A-Z site index | Search | Links | Contact DH | Help

1. Members will recall that MTBE was discussed at the last meeting of the Committee. A number of points were raised and have been dealt with. These were:

(i) Professor Richards has been provided with the section of the Existing Substances Regulations Draft dealing with the metabolism of MTBE.

(ii) Additional papers: some have been circulated; we are waiting for copies of others and will send them next week.

(iii) Members asked about concentrations of MTBE in air in the UK and about compounds formed as MTBE takes part in the combustion process in vehicle engines. A note to be provided by DETR will be sent to Members prior to the next meeting of the Committee.

2. Dr Robin Fielder (Scientific Secretary to the Committee of Carcinogenicity of Chemicals in Food, Consumer Products and the Environment) has considered the carcinogenicity of MTBE and has supplied the attached, very useful, note (see Annex A). A short summary of the aspects with respect to drinking-water is provided at Annex B.

3. In the light of discussions at the last meeting taking into account the further information supplied, Members are asked to consider, again, the draft statement attached at Annex C. This has been slightly modified, changes shown in bold typeface, to make clear that the Committee has looked only at the concerns regarding MTBE as a contributor to air pollution.

4. Members are asked whether they can agree the revised statement.

Secretariat

June 2000

ANNEX A

CARCINOGENICITY OF MTBE

I note from the COMEAP review that Stan Venitt and Anthony Dayan were asked to comment on the carcinogenicity of MTBE. I would be interested in their views.

I accept that MTBE can be regarded as non-genotoxic. The weight of evidence in vitro is negative although there are a few positives in mammalian cells. However it is clearly negative in vivo in bone marrow and liver and I have no problems with concluding that ‘non-genotoxic’ mechanism are responsible for the induction of tumours in animals by MTBE.

MTBE clearly induces tumours in rats and mice at a variety of tissues (kidney, liver, testes, possibly thyroid and also lymphoma/leukaemias). It is the multisite nature of these that is probably the biggest concern – on their own they can be easily ‘discounted’ as irrelevant to human or high dose effects. However they occur only at high doses that are associated with toxicity on the target tissues, and often at sites where there was some spontaneous incidence. Overall I can accept that current approaches to assessing a tolerable level can be used although I admit there are uncertainties in the exact mechanisms involved.

Considering the tumour sites in turn:

Liver tumours CD mice

The only effects seen that was statistically significant was the increase in hepatocellular adenomas in female mice at 8000 ppm. This was above the minimal dose producing liver hypertrophy. The apparent increase in hepatocellular carcinoma seen in males at the top dose (16% cf 4% in controls) was not statistically significant and there was much variability in the control incidence. It is reasonable to assume that prevention of hepatocellular proliferation (which occurred at the lower dose levels) would prevent tumour formation.

Kidney tumours in male rats

Very plausible mechanism involving binding to - a 2u globulin; although still some gaps in the data there is enough to assume that this is the mechanism in play. It is widely accepted that this is irrelevant to humans and this tumour type can be discounted.

Leydig cell testicular tumours in rats

I do not think that anything can be read into the apparent increase in such tumours in Fisher 344 rats as this is almost certainly due to the relatively low value in the controls (60%); this strain frequently has 90% plus incidence in the controls.

In the Sprague Dawley rats there was an increase in interstitial cell (Leydig cell) adenomas at 1000 mg/kg but not at 250 mg/kg in the oral study; interstitial cell hyperplasia was seen at both 250 and 1000 mg/kg.

This is clearly a high dose effect related to interstitial cell proliferation possibly due to the effects on plasma hormone (corticosterone up, testosterone down but such effects were inconsistent with no effects being seen on plasma LH levels.

Although the significance for man cannot be discounted the NOEL for interstitial cell hyperplasia can be used in the risk assessment.

Parathyroid in Fisher 344 rats

An apparent increase was seen in parathyroid adenomas in male rats only, but the effects were not dose related (8% at 3000, 2% at 8000 ppm cf 0% at 400 ppm and in the controls. There was a clear dose related increase in parathyroid gland hyperplasia from the lowest dose 400 ppm. These proliferative changes in the parathyroid of male Fisher rats may be secondary to the fall in plasma calcium after chronic renal failure. They are very probably a necessary precursor event for tumour formation.

Haemolymphoreticular neoplasms in Sprague Dawley rats

A dose related increase in what are termed ‘haemolymphoreticular’ neoplasms was reported in female Sprague Dawley rats only in the ingestion study (12% at 250 mg/kg, 20% at 1000 mg/kg cf 3% in the controls). However it is most unusual to combine lymphomas and leukaemias in this way and most would consider this unjustified, and the results uninterpretable. No clear data were given in ‘historic’ controls but it was stated that incidence (combined) could be up to 10%. There may thus be an increase at 1000 mg/kg but in view of the limitations I do not think any conclusions can be drawn.

Overall conclusions

MTBE is a non-genotoxic animal carcinogen. The induction of a range of tumours is of concern, but all are high dose effects and the significance of some can be discounted (the male rat kidney tumours because of the mechanism being irrelevant to humans, the ‘haemolympho-reticular’ neoplasms because of the failure to consider lymphomas separately from leukaemias. In the other areas it would seem entirely reasonable to use the no effect level for proliferative effect in the target organ on which to base a NOAEL for risk assessment purposes (together with the normal default uncertainty factors).

