Note of the meeting held at the Food Standards Agency, on Monday 25 September 2000, Room 137B, Skipton House, Elephant and Castle, London.

Present

CONTENTS

ITEM 1: APOLOGIES FOR ABSENCE

1. Apologies were received from Professor I Kimber, Dr M Joffe and Dr P Hindmarsh.
   
   

2. The Chairman welcomed all those in attendance. Members were invited to comment on the minutes of June 2000 meeting (PEG/MIN/2).
   
   
3. Members suggested that the wording in line 5 of paragraph 12 be amended to read "did not support the hypothesis that phytoestrogens were responsible". Members asked if the COT would reconsider this study again if more data became available. The Chairman replied that the COT would review any new significant data.
   
   
4. Members commented that paragraph 24 should be amended to read that all splice variants of the beta-receptor had been identified at the protein level but that the biological relevance of a number of the splice variants had yet to be determined.
   
   
5. Members indicated that reference should be made in paragraph 26 to a retrospective study of adults fed soya formulae as infants which is currently in progress at the Institute of Child Health. Members added that it was possible that similar studies were being carried out in Japan. The Secretariat indicated that a retrospective study from the US had been submitted in response to the request for submissions of evidence but had not received, or were aware of, any studies of this type from Japan.

ITEM 3: MATTERS ARISING

6. The Chairman reminded Members of the Agency’s request that any commercial or other interests in agenda items be declared. Members were advised that all discussion papers would be placed on the Food Standards Agency website. Members commented that it was important to highlight that the papers were working documents and did not necessarily represent the final view of the Group. The secretariat informed members that a disclaimer to this effect was added as a footnote at the bottom of each page.
   
   
   

7. Professor Bingham and Professor Rowland introduced PEG/2000/16 and Annex 1 of PEG/2000/16.
   
   

   Menopause

   
   
8. It was explained that the symbols used in Table 2 (PEG/2000/16, Annex 1) summarising effects of phytoestrogen-containing foods on menopausal symptoms represent: (+) a significant correlation, (+/-) a weak association and (-) non-significant correlation. A recently published abstract of an intervention study was tabled for discussion. The study showed that larger reductions in the number of hot flushes were observed following placebo rather than the soya-based treatment. Members commented that the subjects selected for inclusion in this study appeared to suffer severe symptoms and it was possible that soya-based supplements may not be effective in reducing menopausal symptoms of that severity.
   
   
9. Members commented that although a number of studies suggested that phytoestrogens appeared to produce positive effects, often these were not statistically significant due to the strong placebo response. Members felt that most of the published data was unconvincing and suggested that the significant variability in response observed between studies may be due to the differing composition of supplements used in the study.
   
   
10. Members concluded that the supporting evidence was often inconsistent and did not conclusively show that either soya-based or phytoestrogens have a significant therapeutic benefit on menopausal symptoms. Members commented that a meta-analysis of the published data may help clarify the issue but noted that this type of analysis had not been carried out.
    

    Osteoporosis

    
    
11. Members agreed that studies examining the effects of phytoestrogens on bone density in rodents suggested beneficial effects. However no beneficial effects were evident when a similar type of study was conducted in primates. Members noted that relatively low doses of phytoestrogens were used in the primate study and the lack of effect may have been due to insufficient dosage.
    
    
12. A year-long study showing reduction in bone loss in the spine but not the hip of pre- and peri- but not post-menopausal women following isoflavone treatment was discussed. Members noted that different effects occurred at different sites in bone. It was explained that these differential effects could be due to the fact that there is greater bone turnover in the spine compared to the hip. Members also noted that isoflavone supplementation did not significantly reduce bone loss in post-menopausal women.
    
    
13. Although a few of the studies showed beneficial effects, no effects on bone mineral density was observed in others. Members commented that although a few of the human studies looked promising, overall the data remained equivocal. Members added that only a small number of short-term human studies had been published and long-term large studies were required to provide conclusive results.
    
