COC Meeting 17th November 2005

Minutes

Present:

Chairman:	Members:	Secretariat:
Professor P Blain CBE (Items 1-5 and 7)	Dr C Allen	Ms F Pollitt (Scientific- DH)
	Professor P Farmer	Dr D Gott (Scientific-FSA)
Professor D Phillips (Items 6 and 8-11)	Mrs R Glazebrook	Mr K N Mistry (Administrative)
	Professor D Harrison
	Ms D Howell
	Dr S Kennedy
	Dr B Miller
	Dr R Roberts
	Professor D Shuker
	Professor P Vineis
	Dr N Wallis

In Attendance:
Dr K Fletcher	(DH Tox Unit)
Mr K Okona-Mensah	(DH Tox Unit)
Dr K O'Leary	(DH Tox Unit)
Dr N Rajapakse	(FSA, item 3)
Dr S Creton	(FSA, item 4)
Dr K Hickling	(Astra-Zeneca, item 4)
Mr C Salole	(DTI, item 5)

Assessors:	Observers:
Dr R Fielder (HPA)	Dr J Pritchard (HPA)
Ms M Meldrum (HSE)
Mr S Samuels (PSD)
Dr H Stemplewski (MRHA)

Contents		Paragraph(s)
Item 1:	Apologies/announcements	1-4
Item 2:	Minutes of meeting held on 14 July 2005 (CC/MIN/2005/2)	5
Item 3:	Matters arising not covered by later agenda items:
	3.1 PFOA	6-9
Item 4:	Carcinogenicity of Furan (CC/05/22)	10-16
Item 5:	Draft working paper on nanotechnology (CC/05/18)	17-20
Item 6:	Review of the recent data on the quantitative relationship between alcohol consumption and squamous cell carcinoma (CC/05/20)	21-26
Item 7:	Statement of joint meeting 9 June 2005. Use of target organ mutagenicity data in carcinogenic risk assessment (CC/05/19)	27
Item 8:	Horizon Scanning 2005	28-40
Item 9:	Papers for information:
	9.1 Statement on the review of the possible associations between childhood leukaemia and residence near sources of traffic exhausts and petrol fumes (CC/05/21)	41
	9.2 Opinion of the EFSA Scientific Committee on a Harmonised Approach for Risk Assessment of Compounds which are both Genotoxic and Carcinogenic (CC/05/24)	42
Item 10:	Any other business	43
Item 11:	Date of next meeting	44

ITEM 1: APOLOGIES FOR ABSENCE/ANNOUNCEMENTS

1. Apologies for absence were received from Professor A Boobis, Dr P Carthew, Mr J Battershill, Dr Lesley Hetherington, Dr Diane Benford, Mr Ronnie Alexander and Ms Alison Gowers.

Announcements

2. The Chairman welcomed Mr K Okona-Mensah (DH Tox Unit), Dr K Fletcher (DH Tox Unit), Dr K O'Leary (DH Tox Unit), Dr J Pritchard (HPA), Dr D Gott (FSA), Dr S Creton (FSA), N Rajapakse (FSA) and Dr Kevin Hickling (Astra Zeneca).

3. The Chairman welcomed Mrs M Meldrum to the meeting as the new HSE assessor.

4. Members were reminded of the need to declare any relevant interests before discussion of items.

ITEM 2: MINUTES OF THE MEETING OF 14 JULY 2005 (CC/MIN/2005/2)

5. The minutes were agreed subject to one correction: page 6, paragraph 20, line 2 should read mice not rats.

ITEM 3: MATTERS ARISING NOT COVERED BY LATER AGENDA ITEMS

3.1 PFOA (CC/05/23)

6. The COC discussed the carcinogenicity of PFOA (CC/05/16) in July 2005, following a referral from COT. A two-year dietary study in Sprague Dawley rats (Sibinski, 1987) showed that PFOA induced dose-related, non-neoplastic liver effects including megalocytosis, cystoid degeneration and portal mononuclear infiltration, increases in Leydig cell adenomas, and proliferative lesions of the pancreatic acini. In a second study, rats were administered a single high dietary dose for 24 months and reported increased incidences of hepatocellular adenomas, Leydig cell adenomas and pancreatic acinar cell adenomas (Biegel et al 2001).

