Registration Dossier

Toxicological information

Endpoint summary

Currently viewing:

Administrative data

Description of key information

Skin sensitisation (Landsteiner method, guinea pig): not sensitising

Respiratory sensitisation:

Based on a weight of evidence argument, aluminium oxide does not require classification as a respiratory sensitizer.

Key value for chemical safety assessment

Skin sensitisation

Link to relevant study records
Reference
Endpoint:
skin sensitisation: in vivo (non-LLNA)
Type of information:
experimental study
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline available
Principles of method if other than guideline:
Landsteiner-Draize Method. No more information.
GLP compliance:
no
Type of study:
other: Landsteiner / Draize method
Justification for non-LLNA method:
The study was conducted prior to the current requirement in Regulation (EC) 1907/2006 to perform a LLNA study (OECD 429) as the preferred in vivo skin sensitisation study.
Species:
guinea pig
Strain:
other: albino SPF
Sex:
male
Details on test animals and environmental conditions:
TEST ANIMALS
- Source: Central Institute for the Breeding of laboratory Animals, TNO, Zeist, The Netherlands
- Age at study initiation: young adult
- Weight at study initiation: 276 - 348 g
- Housing: animals were individually housed in suspended stainless steel cages, fitted with wire mesh floors and fronts
- Diet : all animals were fed stock diet (produced by Hope Farms, Woerden, The Netherlands), ad libitum. The composition of the diet and Al content are not provided (not a critical omission).
- Water: tap water ad libitum
- Acclimation period: 3 days

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 24 ± 1°C
- Humidity (%): 45 ± 5%
- Photoperiod (hrs dark / hrs light): Artificial light was used 12 hours daily from 6 a.m. to 6 p.m.

Randomization: Animals were randomly assigned to 4 treatment groups – two test groups and two control groups with 8 animals in each group.

Route:
intradermal
Vehicle:
not specified
Concentration / amount:
33.3%
Day(s)/duration:
10 single injections (3 times weekly for 3 weeks)
Adequacy of induction:
other: Preliminary observations showed that one intradermal injection of 0.1 mL of a 33.3% aqueous suspension induced a mild skin reaction.
No.:
#1
Route:
intradermal
Vehicle:
not specified
Concentration / amount:
33.3%
Day(s)/duration:
single injection
Adequacy of challenge:
other: Preliminary observations showed that one intradermal injection of 0.1 mL of a 33.3% aqueous suspension induced a mild skin reaction.
No. of animals per dose:
8
Details on study design:
RANGE FINDING TESTS: Preliminary observations showed that one intradermal injection of 0.1 mL of a 33.3% aqueous suspension induced a mild skin reaction.

MAIN STUDY
A. INDUCTION EXPOSURE
- No. of exposures: 10 (test group), 1 (control group)
- Exposure period: 3 times weekly during 3 weeks
- Test groups: 0.05 mL of the 33.3% dilution at the first injection, 0.1 mL of the 33.3% dilution at all following injections (2 - 10)
- Control group: single injection of 0.05 mL of the 33.3% dilution
- Site: right flank
- Frequency of applications: 3 times weekly
- Duration: 3 weeks
- Concentrations: 33.3%

B. CHALLENGE EXPOSURE
- No. of exposures: 1 (test and control group)
- Day(s) of challenge: 2 weeks after the 10th injection of the induction phase
- Exposure period: single injection
- Test groups: 0.05 mL of the 33.3% dilution
- Control group: 0.05 mL of the 33.3% dilution
- Site: right flank
- Concentrations: 33.3%
- Evaluation (hr after challenge): 24
Positive control substance(s):
no
Positive control results:
not applicable
Reading:
1st reading
Hours after challenge:
24
Group:
negative control
Dose level:
33.3%
No. with + reactions:
8
Total no. in group:
8
Reading:
1st reading
Hours after challenge:
24
Group:
test group
Dose level:
33.3%
No. with + reactions:
8
Total no. in group:
8
Reading:
1st reading
Group:
positive control
Dose level:
not examined
No. with + reactions:
0
Total no. in group:
0

Test groups (8 animals)

AK 43/79

After the challenge dose of AK 43/79, all animals from this group developed a mild skin reaction.