    
14. Members discussed the information presented in Annex 2 of PEG/2000/16 on ipriflavone, a synthetic isoflavone marketed for treatment of osteoporosis. It has been claimed that this compound does not interact with estrogen receptors and would therefore, not produce the negative effects associated with the breast cancer drug, tamoxifen. Members commented that ipriflavone metabolites had been shown to possess estrogenic activity in vitro. The secretariat added that interaction with the estrogen receptor alpha, not beta had been assessed. It was therefore incorrect to assume that ipriflavone did not interact with estrogen receptors.
    
    Cardiovascular Disease
    
    
15. Members were informed that a considerable body of evidence suggested that the consumption of soya can have beneficial effects on LDL and total cholesterol levels. Although attempts had been made to extrapolate these effects to isoflavones in soya, purified isoflavones did not appear to produce the same beneficial effects and it is probable that the beneficial effects are due to constituents in soya other than isoflavones. Members concluded that while soya may have a protective effect against cardiovascular disease, there is little evidence to suggest that this effect is associated with the isoflavone component of soya. This is also the conclusion reached by the US FDA in a recent review (Annex 3 PEG/2000/16). Members agreed this conclusion should be clearly stated in the final report.
    
    
16. Members asked if there was any evidence to suggest that lignans have beneficial cardiovascular effects. It was noted that very little work has been carried out on lignans. Members added that it would be very difficult, if not impossible to design and conduct epidemiology studies on lignans due to the prevalence of these compounds in fruit and vegetables.
    
    Breast Cancer
    
    
17. 17. There was general agreement that the quality of studies examining the effect of isoflavones on breast cancer was poor. Confounding factors, such as the phytoestrogen content of foods other than soya had not been accounted for. Members noted that a small number of studies had examined the association between urinary isoflavone excretion and breast cancer. Members commented that the results of these studies are inconclusive and added that urinary excretion is not a good biomarker of phytoestrogen intake due to the significant inter-individual variability in the metabolism of these compounds.
    
    
18. Members were informed that a meta-analysis (as yet unpublished) using 10 studies indicated that soy intake was not significantly correlated to reduced breast cancer incidence in Asian women.
    
    Prostate Cancer
    
    
19. Epidemiological studies have shown a negative association between prostate cancer and soy consumption and also between lignan excretion. Members commented it was unclear if this is due to phytoestrogen intake per se or a reflection of other constituents in the "whole food". Members were informed that a FSA funded prospective study examining prostate cancer and diet was currently in progress, however the trial had been delayed due to problems recruiting adequate numbers of volunteers and the results were not available for review by the Working Group.
    
    
20. Members commented that it was important that studies investigate and distinguish the effect of phytoestrogen treatment on tumour prevention and regression. They noted that although the rate of carcinoma-situ is similar between Japanese and Western men, the rate and incidence of progression to full clinical cancer is significantly lower in Japanese compared to Western populations. A polymorphism in the estrogen receptor gene has been associated with the progression of prostate cancer. This indicates that genetic background plays an important role in this disease, and it’s important that ethnic background is taken into consideration during the interpretation of these data.
    
    Lung Cancer
    
    
21. Members were informed that all of the epidemiological studies had been carried out in Chinese and Japanese populations and the results were inconclusive. Members noted that confounding factors such as smoking had not been taken into account in these studies and no data were available for Western populations.
    
    Colon Cancer
    
    
    
22. Members concluded there was no firm evidence to suggest that phytoestrogens have a beneficial effect on colon cancer. However the group noted a study which reported that the incidence of colon cancer was lower in women receiving HRT. Members mentioned the conflicting data from rodents showing that genistein increased the tumour incidence in rats.
    