7. It has been proposed that PFOA induces liver, Leydig cell and pancreatic acinar cell tumours via PPAR-alpha activation (Klaunig et al 2003). However, at the July meeting, the Secretariat had informed the committee that, in a study with PPAR-alpha null mice, PFOA had caused increased liver weight but Wyeth-14,643 had not (Yang et al 2002). The COC had concluded that it was not possible to propose modes of action (MOAs) for the liver tumours reported. Members were not provided with the relevant paper in July and it was provided as Annex 1 to CC/05/23.

8. The COC conclusions were included in the 'First draft working paper on the TDI for perfluorooctanoic acid' considered by COT at its October meeting. The secretariat informed members that comments had been received from a COT member. These questioned whether the single experiment in mice by Yang et al (2002) provided sufficient reason to discount significant evidence in rats supporting the proposed MOA and to conclude that it was not possible to propose MOAs for the liver tumours. COC members were asked to consider the publication by Yang et al (2002) and to advise whether they are still content to conclude that it is not possible to support the proposed MOA.

9. A member commented that, if the COC had concluded at the last meeting that the MOA was peroxisome proliferation, the conclusion would have been thrown into doubt by the Yang et al (2002) paper. However, the COC had decided that it was not possible to identify the MOA and this conclusion should still stand. PFOA showed indications of oestrogenicity, possibly via aromatase induction, and thus one could not be certain that all toxic effects were mediated via peroxisome proliferation. The conclusion that it would be acceptable to use a threshold approach for risk assessment purposes also still stands. The Yang et al (2002) paper was interesting but lacked critical data on liver cell proliferation and liver pathology. The secretariat was asked to contact the authors to ask whether any studies had been done on these endpoints.

ITEM 4: CARCINOGENICITY OF FURAN (CC/05/22)

10. Members were informed that furan is an industrial chemical which occurs as a contaminant in food. The COT considered it in February 2005 and asked for further advice from the COM and COC on the existing evidence for genotoxicity and carcinogenicity, and on the need for further research.

11. Furan is rapidly and extensively absorbed and distributed following administration to rats. A key reactive metabolite is cis-2-butene-1,4 dial. CC/05/22 included details from two NTP studies which indicate that it can cause cholangiocarcinoma in rats, hepatocellular tumours in rats and mice, phaeochromocytomas of the adrenal gland in mice and mononuclear cell leukaemia in rats. At least two possible modes of action for furan-induced liver changes have been described: altered cell proliferation and oxidative stress mechanisms. The committee was asked to consider whether a threshold approach was appropriate for furan induced tumours; whether sufficient evidence exists to permit a mode of action analysis of carcinogenicity resulting from furan and or its metabolites; and to consider what further research, if any, would help to determine whether it was appropriate to assume a threshold mode of action?
12. The committee was told that the COM had discussed the mutagenicity of furan in October 2005 and had concluded that it should be regarded as an in vitro mutagen, but that there was insufficient evidence to reach a conclusion on the available in vivo mutagenicity data.

13. Recent work conducted by a group from the University of Birmingham in collaboration with AstraZeneca had examined the role of oxidative stress in the induction of cholangiocarcinoma in the rat. Dr Kevin Hickling from AstraZeneca gave a presentation of the findings to the committee. He outlined what is known about cholangiocarcinomas in humans, specifically that many oncogenes, including p53, are up regulated. In rats, these tumours are induced by potent genotoxins and also by proposed non-genotoxic mechanisms (coumarin) when the same oncogenic changes are apparent. The 'furan' model of tumour formation is the best characterised. Dr Hickling went on to describe the model used. Administration of doses of 4 or 13 mg/kg/day furan for 10 days resulted in characteristic intestinal metaplasia, diffuse centrilobular toxicity and a distinct focal response in the portal tracts. Notably, only particular lobes are affected, possibly because of increased portal drainage to these lobes, close proximity to the intestine or as a reflection of the capacity to respond to tissue damage. Administration of furan for 1 month is sufficient to cause tumours by 12 months. Areas expressing CYP2E1 correlate with areas of necrosis and p53 activation, although it is noted that CYP2E1 is down regulated on continued administration. In areas where the periportal damage occurs, ductal cells develop as early as day 3 although only in the target lobes. This lesion spreads, becomes surrounded by basement membrane and is infiltrated by intestinal metaplasia and capilliarisation of the damaged area. These changes are associated with dysregulation of HPC (hepatic progenitor cell) repair, biliary markers and connexin 43 and 32 expression. If furan administration is discontinued, the properties of intestinal metaplasia are maintained. High, cytotoxic dose are required to elicit this response.

