Registration Dossier

Administrative data

Key value for chemical safety assessment

Genetic toxicity in vivo

Description of key information

The Registrant has undertaken a thorough evaluation of all available data and has compared the data against the classification criteria as laid down in the Guidance on the Application of the CLP criteria (ECHA, 2017) in a weight-of-evidence analysis. The outcome of this weight-of-evidence analysis can be summarised as follows:

            No evidence for in vitro mutagenicity in bacteria

            Equivocal evidence for in vitro clastogenicity/aneugenicity

            No evidence for in vitro mutagenicity in mammalian cells

            No evidence for in vivo mutagenicity in transgenic rodents

            No evidence for site of in vivo contact genotoxicity after inhalation

            No evidence for in vivo clastogenicity, positive findings stem largely from unreliable studies with unphysiological route of exposure

            Positive findings were largely obtained from studies published by one and the same working group of E. Rojas and M. A. Altamirano-Lozano (University of Mexico City), whose study design and reporting shows recurring deficiencies

The weight-of-evidence analysis of the entire genotoxicity database does not show any clear evidence of germ cell mutagenicity.

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (negative)

Additional information

The endpoint genetic toxicity is not addressed by substance-specific information but rather by read-across of data available for soluble tri-, tetra- and pentavalent vanadium substances. This is based on the assumption that once inorganic vanadium compounds become bioavailable, this will be in tetra- or pentavalent vanadium forms. Based on the available weight-of-evidence and considering guideline-conform studies conducted under GLP both in vitro as well as in vivo, vanadium substances, including sodium metavanadate, should be considered void of genotoxicity. A detailed evaluation of the relevance, reliability and adequacy of each study is presented in the individual study records.

 

Read-across:The read-across approach based on dissolved vanadium is based on the assumption that once inorganic vanadium compounds dissolve or become bioavailable, this will be in tetra- or pentavalent vanadium forms. In bioaccessibility tests of tetra- and pentavalent vanadium substances, tetra- and pentavalent forms dissolved completely within 2h in various media selected to simulate relevant human-chemical interactions (i.e. PBS mimicking the ionic strength of blood, artificial lung, lysosomal, and gastric fluid as well as artificial sweat).Pentavalent vanadium substances are released and retained as pentavalent forms in physiological media, with the exception of artificial lysosomal fluid in which tetravalent V dominates after 2h and is the only form present after 24h. Thus, it can be assumed that vanadium speciation in body fluids is controlled by the conditions of the respective medium but not by the vanadium source. Thus, read-across of genetic toxicity data from soluble tetra- and pentavalent vanadium substances is justified.

 

In vitro gene mutation assays

Testing in bacteria reverse mutation assays (NTP 2002, Wolf 2006), although of limited relevance for metals (HERAG, 2007), yielded negative results:

- vanadium pentaoxide was not mutagenic in Salmonella typhimurium strain TA97, TA98, TA100, TA102, or TA1535, both in the presence as well as absence of metabolic activation (rat or hamster liver S9 enzymes).

- sodium polyvanadate (Na2V6O16) is non-mutagenic in Salmonella typhimurium strain the TA97a, TA98, TA100, TA102 and TA1535, both in the presence as well as absence of metabolic activation up to the limit of toxicity.

Guideline-conform in vitro mammalian cell gene mutation tests according to OECD 476 conducted by Loyd (2010a,b,c) under GLP are considered as key studies and will be used for classification. In these studies,soluble tri-, tetra- and pentavalent vanadium substances, including divanadium pentaoxide, did not induce any mutations at the HPRT locus of L5178Y mouse lymphoma cells when tested under the conditions employed in these studies, including treatments up to toxic and/or precipitating concentrations in two independent experiments in the absence or presence of a rat liver metabolic activation system (S9).

 

In vivo gene mutation assays

The lack of significant induction of cII mutant frequencies in the lungs of the transgenic Big Blue mice exposed to tumorigenic concentrations of divanadium pentaoxide by inhalation for up to 8 weeks suggests that divanadium pentaoxide is unlikely to act via a mutagenic mode of action.

 

Inhalation of aerosols of particulate divanadium pentaoxide for 4 or 8 weeks did not result in significant changes in levels of Kras codon 12 GAT or GTT mutation. The data support the idea that the accumulation of additional Kras mutants is not an early event, and/or that the proliferative advantage of Kras mutant clones requires either longer expression times or larger cumulative divanadium pentaoxide exposures. Furthermore, the data do not provide support for either a direct genotoxic effect of divanadium pentaoxide on Kras in the context of the exposure conditions used, or early amplification of pre-existing mutation as being involved in the genesis of divanadium pentaoxide-induced mouse lung tumours.

