways of assessing omega-3 consumption, only eight of these were found to be statistically significant.

Three studies showed decreased risk of breast cancer with omega-3 consumption, one for colorectal cancer, one for lung cancer and one for prostate cancer. But for each type there were also significant associations for decreased risk, and more estimates that did not identify any association.

Indeed, commenting at the time, Josephine Querido, science information officer at Cancer Research UK, said: "The jury is still out as to whether eating more omega-3 fatty acids will reduce your risk of developing cancer.”

A study published in the June issue of the Journal of the National Cancer Institute (Vol. 97, no 12) concluded from data from 1 million participants in the European Prospective Investigation into Cancer and Nutrition (EPIC) trial that people eating less than 14g of fish a day were 40 per cent more likely to develop colorectal cancer than those eating more than 50g per day.

However the researchers were unable to differentiate between fatty fish, which contains the majority of omega-3 fatty acids, and other fish.