2-ethoxy-2-methylpropane

Carcinogenicity

Description of key information

Induction of hepatocellular adenomas has been recorded in male (but not female) rats following chronic exposure to high atmospheric concentrations of ETBE but not after chronic exposure via drinking water. A mode of action involving increased cell proliferation associated with receptor-mediated enzyme induction appears plausible. The genotoxic potential of ETBE has been adequately investigated in a range of in vitro and in vivo studies, and DNA reactivity or other expressions of genetic activity can be discounted. Based on all available data, it is concluded that ETBE is not a cancer concern to humans.

Key value for chemical safety assessment

Carcinogenicity: via oral route

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed

Dose descriptor:

NOAEL

550 mg/kg bw/day

Study duration:

chronic

Species:

rat

Carcinogenicity: via inhalation route

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

NOAEC

6 375 mg/m³

Study duration:

chronic

Species:

rat

Justification for classification or non-classification

In accordance to Directive 67/548/EEC and EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008, classification is not necessary for carcinogenicity.

Additional information

Limited information is available from a poorly reported rat oral (gavage) carcinogenicity study on ETBE (Maltoni 1995, 1999; Belpoggi, 2002) that has notable shortcomings in study design. It is also the case that infection with Mycoplasma pulmonis has been proposed as a possible source of confounding in investigations conducted by these same authors on MTBE (Schoeb et al., 2009), with recommendations that those findings be subject to independent audit and peer review in order to determine their reliability (Ward and Alden, 2009). Since similar limitations apply also to the investigation of ETBE by these authors, the results are considered to have negligible value for the purposes of hazard identification. Higher quality information is available from a 2-year drinking water study in which groups of rats ingested ETBE via drinking water at received doses up to 542 mg/kg bw/d (males) to 560 mg/kg bw/day (females). The study design was equivalent to that of OECD Guideline 453 and was GLP compliant. There was no increase in tumours in any tissue, with an NOAEL for carcinogenicity of 542 mg/kg bw/day.

Information is also available from a rat 2-year inhalation study (OECD Guideline 453, GLP compliant) using exposure levels of 500, 1,500 and 5,000 ppm (v/v) (equivalent to 2,100, 6,300 and 21,000 mg/m3). Using REACH guidance, such exposures correspond to an internal dose of 304, 913 or 3045 mg/kg bw/day (assuming an 6 hour rat respiratory volume of 0.29 m3/kg bw and 50% inhalation absorption). High dose males showed a significantly increased incidence of liver tumours (primarily hepatocellular adenoma, with one hepatocarcinoma) together with an increased incidence of altered cellular foci of the liver (a preneoplastic lesion), however no comparable findings occurred in females. Relative liver weight was increased by around 50% in males and by approx. 30% in females, potentially indicating the induction of drug metabolising enzymes in response to the high body burden of ETBE experienced by these animals.

ETBE Mode of Action discussion

The incidence of hepatocellular adenomas was significantly increased among male F344/DuCrl rats administered ETBE by inhalation at an atmospheric concentration of 5000 ppm (21,200 mg/m3 6h/day, 5 days/week for 104 weeks (Japan Bioassay Research Center, 2010b). There was no increased incidence of these tumours at lower dose levels in male rats (500 and 1500 ppm) (2,120 and 6360 mg/m3) and there were no liver tumours recorded in female rats at any dose level. The incidences of adenomas in male rats were: 0/50, 2/50, 1/49 and 9/50 in the 0, 500, 1500 and 5000 ppm concentration levels respectively. There was, in addition, a single hepatocellular carcinoma recorded at 5000 ppm in male rats. The NOAEL for tumours was judged to be 1500 ppm, which is confirmed by BMD analyses on adenomas and carcinoma combined which give BMDL values of 1505 ppm for the multistage model and 1679 ppm for the gamma model (EPA BMDS version 2.1.1).