 

In vitro clastogenicity assays

Trivalent vanadium substance:

In the key study by Lloyd (2010):

- V2O3did not induce micronuclei in cultured human peripheral blood lymphocytes when tested up to toxic concentrations for 3+21 hours in the absence of S9

- V2O3showed evidence of inducing micronuclei when tested for 3+21 hours in the presence of S9 (but primarily at precipitating concentrations, therefore considered of questionable biological relevance)

- V2O3induced micronuclei in cultured human peripheral blood lymphocytes when tested for 24+24 hours in the absence of S9

 

Tetravalent vanadium substance

In the key study by Lloyd (2010):

- VOSO4 did not induce micronuclei in cultured human peripheral blood lymphocytes when tested up to toxic concentrations for 3+21 hours in the presence of S9

- VOSO4 induced micronuclei in cultured human peripheral blood lymphocytes and human lymphoblastoid TK6 cells when tested for 3+21 hours and for 24+24 hours in the absence of S9

Rodriguez-Mercado et al. (2003) also examined the chromosome aberration and sister chromatid exchange in human lymphocytes after V2O4exposure; both experiments suggest a clastogenic effect. However the toxicological significance could not be properly assessed due to reporting and experimental deficiencies.

 

Pentavalent vanadium substance

In the key study by Lloyd (2010),

- divanadium pentaoxide showed evidence of inducing micronuclei in cultured human peripheral blood lymphocytes when tested up to toxic concentrations for 3+21 hours in the absence and presence of S9 (at the two highest concentrations with observed precipitation and cytotoxicity levels > 25%, therefore considered of questionable biological relevance) and for 24+24 hours in the absence of S9 (at cytotoxicity levels > 10%).

- Divanadium pentaoxide showed evidence of inducing micronuclei in cultured human lymphoblastoid TK6 cells when tested up to toxic concentrations for 3+21 hours in the absence and presence of S9 and for 24+24 hours in the absence of S9 (with observed precipitation and at cytotoxicity levels > 15%, and therefore considered of questionable biological relevance).

- Vanadium pentaoxide was tested for a 24-hour treatment period in the in vitro micronucleus assay in Syrian hamster embryo (SHE) cells at the following concentrations: 10, 15, 20, and 25 µg/mL. In this test, vanadium pentaoxide did not show any genotoxic potential(Gibson, 1997).

 

It is not evident that the positive responses observed in the studies by Lloyd (2010) are true effects based on V2O3, VOSO4 or V2O5induced chromosome damage at physiologically relevant concentrations since both cytotoxicity as well as precipitation were observed in treatment groups that tested positively. Negative as well as positive results were obtained in clastogenicity assays in vitro with soluble vanadium substances. However, these in vitro studies of effects upon eukaryotic cells in vitro employed high concentrations of soluble vanadium producing significant levels of cytotoxicity and only weak genotoxic responses. A central issue that requires resolution is whether such in vitro results have physiological relevance by virtue of the mechanisms involved or the concentrations required to produce effects.

 

In vivo clastogenicity studies

Altogether, there are several reliable and unreliable studies that report the in vivo induction of micronuclei or chromosomal aberrations as well as the lack thereof that are summarized in the following table:

In vivo genotoxicity test

Route - exposure

Substance - Valency

Dose

(mg V / kg bw d)

Result - Reliability

Reference

Micronuclei in bone marrow - rats

p.o. – once, examined after 24h and 48h

Divanadium pentaoxide - V

i) 16.8

ii) 33.6

iii) 50.4

iV) 67.2

Negative – RL = 1

Beevers, 2011

Micronuclei in peripheral blood - mice

Inhalation – 3 mo

Divanadium pentaoxide - V

0.5 – 9 mg/m³

Negative – RL = 1

NTP, 2002

Micronuclei in peripheral blood lymphocytes & bone marrow - mice

 

p.o. via drinking water - 5 wk

Vanadyl sulphate - IV

0.39 – 30

0.10 – 0.40

Negative –RL = 2

Villani et al, 2007

Micronuclei in peripheral blood lymphocytes & bone marrow - mice

 

p.o. via drinking water - 5 wk

Sodium orthovanadate - V

0.06 – 5.49

20.8– 33

Negative–RL = 2

Positive –RL = 2

Leopardi et al, 2005

BrdU-incorporation assay

i) Aneuploidy in sperm

ii) Micronuclei in bone marrow – mice

i.p. – once

i) examined after 22 d

ii) examined after 24h

 