 

AK 44/79

The challenge dose of AK 44/79 provoked mild reaction in 6 animals and moderate reaction in 2 animals.

 

Control group (8 animals)

AK 43/79 (controls)

After the challenge dose of AK 43/79, the control animals showed a mild reaction in 7 and a moderate reaction in 1 animal. 

 

AK 44/79 (controls)

The challenge dose of AK 44/79 provoked 7 mild reactions and 1 moderate reaction.

 

No significant differences were reported between both test compounds and the control animals with respect to the degree and incidence of erythema and oedema.

Individual data are provided in Table 1 and 2 of the report.

 

Table 1. Individual positive reactions (marked with + signs) observed during the induction and challenge period with AK 43/79

Test animals

Control animals

No.

Individual direct reaction

challenge

No.

Reaction to injection at the same time as the challenge

1

2

3

4

5

6

7

8

9

10

2278

+

+

+

+

+

+

+

+

+

++

+

2286

+

2279

+

+

+

+

+

+

+

++

++

++

+

2287

+

2280

+

+

+

+

+

+

+

+

+

++

+

2288

+

2281

+

+

+

+

+

+

+

++

++

++

+

2289

++

2282

+

+

+

+

+

+

+

+

++

++

+

2290

+

2283

+

+

+

+

+

+

+

+

+

++

+

2291

+

2284

+

+

+

+

+

+

+

+

++

++

+

2292

+

2285

+

+

+

+

+

+

+

+

+

++

+

2293

+

2298

 

+

+

+

+

++

++

++

++

++

+

2306

+

2299

+

+

+

+

+

++

++

++

++

++

+

2307

+

2300

+

+

+

+

+

+

++

+

++

++

++

2308

++

2301

+

+

+

+

+

+

++

++

++

++

+

2309

+

2302

+

+

+

+

+

++

++

++

++

++

+

2310

+

2303

+

+

+

+

+

++

++

++

++

++

+

2311

+

2304

+

+

+

+

+

+

++

++

++

++

+

2312

+

2305

+

+

+

+

+

++

++

++

++

++

++

2313

+

 Degree of reaction: + = mild; ++ = moderate

Interpretation of results:
other: CLP/EU GHS criteria not met, no classification required according to Regulation (EC) No 1272/2008
Conclusions:
Based on the reaction to the challenge dose, it can be concluded that Al oxide AK 43/79 and Aluminium oxide AK 44/79 has no sensitisation potential under the experimental conditions.
Executive summary:

The skin sensitisation potential of two samples aluminium oxide, namely aluminium oxide TBH: AK 43/79 and aluminium oxide TOF: AK 44/79, was assessed in guinea pigs (male albino SPF, 8 animals in each group) using the Landsteiner/Draize method (Central Institute for Nutrition and Food Research, Germany) at the request of Degussa AG, Germany (1979). Both compounds were administered by intra-dermal injections. A 33.3% aqueous suspension was used in both the induction and challenge phases. During the induction phase, the test animals received 10 intra-dermal injections of the test suspension, 3 times per week over a 3 week period. The test suspensions were administered to different shaved spots on the right flank of the animals within an area of 3 x 4 cm. The injected volume was 0.05 mL for the first injection and 1.0 mL for subsequent. The control group received a single injection during this phase. The injection sites were examined 24 hours after the injection and the diameter, colour and thickness of any lesions were used as criteria for the intensity of the reaction. In the induction phase,following 1 to 7 injections of AK 43/79, all animals showed mild reactions. Two animals showed moderate reactions after the 8th injection, an additional 2 animals showed moderate reactions after the 9th injection and all 8 animals showed a moderate reaction after the 10th injection. Data were provided on any skin reactions on the single injection received by the control animals. For AK 44/79, all reactions were mild until after the 6th injection when 5 animals showed a moderate reaction. All animals showed moderate reactions