In a two-year drinking water study conducted at exposure levels of up 10,000 ppm within the same laboratory, with the same strain of rats and the same pathologists, there were no significant increases in the incidences of any tumours (Japan Bioassay Research Center, 2010a). This is the background against which modes of action must be considered. However, the actual doses of ETBE received by the rats in the drinking water study would appear to be no more than 542 mg/kg bw/day by male rats and 560 mg/kg bw/day by female rats. These doses are appreciably lower than would have been experienced in the high exposure group of the inhalation study. The internal doses to the rats exposed to ETBE concentrations of 0, 500, 1500 or 5000 ppm in the two-year inhalation experiment were calculated using REACH default assumptions as 0, 304, 913 and 3045 mg/kg bw/day (assuming a 6h rat respiratory volume of 0.29 m3/kg bw and 50% absorption by inhalation; REACH TGD Table R.8-2), while 0, 260, 780 and 2610 mg/kg bw/day in males and 0, 280, 850 and 2850 mg/kg bw/day in females can be derived based on a 6 hr exposure, 50% uptake and the sex-specific long-term defaults given in REACH TGD Table R.8-17. Two points should be made regarding the use of these defaults. Firstly, the respiratory volume per hour that has been used (45 L/h/Kg) is very high in comparison with another default (Gold et al., 1984) that is commonly used and is also quoted in the REACH document. This is 6 L/h for a 500 g male rat. Secondly, if actual values for parameters are available then it is preferable to use these. A pulmonary retention of 50% is likely close to 2-fold higher than the actual value for in rats. Human data indicate 26% pulmonary retention for ETBE (Nihlèn et al., 1998). These alternative parameters would reduce in the retained inhalation doses towards those of the drinking-water study (but remaining higher).

Additional to these well-established testing protocols, ETBE has been subjected to a number of medium term two-stage model predictive assays in which a significant result is considered to demonstrate promoter activity. These included a multi-organ test in which rats were administered a cocktail of potent initiators, followed by ETBE doses of 0, 300 or 1000 mg/kg bw/day by gavage, as well as organ-specific models in which initiators specific for liver, kidney, colon and urinary bladder tumours were administered, followed by ETBE doses of 0, 100, 300, 500 or 1000 mg/kg/day by gavage. Studies were also conducted for promotion of thyroid and forestomach tumours by ETBE.

The multi-organ model (DIMS, 2008a) demonstrated the following:

* at an ETBE dose of 1000 mg/kg bw/day only, there was a significantly increased incidence of hepatocellular adenoma, but not of hepatic glutathione S-transferase placental form (GST-P)-positive cell foci;

* at both ETBE dose levels, the incidences of renal tubule atypical hyperplasia and adenomas were comparable with the control values;

* no increase in tumour incidence was observed in the urinary bladder at either dose level of ETBE, but papillomatosis (extensive papillary hyperplasia) was significantly increased at an ETBE dose of 1000 mg/kg bw/day;

* in the forestomach, the combined (but not the independent) incidences of focal squamous cell hyperplasia and papillomas were significantly increased in both the 300 and 1000 mg/kg bw/day groups, neither lesion being observed in the controls; and

* thyroid follicular cell hyperplasia and adenoma incidences were significantly increased at an ETBE dose of 1000 mg/kg bw/day. In addition, the incidence of thyroid follicular cell adenomas and carcinomas combined was significantly increased at 300 mg/kg bw/day.

These results tended to be confirmed in the organ-specific medium-term assays. Thus:

* at an ETBE dose of 1000 mg/kg bw/day only there were increased incidences of hepatocellular adenomas and of hepatocellular adenomas and carcinomas combined; foci of altered hepatocytes were present in all rats of all groups, including the controls; (DIMS, 2008b);

* while the incidences of renal tubule adenomas were significantly higher in the 500 and 1000 mg/kg bw/day dose groups, renal tubule cell carcinomas were significantly higher only in the 300 mg/kg bw/day group and the incidences of adenomas and carcinomas combined were comparable with the control values; renal tubule atypical hyperplasia was observed in all rats of all groups, including the controls; (DIMS, 2008b);

* no significant increase in tumour incidence was observed in the urinary bladder at any dose level of ETBE, although a single carcinoma was observed at 1000 mg/kg bw/day and papillomatosis was observed in 4/30 rats at an ETBE dose of 1000 mg/kg bw/day (DIMS, 2010a);

* in the colon, incidences of focal atypical hyperplasia, adenocarcinomas, mucinous adenocarcinomas and (combined) carcinomas and adenomas (the latter were significantly decreased in the 100 and 300 mg/kg bw/day groups) were comparable in all ETBE treated groups with the controls (DIMS 2010b).