14. Members considered that a more precise, sequential mode of action was needed. It was proposed that the cytotoxic intermediate generated by CYP2E1 was stimulating the regenerative response via oval cell proliferation and/or biliary progenitor cells. The intestinal infiltration may be a secondary response. Phenotypic changes were evidently being induced by the chronic inflammatory response, although it wasn't apparent exactly how Furan was 'flicking the switch' leading to the dedifferentiated response. It was unfortunate that no data were available on the response at low doses ie below those used in the NTP study, where cholangiocarcinoma occurred at all dose levels. It was noted that Furan has a low boiling point (310C) which is likely to affect the outcome of in vitro studies and that the dial metabolite is known to be a potent protein binder and depletes glutathione; its effects on O6-methyl guanine methyl transferase were not known.

15. In conclusion, the committee considered that it was reasonable to hypothesise that there was a threshold for the mechanism for induction of cholangiocarcinoma but this needed further investigation before a decision could be made. It was suggested that data from carcinogenicity studies with lower doses were required, as were in vivo mutagenicity data, and information on metabolite production and their contribution to an oxidative stress mechanism.

16. It was noted that the MOA for cholangiocarcinoma may not be applicable to the other tumours seen. COC members commented that the strain of rat used in the NTP study was known to have a high incidence of mononuclear cell leukaemia and that the phaeochromocytomas in the mouse may arise from a neuroendocrinological mechanism.

ITEM 5: Draft working paper on nanomaterial toxicology (CC/05/18)

17. The risk assessment of nanomaterials was identified by COT/COC/COM as an area of interest during the 2004 horizon scanning discussions. The Committees' interest in this area was prompted by the publication of the Royal Society review of nanotechnology (http://www.nanotec.org.uk/). Initial views were sought from the COC, COM and COT earlier in the year with a view to producing a statement by the end of 2005 for submission to Defra. Members were asked to consider the draft working paper (CC/05/18) which had already been considered by the COM and COT. The document contains comments suggested by the COT at its October 2005 meeting and the secretariat pointed out that these changes were tracked in the working draft. A revised flow diagram, produced by a COT/COC member, was attached as an addendum to the paper.

18. Members questioned the definition of nanomaterials as reported to have been defined by the Royal Society and asked that it be checked. They also expressed reservations about the suggestion that fluorophores be incorporated into nanomaterial manufacture as an aid to detection as these would probably be toxic. Members commented that, in paragraph 9, the final sentence should include hepatocytes as well as neurones as these could suffer storage disorders. They did not have any comments on the newly drafted paragraph (paragraph 11) from the COM. Members also noted that the final paragraph should be 16 not 14 and suggested that the first paragraph be reworded to make the first sentence clearer and to indicate that there were two key concerns: the intrinsic toxicity of the nanomaterial itself and the fact that familiar materials (eg paint) will be delivered by this method in future. Members considered that point II should be more tentative and indicate the need for a watching brief.

19. Members were informed that a cross governmental report should be available shortly which was expected to highlight research areas where there are gaps in knowledge on nanomaterials. Members were informed also about an ILSI recommended test strategy, and an EU Scientific Committee report on the new emerging health risks of nanomaterials which is currently out for comment (Post meeting: this is now published at (http://europa.eu.int/comm/health/ph_risk/committees/04_scenihr/docs/scenihr_o_003.pdf). Both reports made similar recommendations to the COT/COC/COM draft working paper. The Chairman suggested that the committee should review these reports at a later date.

20. Members asked if there was any data on exposure of workers to nanomaterials and were told that these are limited. The only data were on ultrafine carbon black where no significant exposure had been observed. Particles agglomerated rapidly so workers at any distance from source were not exposed to the individual nanoparticles.

