Registration Dossier

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

59.37 mg/m³

Most sensitive endpoint:

repeated dose toxicity

Route of original study:

Oral

DNEL derivation method:

ECHA REACH Guidance

Overall assessment factor (AF):

75

Modified dose descriptor starting point:

NOAEC

Value:

4 453 mg/m³

Explanation for the modification of the dose descriptor starting point:

See discussion section (Hazard via inhalation route: systemic effects following long-term exposure)

AF for dose response relationship:

1

Justification:

Default ECHA AF; NOAEL from a well-conducted combined repeated dose and reproductive/developmental toxicity study.

AF for differences in duration of exposure:

6

Justification:

Default ECHA AF for subacute (28-day) to chronic extrapolation. Male animals were dosed for 28-days in total, while low- and mid-dose females received treatment for a longer period of time (incorporating the gestation period and proceeding up until postpartum day 5)

AF for interspecies differences (allometric scaling):

1

Justification:

Default ECHA AF of 4 for rat for toxicokinetic differences in metabolic rate (allometric scaling) is not required

AF for other interspecies differences:

2.5

Justification:

Default ECHA AF for remaining toxicokinetic differences (not related to metabolic rate) and toxicodynamic differences

AF for intraspecies differences:

5

Justification:

Default ECHA AF for (healthy) worker

AF for the quality of the whole database:

1

Justification:

Default ECHA AF; the human health effects data are reliable and consistent, and confidence in the database is high. Read-across from the structurally-similar compound, palladium(II) dihydroxide was used to fill the reproductive/developmental and repeated dose toxicity (oral) endpoints. No AF is considered necessary for the use of read-across in this instance since the DNEL was derived on the basis of palladium metal itself

AF for remaining uncertainties:

1

Justification:

Not required

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

medium hazard (no threshold derived)

Most sensitive endpoint:

sensitisation (skin)

Hazard assessment conclusion:

medium hazard (no threshold derived)

Most sensitive endpoint:

sensitisation (skin)

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

16.84 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

Route of original study:

Oral

DNEL derivation method:

ECHA REACH Guidance

Overall assessment factor (AF):

75

Modified dose descriptor starting point:

NOAEL

Value:

1 263 mg/kg bw/day

Explanation for the modification of the dose descriptor starting point:

See discussion section (Hazard via dermal route: systemic effects following long-term exposure)

AF for dose response relationship:

1

Justification:

Default ECHA AF; NOAEL from a well-conducted combined repeated dose and reproductive/developmental toxicity study.

AF for differences in duration of exposure:

6

Justification:

Default ECHA AF for subacute (28-day) to chronic extrapolation. Male animals were dosed for 28-days in total, while low- and mid-dose females received treatment for a longer period of time (incorporating the gestation period and proceeding up until postpartum day 5)

AF for interspecies differences (allometric scaling):

1

Justification:

The default ECHA AF of 4 for rat for toxicokinetic differences in metabolic rate (allometric scaling) is considered unnecessary as the compound is inorganic and is consequently not metabolised to any relevant extent. Moreover, ECHA guidance notes that “allometric scaling is an empirical approach for interspecies extrapolation of various kinetic processes generally applicable to substances which are renally excreted”, while systemically available palladium is excreted predominantly via the biliary/faecal route

AF for other interspecies differences:

2.5

Justification:

Default ECHA AF for remaining toxicokinetic differences (not related to metabolic rate) and toxicodynamic differences

AF for intraspecies differences:

5

Justification:

Default ECHA AF for (healthy) worker

AF for the quality of the whole database:

1

Justification:

Default ECHA AF; the human health effects data are reliable and consistent, and confidence in the database is high. Read-across from the structurally-similar compound, palladium(II) dihydroxide was used to fill the reproductive/developmental and repeated dose toxicity (oral) endpoints. No AF is considered necessary for the use of read-across in this instance since the DNEL was derived on the basis of palladium metal itself.