In summary there was evidence presented for promoter activity (i.e., probably cell proliferation) in liver and thyroid, although no attempt was made to confirm the latter. No evidence for promoter activity was forthcoming in the case of colon, urinary bladder, kidney or forestomach. Thus, the demonstration in a two-year inhalation study of hepatocellular adenoma induction in male rats exposed to a very high concentration of ETBE (that was not confirmed in a two-year drinking-water study) finds some support in the medium-term studies.

Various possible modes of action for hepatocellular tumours in rats have been listed (Cohen, 2010). These are reproduced below.

I DNA Reactivity
II Increased cell proliferation	A. Receptor mediated	1. PPARα (peroxisome proliferation) 2. Enzyme induction (CAR, PXR, AhR) 3. Oestrogens 4. Statins 5. Cytotoxicity 6. Other
	B. Non-receptor mediated	1. Cytotoxicity 2. Infectious 3. Iron (copper) overload 4. Increased apoptosis (e.g., fumonisin B1) 5. Other

CAR, constitutive androstane receptor; PXR, pregnane X receptor; AhR, arylhydrocarbon receptor

 

Procedures by which proposed MOAs can be analysed have been published on a number of occasions. However, most requirements for a mode of carcinogenic action accept the position that if a carcinogen is mutagenic, then this should be assumed to be its MOA. Although this is not necessarily true and the consequence (linear dose-response over all low dose extrapolations) is being less readily accepted now than previously, it is a starting point. The in vitro and in vivo genotoxicity and mutagenicity test data referring to ETBE have been described and evaluated, leading to the conclusion that this chemical is not a mutagen or genotoxin (McGregor, 2007). Consequently, other MOAs should be examined and this will be done using the IPCS guidance originally published by Sonich-Mullin et al. (2001), but extended by (Boobis et al., 2006) to include human relevance. These frameworks are basically very similar to the procedures described by Meek et al. (2003) and EPA (2005).

Postulated mode of action (theory of the case)

Clearly, many of the possibilities listed above can be dismissed. The most important of these to be dismissed is DNA reactivity, there being no evidence for ETBE being genotoxic at all, even if non-DNA reactivity is included in this term. In contrast, the most likely of these suggested MOAs that is identifiable is increased cell proliferation associated with receptor-mediated enzyme induction. The involvement of both liver and thyroid effects in the database suggests that constitutive androstane receptor (CAR) participation could be involved, as has been suggested for a number of chemicals.

Key events in experimental animals

A sequence of key events following the receptor-mediated enzyme induction pathway is:

* induction of liver enzymes, including cytochrome P450 (CYP) isozymes, by activation of nuclear receptors;

* this is followed by the induction of liver cell proliferation and hypertrophy;

* with continued exposure, the increase in liver growth continues, and foci of altered hepatocytes (FAH) appear;

* such foci ultimately progress to neoplasia (Bannasch et al., 2003).

Evidence from experiments with ETBE or its metabolites should now be examined in order to judge the participation of these key events.

1) Activation of nuclear receptors is anticipated to be an early step, but no data have been presented that demonstrate an interaction of ETBE with any nuclear receptors. The absence of these data is problematic because it is assumed to be the first key event, but

2) Induction of CYP enzymes has been demonstrated. Oral (gavage) administration of ETBE at 2 ml [1500 mg]/kg bw to male rats for 2 days resulted in increased activities of CYP2B1-dependent 16-beta-testosterone hydroxylase, CYP2B1/2-dependent pentoxyresorufin-dealkylase (PROD), by about 7-fold, and CYP2E1-dependent p-nitrophenol hydroxylase, by about 2-fold, but not ethoxyresorufin dealkylase (EROD), while other microsomal enzyme activities - although higher at this dose level of ETBE - were not significantly altered. Neither ETBE nor TBA administered at intraperitoneal doses of 200 or 400 mg/kg bw/day for 4 days in the same study had any effect on these enzyme activities (Turini et al., 1998).

