|
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||
|
 |
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
Present:
ITEM 1: APOLOGIES FOR ABSENCE/ANNOUNCEMENTS 1. Apologies for absence were received from Professor P Blain (Chair), Ms D Howel, Professor D Shuker, Mr R Alexander, Dr R Fielder, Dr S Samuels and Dr H Stemplewski.
2. The Chairman welcomed Mr K Okona-Mensah (DH Tox unit), Dr N Rajapakse (FSA), Dr S Bull (HPA), Ms A Gowers (EA), Dr R Shillaker (PSD), Dr K Fletcher (DH Tox unit), Miss I Mills, Dr J Pritchard (HPA) and Dr L Hetherington (DH). 3. Members were reminded of the need to declare any relevant interests before discussion of items. ITEM 2: MINUTES OF THE MEETING OF 21 APRIL 2005 (CC/MIN/2005/1) 4. The minutes were agreed subject to minor editorial changes. ITEM 3: MATTERS ARISING NOT COVERED BY LATER AGENDA ITEMS
5. The ILSI Human Relevance Framework (HRF), which extends the IPCS Mode of Action approach by incorporating information on kinetic/dynamic factors relevant to tumourigenesis in humans, was reviewed by the COC at the April 2005 meeting. The secretariat reported that, after discussion, it had made a number of suggestions for additional work using Group 1 and 2A IARC carcinogens, as outlined in CC/05/12. Members concurred with the principles behind the suggestions but considered that further thought was required about the practicality of undertaking the work. It was noted that USEPA was undertaking some HRF evaluations of pesticides which would be useful as background information. Members noted that Dr J Rice had undertaken an evaluation of rodent/human target organ concordance for the IPCS which would be published in the near future. 6. Further projects from USNTP and ILSI HESI were underway which evaluated
the concordance between precursor events seen in 90 day studies and tumours
seen in carcinogenicity bioassays in rodents. An expanded IPCS HRF document
had been provided to the secretariat. This would be forwarded to members
and the secretariat would discuss further action with colleagues in the
DH Toxicology Unit.
7. Proquinazid (6-iodo-2-propoxy-3-propyl-3H-quinazolin-4-one) is a novel fungicide being considered by the ACP under the Plant Protection Directive (91/414/EEC). The data holder is DuPont Chemicals. The Chairman asked for any declarations of interest. Professor Farmer declared a non-personal, non-specific interest. The Chairman agreed that he could take part in the discussion. 8. The secretariat informed members that the Advisory Committee on Pesticides (ACP) had discussed an application for the use of proquinazid in pesticide products on 13 January 2005 and had deferred making a decision pending advice from the COM and COC on the relevance to human risk assessment of cholangiocarcinomas reported in a 2-year chronic toxicity and carcinogenicity study in rats. The data on proquinazid were provided in CC/05/11. Annex 1 of CC/05/11 comprised the relevant extracts from the scientific evaluation and assessment document prepared by Defra's Pesticide Safety Division (PSD). Annex 2 gave the advice from the COM, which had considered proquinazid on 26 May. Other annexes included key extracts from the two-year rat carcinogenicity study, 18-month mouse carcinogenicity study and one-year dog study on proquinazid. The key sections from the Scientific Advisory Panel review were at Annex 7, Dr Greaves' memorandum of 20 May 2005 at Annex 8 and a Mode of Action Assessment of 8 June 2005 at Annex 9. 9. The committee was updated on the current status of the additional work which had been requested by COM. Members were told that an additional mouse lymphoma assay had been submitted which reported negative results. The COM would need to see the full report before reaching a final conclusion. COM members had considered that the recommendation of an extended exposure treatment of human lymphocytes for clastogenicity might be deferred if the COC accepted that there was a plausible non-genotoxic Mode of Action (MOA) for the cholangiocarcinomas seen in rats. 10. DuPont had provided a MOA for the cholangiocarcinomas which postulated that they are consequent on chronic severe liver toxicity resulting in a regenerative demand which exceeds the capacity of resident hepatocytes, and leads to oval cell hyperplasia to meet this demand and cholangiofibrosis from proliferation and metaplasia of pluripotent oval cells. The secretariat noted that CC/05/10 provided background information on the pathogenesis of cholangiocarcinoma in humans and rats. This had been provided to the company for comment and the response had raised the question of whether oval cells also play a role in human cholangiocarcinoma. The secretariat had retrieved abstracts of recent papers which indicated that oval progenitor cells have been identified in the human liver. Both DuPont's response and the abstracts had been tabled. 