AF for remaining uncertainties:

1

Justification:

Not required

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

high hazard (no threshold derived)

Most sensitive endpoint:

sensitisation (skin)

Hazard assessment conclusion:

high hazard (no threshold derived)

Most sensitive endpoint:

sensitisation (skin)

Hazard assessment conclusion:

medium hazard (no threshold derived)

Hazard via inhalation route: systemic effects following long-term exposure

Justification for route to route extrapolation

As no relevant data on effects of repeated exposure to palladium dichloride in humans or laboratory animals are available, route-to-route extrapolation to calculate an inhalation DNEL from a reliable combined repeated dose and reproductive/developmental oral toxicity study on another read-across category member, palladium(II) dihydroxide, was considered a suitable alternative (particularly as first pass effects are not expected to be significant for an inorganic compound).

The oral NOAEL for palladium(II) dihydroxide was 1000 mg/kg bw/day. This equates to NOAELs of 758 and 1263 mg/kg bw/day for palladium and palladium(II) dichloride, respectively (based on MWt ratios).

In the absence of data allowing quantitative comparison between absorption following oral and inhalation exposure, this derivation utilised the REACH default assumption that the absorption percentage for the oral route is half that of the inhalation route, and a default factor of 2 is proposed for absorption differences in the case of oral-to-inhalation extrapolation.

Corrected inhalatory NOAEC (worker, 8 h exposure/day) = oral NOAEL*(1/sRv[rat])*(ABS[oral-rat]/ABS[inh-human]) *(sRV[human]/wRV)

= 1263 mg/kg bw/day*(1/0.38 m3/kg bw/day)*(1/2)*(6.7 m3 [8h]/10 m3 [8h]) =1113 mg/m3

It is noted that the standard respiratory rate conversion figure (0.38 m3/kg bw/day) already incorporates a factor of 4 for allometric scaling from rat to human. An assessment factor (AF) for allometric scaling is not considered to be justified in this scenario, given that the metabolism of inorganic metal cations is conventionally assumed not to occur to any relevant extent. Moreover, ECHA guidance notes that “allometric scaling is an empirical approach for interspecies extrapolation of various kinetic processes generally applicable to substances which are renally excreted, but not to substances which are highly extracted by the liver and excreted in the bile. It appears that species differences in biliary excretion and glucuronidation are independent of caloric demand (Walton et al. 2001)” (ECHA, 2012a). Oral toxicokinetic studies have demonstrated that systemically available palladium is excreted predominantly via the biliary/faecal route.

It is therefore appropriate to increase the corrected inhalatory NOAEC by a factor of 4.

Dose descriptor starting point (after route to route extrapolation) = Corrected inhalatory NOAEC (worker, 8 h exposure/day)*4 =1113*4 = 4453 mg/m^{3 .}

Justification and comments

In an OECD Test Guideline 422 combined repeated dose and reproductive/developmental toxicity screening study in rats, involving the gavage administration of palladium dihydroxide for at least 28 days, the systemic NOAEL was the highest tested dose (1000 mg/kg bw/day). Although some treatment related microscopic findings (mucosal discoloration in the non-glandular stomach, ileum, cecum, colon and/or rectum) were noted at dose levels of 300 and 1000 mg/kg bw/day, these were considered to result from direct (local) contact with the test substance rather than systemic toxicity(Torok-Batho, 2015).

Hence, the systemic repeated dose NOAEL of 1000 mg/kg bw/day was taken as the critical point of departure for calculating the long-term systemic DNELs for palladium dichloride, and is considered protective of fertility and developmental toxicity.

The substance DNEL (59.37 mg/m³) equates to a palladium exposure of 35.63 mg/m³.

Hazard via inhalation route: systemic effects following acute exposure

Justification and comments

DNELs for acute toxicity should be calculated if an acute toxicity hazard, leading to classification and labelling (i.e. under EU CLP regulations) has been identified and there is a potential for high peak exposures (this is only usually relevant for inhalation exposures).

There are no data in relation to acute inhalation exposure to palladium dichloride. In an early acute oral toxicity study in rats with palladium dichloride dihydrate, the LD50 value was found to be approximately 576 mg/kg bw (479 mg/kg bw as palladium dichloride) (Holbrook et al., 1975). The compound is classified in Category 4 for acute oral toxicity according to EU CLP criteria. In a more recent guideline (OECD TG 401) acute oral toxicity study in rats with palladium dichloride, the LD50 value was determined to be >2000 mg/kg bw (Allen, 1994a). This suggests that the classification of palladium dichloride for acute oral toxicity, based on the early Holbrook et al. data, is a very health precautionary approach. An oral N(L)OAEL (for sub-lethal effects) could be modified into an inhalation N(L)OAEC using route-to-route extrapolation. However, ECHA (2012a) guidance on DNEL calculation notes that this “procedure introduces significant uncertainties especially in relation to what inhalation time-frame this extrapolated N(L)OAEC would represent, and the procedure is therefore discouraged”.

