Acrylaldehyde

Carcinogenicity

Description of key information

“There is evidence that acrolein is not an oral carcinogen. No dermal studies allowing assessment of the carcinogenic potential of acrolein were available. The available data do not allow a conclusion with regard to possible carcinogenicity upon exposure by inhalation. The limited inhalation data available did not indicate carcinogenicity after inhalation. It can not be excluded however, that respiratory tumours may be induced at non-cytotoxic concentrations. In analogy with other aldehydes as formaldehyde and possibly aceetaldehyde it is most unlikely that carcinogenic effects occur when irritation, as indicator for cytotoxicity, is avoided, but carcinogenic activity at non-cytotoxic exposure levels cannot be fully excluded.” quotations from European Union Risk Assessment Report Acrylaldehyde (EU, 2001)

Justification for classification or non-classification

There in no clear evidence on intrinsic carcinogenic activity of acrolein. Acrolein has been the subject of a risk assessment carried out under Community Regulation (EEC) No 793/93 (EU, 2001). This assessment came to the following conclusion: “There is evidence that acrolein is not an oral carcinogen. No dermal studies allowing assessment of the carcinogenic potential of acrolein were available. The available data do not allow a conclusion with regard to possible carcinogenicity upon exposure by inhalation. The limited inhalation data available did not indicate carcinogenicity after inhalation. It can not be excluded however, that respiratory tumours may be induced at non-cytotoxic concentrations. In analogy with other aldehydes as formaldehyde and possibly acetaldehyde it is most unlikely that carcinogenic effects occur when irritation, as indicator for cytotoxicity, is avoided, but carcinogenic activity at non-cytotoxic exposure levels cannot be fully excluded.” quotations from EU, 2001, p77, 83, 87 and p90

However, there is no evidence on inhalative carcinogenic activity of acrolein at non-cytotoxic exposure levels. At very low exposure levels, the amount of intracellular gluthation is expected to last out for detoxification of acrolein. An adaptive response of respiratory glutathione levels to repeated acrolein exposures has been shown in an inhalative toxicokinetik study in rats. In reliable animal studies on oral carcinogenicity of acrolein neither a carcinogenic activity at the site of exposure nor a systemic carcinogenic activity was seen.

Further on, based on theoretical considerations, a systemic carcinogenic activity of acrolein is highly unlikely, because as seen in vivo studies on metabolism and toxicokinetic, acrolein has such a high reactivity that it binds primarily at the application site and will not be systemically distributed. Thus, acrolein may never reach the target tissues of whole animals other than those at the site of exposure.

In conclusion, there is no evidence on a carcinogenic potential of acrolein relevant to humans.

Thus, acrolein does not comply with the classification requirements regarding carcinogenicity outlined in regulation (EC) 1272/2008 or the former directive on classification and labelling 67/548/EWG.

Additional information

Acrolein has been the subject of a risk assessment carried out under Community Regulation (EEC) No 793/93 (EU, 2001). The key information on carcinogenicity of acrolein as stated are quotations taken from this EU Risk Assessment. The evaluation whether the available studies on carcinogenicity by oral exposure are sufficient to come to a conclusion on this topic within the EU Risk Assesement and further assessments carried out under other international and national programmes published after finalisation of the EU Risk Assessment Report 2001 (World Health Organization, International Programme on Chemical Safety (IPCS), Concise International Chemical Assessment Document of Acrolein, CICADS 43 (WHO, 2002); United States Environmental Protection Agency, Toxicological Review of Acrolein (US-EPA, 2003); United States Agency for Toxic Substances and Desease Registry, Toxicological Profile for Acrolein (US ATSDR, 2007); United States Environmental Protection Agency, HED Risk Assessment for Reregistration Eligibility Decision (RED) Document for Acrolein (US-EPA, 2008)) is divergent. Here the interpretation of the data as given in the EU Risk assessment Report, 2001 is followed:

1. European Union Risk Assessment Report of Acrolein (EU, 2001)

“There is evidence that acrolein is not an oral carcinogen.” “No dermal studies allowing assessment of the carcinogenic potential of acrolein were available.” “The available data do not allow a conclusion with regard to possible carcinogenicity upon exposure by inhalation.” “The limited inhalation data available did not indicate carcinogenicity after inhalation. It can not be excluded however, that respiratory tumours may be induced at non-cytotoxic concentrations. In analogy with other aldehydes as formaldehyde and possibly acetaldehyde it is most unlikely that carcinogenic effects occur when irritation, as indicator for cytotoxicity, is avoided, but carcinogenic activity at non-cytotoxic exposure levels cannot be fully excluded.” quotations from EU, 2001, p77, 83, 87 and p90