11. Members discussed the hepatic pathology in the rat study in order to identify any questions to put to the company. There was considerable support for the view that the lesions described as cholangiocarcinomas were not neoplastic but were a florid inflammatory response to severe, chronic hepatocellular damage with marked reactive epithelial changes. It was noted that the lesion in rats differed histopathologically from that in human cholangiocarcinoma. In particular, there was no evidence of dysplasia in the rat tissue. It was also noted that it was possible to distinguish between cholangiocarcinoma in rats caused by genotoxic agents and that caused by non-genotoxic agents. Non-genotoxic agents such as thioacetamide cause widespread, florid oval cell proliferation which is only seen at high doses. In comparison, genotoxic agents which induce cholangiocarcinoma cause a dysplastic response, often without an extensive inflammatory reaction. Also, mucinous metaplasia was not usually seen with genotoxic carcinogens. The pattern of changes seen with proquinazid fitted that seen with non-genotoxic agents. The COC agreed to ask the company for information on the extent and completeness of the pathology review by the Scientific Advisory Panel and the most recent review of liver slides from the rat carcinogenicity study by Dr Greaves. Members also agreed to ask about the steps identified in the proposed MOA and, in particular, the evidence that regenerative demand exceeded the capacity of resident hepatocytes.
12. The chairman welcomed the representatives from DuPont (Dr SR Frame, Mr G Stewart) and a consultant, Dr P Greaves (University of Leicester). 13. Mr Stewart (Regulatory Affairs, DuPont UK) explained that Drs Frame and Greaves would make presentations and then take questions from COC members. Copies of overheads had previously been sent to members. He noted that proquinazid had been submitted for UK evaluation for inclusion both in Annex 1 of Directive 91/414/EEC and for national approval in the UK. The UK Pesticide Safety Directorate (PSD) was acting as the EU rapporteur. A dossier of information had been provided to PSD in January 2004. The ACP had reviewed the submission in January 2005. Proquinazid provides up to 8 weeks control of powdery mildew on cereals at a lower dose than other fungicides. 14. Dr Frame (Research Fellow and Manager, Pathology, Haskell Laboratory for Health and Environmental Sciences) gave an overview of the toxicology of proquinazid. He noted that proquinazid was uniformly negative in a battery of in vitro and in vitro genotoxicity studies. A mouse lymphoma assay (MLA) had been initiated following the COM recommendation. It had been conducted using the Mictotitre® fluctuation technique. Two independent assays had been undertaken: one for 3 hours with and without S9, one for 24 hours without S9 and 3 hours with S9. There had been no statistically significant increases in mutant frequency with any of the treatments and he concluded that proquinazid is negative in the MLA. In discussing the cholangiocarcinomas seen in female rats, Dr Frame noted that the tumourigenic doses exceed the MTD, were late onset and that metastasis did not occur in any animals. 15. Dr Greaves explained that a Scientific Advisory Panel (SAP), of which he was a member, had reviewed aspects of the liver pathology in 2003 and a report had been finalised in November 2003. He had reviewed the slides again recently before the submission for the COC was compiled. The SAP considered that the lesions seen in the female rats did not indicate a carcinogenic risk for humans. The Panel's rationale had been that proquinazid was not genotoxic, that cholangiocarcinomas of the "intestinal" type occurred only in female rats fed 600 ppm or more at the end of the two-year study, the tumour type was linked to chronic hepatocellular injury with "cholangiofibrosis" and there was no risk of the lesion occurring in the absence of severe hepatic toxicity. He showed a number of representative sections. Features of the histopathology of the lesions seen at 1200 ppm included: cellular degeneration, inflammatory infiltration, extensive necrosis, giant nuclei in hepatocytes and fibrotic changes termed "cholangiofibrosis" by the original reporting pathologist. The lesions showed evidence of mucin production and were therefore termed "intestinal type". Some of the changes referred to as cholangiofibrosis were not related to the portal tract but occurred in the hepatic parenchyma. The appearance of the lesions termed "cholangiocarcinoma" was quite different from the usual appearance of human hepatocelluar carcinomas. The necrosis observed in the slides was extensive and seemed to be associated with an inflammatory response. COC pathologists noted that there no evidence of increased mitotic activity in the lesions and aberrant mitoses were not identified. 