|
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
 |
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
Present:
ITEM 1: APOLOGIES FOR ABSENCE/ANNOUNCEMENTS 2. Apologies for absence were received from the members Professor A Boobis, Dr P Carthew, Professor D Phillips and the assessor Dr H Stemplewski (MHRA). Announcements 3. The Chairman thanked Professor Phillips for acting as the Chair at the 6 November 2003 meeting during his absence. The Chairman welcomed Mr Ken Okono-Mensah, Dr Sarah Bull and Dr Karin Fletcher from the DH Toxicology Unit and Alison Gowers and Dr Heather Walton from DH who would be attending for item 8. 4. Members were reminded of the need to make any relevant declaration of interests before the discussion of items. ITEM 2: MINUTES OF THE MEETING OF 6 NOVEMBER 2003 5. The minutes were agreed subject to minor editorial changes. ITEM 3: MATTERS ARISING NOT COVERED BY LATER AGENDA ITEMS
6. At the COC meeting on the 6 November 2003, the Committee had considered whether to revise its opinion on the carcinogenicity of 1,3-dichloropropanol (1,3-DCP) in the light of new advice from the COM. This was in response to new data on the mutagenicity of 1,3-DCP. The COM had considered that negative results in an in-vivo bone marrow assay for clastogenicity and in an in-vivo liver assay had provided adequate re-assurance that the mutagenic activity seen in vitro was not expressed in vivo. Background to the COC consideration was given in paper CC/03/42. 7. The COC at its meeting on 6 November 2003 considered the possible non-genotoxic mechanisms whereby 1,3-DCP could induce the tumours seen in the animal bioassays. They reaffirmed their earlier view that the tumours of the kidney and thyroid could have been induced as secondary events to sustained cell proliferation. Regarding the liver tumours, members agreed that there was evidence of a hepatotoxic effect at doses below those producing a significant increase in combined hepatocellular adenoma and carcinoma. The COC that agreed in view of this, and the fact that negative results were obtained in the liver UDS assay, it could be concluded that the liver tumours were induced by a non-genotoxic mechanism. 8. However, the Committee could not discount a possible genotoxic mechanism for the increased incidence of tumours seen in the tongue of rats. These were seen at the top dose, which exceeded the maximum tolerated dose (MTD). Members considered that the tumourigenicity in the tongue might also have been due to sustained irritancy following administration of 1,3-DCP via drinking water. The tongue was not routinely evaluated histologically in bioassays at that time (1986) and there were no data available on irritancy at that site. It was suggested that potential bacterial metabolism of 1,3-DCP to epichlorohydrin might have been responsible for the tongue tumours. Members considered that further genotoxic investigation of 1,3-DCP in the tongue in rats was needed and suggested 32P-post labelling studies might be valuable. 9. Paper CC/04/4 presented a review of the carcinogenicity of epichlorohydrin with particular reference to the oral cavity in rats. The carcinogenicity of epichlorohydrin was last reviewed by IARC in 1999. It was concluded that there was sufficient evidence in experimental animals for the carcinogenicity of epichlorohydrin. The overall evaluation was Group 2A (probably carcinogenic to humans). Results from two bioassays in rats were available using the oral route. In one drinking water study the only comment regarding cancers of the oral cavity was that squamous cell carcinomas of the oral cavity were seen in 2/12 (16.7%) of the surviving rats from the high dose group. There was no mention of tumours of the tongue. The other oral study involved gavage administration and there was no increase in tumours in the oral cavity or tongue. In an inhalation study in rats the only treatment related increase in tumours was in the nasal cavity (site of initial contact). The Committee was asked whether these data on epichlorohydrin warranted a reconsideration of the advice on the need for a 32P-post labelling study in the tongue with 1,3-DCP. Additionally, the COC was asked whether they could recommend alternative approaches to investigate if epichlorohydrin was formed in the oral cavity and the potential mechanism of the tumourigenicity of 1,3-DCP to the tongue in rats. 10. Members agreed that these new data did not help evaluate the mechanism
of carcinogenicity of 1,3-DCP in the tongue of rats. The COC considered
that chronic irritation by 1,3-DCP administered in drinking water was
a plausible hypothesis for the induction of tumours in the tongue, but
could not discount a possible genotoxic mechanism for these tumours. Members
felt that investigation of the potential metabolism of 1,3-DCP to epichlorohydrin
was not warranted at this juncture. Thus, the COC reiterated its previous
advice that information on contact-irritancy, cell proliferation and formation
of adducts using 32P-post labelling in animals treated with
suitably high doses of 1,3-DCP was required.