    Stomach Cancer
    
    
23. No studies have been carried out in Western populations, however studies in Far-Eastern populations show higher rates of stomach cancer which have been associated with the intake of fermented soya products, such as miso. Members felt these studies did not clearly implicate phytoestrogens and added that fermented soy products are high in salt, which may contribute to the higher incidence of stomach cancer.
    
    Endometrial & Ovarian Cancer
    
    
24. Members agreed that there is little supporting evidence to suggest that phytoestrogens have beneficial effects on the incidence of endometrial or ovarian cancers. The single study reported to date showed no differences in the rates of these cancers in Eastern and Western populations.
    
    
    
25. The Chairman invited members to comment on conditions not covered where phytoestrogens may exert beneficial effects.
    
    
26. Alzheimer’s Disease: There are no reports on the beneficial effects of phytoestrogens and Alzheimer’s but members noted that there have been reports linking tofu consumption with adverse effects. The Secretariat explained that adverse effects of phytoestrogens would be reviewed at the next meeting.
    
    
27. Breast enhancement: Supplements containing phytoestrogens from hops are being marketed as a "natural" alternative to surgical breast enhancement. The Secretariat indicated that FSA funded research showed that although hop phytoestrogens had been shown to be estrogenic in vitro, these phytoestogens were not estrogenic when tested in a rodent model. The Secretariat replied that no submissions of evidence had been received to support these claims. Members acknowledged there was no published scientific evidence to support these claims. They added that if these supplements were shown to increase breast size by the induction of cell proliferation, they would have concerns about the potential health implications of these supplements. Members were informed a trial was currently in progress to investigate if isoflavones or soy had any effects on mammographic density. The results of this study would be available for review at the Phytoestrogen Workshop in November. The Chairman suggested that a statement should be made which stressed that no scientific evidence was available to support these claims.
    
    
28. Antioxidant effects: Members were informed that phytoestrogen dietary supplements were being sold with claims for beneficial antioxidant activity. The Secretariat suggested that the results from an in vitro study (PEG/2000/17) indicated that levels of isoflavones necessary to give antioxidant activity were in excess of the levels expected in a phytoestrogen-rich diet.
    

    Submissions of Evidence
    

29. The Chairman invited Members to comment on the submissions of evidence received by the group (PEG/2000/18). The Chairman reminded Members that submissions received by the Group would be acknowledged in the final report. However if the submission had been received in confidence, the detail included could not be released into the public domain. The Chairman stated that all submissions would be fully reviewed by the group who would assess the validity of the data and reach their own conclusions based on the data.
    
    Final report
    
    
30. Members commented that much of the data on beneficial effects had been obtained from abstracts and that these did not provide sufficient information to determine how the study had been conducted. The Chairman informed Members that only data that had been peer reviewed should appear in the final report. In addition, important studies published after preparation of the discussion papers would also be included.
    
    
31. Members suggested that the final report include a caveat stating that human studies conducted on adults may not accurately represent the response in children and that inter-individual differences in metabolism could make other sub-groups more susceptible. In this respect, evidence was available that bone mineral density in adolescent boys correlated with polymorphisms in the estrogen receptor. Thus, genetic alleles may give rise to other susceptible grou
    
    

ITEM 5: DEFINITIONS OF PHYTOESTROGEN & ISOFLAVONE

32. Members were invited to comment on the suggested definitions for ‘phytoestrogen’ and ‘isoflavone’ (PEG/2000/19). Members agreed that ‘in the body’ should be replaced by ‘in vivo’ and that the definition of phytoestrogen should be restricted to interaction with estrogen receptors. Members suggested that the isoflavone definition be restricted to the chemical name and class. Members agreed to consider the definitions for discussion at the January 2001 meeting.
    
    

ITEM 6: PAPERS FOR INFORMATION

33. Members were informed that articles provided by the FSA press office were the source of the papers for information. Members commented that the articles were informative and useful.
    

34. Members were informed that the next meeting will be held on the 14^th November 2000 in Cambridge.
    
    
    top of the page