This, provided Stan Venitt and Anthony Dayan had no concerns, I would believe that the above approach is defensible without reference to COC.

ANNEX B

MTBE AND DRINKING-WATER

In the United States, instances of significant contamination of drinking-water supplies with MTBE have occurred due to leaks from underground and above-ground petroleum storage tank systems and pipelines affecting groundwater sources. At present, MTBE is not included in US national drinking-water regulations. In December 1997, US Environmental Protection Agency published a "Drinking-water Advisory" which reviewed the toxicity data then available and recommended that "keeping levels of contamination in the range of 20 to 40 microg/l or below to protect consumer acceptance of the water resource would also provide a large margin of exposure (safety) from toxic effects." The document summarised several small studies which found a range (in humans) of individual detection thresholds for MTBE in water of 15-180 microg/l (odour) and 24-135 microg/l (taste), and concluded that 20-40 microg/l was an approximate "threshold" for these aspects.

The toxicity of MTBE via drinking-water has not been assessed by DH Advisory Committees, and there is no WHO Guideline for drinking-water. In the UK as in other EU member states, there is no statutory limit for MTBE in drinking-water, although excessive concentrations would result in changes in taste and odour, probably leading to consumer complaints, and to breaches of the statutory requirements regarding taste and odour. In England and Wales, only one incident affecting drinking-water has been recognised and reported. This incident occurred in 1990, when consumers complained to Anglian Water of a strong sweet taste and odour in the supply. This proved to be caused by MTBE from an airfield adjacent to a borehole. The source was taken out of supply, activated carbon filters were installed to remove MTBE, and the supply was reinstated.

There is no systematic routine monitoring of water sources or supplies for MTBE. The Environment Agency and the Institute of Petroleum have jointly funded Komex Europe in a project to: collate information on known ether oxygenate occurrence in groundwater in England and Wales from Environment Agency records, water companies, oil companies and additional bodies and sources of information; review and evaluate the presence and behaviour of MTBE in UK groundwater, and; recommend further research with regard to the scope and practice of MTBE monitoring and analysis. The final report of the project is in preparation.

ANNEX C

COMEAP STATEMENT: METHYL TERTIARY-BUTYL ETHER (MTBE)

1. Methyl tertiary butyl ether (MTBE) is used as an oxygenating agent in petrol as it increases the octane rating of the fuel and reduces production of some air pollutants. In the UK, petrol typically contains up to 5% MTBE, the EC regulatory limit being 15% (from January 1^st 2000: EC Directive 98/70/EC).

2. We are aware that the use of MTBE in the United States has given rise to two groups of complaints:

3. We are aware of the considerable volume of toxicological research that has been done on the possible effects of MTBE in experimental animals. We comment first on the results of these studies and then turn back to consider the reports of effects in man.

4. In terms of both acute and chronic dosing in experimental animals MTBE is a compound of low toxicity. Whilst there is evidence that large doses can produce cancer in animals, there is no indication that this occurs by mechanisms relevant to humans; effects are unlikely to be mediated by a genotoxic process. This being so, there is no case for assuming that inhalation of low concentrations of MTBE would be associated with a risk of cancer. Studies designed to investigate the possibility of reproductive and developmental toxicology are also reassuring.

5. Attention has been focussed on animal studies that have revealed neurotoxic effects at very high dose levels. We note the points made by the HEI panel who concluded that motor activity in rats was affected on exposure to 800 ppm MTBE and that at higher levels of exposure sedation and ataxia occurred. These are very high concentrations: 800 ppm is 3142 mg/m³. Ambient concentrations are likely to be < 10 µg/m³. Levels at petrol stations are significantly higher than this, reaching a peak of about 30 µg/m³ during refuelling. Details are provided on pages 10 and 11 of the HEI report and in the table attached at Appendix IV. Given that concentrations lower than 800 ppm were not studied in animal models and that effects were found at this level, it is not possible to identify a No Observed Adverse Effect Level as far as neurotoxic effects are concerned. We also note that the HEI report records exposure of workers involved in handling levels of MTBE of > 1000 ppm. This as a cause for concern and we return to this point in our conclusions.

6. Evidence of effects in populations exposed to petrol treated with MTBE has been reviewed in detail by the Health Effects Institute in their report: The Potential Health Effects of Oxygenates added to Gasoline (published in 1996). The community studies are difficult to evaluate because of problems of design and statistical power and also because of confounding produced by awareness that MTBE had been added to gasoline. We also note that experimental studies involving volunteers have failed to produce effects similar to those described in some community-based studies. No evidence of effects on psychomotor or psychometric functioning was produced in the volunteer studies. This is reassuring in view of the findings of neurotoxic effects in rats exposed to very high concentrations of MTBE.

7. Our conclusions from this review of the evidence and more recent literature are similar to those of the Health Effects Institute report:

COMEAP Secretariat

June 2000

Top

copyright: © | last updated 17 July 2002