Registration Dossier

Administrative data

Key value for chemical safety assessment

Additional information

The genotoxic potential of furfural was reviewed and reported in the EU RAR (2008) and also an EFSA review (2011). 

 

Non-human information

 

In vitro data

The key studies are considered to be bacterial mutation assays (Mortelmans et al., 1986, Haddouk, 1999), a mammalian cell gene mutation assay (McGregor et al 1988) and mammalian cell cytogenetic assays (Irwin, 1990; Gudi and Schadley, 1996).  These are recognised core assay types for investigating mutagenicity in vitro.

Furfural was tested using a pre-incubation Ames assay in S. typhimurium strains TA1535, TA1537, TA 100, and TA98 in both the presence and absence of rat and hamster liver auxiliary metabolic activation (S9). A range of doses up to 10000 ug/plate was used. Furfural was negative in this assay (Mortelmans et al., 1986). Haddouk (1999) used the same Salmonella strains, but also included strain TA 102. Both plate and pre-incubation protocols were used and tests were carried out in the absence and presence of rat liver S9, and doses up to 5000 ug/plate. Furfural was again negative in the assay.

Further Ames test studies reporting mainly negative, but some positive Ames assay results are included in an EFSA review (EFSA, 2011). However, the positive results are primarily from early studies, and the more recent studies have given negative results in the same bacterial strains. The weight of evidence is that furfural is not genotoxic in bacterial mutation assays.

McGregor et al (1988) tested furfural in a mammalian cell gene mutation assay using mouse lymphoma (L5178Y) cells in the absence S9 at a range of doses limited by toxicity. A positive response was reported.

In a mammalian cell cytogenetic assay furfural was examined in cultured CHO cells at a range of doses up to 500ug/ml in the absence of rat liver S9, and 1230 ug/ml in the presence of S9 (Irwin, 1990). Cells were harvested at 23.5 hours following exposure – S9, and at 20 hours following a 2 hour exposure +S9, and evaluated for chromosomal damage. Increases in the percentage aberrant cells were seen and the report concluded a positive result in the absence of S9 and a weakly positive result in the presence of S9. Any cytotoxicity induced was not given in this report. In a more recent and fully reported study, furfural was again examined in CHO cells in both the absence and presence of rat liver S9 at a range of doses limited by toxicity (Gudi and Schadley, 1996). In the absence of S9, exposure was for either 6, 20 or 44 hours with harvesting at 20, 20 or 44 hours after initiation of treatment respectively. In the presence of S9, exposure was for 6 hours, with harvesting at either 20 or 44 hours. Increases in the percentage aberrant cells were observed in all portions of the study and a positive result both in the absence and presence of S9 was reported. 

In summary, furfural shows no significant genotoxicity in bacterial mutation assays, but is positive in both mammalian cell gene mutation and cytogenetic assays. The activity observed in the cytogenetic assays is present in both the absence and presence of S9.

Additional data for furfural from non-core genotoxicity assays in vitro include positive results for sister chromatid exchanges in CHO cells and isolated human lymphocytes, and negative results for DNA repair (unscheduled DNA synthesis) in rat nasal epithelial cells and human liver slices (EU RAR, 2008; EFSA, 2011). These non-core studies are considered to be of limited value in a weight of evidence evaluation compared with the core genotoxicity assays above.

 

In vivo data

The key studies are considered to be those examining for chromosomal aberrations in mouse bone marrow (Irwin 1990), DNA repair in mouse and rat hepatocytes (Lake et al., 2001) and gene mutations (transgenic assay) in mouse liver (Steenwinkel et al., 2003). These are recognised core assay types for investigating mutation in vivo.

For the chromosomal aberration assay, furfural was administered by a single intraperitoneal injection to male B6C3F1 mice (8 per group) at a range of doses up to 200 mg/kg and the bone marrow examined at 17 hours after dosing (Irwin, 1990). Furfural gave a negative response. 