3) Induction of cell proliferation, which is usually transient, in the liver has not been investigated in rat liver, although mitogenesis has been recorded in rat kidney and mouse liver (Medinsky et al., 1999). However, there have been a number of experiments conducted in Japan in which treatment of F344/DuCrj rats with ETBE was preceded by short treatments with known, potent carcinogens that could be defined as initiators in a two-stage model of carcinogenesis. Two of these models involved liver and ETBE treatment at 1000 mg/kg bw/day was shown to increase the incidence of liver tumours. This result is consistent the ETBE-only inhalation treatment for two-years and encourages the conclusion that ETBE could be acting as a promoter. This, however, begs the question, what was the initiator in that case? No evidence for promoter activity was forthcoming in the case of colon, urinary bladder, kidney or forestomach, whereas there was an increased incidence of thyroid follicular cell tumours, even at an ETBE dose of 300 mg/kg bw/day. This latter result also is problematic because increased tumour incidences were not observed in the thyroid when ETBE was administered by inhalation or in drinking water for two years.

4) Hypertrophy and hyperplasia are expected to develop in the enlarging liver. Hypertrophy of centrilobular hepatocytes was observed in male rats given ETBE by gavage at 1000 mg/kg bw /day for ten weeks the F0 and F1 generations of a multigeneration study (Centre International de Toxicologie , 2004a), at 400 mg/kg bw/day in a 26 week gavage study (Chemicals Evaluation Research Institute, 2008a) and at 5000 ppm in male and female rats in a 90-day inhalation study (Mitsubishi Chemical Safety Institute Ltd., 2008c). However, no microscopic changes were reported in rat liver in two other studies in which ETBE was administered by inhalation at concentrations of up to 5000 ppm for 13 weeks (Medinsky et al., 1999) or 4000 ppm for 4 weeks (White et al., 1995). There have been no reports of hyperplasia.

5) Liver weight increases are induced by treatment. In a multigeneration study, absolute liver weights were increased by 17% and 27% in the F0 and F1 generation males given ETBE by gavage at 1000 mg/kg bw for 10 weeks and by 14% in the F1 generation males given ETBE at 500 mg/kg bw (Centre International de Toxicologie, 2004a). In male rats given ETBE by gavage at 400 mg/kg bw/day for 26 weeks, body weight-relative liver weight was increased (Chemical Evaluation and Research Institute, 2008a). Inhalation of ETBE by male F344 rats for 13 weeks at concentrations of 1750 or 5000 ppm resulted in significant liver weight increases of 22% and 32%, respectively (Medinsky et al., 1999) and liver weights of male Sprague-Dawley rats were significantly increased by16% after 4 weeks inhalation of 4000 ppm while there was no effect of 2000 ppm ETBE (White et al., 1995).

6) Foci of altered hepatocytes (FAH) develop. The incidences of acidophilic and basophilic cell foci were significantly increased in the 5000 ppm group of the two-year inhalation study of ETBE (Japan Bioasssay Research Center, 2010b). The incidences and severity of acidophilic cell foci in the 0, 500, 1500 and 5000 ppm groups, respectively, were 31/50 (slight),; 28/50 (slight); 36/49 (slight); and 39/50(30 rats slight, 9 rats moderate). The corresponding incidence and severity of basophilic cell foci were 18/50 (slight); 10/50 (slight); 13/49 (slight); and 33/50(31 rats slight, 2 rats moderate). Acidophilic and basophilic cell foci are considered to be preneoplastic lesions. Thus, no increase in the incidence or severity of these FAHs occurs at 1500 ppm, but only at 5000 ppm. It is remarkable that no FAH developed in the two-year drinking water study at any exposure level.

7) Hepatocellular adenomas develop from within FAH. Whether there was a development of adenomas from FAH or not has not been ascertained and cannot be unless a chronic experiment is conducted with sequential, carefully scheduled killing times. This is an uncommon investigation and in the interests of animal welfare it would difficult to justify.