16. Dr Greaves confirmed that the SAP had reviewed virtually all the lesions diagnosed as cholangiocarcinomas and he had reviewed a proportion of them again recently. In answer to a question, he agreed that, if chronic exposure to proquinazid was sufficiently high to induce similar severe liver toxicity in humans, there was a risk that the cholangiocarcinoma lesion identified in rats could occur. He noted that liver toxicity was seen in female rats receiving 300 ppm proquinazid but was less severe than at 600 ppm. There had been individual variation in severity at 600 ppm. In answer to another question, DuPont representatives noted that the pigment seen in the liver sections had been shown to be PAS positive and had the appearance of lipofuscin. It was thus assumed to be lipofuscin. There was only minimal deposition of stainable iron. 17. The committee questioned the validity of the step in the MOA proposed by DuPont which proposed that regenerative demand exceeded the capacity of resident hepatocytes, thus triggering an oval cell response. It was agreed that oval cell proliferation was seen in rats when a regenerative stimulus was supplied but hepatocyte proliferation inhibited, for example by 2AAF. 18. The representatives from DuPont and Dr Greaves withdrew and members finalised their advice. 19. The committee concluded that it was uncertain that the lesions termed "cholangiocarcinomas" were truly neoplastic but accepted the conventionality, used by the study pathologist, of classifying them as neoplastic. However, it noted that the pathology seen with proquinazid lacked the characteristics of that caused by a genotoxic carcinogen. 20. Members agreed that the MOA proposed by DuPont for the putative cholangiocarcinomas was plausible. They noted that the lesions were only seen at dose levels which caused severe liver toxicity and which exceeded the MTD. They further agreed that the lesions were relevant to humans in that they were a potential hazard. However, whether they were a risk in humans depended on the level of exposure. 21. Members agreed that the No Observed Adverse Effect Level for cholangiocarcinoma was 16 mg/kg bw/day (300 ppm in female rats). They were informed that PSD had proposed an acceptable daily intake for proquinazid of 0.01 mg/kg bw, based on the overall No Observed Adverse Effect Level of 1.2 mg/kg bw/day and a safety factor of 100. Members confirmed that this would give an adequate margin of safety (1600) for cholangiocarcinoma. 22. The secretariat explained that the draft joint COM/COC statement will be expanded to include the COC conclusions. The company would be allowed 20 working days to consider the draft statement. Then, it will be finalised by post and the finalised statement forwarded to the ACP. ITEM 5: CARCINOGENICITY OF PFOS/PFOA (CC/05/16) 23. Members were informed that perfluorooctane sulfonate (PFOS) is used in a wide range of industrial and consumer products as a precursor material. PFOS is the predominant degradation material resulting from PFOS-related substances. Perfluorooctanoic acid (PFOA) is used as an industrial emulsifier and is also produced by environmental degradation of some PFOS precursors. 24. The Environment Agency recently concluded that PFOS meets the criteria for classification as a Persistent, Bioaccumulative and Toxic (PBT) substance. The half-life for humans is between 4 and 9 years for PFOS, and approximately 4.5 years for PFOA although potential re-exposure of individuals in the kinetics studies limited the value of these estimates. PFOS and PFOA have the potential to enter the food chain and the results from a Food Standards Agency (FSA) commissioned analysis of the 2003 Total Diet Study samples for PFOS and PFOA are due to be published in the latter half of 2005. 25. The COT had advised that the toxicology data indicate that PFOS and PFOA should be considered as separate entities. The COT had also recommended that the COM and COC be consulted for advice on the mutagenicity and carcinogenicity of PFOS and PFOA. 26. The COM assessed the available mutagenicity data for these substances in May 2005. The Committee agreed that PFOS should be regarded as not mutagenic. PFOA induced a positive mutagenic response in the in vitro chromosomal aberration assay in CHO cells with metabolic activation. It was unclear to what extent the results were due to cytotoxicity. PFOA did not induce chromosomal aberrations in cultured human whole blood lymphocytes nor in mouse bone marrow erythrocytes. Overall, the COM concluded that the results from the in vitro chromosomal aberration assays were likely to represent a cytotoxic response. However, a plausible in vitro mechanism for the positive response was required to reassure the Committee about this conclusion. 27. The COC discussed the carcinogenicity and epidemiology studies relating to each of these substances. Advice from COC will be forwarded to the COT later this year.