Sodium orthovanadate - V

i) 1.4 - 7

ii) 0.3 - 7

i) Positive –RL = 2

ii) Negative – RL = 2

Attia et al, 2005

Micronuclei in bone marrow - mice

i.p. – once, examined after 30h

Ammonium metavanadate - V

10.19

Negative – RL = 3

Wronska-Nofer et al, 1999*

Sister chromatid exchange

i.p. – once, examined after 24h

Divanadium pentaoxide - V

3.2 – 12.9

Negative – RL = 3

Altamirano-Lozano et al, 1993*

Micronuclei in bone marrow - mice

i.p. – once, examined after 18h

Sodium orthovanadate - V

1.4 – 6.9

Positive – RL = 3

Mailhes et al, 2003*

Micronuclei in bone marrow - mice

p.o. – once, examined after 4-48 h

i) Ammonium metavanadate – V

ii) Sodium orthovanadate – V

iii) Vanadyl sulphate - IV

i) 21.8

ii) 20.8

iii) 31.2

Positive – RL = 3

Ciranni et al, 1995*

chromosome aberration in bone marrow, mice

p.o. – once, examined after 24-36 h

i) Ammonium metavanadate – V

ii) Sodium orthovanadate – V

iii) Vanadyl sulphate - IV

i) 21.8

ii) 20.8

iii) 31.2

Positive – RL = 3

Ciranni et al, 1995*

chromosome aberration in bone marrow, mice

i.p.- duration and frequency not reported

Divanadium pentaoxide - V

i) 5.75

ii) 11.5

iii) 23

as mg V2O5/kg

Negative – RL=3

Altamirano-Lozano et al. 1996*

Micronuclei in peripheral blood - mice

i.p. – once, examined after 24, 48, 72h

Divanadium pentaoxide - V

40 mg/kg bw

Positive – RL=3

García-Rodríguez et al. 2016*

Micronuclei in peripheral blood - mice

inhalation - 1 h/day; 2 days/week, 4 weeks

Divanadium pentaoxide - V

1.4 mg/m³

Positive – RL=3

Rojas-Lemus et al. 2014*

* These references were rated unreliable according to the criteria by Klimisch et al. (1997) and the ECHA Guidance Chapter R.4: Evaluation of available information (December 2011, Version 1.1). Please find the list with all acquired (evaluated according to Klimisch et al.: A systematic approach for evaluating the quality of experimental and ecotoxicological data, Reg.Tox. and Pharm. 25, 1-5 (1997) and sorted by REACH Annex VII – X endpoints) in Chapter 13.

 

Whereas reliable in vivo data indicate a negative potential for tetravalent V, unreliable in vivo data appear to report the contrary.

 

The in vivo genetic toxicity of V2O5 was assessed within the framework of an NTP (2002) programme by testing the ability of the chemical to induce increases in the frequency of micronucleated erythrocytes in mouse peripheral blood. Female and male mice were exposed for 90 days to a vanadium pentaoxide aerosol by inhalation (at 0.5 – 9 mg/m³) before blood was sampled and analysed. As a result, divanadium pentaoxide, administered to male and female mice, did not increase the frequency of micronucleated normochromatic erythrocytes in peripheral blood. Furthermore, the genetic toxicity of divanadium pentaoxide was assessed by testing the ability of V2O5to induce an increase in the frequency of micronucleated erythrocytes in bone marrow of male rats. While, vanadium tissue levels increased in all sampled tissues with increased V2O5dosage levels, and thus, vanadium reached the target tissues bone marrow and testes in a dose-dependent manner, V2O5 administered orally by gavage to male rats, did not induce micronuclei in polychromatic erythrocytes of bone marrow treated up to the maximum tolerated dose of 120 mg/kg/day(Beevers, 2011).

 

Thus, for pentavalent V, reliable (RL=1) and GLP-conform in vivo data indicate a negative potential. However, reliable (RL=2) in vivo data also indicate positive threshold effects, but in these studies, the toxicological significance could not be properly assessed due to reporting and experimental deficiencies.

 

in vitro assays on DNA damage

In accordance with regulation (EC) 1272/2008, Annex I: 3.5.2.3.3 test systems assessing heritable alterations of DNA (such as in vitro gene mutation in the hprt locus) and systems investigating the induction of unspecific and transient DNA damage (such as comet assay) should be assessed differently: “Classification for heritable effects in human germ cells is made on the basis of well conducted, sufficiently validated tests, preferably as described in Regulation (EC) No 440/2008 adopted in accordance with Article 13(3) of Regulation (EC) No 1907/2006 (‘Test Method Regulation’) such as those listed in the following paragraphs.”. In vitro DNA damage tests are neither validated nor were these tests conducted in accordance with an accepted guideline. The information gained from such systems should therefore be evaluated with great care and regarded to contribute to the overall evidence to a much lesser extent than information gained from tests using standardised, validated and accepted guidelines.