ECHA (2015a,b) guidance on requirements for acute toxicity testing notes that “inhalable particles…are generally smaller than 100 μm in diameter. Particles larger than 100 μm are less likely to be inhalable”. In a guideline (OECD TG 110) granulometry screening test, the proportion of palladium dichloride <100 μm was 91.8% (O’Connor, 2011). Dustiness testing, a more energetic particle size distribution measurement, with various palladium compounds (not including palladium dichloride) returned mass median aerodynamic diameter (MMAD) values in the range of 24.1-38.2 μm. An MMAD value <100 μm indicates that a significant proportion of a substance is likely to be inhalable. Respiratory tract deposition modelling with the dustiness data for each of the palladium species yielded output values in the range of 40.0-52.5, 0.12-0.41 and 0.08-0.64% for the nasopharyngeal (head), tracheobronchial (TB) and pulmonary regions of the respiratory tract, respectively. Hence, for all of the tested palladium species, very little airborne substance (<1%) is expected to deposit in the lower regions of the human respiratory tract, i.e. the TB or pulmonary regions via oronasal normal augmenter breathing. Similar behaviour is anticipated for palladium dichloride. Most of the inhaled fraction is likely to be retained in the head region and therefore would be cleared by ingestion, along with that deposited in the TB region, and oral bioavailability will again predominantly determine systemic uptake. Less than 1% is likely capable of reaching the alveoli. Consequently, inhalation is not anticipated to be a significant route of exposure for palladium dichloride.

Further, given that the long-term systemic inhalation DNEL is high (above 10 mg/m3), setting acute DNELs is unnecessary, based on the high-level principle referenced in ECHA (2012a). This criterion states that “As a rule of thumb, a DNELacute should be set for acutely toxic substances if actual peak exposure levels significantly exceed the long-term DNEL”. This is typically inferred to mean several fold exceedance for task-based (e.g. 15 minute TWA) situations. The foreseeable industrial situations are highly unlikely to result in airborne peak exposures well above 10 mg/m3as these would not be tolerated in the workplaces (due to the general standards applicable to control of particulates). Consequently, no worker-DNEL for acute systemic toxicity has been calculated.

“A qualitative risk characterisation for this endpoint could be performed for substances of very high or high acute toxicity classified in Category 1, 2 and 3 according to CLP…. when the data are not sufficiently robust to allow the derivation of a DNEL” (ECHA, 2012b). However, palladium dichloride is only classified in Category 4, so a qualitative assessment is not required.

Hazard via inhalation route: local effects following long-term exposure

Justification and comments

There are no data in relation to respiratory tract irritation or sensitisation in humans or laboratory animals. Consequently, no worker-DNELs for respiratory tract irritation/corrosion or sensitisation have been calculated.

However, according to ECHA (2012b) guidance (Part E), “since sensitisation is essentially systemic in nature, it is important for the purposes of risk management to acknowledge that skin sensitisation may be acquired by other routes of exposure than dermal. There is therefore a need for cautious use of known contact allergens in products to which consumers or workers may be exposed by inhalation”. Palladium dichloride is classified as a strong skin sensitiser (Category 1). Therefore, consider recommended Risk Management Measures/Operational Conditions (RMMs/OCs) in Table E.3-1 of ECHA (2012b).

Hazard via inhalation route: local effects following acute exposure

Justification and comments

There are no data in relation to respiratory tract irritation or sensitisation in humans or laboratory animals. Consequently, no worker-DNELs for acute local effects in the respiratory tract have been calculated.

However, according to ECHA (2012b) guidance “since sensitisation is essentially systemic in nature, it is important for the purposes of risk management to acknowledge that skin sensitisation may be acquired by other routes of exposure than dermal. There is therefore a need for cautious use of known contact allergens in products to which consumers or workers may be exposed by inhalation”. Palladium dichloride is classified as a strong skin sensitiser (Category 1). Therefore, consider recommended RMMs/OCs in Table E.3-1 of ECHA (2012b).