11. The minutes of this item now published - see paragraph 54. ITEM 4: OLFACTORY NEUROBLASTOMA: EVIDENCE FOR ELEVATED INCIDENCE IN DENTAL WORKERS. 1ST DRAFT STATEMENT (CC/04/1) 12. The COC was reminded that the Chief Medical Officer for England Sir Liam Donaldson had previously asked the Committee for a view on research by Professor Valerie Lund (from the Institute of Laryngology and Otology, University College London). The COC concluded that the finding by Professor Lund of 4 dentists/dental nurses with olfactory neuroblastoma (ONB) by the Institute of Laryngology and otolaryngology was likely to be a statistically significant association. Additional epidemiological data were needed to substantiate this observation. No definite conclusions on the potential association between dentists/dental nurses and ONB could be reached at that time. 13. The Committee had agreed at their November meeting that it might be feasible to collate information on individual cases from centres of excellence and pathology departments from the UK, Europe and elsewhere. Professor Lund had offered to provide as much information on this aspect as possible. European Chief Dentists had been alerted to the COC discussion and the British Dental Association had indicated a willingness to provide technical help. 14. Members were informed that Professor Lund had visited the University of Virginia in the USA, a centre of excellence on head and neck cancers, to see whether any further useful information could be obtained on the possible association between ONB and dentists/dental nurses. There was a possibility of future collaborative work. The COC would be informed of any significant findings in due course. 15. The draft COC statement would be amended in light of members' comments
and forwarded to Sir Liam Donaldson for information. ITEM 5: PROSTATE CANCER: ADDITIONAL DATA AND 1ST DRAFT STATEMENT (CC/04/2) 16. The COC considered a review from the DH Toxicology Unit at the November 2003 meeting. Members agreed that a first draft statement could be prepared but asked for some additional information on a number of aspects. This information was presented in paper CC/04/2. 17. The Committee considered data which reported evidence that low concentrations of cadmium chloride could interact with the androgen receptor in vitro and could also produce an androgenic response (eg increased prostate weight) in vivo in rats given relatively low intraperitoneal doses. This suggested a plausible mechanism by which cadmium might be associated with prostate tumours in rats. However, members agreed that evidence from occupational studies regarding prostate cancer showed no association with exposure to cadmium in the majority of studies including relatively large cohort studies. The COC noted the possibility that cadmium might induce androgen imbalance and could potentially have a role in prostate cancer. Thus members agreed with the suggestion in the 1998 HSE review that this possibility should be monitored and any relevant information considered in the future. 18. The Committee was provided with additional epidemiological data from a review prepared for the HSE (1998) of occupation and prostate cancer. The review noted that a number of studies had indicated an elevated risk of prostate cancer for farmers and farm related workers but other studies did not confirm this observation. The review prepared for the HSE concluded that farmers/farm workers were not clearly identified as occupational groups with an elevated risk for prostate cancer but continued monitoring was appropriate. The review prepared for the HSE also suggested that exposure to pesticides might be a potential risk factor. Members agreed with these conclusions. Members were aware that the rubber industry had queried, via HSE, the conclusions reached on prostate cancer and the rubber industry in the DH Toxicology Unit paper, which had been published on the COC internet site. The Committee agreed to review available literature at the next meeting. 19. The COC considered a systematic review on occupational related pesticide exposure and cancer (Van Maele-Fabry G and Willems JL Occupational and Environmental Medicne vol 60, p634 - 642, 2003). This meta-analysis produced an overall relative risk of 1.13 (95% C.I. 1.04-1.22) for prostate cancer in workers exposed to pesticides in pesticide related occupations (from 11 cohort, 4 Proportional Mortality Ratio, and 7 case control studies). Members noted that for all studies excluding proportional mortality ratio the relative risk was 1.09 (95% C. I. 1.00 - 1.19) and that the risk estimates were for all farming occupations and not just for pesticide applicators. North American studies tended to show higher prostate cancer risk than European studies. The Committee considered that pesticide exposure was likely to be lower in Europe. The separate risk ratio for pesticide applicators (relative risk 1.64 95% C.I. 1.23 - 2.38) was greater than the overall risk ratio for all studies. 