ITEM 6: Review of the recent data on the quantitative relationship between alcohol consumption and squamous cell carcinoma of the oesophagus (CC/05/20)

21. The COC considered the possible quantitative relationship between alcohol and oesophageal cancer in 1995 as part of the review of alcohol and cancer. Several studies indicated that there was a quantitative relationship between alcohol intake and squamous cell carcinoma (SCC) of the oesophagus but a threshold level could not be defined. Therefore, the COC estimated levels of alcohol intake for which there was convincing evidence of an increased risk of SCC. A 2004 overview paper on oesophageal cancer suggested that further consideration of alcohol-induced SCC was necessary. Following consultation with a former COC member, it was decided that a review should be conducted to evaluate new data (post 1995) on the quantitative relationship between alcohol and SCC and this was provided in CC/05/20. Members were informed that the paper also considered the risk in smokers who drink heavily and the possible existence of any susceptible groups.

22. The secretariat explained that the 24 new studies were outlined in Annex 1 to CC/05/20. Of the six cohort studies, only one could potentially provide useful data (Boehing et al 2002). In contrast, the case control studies yielded useful information. The data for women appear to be consistent with the 1995 COC paper and a clear dose-response relationship is seen in men. However, the data suffered from the same limitations as were noted in 1995 including a lack of data in women, poor study design, and imprecise estimates of alcohol intake. It had also been suggested that the use of categorical analysis represents a significant flaw. Members were informed that an updated version of the dose-response relationship, which takes into account the findings of Kinjo et al (1998) is not tabled as stated in CC/05/20. In fact, Kinjo et al had measured the frequency, rather than the amount, of alcohol consumption and this prevented quantification in terms of grams ethanol consumed per day.

23. Members were asked to consider whether the new data indicated that a change was needed to the 1995 conclusions on the quantitative relationship between alcohol and SCC; whether it was possible to set a PAR for alcohol-induced SCC in the UK; and whether the data provided evidence of potentially susceptible groups.

24. Members considered that the new data strengthened the overall picture. An increased risk was apparent at intakes above 30 g/day but it was not possible to identify a lower level of consumption below which there is no increase in risk. The reasons for this were lack of data and the intrinsic weakness of the data, in particular the fact that the 'non-drinkers' group may include light drinkers. Also, confounding by other risk factors, such as dietary intake, lack of vitamins and, possibly, Barrett's oesophagus, should have been considered in the studies. Members noted that the papers included in Figure 1 were not comparable as some were original studies, some review papers and some studies which included other forms of cancer apart from SCC. It was suggested that it may only be necessary to consider the meta-analysis data, as these give the best estimates.

25. A member offered to contact Dr Boeing to establish whether an updated analysis of the EPIC study was expected.

26. Members considered that the data provided evidence of potentially susceptible groups. They noted the multiplicative effects of alcohol and smoking on the incidence of SCC and that smoking increased the risk of SCC at even moderate doses of alcohol. There was some scepticism about gender and ethnic differences, which it was considered may reflect temporal diffferences in drinking habits rather than a real difference in susceptibility.

ITEM 7: STATEMENT OF JOINT MEETING 9 JUNE 2005. USE OF MUTAGENICITY DATA IN CARCINOGEN RISK ASSESSMENT(CC/05/19)

27. The COM and COC held a joint, open meeting in June on the use of data derived from in-vivo target organ mutagenicity studies in carcinogen risk assessment. Members were informed that a brief statement had been drafted to record the main outcomes of the meeting and that a full report was being prepared for publication in a peer review journal. Comments provided by COM members had been incorporated into the draft statement. Members were invited to provide any further comments. Clarification on two points were addressed and agreed.

ITEM 8: HORIZON SCANNING (CC/05/17)

28. The annual horizon scanning paper was presented for members' comments. A number of items had been selected following literature searches and these were presented in CC/05/7, together with a review of the outcome of the 2004 horizon scanning exercise.

29. CC/05/17 noted that one item arising from the 2004 exercise was whether it was possible to give fuller advice than at present on the carcinogenic risk resulting from a single exposure to a carcinogen. Earlier in the year, the committee had considered some limited epidemiology data relating to short term exposure to carcinogens identified in a paper by Calabrese EJ and Blain RB. Certain Government departments and agencies were keen for the COC to give further guidance on the health risks of single exposure to carcinogens and CC/05/7 provided a brief outline of what further work might be undertaken. However, Members were not optimistic that further work would enable them to give better advice than at present. It was decided that the secretariat would consider whether it was practical to take this forward and, if so, how.

30. The COC has considered a few MOA assessments in brief. Members were invited to submit suggestions for candidate chemicals for further assessment in writing after the meeting.