Hazard via dermal route: systemic effects following long-term exposure

Justification for route to route extrapolation

As no relevant data on effects of repeated exposure to palladium dichloride in humans or laboratory animals are available, route-to-route extrapolation to calculate a dermal DNEL from a combined repeated dose and reproductive/developmental oral toxicity study on a read-across category member, palladium(II) dihydroxide, was considered a suitable alternative (particularly as first pass effects are not expected to be significant for inorganic compounds).

The oral NOAEL for palladium(II) dihydroxide was 1000 mg/kg bw/day. This equates to NOAELs of 758 and 1263 mg/kg bw/day for palladium and palladium(II) dichloride, respectively (based on MWt ratios).

Estimation of dermal absorption is based on relevant available information (mainly water solubility, molecular weight and log Pow) and expert judgement. Palladium dichloride, with water solubility of 4.03 g/L (Gregory, 2014), may be able to cross the lipid-rich environment of the stratum corneum to a “moderate to high” extent (ECHA, 2014). In the light of the limited available experimental data, ECHA guidance indicates a default value of 100% dermal absorption (ECHA, 2014). However, guidance on the health risk assessment of metals indicates that molecular weight and log Pow considerations do not apply to these substances (“as inorganic compounds require dissolution involving dissociation to metal cations prior to being able to penetrate skin by diffusive mechanisms”) and tentatively proposes dermal absorption figures: 1.0 and 0.1% following exposure to liquid/wet media and dry (dust) respectively (ICMM, 2007). However, although palladium dichloride is not classified as a skin irritant, moderate skin irritation was observed in rabbits (Allen, 1994b). Such irritant potential may disrupt skin barrier function, facilitating dermal penetration. As such, it is considered health precautionary to take forward the ECHA default dermal absorption value of 100%.

In the absence of data allowing quantitative comparison between absorption following oral and dermal exposure, and noting that, in general, dermal absorption will not be higher than oral absorption, no default factor (i.e. factor of 1) is required for oral-to-dermal extrapolation, in line with ECHA (2012a) guidance.

Justification and comments

In an OECD Test Guideline 422 combined repeated dose and reproductive/developmental toxicity screening study in rats, involving the gavage administration of palladium dihydroxide for at least 28 days, the systemic NOAEL was the highest tested dose (1000 mg/kg bw/day). Although some treatment related microscopic findings (mucosal discoloration in the non-glandular stomach, ileum, cecum, colon and/or rectum) were noted at dose levels of 300 and 1000 mg/kg bw/day, these were considered to result from direct (local) contact with the test substance rather than systemic toxicity(Torok-Batho, 2015).

Hence, the systemic repeated dose NOAEL of 1000 mg/kg bw/day was taken as the critical point of departure for calculating the long-term systemic DNELs for palladium dichloride, and is considered protective of fertility and developmental toxicity.

The substance DNEL (16.84 mg/kg bw/day) equates to a palladium exposure of 10.10 mg/kg bw/day.

Hazard via dermal route: systemic effects following acute exposure

Justification and comments

DNELs for acute toxicity should be calculated if an acute toxicity hazard, leading to classification and labelling (i.e. under EU CLP regulations) has been identified and there is a potential for high peak exposures (this is only usually relevant for inhalation exposures).

No acute dermal toxicity study was conducted on palladium dichloride. In an early acute oral toxicity study in rats with palladium dichloride dihydrate, the LD50 value was found to be approximately 576 mg/kg bw (479 mg/kg bw as palladium dichloride) (Holbrook et al., 1975). The compound is classified in Category 4 for acute oral toxicity according to EU CLP criteria. In a more recent guideline (OECD TG 401) acute oral toxicity study in rats with palladium dichloride, the LD50 value was determined to be >2000 mg/kg bw (Allen, 1994a). This suggests that the classification of palladium dichloride for acute oral toxicity, based on the early Holbrook et al. data, is a very health precautionary approach.