20. A study by Morrison et al American Journal of Epidemiology vol 137, p270-280, 1993 on Farming and prostate cancer mortality found a relative risk of 2.23 (95% C.I. 1.30-3.84) associated with 250 or more acres sprayed with herbicides. Members noted that more deprived workers might carry out most of the spraying and therefore would be more exposed to herbicides. This study suggested a positive dose response relationship. The risk associated with larger areas sprayed with herbicides was greater than the risk associated with smaller areas sprayed with herbicides. The COC agreed that these two studies provided more evidence for prostate cancer risk for farm workers most exposed to pesticides and herbicides than farmers as a whole, and that this should be reflected in the statement. 21. Members were informed at the November 2003 meeting that there was some published information (Ritchie et al Annals of Epidemiology, 13 (8), p571-572, 2003), which attempted to correlate estimated exposure of farmers/farm workers to pesticides. The Committee considered the abstract of this biomonitoring investigation of pesticide exposure. Members were aware that appropriate biomarker studies could assist in evaluating data where there were doubts concerning self reported exposure to pesticides. But, the COC agreed that this abstract provided insufficient information to draw any conclusions. 22. At the November 2003 meeting the Committee discussed the evidence relating to zinc and thought it unlikely that there were many individuals in the UK regularly consuming 100 or more mg zinc per day. Members felt that there may be sufficient numbers of individuals consuming such high levels of dietary supplements containing zinc in the EPIC cohort (European Prospective Investigation of Cancer) and asked the secretariat to respond to this query. Information from the EPIC-Norfolk cohort (for first 1860 individuals entering the cohort in 1993/4) reported that 5% of subjects were consuming zinc supplements (the intake from supplements was 4.9 (+ or - 4.1) mg/d). Thus the number of individuals consuming more than 50 mg zinc/day from supplements was likely to be very small. The Committee agreed that it was not possible to identify sufficient numbers of individuals for study from the EPIC cohort. Members agreed that it was not feasible to undertake any further epidemiological investigation of dietary zinc intake and prostate cancer in the UK. Members noted the correspondence (Correspondence in the Journal of the National Cancer Institute, vol 96, No. 3, p239-241, 2003) on a paper by Leitzmen et al on zinc supplement use and risk of prostate cancer (Leitzman et al National Journal of Cancer, vol 95, p1004-7, 2003). 23. The COC was informed that some additional work on PAHs with regard to prostate cancer was ongoing and that this would be submitted to the COC at the June 2004 meeting. 24. The draft statement would be amended in the light of members' comments. ITEM 6: ORGANOCHLORINES AND RISK OF BREAST CANCER: ADDITIONAL INFORMATION AND FIRST DRAFT STATEMENT (CC/04/3) 25. The COC considered a review paper on organochlorine insecticides (OCls) and risk of breast cancer at its September 2003 meeting. Tabulated and graphical summaries of the epidemiological data were considered at the November 2003 meeting. A review of additional data on the oestrogenicity of organochlorines was also considered. Members had agreed that sufficient information had been provided for a first draft statement and non-technical summary to be prepared. Members had asked for the graphs to be revised using a log scale for odds/risk ratio and for a footnote regarding interpretation of the data. The Committee had agreed that the available data were consistent with the conclusion reached in 1999, namely that it was most unlikely that current exposures to organochlorine compounds would represent a significant risk for breast cancer. 26. Some additional epidemiology had been published since the November 2003 meeting and these studies were made available to the Committee. A cut-off date of January 2004 had been proposed for the current review. A meta-analysis by Lopez-Cervantes et al (Environmental Health Perspectives volume 112, p207-214, 2004) evaluated p,p'DDE and beast cancer and found no evidence for an association. The summary risk ratio was 0.97 (95% C.I. 0.87-1.09) and was derived from 22 case control studies (9 were nested analyses from prospective studies). There was no clear evidence for heterogeneity. Two further small case-control studies showed no evidence for an association between OCls and would be included in the statement. 27. A small case-control study published in 2002 had recently been retrieved. The authors (Mathur V et al Environmental International, 28, p331-336, 2002) reported elevated levels of a range of OCls in the blood of breast cancer cases. Members considered that some of the differences in the blood levels of OCls were quite large, up to ten fold. Similar studies had not shown such effects and members agreed that the weight of evidence suggested no concern over a reported positive effect. 28. Some additional information on potential mechanisms had also been identified but overall provided little useful information. Members were also informed of an unpublished paper (accepted for publication in Environmental Health Perspectives) from Tinwell and Ashby that showed effects of mixtures of potential xeno-oestrogens (not OCls) was likely to be less than additive. 29. A further hypothesis paper was included as an addendum to CC/04/3. It suggested that free radical mediated oxidative damage could occur as a result of OCl residues in breast tissue. Members considered that this was a general observation of what would be expected to occur during carcinogenesis, and did not necessarily demonstrate a specific mechanism for cancer induction by OCls. 30. The Committee agreed the revised tables and graphs. Members had a