In Detail:

Carcinogenicity: Oral:

“Fischer 344 rats were given acrolein in drinking water at concentrations of 100 (5 days/week, 124 weeks), 250 (5 days/week, 124 weeks), or 625 mg/l (5 days/week, 104 weeks). The number of animals amounted to 20/sex/group in the high dose and the untreated control group, and to 20 male rats in low and mid dose groups. Surviving rats were killed at 123 to 132 weeks. Major organs and tissues were examined histologically. A slightly higher incidence of adenomas of the adrenal cortex was found in females of the high dose group, the incidence amounting to 5/20 in high dose females and to 1/20 in untreated control females. No treatment-related increased incidence of other neoplastic lesions was recorded. In the same study no increased incidences of any tumor (including adrenal cortex adenomas) were induced by acroleinoxime, acrolein diethylacetal or allyl alcohol, compounds considered to be converted to acrolein (Lijinsky and Reuber, 1987, Lijinsky, 1988). The results of this study were re-evaluated by a pathology working group especially organised to this end. The working group concluded that the slightly elevated incidence of adrenal cortex adenomas (i.e., pheochromocytomas) found in the treated females was well within the limits for historical controls, and was not of biological significance. Furthermore, it was concluded that there was no evidence of any carcinogenic effect of acrolein on the adrenal glands of female rats in the Lijinski/Reuber study (see Parent et al.,1992).” quotation from EU, 2001, p75-76

“Sprague-Dawley rats were given daily by gavage 0.05 mg/kg, 0.5 mg/kg or 2.5 mg/kg per kg bw acrolein (distilled acrolein, stabilised with 0.25% hydroquinone) for 102 weeks. The study was conducted according to OECD 453 “Combined chronic toxicity/carcinogenicity studies”. The number of animals amounted to 70/sex/group. Interim kills were included at 13 weeks (high dose; n=5/sex) and 1 year (n=10/sex/group). Examinations included daily observations, measurement of various clinical, hematological and urine parameters at 3, 6, 12, 18 and 24 months All animals were subject to necropsy, recording of organ weights, and extensive microscopic examination of tissues. Male rats showed a dose-related reduction in survival during the first year, statistically significant in the high dose group and marginally significant in the mid dose group; this trend did not persist to the end of the study. In female rats a dose-related reduction in survival occurred during the first year persisting to the end of the study; survival was statistically significantly reduced in the high dose group and marginally in the mid dose group. Creatinine phosphokinase levels were decreased in all dose groups at almost all time intervals but only occasionally statistically significant (no data/figures presented!). No other treatment-related effects were seen including neoplastic or non-neoplastic lesions (Parent, R.A. et al., 1992a).” quotation from EU, 2001, p76 (pointer to summary table not include)

“CD-1 mouse were given daily acrolein (distilled acrolein, stabilized with 0.25% hydroquinone) by gavage at dose levels of 0.5 mg/kg, 2.0 mg/kg or 4.5 mg/kg per kg bw for 18 months. The study was conducted according to OECD 451, Carcinogenicity studies”. The number of animals amounted to 70/sex/group, and in the high dose group to 75/sex. Clinical observations were performed daily for the first 4 weeks and weekly thereafter; blood smears were taken at 12 and 18 months. Mice were subject to necropsy, and extensive microscopic examination of tissues. Observations included a statistically significant reduced survival rate and decreased body weights gain for males at 4.5 mg/kg; decreased body weight gain (not statistically significant) for females at 2.0 mg/kg and 4.5 mg/kg; no other treatment-related effects (including neoplastic and nonneoplastic lesions) were observed (Parent et al. 1991)." quotation from EU, 2001, p76 (pointer to summary table not include)

"In summary, oral (gavage, drinking water) administration of acrolein did not induce carcinogenic effects in rats or mice (Lijinski, 1988; Lijinski and Reuber, 1987; Parent et al., 1991, 1992a)." quotation from EU, 2001, p76

"From the results it can be concluded that acrolein is not an oral carcinogen.” quotation from EU, 2001, p76

Carcinogenicity: Dermal:

“There are no dermal or subcutaneous studies that permit the assessment of the dermal carcinogenic potential of acrolein. The exposure time in the study of Salaman and Roe in which acrolein was applied to the backs of mice once a week for 10 weeks, was too short and the number of animals per group too small to assess the carcinogenic potential of acrolein (Salaman and Roe 1956).” quotation from EU, 2001, p76