28. One dietary carcinogenicity study in Sprague-Dawley rats is available in which dose levels of 0, 0.5, 2, 5 and 20 ppm were administered in the diet for 104 weeks. Interim sacrifices were made at 4, 14 and 52 weeks. Survival was considered to be adequate in this study. Non-neoplastic effects reported in the liver included increased absolute and relative liver weight, hepatocellular cystic degeneration and hepatocellular hypertrophy (often associated with vacuolation). No signs of hepatotoxicity were evident 52 weeks after cessation of a 52 week high-dose treatment. The NOAEL for non-neoplastic liver pathology was 2ppm. There were no signs of hepatotoxicity in the recovery group. 29. The incidence of hepatocellular adenomas was significantly increased at 20ppm in both males and females. There was a single hepatocellular carcinoma in the female high dose group. The incidence of thyroid follicular cell adenoma was significantly increased in the male high-dose recovery group, but not in the male and female high dose groups fed PFOS for 104 weeks. 30. The committee considered that the time-points chosen for investigating treatment-related changes to hepatic palmitoyl-CoA oxidase activity were not optimal for this endpoint. Members considered that appropriately chosen time-points (7-, 14-, and 21-days) may have provided the information required to conclusively show whether or not PFOS was a peroxisome proliferators. 31. A dietary carcinogenicity study in Sprague Dawley rats is also available in which N-ethylperfluorooctanesulfonamido ethanol (N-EtFOSE) was administered at dose levels of 3, 30 and 100ppm in the diet for 104 weeks. This compound has been shown to degrade to PFOS. No significant treatment-related effects were observed on 2-year survival rates. However, survival in all groups including the controls was relatively poor. There was evidence of hepatocellular hypertrophy in high dose males and females. The incidence of hepatocellular adenomas was slightly higher in high-dose male and female groups than in controls. This difference was statistically significant in the high-dose males. A single hepatocellular carcinoma was observed in a high-dose female. 32. Two limited human epidemiological studies (a retrospective mortality study and an 'episodes of care' analysis) had been conducted in occupationally exposed populations. Cohorts were relatively small and also relatively young. In the retrospective cohort mortality study, when restricted to workers with at least one year of employment and high exposure to PFOS, standardised mortality ratios were below one for all causes of death and all malignant neoplasms. There were three deaths from malignant neoplasms of the bladder (0.63 expected) in males with over 5 years in high-exposure jobs. This excess was statistically significant (SMR 16.12 (95% CI 3.32-47.14)). Members questioned the adequacy of exposure assessment by using job categories. It was noted that there had been potential exposure of the workers to benzidine, a known bladder carcinogen. Members advised that, overall, it was not possible to draw definite conclusions from this study. Further evaluation across all PFOS manufacturing sites would have provided more appropriate information. Members considered that the episode of care analysis was unusual in design and uninformative. 33. In answer to the specific questions set before the committee, Members agreed that there was equivocal evidence for carcinogenicity limited to hepatocellular adenoma in the animal studies. The NOAEL for tumourigenicity was 5 ppm. Members agreed that a threshold approach could be used for risk assessment but were not convinced that adequate evidence had been provided for a mode of action incorporating peroxisome proliferation.