after the 7th to 10th injections.  Two weeks after the last injection, guinea pigs from both test groups and the control group received the challenge dose in the amount of 0.05 mL per animal. The reaction sites were examined 24 hours after the injection.After the challenge dose of AK 43/79, all animals exposed during the induction period developed a mild skin reaction. A mild reaction was also found in 7 and a moderate reaction in 1 of the animals in the control group. The challenge dose of AK 44/79 provoked a mild reaction in 6 animals and moderate reaction in 2 animals exposed during the induction period. In the control animals, 7 mild reactions and 1 moderate reaction were observed. No significant differences were observed between the test and control animals with respect to the degree and incidence of erythema and oedema. Under the conditions of this test, aluminium oxide AK 43/79 and aluminium oxide AK 44/79 are not skin sensitizers (Landsteiner/Draize test, guinea pig).

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:

Human studies:

One small study of relatively weak design (Hosovski et al. 1998) observed higher rates of skin hypersensitivity, increased aluminium dermal responsiveness and increased serum IgA and IgG levels among aluminium smelter workers compared with an “unexposed” control group. Thomas et al. (2003) found no positive patch-test reactions to ceramic alumina disks in a sample of 251 consecutive cases visiting a dermatology clinic. Although limited in amount, the evidence from human studies supports a low sensitisation potential for aluminium oxide on dermal exposure.

 

Three new case reports and studies presented at the scientific conferences (available ahead of publication) have reported single cases of sensitization which might be due to Al after exposure to Al-containing deodorants (Garg et al., 2011; Gies et al., 2010 or Watson et al., 2012).

Garg et al. (2011) described in brief the case of a 28-year-old woman who presented at the hospital with eczema in both axillae. The patient complained her condition occurred after the use of deodorants that contained aluminium chlorohydrate. Patch testing was performed to the British Contact Dermatitis Society (BCDS) baseline series (the series includes approximately 30 substances and mixtures which account for around 85% of all contact dermatitis reactions), fragrance series and AlCl3. Allergens were supplied by Trolab®, Reinbeck, Germany and Chemotechnique Diagnostics®, Vellinge, Sweden. The results were scored and evaluated according to the International Contact Dermatitis Research Group (ICDRG) criteria. The patient had the 3+ reaction with 2% AlCl3 (in petrolatum (pet.), but there was no skin reaction after contact with the 8 mm Aluminium Finn Chambers® (Epitest Ltd Oy, Tuusula, Finland). In addition, the patient responded with a 2+ reaction to quaternium-15 (1% in pet.) at day 2 and at day 4 she had a 1+ reaction to benzophenone 4 (10% in pet.), farnesol (2.5% pet.) and citral (1% pet.). After she avoided all Al-containing deodorants, her eczema cleared completely. Limited details provided by the authors regarding the patient’s overall health, dietary habits, possible occupational exposures or other factors precluded rigorous evaluation of the case. It should be noted that together with Al salts these deodorants contained a mixture of other chemicals which might be responsible for or contribute to the patient’s complaints of local irritation and sensitization. In the absence of the positive reaction to the Al in Finn Chambers and the established patch testing for multi-chemical sensitivity, the clinical relevance of the reported Al allergy is unclear. A Reliability Score is 3 (not reliable) for this weight of evidence study.

Watson et al. (2012) carried out a clinical study to evaluate whether the use of Al- based antiperspirants while receiving external beam radiotherapy for stage 0, I, or II breast cancer could exacerbate local irritation to axilla skin. A total of 198 participants were randomized to either the experimental group (antiperspirant) or a control group (standard care-wash only). The skin reactions in both groups were measured weekly and at two weeks after cessation of treatment using the National Cancer Institute Common Toxicity Criteria Adverse Events (version 3) toxicity grading criteria. The skin reaction data were analyzed using the generalized estimating equation. No statistically significant differences were observed in axilla skin reactions between the antiperspirant group and the control group. The authors concluded that using antiperspirant routinely during external beam radiotherapy for Stage 0, I, or II breast cancer had no influence on the intensity of the skin reaction. However, the full report is currently not available for review thereby precluding an assessment of reliability. A Reliability Score is 4 (not assignable) for this weight of evidence study.