8) There is evidence that the mode of action of ETBE hepatotumorigenicity in rats may be related to induction of oxidative stress, 8-OHdG formation, subsequent cell cycle arrest, and apoptosis, suggesting regenerative cell proliferation predominantly via activation of constitutive androstane receptor (CAR) and pregnane-X-receptors (PXR ) nuclear receptors by a mechanism similar to that of the non-genotoxic carcinogen phenobarbital (PB) and differentially by activation of peroxisome proliferator-activated receptors (PPARs). Male rats were administered ETBE at doses of 0, 150 and 1000 mg/kg bodyweight twice a day by gavage for 1 and 2 weeks using PB as a positive control. Significant increase of P450 total content and hydroxyl radical levels by low, high doses of ETBE and PB treatments at weeks 1 and 2, and 8-OHdG formation atweek 2, accompanied accumulation of CYP2B1/2B2, CYP3A1/3A2 and CYP2C6, and downregulation of DNA oxoguanine glycosylase 1, induction of apoptosis and cell cycle arrest in hepatocytes, respectively. Up-regulation of CYP2E1 and CYP1A1 at weeks 1 and 2, and peroxisome proliferation at week 2w ere found in high dose ETBE group. Results of proteome analysis predicted activation of upstreamregulators of gene expression altered by ETBE including CAR, PXR and PPARs (Kakehashi, 2013). Such a MOA for ETBE hepatotumorigenicity in rats is unlikely to be relevant to humans.

Concordance of dose-response relationships

CYP enzyme induction by ETBE was achieved with a dose of 1500 mg/kg bw/day (for 2 days), there being no enzyme induction at 400 mg/kg bw/day by either ETBE or its primary, major metabolite, TBA. The lowest doses at which liver weight increases were observed were 400 mg/kg bw/day by gavage and 1750 ppm by inhalation. The demonstration of promoter activity in the liver of rats by ETBE also was achieved with a dose of 1000 mg/kg bw/day, but not the next lower dose of 500 mg/kg bw/day. These dose levels are in concordance with the lack of hepatic tumour induction in a two-year study with a high oral (drinking-water) dose level of about 540 – 560 mg/kg bw/day but the observation of a significant increase in hepatocellular adenoma incidence at the highest exposure level (5000 ppm, 6h/day, 5 days/week) used in a two-year inhalation study. Using the REACH defaults from Table R.8-2 of the TGD, this corresponds to a received dose of about 3045 mg/kg bw/day. The next lower exposure concentration (1500 ppm) corresponds to a received dose of about 913 mg/kg bw/day. It is believed that these estimates might be high because an assumption in the REACH calculation is that 50% of the inhaled material remains within the rats. No estimate of the proportion remaining in rats appears to have been made, but calculations based on human experiments suggest that about 26% is more likely. Should this be the case, then these inhalation dose estimates should be halved to about 500 and 1500 mg/kg bw/day. In either event the demonstration of tumour incidence increase only at the high inhalation exposure level is consistent with the data coming from other studies with ETBE on enzyme induction and promotion.

Temporal association

While there are no data on receptor binding (if it occurs), CYP enzyme induction is a very early event, liver hypertrophy can be demonstrated in a few weeks and tumour promotion (in medium-term assays) occurs within about 20 weeks. The inhalation report does not provide individual animal data, but if it can (reasonably) be assumed that tumour diagnosis could only occur after the death of the animal, then the first tumour in the high dose group could not have been discovered earlier than 80 weeks into the experiment (derived from Figure 1, Survival Animal Rate, Male). Thus, there is complete temporal association of the events that can be assessed.

Strength, consistency and specificity of association of tumour response with key events

ETBE is metabolised to a relatively longer-lasting metabolite in blood, tertiary-butyl alcohol (TBA) and a short half-life metabolite, acetaldehyde, which is rapidly incorporated into various metabolite pathways. This oxidation of ETBE is CYP enzyme-mediated. An extremely high dose of ETBE (2 ml/kg bw on 2 consecutive days) was shown to induce CYP2B1 and CYP2E1 in rat liver, whereas neither ETBE nor TBA had a significant effect on the activities of these enzymes at dose levels of 400 mg/kg bw/day for 4 days by intraperitoneal injection (Turini et al., 1998). This suggests that ETBE would have minimal effects on the expression of hepatic cytochrome enzymes in rats, except at very high dose levels. While hepatocellular hypertrophy has been shown to occur at relevant doses, there has been no demonstration of hyperplasia, although it can probably be assumed in the medium-term assays in which ETBE exposure is preceded by exposure to known, potent carcinogens.

The above descriptions have concentrated on effects in male rats, yet some of the changes also occur in mice where in addition to hypertrophy increased hepatocellular DNA synthesis was observed in a 13 week study (Mendinsky et al., 1999). No carcinogenicity study has been performed in mice. This situation does not constitute a lack of specificity, but an area of uncertainty. The data that are most suggestive of possible inconsistency is the absence of an hepatocarcinogenic effect in male or female rats treated with ETBE in their drinking-water, but this may be explained by the difference in received doses when compared with the inhalation experiment.

Biological plausibility and coherence

The induction of CYP enzymes could signal that interaction of a nuclear receptor and ETBE has occurred, although it is not clear whether CYP induction is on the pathway leading to neoplasmia or is a surrogate for a wider pleiotrophic response (Ueda et al., 2002). It is known that the well-studies CAR activation is involved in the epigenetic alteration of a large number of different genes, many of which may be involved in tumorigenesis (Phillips & Goodman, 2009; Phillips, Burgoon & Goodman, 2009).

Other modes of action

Liver tumours are common in long-term studies in rats and mice, and a number of potential MOAs other than enzyme induction have been suggested that may lead to the development of these tumours. These include 1) genetic activity, including DNA reactivity, 2) cytotoxicity and proliferative regeneration, 3) estrogenic stimulation and 4) alternative receptor-mediated mechanisms.

The genotoxic potential of ETBE has been adequately investigated in a range of in vitro and in vivo studies, none of which has shown a significant or questionable result. Therefore, DNA reactivity or other expressions of genetic activity can be discounted as possible alternative MOAs.

It is assumed that observation of a significant increase in the incidence of hepatocellular adenomas in male rats at the highest exposure concentration of ETBE in the two-year inhalation study is real and not a statistical outlier. No similar observation (not even of preneoplastic foci) was made in the drinking-water study; therefore, the response is clearly a very high dose phenomenon and there was no indication of hepatotoxic effects, such as peroxisome proliferation or chronic degeneration, in the toxicity studies performed on ETBE that might suggest cycles of degeneration and regenerative hyperplasia. Furthermore, none of the studies of general toxicity or toxicity to reproduction (reviewed in McGregor, 2007) has suggested that there might be perturbation of estrogenic hormone homeostasis that could result in a mitogenic stimulus to the liver.

Uncertainties, inconsistencies, and data gaps

An uncertainty alluded to above is whether the increase in hepatocellular tumours recognised only in the high dose group male rats of one experiment is the result of exposure to ETBE or an observation with a trivial cause, given that no hint of liver neoplasia was forthcoming from a drinking water study in the same laboratory. While the dose received by the rats was clearly higher in the inhalation study, all ETBE absorbed from the gut would pass first through the liver, whereas, upon inhalation, only a modest proportion would do so. The absence of recognised hyperplasia constitutes another area of uncertainty for the proposed MOA, although the medium term multi-stage models would seem to suggest that cell proliferative activity does occur. The data that are available appear to present a picture consistent with the MOA. However, there are two important data gaps: first, there is an absence of any studies for interactions of ETBE or its major, longer half-life metabolite, TBA with CAR or any other nuclear receptor; and second, there is no long-term study of ETBE effects in mice where phenobarbital, a CAR activator, is more active than it is in rats (IARC, 2001).

Assessment of postulated mode of action

A case has been proposed to explain the induction of hepatocellular adenomas in male rats exposed to a particularly high atmospheric concentration of ETBE. Some weaknesses have been described above, among them being an absence of a companion data set that might be applied to male mice. Also lacking are data from nuclear receptor interaction studies. Notwithstanding these limitations, existing data show good consistency with what would be expected of this MOA.

Bannasch, P., Haertel, T. & Su, Q. (2003) Significance of hepatic preneoplasia in risk identification and early detection of neoplasia. Toxicologic Pathology, 31:134–139.

Boobis AR, Cohen SM, Dellarco V, McGregor D, Meek ME, Vickers C, Willcocks D, Farland W. (2006). IPCS Framework for analysing the relevance of a cancer mode of action for humans. Crit Rev Toxicol 36:781–792.

Cohen, S. (2010) Evaluation of possible carcinogenic risk to humans based on liver tumors in rodent assays: the two-year bioassay is no longer necessary. Toxicol. Pathol., 38:487-501.

DIMS (2008a) Medium-term multi-organ carcinogenesis bioassay of 2-ethoxy-2-methylpropane (ETBE) in rats. DIMS Institute of Medical Science, Inc., Ichinomiya, Japan. Study no. 0635. 105 pp. (translation from Japanese).

DIMS (2008b) Medium-term kidney and liver carcinogenesis bioassay of 2-ethoxy-2-methylpropane (ETBE) in rats. DIMS Institute of Medical Science, Inc., Ichinomiya, Japan. Study no. 0721. 66 pp. (translation from Japanese).

DIMS (2010a) Medium-term urinary bladder carcinogenesis bioassay of 2-ethoxy-2-methylpropane (ETBE) in rats. DIMS Institute of Medical Science, Inc., Ichinomiya, Japan. Study no. 0828. 82 pp. (translation from Japanese).

DIMS (2010b) Medium-term colon carcinogenesis bioassay of 2-ethoxy-2-methylpropane (ETBE) in rats. DIMS Institute of Medical Science, Inc., Ichinomiya, Japan. Study no. 0827. 63 pp. (translation from Japanese).

EPA (US Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment. Washington, DC: Risk Assessment Forum. EPA/639/P-03/001F

IARC (2001) Some thyrotropic agents. Lyon, International Agency for Research on Cancer, IARC Press, pp. 161–288 (IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 79).

Medinsky, M.A., Wolf, D.C., Cattley, R.C., Wong, B., Janszen, D.B., Farris, G.M., Wright, G.A., and Bond, J.A. (1999). Effects of a thirteen-week inhalation exposure to ethyl tertiary butyl ether on Fischer-344 rats and CD-1 mice. Toxicol. Sci. 51:108–118.

Meek ME, Bucher JR, Cohen SM, Dellarco V, Hill RN, Lehman-McKeeman LD, Longfellow DG, Pastoor T, Seed J, Patton DE. (2003). A framework for human relevance analysis of information on carcinogenic modes of action. Crit Rev Toxicol 33:591–653

Phillips, J.M. & Goodman, J.I. (2009) Multiple genes exhibit phenobarbital-induced constitutive active/androstane receptor-mediated DNA methylation changes during liver tumorigenesis and in liver tumors. Toxicological Sciences 108:273–289.

Phillips, J.M., Burgoon, L.D. & Goodman, J.I. (2009) The constitutive active/androstane receptor facilitates unique phenobarbital-induced expression changes of genes involved in key pathways in precancerous liver and liver tumors. Toxicological Sciences, 110:319–333.

Sonich-Mullin C, Fielder R, Wiltse J, Baetcke K, Dempsey J, Fenner-Crisp P, Grant D, Hartley M, Knaap A, Kroese D, Mangelsdorf I, Meek E, Rice J, Younes M. (2001). IPCS Conceptual framework for evaluating a mode of action for chemical carcinogenesis. Regul Toxicol Pharmacol 34:146–152.

Turini, A., Amato, G., Longo, V. and Gervasi, P.G. (1998) Oxidation of methyl- and ethyl-tertiary-butyl ethers in rat liver microsomes: role of the cytochrome P-450 isoforms. Arch Toxicol 72:207-214.

Ueda A, Hamadeh HK, Webb HK, Yamamoto Y, Sueyoshi T, Afshari CA, Lehmann JM, Negishi M. (2002) Diverse roles of the nuclear orphan receptor CAR in regulating hepatic genes in response to phenobarbital. Mol Pharmacol.61:1-6

White, R.D., Daughtrey, W.C., and Wells, M.S. (1995). Health effects of inhaled tertiary amyl methyl ether and ethyl tertiary butyl ether. Toxicol. Lett. 82/83:719–724.

Justification for selection of carcinogenicity via oral route endpoint:
Most reliable study available

Justification for selection of carcinogenicity via inhalation route endpoint:
Most reliable study available

Carcinogenicity: via inhalation route (target organ): digestive: liver