34. The carcinogenicity of PFOA has been investigated in two dietary exposure studies in Sprague-Dawley rats. Ammonium perfluorooctanoate (APFO) was administered to Sprague-Dawley rats at levels of 0, 30, or 300 ppm in the diet for 104 weeks. Dose-related non-neoplastic liver effects included megalocytosis, cystoid degeneration and portal mononuclear infiltration. A significant increase in female mammary fibroadenomas was considered not to be significant by the study authors when compared to historical control data. PFOA also induced an apparent dose-related increase in Leydig cell adenomas, which was not significant compared to historical control incidence. COC members were concerned to note the occurrence of at least two viral infections in the rats used in this study. This limited the value of the results. Biegel et al. (Toxicol. Sci. , 60, 44-55, 2001) investigated a single high dietary dose of PFOA (300ppm for 24 months) and reported increased incidences of hepatocellular adenomas, Leydig cell adenomas and pancreatic acinar cell adenomas. Serum estradiol concentrations were significantly increased in treated animals. Reassessment of pancreas slides from the earlier study confirmed the occurrence of proliferative lesions of the pancreatic acini. Members noted this study was not designed to identify a NOAEL. 35. A hypothesis had been put forward which proposed that PFOA induces liver, Leydig cell and pancreatic acinar cell tumours via PPAR-alpha activation and that, therefore, these tumours are unlikely to be induced in humans (Klaunig et al., Crit. Rev. Toxicol., 33, 655-789, 2003). 36. Members heard that some additional research which had just been retrieved indicated that PFOA had effects on liver weight in PPARα null mice (Yang et al., Biochem Pharmacol., 63, 1893-1900, 2002). Members agreed with the proposal by Klaunig et al. that activation of aromatase and subsequent increases in serum estradiol levels were suggestive that a mode of action (MOA) could be proposed for the Leydig cell tumours. However, they considered that it was not possible to propose MOAs for the liver and pancreatic tumours reported by Biegel et al., 2001. Members agreed that, for risk assessment, it would be acceptable to use a threshold approach, to select an appropriate NOAEL for a precursor event for the most sensitive tumour and to apply an uncertainty factor of 100. Members considered that effects on reproduction, which had lower NOAELs, were likely to be most relevant in the risk assessment of PFOA. 37. In two linked, retrospective cohort studies of mortality in an occupationally exposed population, small increases were reported in death from cancer of the large intestine and from cancer of the prostate in employees with over 1 year definite exposure to PFOA. Members considered that none of the effects reported were significant for risk assessment. It was agreed that the COC's conclusions would be incorporated into a COT statement. ITEM 6: Draft Statement on the review of the possible association between childhood leukaemia and residence near sources of traffic exhaust and petrol fumes (CC/05/14) 38. At the last meeting, members discussed the review paper on the possible association between childhood leukaemia and residence near sources of traffic exhaust and petrol fumes. The paper identified 17 studies (of which 16 were from the published literature) that sought to investigate whether such an association exists. Each study was assessed in terms of its methodology and study design and only two were considered to have adopted an adequate method for the assessment of exposure to petrol fumes and vehicle exhaust. Both these studies reported null results. Based on the available evidence, Members decided that there was no firm basis upon which to conclude that an association exists between childhood leukaemia and residential proximity to sources of traffic exhaust and petrol fumes. 39. The committee considered a draft statement. A number of amendments were identified. The secretariat undertook to circulate a revised draft for comments by post. ITEM 7: PAPERS FOR INFORMATION
40. A number of COC members were contacted for their advice on a MOA evaluation of styrene induced lung tumours in mice. The secretariat was very grateful for the advice received which had been passed onto HSE. Members agreed the proposed MOA, although a number of uncertainties were identified. It has therefore been considered unnecessary to refer styrene to COC for a full evaluation.
41. The COC's reponse on the Draft Opinion of the EFSA Scientific Committee on a Harmonised Approach for Risk Assessment of Compounds which are both Genotoxic and Carcinogenic were submitted for Members' information. ITEM 8: ANY OTHER BUSINESS 42. Dr Benford tabled a short prepublication paper of a carcinogenicity study of aspartame in rats published by the European Ramazzini Foundation (Soffriti et al., Eur. J. Oncol., 10, 2, 00-00, 2005). An increased incidence of lymphomas and leukaemias was reported in rats. There are slight differences in historical control data for the incidences of lymphomas and leukaemias in female rats between the published data (average 13.4%, range 7.0-18.4%) and data presented to the EFSA on 17 June 2005 (average 12.9%, average 4-25%). The incidences of lymphomas and leukaemias in female aspartame-fed rats occurred over the range of the historical control data sets presented in the paper. Members considered that a small increase a tumour incidence over such a wide range of doses was implausible. Members agreed that there may be reasonable explanation for the differing historical control figures but that the variation in control data casts doubt on the quality of the study observations. Members questioned the validity of the practice of allowing the rats to live until a natural death. It was noted the statistical approach used did not fully adjust for age related effects. The study will be evaluated formally by the EFSA. ITEM 9: DATE OF NEXT MEETING 43. 17 November 2005
ACTIONS
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
You
are here: 