The evidence from human studies supports a low sensitisation potential for aluminium oxide on dermal exposure.

 

Animal studies:

Results from the non-guideline compliant study (based on the Landsteiner/Draize test) on the sensitisation potential of aluminium oxide (Degussa AG, 1979) indicate that aluminium oxide is not a skin sensitizer. Results from studies of the dermal irritation properties of aluminium oxide (Degussa AG, 1977; IUCLID, 2000) support a low sensitisation potential for this substance. Data on the physico-chemical characteristics of aluminium oxide and toxicokinetics, and evidence from sufficiently valid results for a closely related compound (aluminium hydroxide, LAB Research Ltd., 2010) and more bioavailable aluminium compounds (aluminium chloride hexahydrate, Basketter et al., 1999; aluminium chloride, Magnusson and Kligman, 1969) support this evaluation (a detailed rationale and justification for the analogue read-across approach is provided in the technical dossier (see IUCLID section 13.2)).

The evidence from animal studies supports a low sensitisation potential for aluminium oxide on dermal exposure.

Overall, the weight of evidence supports a low sensitisation potential for aluminium oxide.

Based on the available data (Degussa AG, 1979), aluminium oxide does not require classification as a skin sensitizer.

 

A study was performed in Guinea pigs (Dunkin Hartley (LAL/HA/BR) using the Magnusson and Kligman method (LAB Research Inc., 2010). The study design was based on OECD TG # 406 (17 July 1992), Commission Regulations (EC) No 440/2008 of 30 May 2008; B.6; and the US EPA OPPTS 870.2600 (EPA 712-C-03-197, March 2003). Methylcellulose (1%), selected based on results from a Preliminary Compatibility Test, was used as the vehicle in this study. Based on the preliminary dose range finding study, 1% (w/v) was used for a first induction stage by intradermal administration. This consisted of three injections to both left and right flanks: an injection with 0.10 mL of Freund's Complete Adjuvant mixed with physiological saline (1:1 v/v); an injection with 0.10 mL of the test item in 1% methylcellulose at the selected concentration; and an injection with 0.10 mL of test item at the appropriate concentration in a 1:1 (v/v) mixture of Freund's Adjuvant and physiological saline. The animals in the control group received three similar injections to each side with the omission of the test item. Again based on the results of a dose range finding study, 100% (w/v) was used for a second induction stage by dermal application. 0.5 mL of the suspension was applied with occlusion for 48 hours. Two weeks after the last induction exposure, two concentrations were used for the occlusive epicutaneous challenge exposure: 0.5 mL of 75% (w/v) suspension was applied to the left flank of the animals and 0.5 mL of 37.5% (w/v) suspension was applied to the right flank. The test item was applied to the flanks of the test and control animals using a 5x5 cm sterile gauze patch saturated with the test item. The patches remained in place, occluded, for 24 hours. After patch removal, residual test item was removed with a swab and observations were made at 24 and 48 hours. No irritation effects (scored according to Draize, 1977) were observed during the dose-range finding study or the induction exposures. In the test group, no positive responses were observed in the treated animal (n=10) with either the 75% (w/v) or 37.5% (w/v) formulations. No positive responses were observed on challenge exposure in the control animals (n = 5). In summary, the Guinea-Pig Maximisation Test was used to determine the skin sensitisation potential of the test item, aluminium hydroxide.Challenge with the test item produced no positive responses in the previously sensitized test animals or in the control animals. The incidence rate was 0% and the net score 0.00. Thus, it was shown that, under the conditions of this test, aluminium hydroxide had no detectable sensitisation potential and does not meet EU criteria for classification for sensitisation.