Carcinogenicity: Inhalation:

“Two long-term experiments using inhalation exposure were available, one with rats and the other with hamsters. However, inhalation experiments of appropriate duration specifically designed to assess the carcinogenicity of acrolein vapour have not been conducted. No treatment-related tumours or metaplasia were found in the lungs of rats (n = 20 per group) exposed to 18.3 mg/m³ (8 ppm) acrolein (nature not reported), 1 h/day and 5 days per week, for 10 or 18 months (Le Bouffant et al., 1980). quotation from EU, 2001, p74-75

“Remark: Study is not suitable for evaluation of carcinogenic potential of acrolein, because of restricted experimental design.” quotation from EU, 2001, p75

“Syrian golden hamsters exposed to 9.3 mg/m³ (4 ppm) acrolein vapour for 52 weeks (7 h/day, 5 days/week) showed inflammatory changes and metaplasia of the olfactory epithelium of the nose (Feron and Kruysse, 1977). After a withdrawal period of 29 weeks the affected mucosa had partially recovered in most animals. No respiratory tract tumours were found after exposure to acrolein vapour except for a small tracheal papilloma in one female hamster, which was considered an incidental finding, unrelated to acrolein treatment. Nasal tumours or treatment-related tumours at other sites were not encountered. In the same study no conclusive evidence of an enhancing (co-carcinogenic) effect of acrolein on the carcinogenicity of benzo[a]pyrene or N-nitroso-diethylamine was found (Feron and Kruysse, 1977). It is noted that the exposure and experimental period, 52 and 81 weeks respectively, were relatively short and did not cover two third of the hamster lifespan.” quotation from EU, 2001, p75 (pointer toHEDSET not include)

“Since both of the inhalation studies did not meet normal requirements for examination of the carcinogenic potential of a chemical, the study results do not permit a definite conclusion regarding the carcinogenic potential of acrolein upon exposure by inhalation.” quotation from EU, 2001, p75

"The limited inhalation data available did not indicate carcinogenicity after inhalation. However, none of the available repeated-dose

inhalation studies meet the generally accepted requirements for adequate carcinogenicity testing. On the basis of the experimental data it cannot be excluded therefore, that respiratory tumours may be induced at non-cytotoxic concentrations. It can be hypothesised that in analogy with other aldehydes such as formaldehyde and possibly acetaldehyde carcinogenic effects will not occur when irritation, as indicator for cytotoxicity, is avoided, but carcinogenic activity at non-cytotoxic exposure levels cannot be fully excluded." quotation from EU, 2001, p87

Carcinogenicity: Humans:

“There are no human data on carcinogenicity.” quotation from EU, 2001, p77

2. Agreement with further International Reports and Studies Published after Finalisation of the EU Risk Assessment Report 2001

No, the evaluation whether the available studies on carcinogenicity by oral exposure are sufficient to come to a conclusion on this topic is divergent. In reference to an identical data base, EU, 2001 as well as WHO, 2002 come to the conclusion that acrolein is not carcinogenic by the oral route, whereas the US reports US-EPA, 2003, US ATSDR, 2007 and US-EPA, 208 come to the conclusions that the data base on oral carcinogenicity is inadequate for an assessment and request further data. Consensus exists within the international reports, regarding the judgement of the two available inhalative carcinogenicity studies as inadequate to serve as a basis for assessment of the carcinogenicity of acrolein following inhalation:

3. Substantial Disagreements in Comparison to further International Reports to European Union Risk Assessment Report 2001

EU Risk Assessment Report 2001: “There is evidence that acrolein is not an oral carcinogen. The available data do not allow a conclusion with regard to possible carcinogenicity upon exposure by inhalation. ” quotation from EU, 2001, p77, 83 and 87

WHO 2002: "Available data are inadequate to serve as a basis for assessment of the carcinogenicity of acrolein following inhalation. Tumours have not been observed in the two relevant identified studies in rats and Syrian golden hamsters. However, these investigations were limited by small group sizes, limited exposure periods, and single dose levels (Feron & Kruysse, 1977; LeBouffant et al., 1980). Available data concerning the chronic toxicity/carcinogenicity of acrolein following oral exposure include three bioassays in which a wide range of end-points was examined following administration in Sprague-Dawley rats (Parent et al., 1992a), CD-1 mice (Parent et al., 1991), and beagle dogs (Parent et al., 1992b) and an earlier study in male F344 rats in which only mortality and histopathology in selected tissues were examined (Lijinsky & Reuber, 1987). In the more extensive of these studies, there have been no increases in the incidence of tumours of any type, although mortality, the cause of which is unclear, was increased