In DNA repair assays, furfural was dosed by oral gavage to male and female B6C3F1 mice (50, 175 or 320 mg/kg) or male F344 rats (5, 16.7 or 50 mg/kg) and hepatocytes prepared at either 2-4 or 12-16 hours after dosing (Lake et al 2001). Groups of three mice or four rats per dose level were used. DNA repair (unscheduled DNA synthesis) was assessed by autoradiography, and furfural was negative in both mice and rats.

Steenwinkel et al. (2003) conducted a transgenic assay (Lac Z) in livers from male mice following dosing at 75, 150 and 300 mg/kg by oral gavage for 28 days. Hepatotoxicity was assessed at this time. After a further 34-35 days, livers were taken for examination of the transgenic locus using eight animals per group. Changes in bodyweight and evidence of hepatotoxicity were seen at the end of the dosing period indicating a sufficiently high dose level had been achieved. There was no significant difference in mutation frequency between the furfural treated groups and the controls, and a negative result was reported.

Additional data for furfural from non-core genotoxicity assays in vivo are summarised by existing reviews and include negative results for SCE induction in mice, both positive and negative results in drosophila and a study examining the nature of oncogenes in tumours induced by furfural that reported differences in only some of the tumours compared with tumours from vehicle treated animals (EU RAR, 2008; EFSA, 2011). These non-core studies are considered to be of limited value in a weight of evidence evaluation compared with the core genotoxicity assays above.

Human information

An examination of lymphocytes from adults in an occupational exposure setting reported a negative result for SCE induction (Gomez-Arroyo & Souza, 1985). Exposure levels of 9545 mg/m3 to 38180 mg/m3 furfural were reported. The workers on the plant were exposed to both furfural and to furfuryl alcohol.

Summary and Discussion of Mutagenicity

Furfural has been examined for mutagenicity both in vitro and in vivo in a range of recognised core assay types, and also in a number of non-standard genotoxicity assays of limited relevance for a weight of evidence assessment of genotoxic activity. 

In vitro, furfural has been examined for gene mutation in a number of studies using the Ames test. It can be concluded to show no significant activity in bacterial mutagenicity assays. However, in both the mammalian cell gene mutation assay and mammalian cell cytogenetic assay, furfural has produced positive results. Overall, there is evidence for genotoxic activity in vitro.

In vivo, furfural has been examined in a range of assays in order to examine further the activity that was seen in vitro. An evaluation in mice showed negative results in the bone marrow for the endpoints of chromosomal aberration and SCE induction. A negative result was also observed for gene mutation in the mouse liver. A negative result for DNA repair, a general marker for DNA damage in vivo, was seen in both the mouse and the rat liver. Furfural has therefore had a thorough evaluation for possible genotoxicity in vivo, including an examination of the major endpoints where activity was seenin vitro (cytogenetic and gene mutation endpoints). The results from these evaluations in vivo were negative, and the data indicate that furfural shows no significant genotoxicity in vivo. 

It is concluded that the available data are sufficient to indicate that furfural has no significant genotoxic activity in animals.

Additional reference

EFSA (2011).  Scientific Opinion on Flavouring Group Evaluation 218, Revision 1 (FGE.218Rev1): alpha,beta-Unsaturated aldehydes and precursors from subgroup 4.2 of FGE.19: Furfural derivatives. EFSA Journal 2011; 9(3):1840

 

 


Short description of key information:
Furfural has been examined for mutagenicity both in vitro and in vivo in a range of recognised core assay types and also in a number of non-standard genotoxicity assays of limited relevance for a weight of evidence assessment of genotoxic activity. 
It is concluded that the available data are sufficient to indicate that furfural has no significant genotoxic activity in animals.

Endpoint Conclusion: No adverse effect observed (negative)

Justification for classification or non-classification

The available data do not lead to classification for genotoxicity according to DSD and CLP.