number of additional comments on the revised statement and also asked
for the DH Toxicology Unit to carry out a further pooled analysis on three
of the papers, which reported data on dieldrin. Members also made a number
of comments on the draft statement and non-technical summary. The tables,
draft statement and non-technical summary would be amended in light of
members' comments. It was agreed that the documents would be finalised
by postal circulation. ITEM 7: TOXICOGENOMICSA/PROTEOMICS/METABONOMICS: UPDATE ON LITERATURE 2003 (CC/04/6, addendum CC/04/6 and CC/04/5) 31. No interests were declared. The COT/COC/COM held a joint symposium on the use of genomics and proteomics in toxicology on 8 October 2001. A statement outlining the conclusions reached had been published on the Committee websites. A full write up of the meeting was published in Mutagenesis. 2003 May;18(3):311-7. There had been a substantial amount of research published, particularly in the last year. Update review papers had also been provided for COT and COM (and were being sent to COC members for information and comment). The objective of the further review is to consider whether there is a need to alter the overall conclusions reached in October 2001. 32. The Chairman asked members to consider the information according to the outline suggested in the paper from the secretariat but asked for general comments on the techniques to be made first. 33. Members noted that toxicogenomic studies in experimental animals resulted in large amounts of data being generated, but often from relatively small and limited experimental designs using too few animals, pooling of tissues, and a limited number of replicates for transcriptome analyses. There were insufficient data from multiple time point analyses, dose-response investigations, and for investigations of reversibility of toxicological effects. However members felt the most important area where there was a need to refine and develop approaches to toxicogenomics concerned the statistical handling of data and bioinformatic approaches to data evaluation. Members commented that many research groups had used a number of statistical techniques to identify gene expression changes that were potentially toxicologically relevant without apparently being aware of their limitations. Members endorsed the review of statistics/bioinformatics requested by the COT. Members commented that future bioinformatic developments currently being evaluated such as "pathway mapping" might be more suitable for the identification of toxicologically relevant changes. Pathway mapping could be applied to screening studies for toxicity and also for experiments to be designed to evaluate specific mechanisms. 34. Members considered the information summarised in CC/04/6 and commented that most data related to gene expression studies in chemically induced tumours or pre-neoplastic lesions that would ultimately form tumours. Investigations of cellular changes that were either reversible or did not proceed to neoplasia (eg resulted in cell death) might provide important information. It was noted that it was difficult to distinguish between gene expression changes induced by chemical exposure from those resulting from cellular responses to damage and/or as part of the neoplastic process. The evaluation of gene expression changes in the early stages of carcinogenesis using two model mouse hepatocarcinogens (oxazepam and Wy-14,643, a peroxisome proliferator) reported a number of differences in gene expression changes between these two chemicals. (Iida M et al Carcinogenesis, 24, 757-770, 2003.) Members considered that this was not an unexpected finding and commented that gene expression differences were likely to be documented between chemicals inducing tumours via the same mode of action. Small differences in experimental design and the stage of chemical induced response evaluated in cDNA microarrays would be expected to result on such differences in gene expression. Members considered that the investigation of gene expression changes in hepatocellular carcinomas in AOX-/- mice and from hepatocellular carcinomas from mice treated with ciprofibrate (a peroxisome proliferator) had been designed to test the hypothesis that genes associated with PPARα activation were relevant to the mechanism of tumourigenicity. The results were consistent with this hypothesis and demonstrated how targeted studies could yield useful information. Members noted it was not possible to compare the results of the gene expression studies of peroxisome proliferators reviewed in CC/04/6 particularly as the data related to different stages in the carcinogenic process. 