in rats and mice (Parent et al., 1991, 1992a)." quotation from WHO, 2002, p31

US-EPA 2003: “Under the Draft Revised Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1999), the potential carcinogenicity of acrolein cannot be determined because the existing “data are inadequate for an assessment of human carcinogenic potential for either the oral or inhalation route of exposure.” There are no adequate human studies of the carcinogenic potential of acrolein. Collectively, experimental studies provide inadequate evidence that acrolein causes cancer in laboratory animals. Specifically, two inhalation bioassays in laboratory animals are inadequate to make a determination because of protocol limitations. Two gavage bioassays failed to show an acrolein-induced tumor response in two species of laboratory animals. The finding of suggestive evidence of an extra-thoracic tumorigenic response in a drinking water study in female rats was not supported in a reanalysis of the data by an independently-convened pathology working group. Questions were also raised about the accuracy of the reported levels of acrolein in the drinking water from this study. A skin tumor initiation-promotion study was negative, and the findings from an intraperitoneal injection study were of uncertain significance. Although acrolein has been shown to be capable of inducing sister chromatid exchange, DNA cross-linking and mutations under certain conditions, its highly reactive nature and the lack of tumor induction at portals of entry make it unlikely that acrolein reaches systemic sites at biologically-significant exposure levels. The observations of positive mutagenic results in bacterial systems occurred at high concentrations near the lethal dose. “ quotation from US-EPA, 2003, p57

US ATSDR 2007:

“ No studies were located regarding the carcinogenicity of acrolein in humans. No increase in tumor incidence was observed in two limited, chronic inhalation studies in animals. These studies either had a very short daily exposure of 1 hour (Le Bouffant et al. 1980) or were carried out for less-than-lifetime duration for the study animals (Feron and Kruysse 1977; Le Bouffant et al. 1980). Dermal application of acrolein to mice for ten weeks did not induce cancer (Salaman and Roe 1956). However, the length of the study is considered too short for proper evaluation of carcinogenicity.No carcinogenic effect was found in rats, mice, and dogs following extensive histopathological examinations after chronic oral exposure to acrolein (Parent et al. 1991a, 1992a, 1992b). An increased incidence of adrenocortical adenomas was reported in female rats after oral exposure to acrolein in another study (Lijinsky and Reuber 1987); however, these findings were dismissed by an independent pathology working group (Parent et al. 1992a). Although Parent et al. (1991a, 1992a, 1992b) provided carcinogenicity data in animals receiving lifetime gavage doses of acrolein, lifetime drinking water studies in animals are needed to further assess lack of carcinogenicity reported in the lifetime gavage studies.”quotation from US ATSDR, 2007, p88

US-EPA, 2008: “The evidence for the carcinogenicity of acrolein is equivocal, with a significant tumor incidence found in a single animal drinking water study. The findings ofthis study were challenged by an independent pathology working group. Another well-designed cancer bioassay in orally-gavaged rats failed to detect significant increases in cancer incidence.” quotation from US-EPA, 2008, p11 "There are no adequate human studies of the carcinogenic potential of acrolein. Collectively, experimental studies provide inadequate evidence that acrolein causes cancer in laboratory animals. Specifically, two inhalation bioassays in laboratory animals are

inadequate to make a determination because ofprotocol limitations. Two gavage bioassays failed to show an acrolein-induced tumor response in two species oflaboratory animals. The finding of suggestive evidence of an extra-thoracic tumorigenic response in a drinking water study in female rats was not supported in a reanalysis ofthe data by an independently-convened pathology working group. Questions were also raised about the accuracy ofthe reported levels of acrolein in the drinking water from this study. A skin tumor initiation-promotion study was negative, and the findings from an intraperitoneal injection study were of uncertain significance." quotation from US-EPA, 2008, p14

In addition, a compilation of conclusions on the carcinogenic potential of acrolein is given:

IARC (International Association for Research on Cancer) 2004: Group 3 = not classifiable as to carcinogenicity to humans

US ACGIH (American Conference of Governmental Industrial Hygienists) 2004: A4 = not classifiable as a human carcinogen

US EPA as in IRIS (Integrated Risk Information System) 2005: Cannot be determined (Potential carcinogenicity cannot be determined because the existing "data are inadequate for an assessment of human carcinogenic potential for either the oral or inhalation route of exposure”)