2005 exercise (priorities are marked high (H) or medium (M))

31. Further information on dibenzo(a,l) pyrene: Members did not support the proposed project.

32. Trends in historical control tumours in rats: Members noted that, in assessing animal carcinogenicity studies, the committee only considered the historical control data for the 2 years prior to the end of study. They did not support the proposed project.

33. Comparison of animal and epidemiology data: Members did not support further work on the T25.

34. Age-related differences in risk of carcinogenesis: Members supported this project (H). It was noted that there was considerable discussion at present as to whether animal carcinogenicity data applied to children. The project should revisit the information which was examined when the committee reviewed the EPA guidelines but should also look independently at the original data.

35. Position paper on ignored concepts of carcinogen risk assessment:
Members were generally critical of this paper and did not favour the proposed project.

36. Trends in cancer incidence: It was agreed that the committee should review the possible chemical aetiology of testicular cancer and of non-Hodgkin's lymphoma (both H). It was also noted that there is interest in the media in the proposal that carcinogens, eg parabens, are absorbed through the armpit and may cause upper outer quadrant breast cancer. Some members considered that this hypothesis should be examined further, although other members favoured a general review of the environmental chemical aetiology of breast cancer. The secretariat was asked to come forward with a proposal for work on breast cancer (H).

37. Comparative risk assessment: Members supported this project (H) but not the proposal to use smoking cigarettes as the comparator. It was suggested that the former CMO's risk scale might be used. Members also agreed with the COM that more should be done on how to get its work across to the public (H). It was suggested that this should be a joint COC/COM/COT project.

38. Hormesis: Members agreed that this should be discussed further but that it should only have medium priority.

39. Paper by Jacobs (2005) on prediction of carcinogenicity results: A member explained that this paper arose from a discussion at the 2004 Society of Toxicology meeting at which there was a debate about whether 3-month rodent studies incorporating measurements of certain additional endpoints could be used to predict the results of rodent carcinogenicity studies. The paper was a response to this proposal. It was suggested that a further opinion be obtained from an absent member and that the committee should keep a watching brief on this issue.

40. Members were asked for suggestions for topics. It was proposed that the COC should review data on in utero exposures to carcinogens eg PAHs and this was agreed (H). It was also suggested that the committee should consider whether computational systems biology could be applied to its work (M). A member asked whether obesity came within the committee's remit and the secretariat agreed to check on this.

ITEM 9 : Papers for information

9.1 Statements on the review of the possible associations between childhood leukaemia and residence near sources of traffic exhaust and petrol fumes (CC/05/21)

41. Members were informed that the technical draft and lay statements had been agreed by post and the statement would be available on the website shortly.

9.2 Opinion of the EFSA Scientific Committee on a Harmonised Approach for Risk Assessment of Compounds which are both Genotoxic and Carcinogenic (CC/05/24)

42. The COC provided comments to the European Food Safety Agency on the draft of this opinion. The opinion had now been finalised and was tabled for Members' information.

ITEM 10: ANY OTHER BUSINESS

43. None at present.

ITEM 11: DATE OF NEXT MEETING

44. 2 March 2006.

ACTIONS

ITEM	ACTION	WHO
1.	PFOA - contact Yang et al to ask if further work done on liver cell proliferation and liver pathology	Sec (NR)
2.	Furan - communicate COC advice to COT	Sec (FSA/JB)
3.	Nanomaterial toxicology - amend statement	Sec (FP/JB)
4.	SCC and alcohol - contact Dr Boehring	PV
5.	Joint meeting - finalise statement	Sec (KF/JB)
6.	Horizon scanning:
	Single exposures - consider whether further work practical	Sec (JB)
	Submit suggestions of chemicals for MOA assessments	Members
	Age related differences in carcinogenicity - prepare paper	Sec (FP)
	Review possible chemical aet. of testicular cancer	Sec (KO'L, FP)
	Review possible chemical aet. of NHL	Sec
	Bring proposal for further work on chem aet of breast cancer	Sec
	Comparative risk assessment - prepare paper	Sec (KO-M, JB)
	Better communication of COC work (joint project with COM/T) - discuss with FSA and take forward	Sec (KO-M, JB)
	In utero exposure to carcinogens - prepare paper	Sec (FP)