35. Members considered that further research on the identification of potential biomarkers of carcinogenesis would be valuable. The gene expression information from stomach tumours from rodents (treated with MNNG) and human stomach tumours was useful for hypothesis generation. There was a concern that studies undertaken in experimental animals utilised high dose levels of limited relevance to human exposures and this would limit the value of investigations aimed at identifying biomarkers of chemical induced carcinogenesis. Members also commented on the need to identify gene expression changes in relevant target cells for chemical induced carcinogenesis. Laser microdisection techniques were beginning to be applied to isolate specific cell types but there was a need for additional information on the impact of this technique on gene expression changes. 36. In response to a question from the chairman on the additional appended papers, members commented that the findings in rodents exposed to main stream smoke and following in-utero exposure to arsenic added useful information and were consistent with existing toxicological assessments available on main stream smoke and arsenic. 37. Overall members observed that there had been a lot of experimental studies of carcinogenesis published using toxicogenomic approaches but felt it was important to reflect on these data with regard to the design and approaches to evaluation of future studies. Members confirmed that the most important area of development concerned bioinformatic evaluation of large complex data sets produced from such experiments. 38. Members were told that a further draft joint statement would be produced
for members consideration. Members were welcome to attend the presentation
of the report on bioinformatics/statistics to the COT. ITEM 8: POLYCYCLIC AROMATIC HYDROCARBONS: HIGH POTENCY PAHS- SUGGESTED BIOMONITORING APPROACH (CC/04/7) 39. The COC recently agreed with a suggested approach of using DNA adducts for ranking PAHs as a suitable surrogate for measuring carcinogenic potency by inhalation. It was noted that the potential for induction of PAH metabolism and differential repair of DNA adducts needed to be considered when interpreting results. However the approach represented a pragmatic method of providing ranking into broad groups. The DH Toxicology Unit had drafted a further paper which presented information on possible biomonitoring approaches for evaluating the potential contribution of high potency PAHs to air pollution related lung cancer. It had been suggested that data from biomonitoring studies might aid in the evaluation of which PAHs were more likely to pose the greatest risk of carcinogenicity from air pollution. The approach utilised the COC agreed position that DNA adducts may serve as a useful approach to ranking PAHs. 40. The Chairman asked members to consider each stage of the proposed approach to biomonitoring studies in turn. 41. Members discussed what might be the most appropriate study population for biomonitoring studies on high potency PAHs. It was agreed that an occupational group would be most likely to have sufficient exposure to high potency PAHs for the detection of specific markers of exposure and uptake. Members observed that studies of total PAH DNA adducts had reported evidence for a non-linear dose response with evidence for a plateau effect at the highest exposure levels for to PAHs. It was noted that one predominant pathway of exposure to PAHs in ambient air was via diesel exhaust emissions and hence an appropriate occupational group with diesel emission exposure (such as garage mechanics or bus drivers) would be suitable. Members heard that Sauvain JJ et al (International Archives of Occupational and Environmental Health, vol 76, 443-455,2003) had published some information on PAH exposures among diesel-exhaust workers and had reported the presence of dibenzo(a,l)pyrene in bus depots, truck repair workshops, and in underground mining. 42. Members heard that the UK air quality standard had been based on the epidemiological study of Canadian aluminium smelters published by Armstrong in 1994 (American Journal of Epidemiology, vol 139, 250-262.) There was no information currently available to compare the PAH exposure to high carcinogenic potency PAHs between aluminium smelters and diesel-exhaust workers. It was agreed that the investigation of exposure levels and uptake of high potency PAHs in workers exposed to diesel-exhaust would be important information in investigating the possible contribution of these particular PAHs to cancer risk associated with exposure to PAHs. 43. Members discussed whether a cross-sectional or short-term follow-up study of exposure to high potency PAHs would be most appropriate to the investigation of high potency PAHs in biomonitoring studies. It was noted that there were a number of factors, which could act as confounding factors in an investigation of high potency PAHs present in air pollution, such as contribution of PAHs from the diet, the genotype status for CYP1A1 and vitamin intakes. It was agreed that a design where subjects acted as their own control might simplify the data collection of any study since it would be more likely that potential confounders could be controlled rather than measured. Potential approaches included pre-and post-shift sampling or sampling during exposure and a few days after exposure (eg weekend). Although, it was not clear whether short term PAH changes would be picked up, or whether other factors, which might affect PAH levels (eg home cooking, smoking and exposure to environmental tobacco smoke), would remain constant. 44. The Committee considered what approach to obtaining PAH samples would be appropriate. Members agreed that bronchoalveolar lavage was invasive and provided a heterogeneous cell sample predominantly composed of macrophages. It was noted that nasal epithelial cell samples could be obtained relatively easily and in sufficient numbers for DNA adduct investigations and was considered by members to be a more pragmatic choice of cell sample. Members also commented that a comparison with peripheral blood lymphocytes would be valuable. 45. Members agreed that DNA adducts could serve as a biomarker of PAH exposure and uptake. It was noted that the COC had previously agreed, on the basis of in-vivo studies in A/J mice using intraperitoneal administration of individual PAHs, that time-integrated estimates of PAH-DNA adducts in lung could serve as a surrogate for carcinogenic potency in the lung. Members agreed that investigations of specific DNA adducts derived from high potency PAHs compared to benzo(a)pyrene would provide useful information on the relative potencies of such compounds in humans. Members considered that development of mass-spectrometry approaches to the identification of individual PAH DNA adducts was the preferred approach. There would be considerable difficulties concerning the synthesis and isolation of appropriate specific PAH-DNA adduct standards. This would be both costly and time consuming. It was considered that a pragmatic approach would be to concentrate on dibenzo(a,l)pyrene and dibenzo(a,h)anthracene as the two highest potency PAHs in the first instance comparing results with benzo(a)pyrene. 46. Members heard that the AMBIPAH project (Mechanism-based approaches to improved cancer risk assessment of ambient air polycyclic aromatic hydrocarbons) funded under the EU Fifth Framework Programme: Quality of Life and Management of Living Resources had initiated some developmental work on standards for high potency PAHs. In addition the project was investigating PAH mechanisms in in-vitro studies using rat lung slices, short-term studies using lacZ transgenic mice and two-stage in-vivo bioassays in SENCAR mice with a range of PAHs including dibenzo(a,h)anthracene, and dibenzo(a,l)pyrene in addition to benzo(a)pyrene. This information would be of value to risk assessment of PAHs when completed. 47. Overall, members agreed the outline proposal for further biomonitoring
work as suggested in CC/04/7. ITEM 9: UTILITY OF GENETICALLY MODIFIED MOUSE ASSAYS FOR IDENTIFYING HUMAN CARCINOGENS (CC/04/9) 48. Members heard that the in-confidence paper had been submitted by MHRA to the secretariat mainly for members' information but a summary of members' comments would be forwarded to MHRA. 49. The Committee agreed that the assays were of uncertain interpretation
and they were not a substitute for an adequate, science-based strategy
to ITEM 10: PAPERS FOR INFORMATION (CC/04/8)
50. Members heard that the Chairman had written to Biobank Project to inform them of the COC's views on the project. The Chairman requested further information on the proposal for a more detailed questionnaire of a subset of individuals who had volunteered to participate in the project and on questions relating to chemical exposure. ITEM 11: ANY OTHER BUSINESS
51. The minutes of this item will be published when the data have been published in a peer-reviewed journal.
52. Members discussed the storage of archive data held on the nitrosamine
dose-response carcinogenicity study undertaken at BIBRA in the early 1990s. 53. Date of next meeting 24 June 2004.
***************************** 3.2 Alcohol and breast cancer 54. The Committee was informed that the research paper on alcohol and breast cancer by the Imperial College submitted to a journal but had been rejected. The authors were advised to revise their paper specifically to include information from more recent studies. A further paper was in preparation (which included some further 20,000 cases) and it is hoped that this revised paper would be submitted to another journal in the near future. This would mean that the overall risk estimate and Population Attributable Risk (PAR) calculations would alter. This should not require the Committee to revise its conclusions, but it would be necessary to ask COC epidemiologists to confirm any alterations to the statement would be acceptable. The statement and non-technical summary would be brought back to the Committee if the COC epidemiologists suggest that a further COC consideration was necessary. **************************
ACTIONS
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
You
are here: <