US NTP (National Toxicology Program) 2005: no data

4. Additional Aspects in further International Reports

WHO 2002: Additionally to the oral carcinogenicity studies referred in the European Union Risk Assessment Report 2001, WHO 2002 refers the subchronic oral toxicity study on dogs (Parent et al. 1992b) on this topic, which is already cited in chapter 7.5. As in the studies on rodents, there have been no increases in the incidence of tumours of any type in dogs. see  WHO, 2002, p9 and p31

“The results of in vitro studies suggest that intracellular glutathione (or other free sulfhydryl groups) may protect against the DNA-damaging effects of acrolein (Eisenbrand et al., 1995).” quotation from WHO, 2002, p26

US-EPA 2003: Furthermore to the discussion of the inherent carcinogenic activity of acrolein in means of the in vivo studies, in US-EPA, 2003, also the probability of a carcinogenic activity in means of toxikodynamic conciderations is discussed: “It is clear from the studies reported that acrolein is highly reactive and cytotoxic. Acrolein has been shown to be mutagenic in some test systems within a narrow range of concentrations. Sensitivity to mutational effects is increased by GSH depleting agents and decreased by addition of metabolic activation, indicating that acrolein is a direct acting agent. While acrolein is capable of alkylating DNA and DNA bases (Maccubbin et al., 1990) and is known to inhibit purified DNA methylase activity from liver and bladder (Cox et al., 1988), it may never reach the target tissues of whole animals other than those at the site of insult. Even in the in vitro assays cited, acrolein is so reactive that special techniques must generally be employed to reduce cytotoxicity to induce positive effects. Parent et al. (1996b) have suggested that the reactivity of acrolein precludes its reaching target cells at a sufficient concentration to initiate the carcinogenic process.” quotation from US-EPA, 2003, p48

In conclusion:”although acrolein has been shown to be capable of inducing sister chromatid exchange, DNA cross-linking and mutations under certain conditions, its highly reactive nature and the lack of tumor induction at portals of entry make it unlikely that acrolein reaches systemic sites at biologically-significant exposure levels.” quotation from US-EPA, 2003, p57

US-EPA 2003: Refers further studies on the carcinogenic potential of acrolein not cited in the European Union Risk Assessment Report 2001. However, all these studies are inadequate for evaluation of the carcinogenic potential of acrolein because of their insufficient reliability.

One injection study: “The earliest reported study investigating the potential carcinogenicity of acrolein was reported by Steiner et al. (1943). Fifteen female partly-inbred albino mice received subcutaneous injections (0.2 mg/kg) of acrolein weekly for 24 weeks. No sarcomas developed at the site of injection. The use of only one dose level and a small number of animals limits any conclusions.” quotation from US-EPA, 2003, p34

One initiation and promotion study:

“Cohen et al. (1992) exposed 30 male Fischer 344 rats/group to acrolein, 2 mg/kg by i.p. injection twice weekly as part of a larger initiation/promotion study. All groups were sacrificed 53 weeks from the start of the study. No increases in tumor incidence were reported in groups exposed to acrolein alone for either 6 or 21 weeks (severe toxicity occurred during the 21 week study). Exposure to acrolein for 6 weeks followed by administration of uracil (3% by weight) for an additional 20 weeks resulted in the induction of 18 papillomas and one carcinoma, a significantly greater incidence (p0.05) than following exposure to uracil alone (8/30). While it appears that acrolein may have some tumor initiating capability, it should be noted that the incidence of papillomas and nodular hyperplasias combined, was significantly greater in the uracil only group compared with the group initiated with acrolein (p0.05).” quotation from US-EPA, 2003, p34

One human study: “Ott et al. (1989) reported a series of nested case-control studies in relation to various work areas, specific chemicals, and chemical activity groups. An odds ratio of 2.6 for nonlymphocytic leukemia was found for workers who had exposure to acrolein during employment. The small number of cases (3) and the likelihood of exposure to other chemicals, however, provide inadequate evidence of acrolein-induced leukemia or of the carcinogenic potential of acrolein.” quotation from US-EPA, 2003, p58

US-EPA 2008: In US-EPA 2008 an metabolite found in a fish metabolism study and a conclusion on the carcinogenicity of this metabolite is mentioned (see US-EPA, 2008, p13). However, this metabolite (glycidol = 2,3-epoxypropane-1-ol (CAS-No. 556-52-5)) has not been found in metabolism studies in mammals or in in vitro studies on mammalian cells. Therefore, it is expected, that this metabolite is of no importance with respect to the carcinogenic activity of Acrolein in mammals including human beings.

5. Additional Information in Newer Studies, not Included in the European Union Risk Assessment Report 2001 or further Cited International Reports None