 

 

in vivo studies using a non-physiological route of exposure

In vivo studies using the intraperitoneal route of administration are not to be considered equally relevant and adequate as studies using a physiological route of exposure. In contrast, an evaluation is to be conducted in-line with (i) the provisions laid down in regulation (EC) 1272/2008 (ii) the test guidelines specified in Article 13(3) of the REACH regulation and (iii) ECHA guidance (Guidance on Information Requirements and Chemical Safety Assessment, Chapter R.4):

 

i.             the CLP Regulation states in Annex I: 3.5.2.3.9 “The relevance of the route of exposure used in the study of the substance compared to the most likely route of human exposure shall also be taken into account.”. For industrial chemicals, the inhalation, dermal and oral route are considered relevant for human exposure. The intraperitoneal injection may only be relevant for substances used in pharmaceutical applications, which is clearly not the case for inorganic vanadium substances. Consequently, only studies via physiological routes should be considered relevant as foreseen in the CLP Regulation.

 

ii.            all recent OECD Test Guidelines for in vivo genetic toxicology testing (OECD 474, 475, 483, 488 and 489) state in section “Administration of doses” that “Intraperitoneal injection is generally not recommended since it is not an intended route of human exposure, and should only be used with specific scientific justification.”. As stated above, the intraperitoneal route may be relevant for pharmaceutical applications, but not for industrial chemicals. None of the studies using the intraperitoneal (IP) route described in the CLH report provides a justification for this route of exposure. Consequently, these studies should be considered non-compliant with the validated and accepted test guidelines and not relevant for the purposes of classification.

 

iii.           the ECHA Guidance on IR & CSA, Chapter R.4: Evaluation of available information (2011) provides guidance on the reliability rating of studies (Chapter 4.2, page 3). Studies are considered not reliable in case organisms/test systems were used which are not relevant in relation to the exposure (e.g. non-physiological pathways of application).

 

In a weight-of-evidence approach, studies using a non-physiological route of administration should therefore be considered to contribute to the overall assessment to a much lesser extent than information obtained respecting the relevant test guidelines and guidance documents.

 

References:

HERAG (2007) Fact sheet 05 - Mutagenicity. EBRC Consulting GmbH / Hannover /Germany. August 2007. [www.metalsriskassessment.org]

 

Justification for selection of genetic toxicity endpoint

Data of the genetic toxicity are available for tri,- tetra,- and pentavalent substances (including V2O3, VOSO4, V2O5).

Justification for classification or non-classification

Based on the available weight-of-evidence, and considering guideline-conform studies conducted under GLP both in vitro as well as in vivo, divanadium pentaoxide should be considered void of genotoxicity.

Divanadium pentaoxide as well as other representative tri-, tetra- and pentavalent vanadium substances have been verified unequivocally in vitro to be void of gene mutation activity both in bacterial as well as mammalian in vitro cell systems.Testing in bacteria reverse mutation assays, although of limited relevance for metals (HERAG, 2007), also yielded negative results. Negative as well as positive results were obtained in clastogenicity assays in vitro. However, these in vitro studies of effects upon eukaryotic cells in vitro employed high concentrations producing significant levels of cytotoxicity and only weak genotoxic responses. A central issue that requires resolution is whether such in vitro results have physiological relevance by virtue of the mechanisms involved or the concentrations required to produce effects. For example, induction of genotoxic effects in cultured cells at dissolved vanadium concentrations in the µM or mM range would have limited relevance to in vivo exposures wherein the concentration of vanadium available for transfer to the soft tissues is in the nM range or lower.

In vitro assays for cytogenetic changes conducted with metals are quite often positive, even with some indications that aneuploidy induction may be common. Conversely, upon in vivo testing, most metals with very few exceptions yield negative results, leading to the conclusion that the in vitro results should not be over-interpreted since unphysiologically high concentrations as obtainable in in vitro test systems do not mirror in vivo physiological conditions. This is verified by (i) a study conducted by NTP (2002) involving in vivo exposure to a V2O5aerosol for 3 months, which failed to elicit any clastogenic effects in mice, and (ii) another reliable GLP-conform study, in which V2O5, administered orally by gavage to male rats up to the maximum tolerated dose of 120 mg/kg/day, did not induce micronuclei in polychromatic erythrocytes of the bone marrow.

V2O5may express some clastogenicity in vitro, this occurs only at unphysiologically high concentrations and via mechanisms that appear to lack physiological relevance, and is not mirrored in corresponding in vivo (RL=1) assays. Based on the available weight-of-evidence, and considering guideline-conform studies conducted under GLP both in vitro as well as in vivo, V2O5 should be considered void of genotoxicity. As a consequence, the current harmonised classification of V2O5 according to Annex VI of Regulation (EC) 1272/2008 with respect to mutagenicity is not supported by the available experimental data. Thus, according to EC Regulation 1272/2008, V2O5should not be considered to have a mutagenic potential, and hence classification or labelling should not be required.