 

The skin sensitisation potential of two samples aluminium oxide, namely aluminium oxide TBH: AK 43/79 and aluminium oxide TOF: AK 44/79, was assessed in guinea pigs (male albino SPF, 8 animals in each group) using the Landsteiner/Draize method (Degussa, 1979). Both compounds were administered by intra-dermal injections. A 33.3% aqueous suspension was used in both the induction and challenge phases. During the induction phase, the test animals received 10 intra-dermal injections of the test suspension, 3 times per week over a 3 week period. The test suspensions were administered to different shaved spots on the right flank of the animals within an area of 3 x 4 cm. The injected volume was 0.05 mL for the first injection and 1.0 mL for subsequent. The control group received a single injection during this phase. The injection sites were examined 24 hours after the injection and the diameter, colour and thickness of any lesions were used as criteria for the intensity of the reaction. In the induction phase, following 1 to 7 injections of AK 43/79, all animals showed mild reactions. Two animals showed moderate reactions after the 8th injection, an additional 2 animals showed moderate reactions after the 9th injection and all 8 animals showed a moderate reaction after the 10th injection. Data were provided on any skin reactions on the single injection received by the control animals. For AK 44/79, all reactions were mild until after the 6th injection when 5 animals showed a moderate reaction. All animals showed moderate reactions after the 7th to 10th injections. Two weeks after the last injection, guinea pigs from both test groups and the control group received the challenge dose in the amount of 0.05 mL per animal. The reaction sites were examined 24 hours after the injection. After the challenge dose of AK 43/79, all animals exposed during the induction period developed a mild skin reaction. A mild reaction was also found in 7 and a moderate reaction in 1 of the animals in the control group. The challenge dose of AK 44/79 provoked a mild reaction in 6 animals and moderate reaction in 2 animals exposed during the induction period. In the control animals, 7 mild reactions and 1 moderate reaction were observed. No significant differences were observed between the test and control animals with respect to the degree and incidence of erythema and oedema. Under the conditions of this test, aluminium oxide AK 43/79 and aluminium oxide AK 44/79 are not skin sensitizers (Landsteiner/Draize test, guinea pig).

 

Basketter et al. (1999) investigated the allergenic potential of 13 metal salts including aluminium chloride hexahydrate (99% purity) in the Local Lymph Node Assay (LLNA). Groups of 4 CBA/Ca mice (7 to 12 weeks of age) were treated with 25 μL of substance or with an equal volume of the vehicle alone, on the dorsum of both ears. The mice were treated once daily for 3 days. Two days later, the mice were injected with 250 μL of phosphate buffered saline (PBS) with 20 μCi of tritiated thymidine (2 Ci mmol-l). The mice were killed 5 hours later and a single-cell suspension of lymph node cells was prepared by mechanical disaggregation. A substance was considered a skin sensitizer the proliferation in the lymph nodes of treated mice was 3-fold or greater than that in the concurrent vehicle-treated controls. Aluminium chloride hexahydrate administered in petrolatum (vehicle) at test concentrations of 5.0%, 10.0% and 25.0% did not induce a lymph node proliferation response compared to concurrent vehicle-treated controls, and therefore the response was judged as negative. Although this study did not provide sufficient detail in the description of the study and its outcome (Klimisch score 3), the results add to the weight of evidence for a low sensitisation potential of the target substances.

 

Additional information

Sun et al. (2011) studied the hyper-susceptibility of guinea pigs in the sensitization assay after intravenous injection of an aluminum hydroxide-containing adjuvant. Forty-two guinea pigs were divided into seven groups (6 animals per group) and sensitized 12 days later with 1.5, 4, 7, 10 or 13 mg/mL Al(OH)3 by i.v. injection. The vehicle control group was injected with an equivalent volume of physiological saline and the positive control group was treated with 1% BSA. