Isopentyl acetate

Carcinogenicity

Description of key information

Data referred to/discussed under IUCLID 7.7.001-7.7.006:

Carcinogenicity, 7.7.005 (Gibel et al. (1974/1975)):

- Test substance: 3-Methyl-1-butanol

- Route of administration: oral, gavage

- Test animals: rats, female/male, 10 wk old

- Dose groups: 25 animals (control group), 15 animals (test group)

- Frequency of treatment: 2 x 0.1 ml/kg per week (which equals 23.1 mg/kg bw/d)

- Study period: until spontaneous death but max. 643 days (when control animals were killed)

The development of malignant tumors/cancer was observed upon treatment with the test substance whereas not in the control group. 4 malign tumours were observed after oral treatment at 23.1 mg/kg bw/d: 1 myeloid leukemia, 2 liver cell carcinoma and 1 forestomach carcinoma (3 benign tumors were also observed). In contrast, no spontaneous malign tumours were observed in the control animals (3 benign tumours were also observed).

In analogy to this finding, malignant tumors/cancer developed in a second experiment upon subcutaneous treatment (Carcinogenicity, 7.7.006 (Gibel et al. (1974/1975)) at 4.6 mg/kg bw/d. The study period was also max. 643 d. Findings: 10 malign tumours including 4 liver sarcoma, 2 liver carcinoma, 1 spleen sarcoma, 1 glandular stomach carcinoma, 2 myeloid leukemia (5 benign tumors were also observed).

There were also other signs of toxicity found (hepatotoxic and haematotoxic effects) in these studies. The mean average survival time was lower in treated animals: 527 (oral treatment), 592 (s.c. treatment) versus 643 in control animals.

Justification for classification or non-classification

No classification as carcinogenic under adoption of the official opinion of MAK (1996) and under consideration of further devaluing arguments of the available carcinogenicity data of Gibel et al. (1974/1975). See the controversial discussion above.

Applicable classification scheme:

REGULATION (EC) No 1272/2008 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 16 December 2008 on classification, labelling and packaging of substances and mixtures (3.6. Carcinogenicity).

Additional information

Criticism on the study results of Gibel et al. (1974/1975):

MAK (1996), epr 7.7.001:

- it would be unusual that no malignant tumours had developed in the control animals

- liver damage and reduced survival would suggest that the doses given to the treated animals had been excessive

- the incidences of liver tumours were suspected to have been a result of massive cytotoxicity

- the leukaemia could not be explained but the incidence of about 8% after both oral and subcutaneous administration would not seem likely to be significantly different from the control values.

- The study would not meet the requirements for a valid carcinogenicity study because of the small numbers of animals, the lack of data for the purity of the substances, the fact that dose-dependency was not investigated and statistical analysis was not carried out.

Other criticism on the study results (from epr 7.7.002 and 7.7.003, the interpretation of Gibel et al. (1974/1975) as published on the ECHA website in the disseminated registration dossier of "3-methylbutan-1 -ol":

Unsuitable test design (only 15 animals used, only one dose applied, dose was above MTD (maximally tolerated single dose), no spontaneous malignacy seen in control group which is unusual for 2 yr old rats)

---

Discussion of the author of this IUCLID data set on the significance of the results (see also epr 7.7.005 and 7.7.006):

Not performed according to official test guideline; not GLP; limitations in study design (e.g. low number of test animals compared to requirement in OECD test guideline 451) and lack of statistical analysis of results; however, carried out in accordance with generally acceptable scientific principles and with possibly significant results as in two separate experiments with different exposure routes (oral and subcutaneous) malignant tumour development was found upon exposure to 3-Methyl-1-butanol (in a reasonably long study period of 1.75 years) whereas not in the respective control groups.

The subcutaneous exposure route may not be relevant for real-life exposure, however, results on the carcinogenicity from this exposure route also yields relevant information on the intrinsic properties of the substance.

The identity of the test substance was sufficiently detailed:

3-Methylbutanol-l; p.a., in addition twice distilled, with info about supplier (VEB Berlin-Chemie).

It appears unlikely that in both experiments, oral and s.c. exposure, the particularly healthy animals which were not naturally prone to cancer had been allocated to the control group whereas the animals with a natural disposition for cancer development had been allocated to the treatment groups. The argument of an unusually low incidence of spontaneous malignancy in the control group is therefore questionable.

For the study part that examined effects of the subcutaneous exposure route the fact that cancer developed at various locations in the body and not at the site of application would point to actual carcinogenicity and not a local effect from high dosage.

The fact that two separate experiments with different exposure routes independently led to the same results (cancer development upon treatment and no cancer development without treatment) may be regarded as meaningful itself even without the possibility to establish a dose-response relationship.

It also needs to be addressed that cancer development could have been a secondary effect from other toxicity effects.

Strong hepatotoxic and also haematotoxic effects occurred as stated by Gibel et al. (1974/1975). If one could find out that cancer development had only been a secondary effect, one could not conclude on carcinogenic properties of the substance. It is, however, not possible to elucidate in retrospect if the observed incidences of cancer were only a side-effect from other toxic effects or at least partially caused by truly carcinogenic properties of 3-Methyl-1-butanol.

The fact that cytogenic or mutagenic effects were not observed in other studies can not rule out the possibility of carcinogenicity based on other mechanisms with a threshold effect.

The following documents were used for guidance to evaluate the data of Gibel et al. (1974/1975) in respect of the significance of the observed carcinogenic effects: Guidance on information requirements and chemical safety assessment Chapter R.7a: Endpoint specific guidance (Version 2.0, ECHA, November 2012); Guidance on the application of the CLP criteria (Version 3.0, ECHA, November 2012).

As a consequence of these evaluations, a LOAEL for tumor/cancer development could be defined for the oral exposure route (relevant in real life in contrary to the subcutaneous route) at the treatment level of 23.1 mg/kg bw/d.

However, this LOAEL is eventually disregarded, adapting to the official opinion of the German MAK Commission (MAK (1996)) where the results of Gibel et al. (1974/1975) were disqualified. MAK evaluated the data quality to be questionable (as laid out above). In the study report of "Study on the oral toxicity of 3-methylbutanol-1 in rats - Administration via the drinking water over 3 months", Project No.: 33S0056/88020, performed by BASF (1990) and owned by BG RCI, was also referred to Gibel et al. (1974/1975). In the report is pointed out that the suspicion of carcinogenicity cannot be ruled out under consideration of the information in Gibel et al. (1974/1975) and that carcinogenity studies may therefore follow in the future. The German MAK Commission evaluated the data apparently different.

A further toxicologist's expert view on the information and results in Gibel et al. (1974/1975) also needs to be considered. The data of Gibel et al. (1974/1975) was evaluated to be questionable with the following arguments:

- Not according to current standard protocols and GLP

- Limited documentation
- Low animal number used
- Low relevance of dose route used, i.e. subcutaneous, bolus
- No characterization of test material used. Information on purity cannot be sufficiently derived by the info given, i.e. twofold  distillation.
- No info on gender specificity of effects
- Only single dose tested, no dose dependency established.
- No MTD determined for correct dose setting
- Control group showed no  tumours, which is unusual.
- No comparison with historical control data possible
- Findings after treatment with the estimated doses (limited doc, the assumed dose level of 23.1 cannot be verified and is not cited in e.g. MAK report) are in strong contradiction to the findings from e.g valid subchronic guideline GLP study, showing only very slight erythrocyte effects. This further questions the reliability of the GIBEL publication.    
- No dose dependency of putative effects demonstrated no statistical analysis performed.
- Acc. to MAK -> study cannot be used for assessment due to these limitations.
- Carcinogenic potential of isopentylacetate/isoamylalcohol unlikely, since no structural alert for cancer, no effects in gentox. studies, no test substance related hyperplasia/preneoplastic lesions in guideline repeated dose studies observed.

- In conclusion, the data of Gibel et al. is not sufficient to make a valid statement on the carcinogenicity of isopentyl acetate but according to the present data situation and under consideration of the above mentioned arguments, there is no confirmation of a possible carcinogenic effect. Also, the results of the 90 d study (Schilling et al. (1997), BG RCI (1990)) devalue the results of Gibel et al. Thus, there was and is no new trigger for a carcinogenicity study.

The author of this IUCLID data set will follow up the data situation on isopentyl acetate and 3-methyl-1-butanol. If new crucial and relevant data on toxicologically concerning repeated dose effects/carcinogenic effects is sighted an adequate test proposal will be made. Under adoption of the official opinion of MAK (1996) and under consideration of further devaluing arguments of the available carcinogenicity data (Gibel et al. (1974/1975)), this step is not regarded as compulsory at this stage.

---

General justification for the use of read-across data of 3-methyl-1-butanol to isopentyl acetate:

Based on the available and adequate data on the metabolism of isopentyl acetate in the body (see under 7.1 Toxicokinetics, metabolism and distribution) it can be assumed that the ingestion of a certain amount of isopentyl acetate would lead to an equal molar amount of 3-methyl-1-butanol and acetic acid.

Acetic acid is not regarded as toxicologically relelevant compared to 3-methyl-1-butanol. Carcinogenic effects of isopentyl acetate should therefore be found to be the same or similar to the effects of 3-methyl-1-butanol. This study on 3-methyl-1-butanol is therefore regarded as meaningful for the description of carcinogenic effects of isopentyl acetate.

All available and adequate data on carcinogenicity was generated by the use of 3-methyl-1-butanol as test substance (data entered under IUCLID 7.7). These data and their interpretation should be regarded together in a weight of evidence approach to evaluate a possible carcinogenic effect of isopentyl acetate.