Skin contact to palladium dichloride during production and/or use is expected to be negligible. Given that the long-term systemic dermal DNEL is relatively high (4.17 mg/kg bw/day), setting acute DNELs is unnecessary, based on the high-level principle referenced in ECHA (2012a). This criterion states that “As a rule of thumb, a DNELacute should be set for acutely toxic substances if actual peak exposure levels significantly exceed the long-term DNEL”. As consumer exposure is expected to be negligible, such peak exposures are not anticipated. Consequently, no worker-DNEL for acute systemic dermal toxicity has been calculated.

“A qualitative risk characterisation for this endpoint could be performed for substances of very high or high acute toxicity classified in Category 1, 2 and 3 according to CLP…. when the data are not sufficiently robust to allow the derivation of a DNEL” (ECHA, 2012b). However, palladium dichloride is only classified in Category 4, so a qualitative assessment is not required.

Hazard via dermal route: local effects following long-term exposure

Justification and comments

In a guideline (OECD TG 404) skin irritation study in rabbits, palladium dichloride produced moderate skin irritation according to the Draize criteria (Allen, 1994b). However, the compound is not classified for skin irritation under CLP. This conclusion was supported by a non-guideline in vivo skin irritation study on palladium dichloride (Campbell et al., 1975). Further, no dose-response data was available from which to derive a DNEL, therefore a qualitative assessment was considered appropriate. At worst this would be considered in the low hazard band according to ECHA (2012b) guidance.

Positive sensitisation responses to palladium dichloride have been observed following human patch testing (Cristaudo et al., 2005; Muris et al., 2014; Rebandel and Rudzki, 1990; Spiewak et al., 2014; Tillman et al., 2013; Yagami et al., 2014). Moreover, in a small number of limited in vivo skin sensitisation studies, palladium dichloride induced effects consistent with those of a strong sensitiser (Wahlberg and Boman, 1990, 1992; Schuppe et al., 1998). This compound is classified for skin sensitisation as Category 1, under CLP. At worst this would be considered in the high hazard band according to ECHA (2012b) guidance.

Therefore, consider recommended RMMs/OCs in Table E.3-1 of ECHA (2012b).

Hazard via dermal route: local effects following acute exposure

Justification and comments

In a guideline (OECD TG 404) skin irritation study in rabbits, palladium dichloride produced moderate skin irritation (Allen, 1994b). However, the compound is not classified for skin irritation under CLP. This conclusion was supported by a non-guideline in vivo skin irritation study on palladium dichloride (Campbell et al., 1975). Further, no dose-response data was available from which to derive a DNEL, therefore a qualitative assessment was considered appropriate. At worst this would be considered in the low hazard band according to ECHA (2012b) guidance.

Positive sensitisation responses to palladium dichloride have been observed following human patch testing (Cristaudo et al., 2005; Muris et al., 2014; Rebandel and Rudzki, 1990; Spiewak et al., 2014; Tillman et al., 2013; Yagami et al., 2014). Moreover, in a small number of limited in vivo skin sensitisation studies, palladium dichloride induced effects consistent with those of a strong sensitiser (Wahlberg and Boman, 1990, 1992; Schuppe et al., 1998). This compound is classified for skin sensitisation as Category 1, under CLP. At worst this would be considered in the high hazard band according to ECHA (2012b) guidance.

Therefore, consider recommended RMMs/OCs in Table E.3-1 of ECHA (2012b).

Hazard for the eyes

Justification and comments

According to ECHA guidance on the application of CLP criteria (ECHA, 2015), “substances can be predicted to be corrosive, if the pH is ≤ 2 or ≥ 11.5” and accordingly the classification “Eye Dam. 1 should be applied”. Further, in a guideline (US CFR) eye irritation study, palladium dichloride produced irreversible effects on the eye in rabbits (Hysell et al., 1974). Based on extreme pH (<2) and the results of the in vivo study, palladium dichloride is classified in Category 1 under EU CLP.

No dose-response data was available from which to derive a DNEL, therefore a qualitative assessment was considered appropriate. Substances classified for serious eye damage (Category 1 in CLP) should be allocated to the “moderate hazard band on the basis that exposure to such corrosives, eye damaging or irritant substances should be well-controlled”. Therefore, consider recommended RMMs/OCs in Table E.3-1 of ECHA (2012b).

Hazard assessment conclusion: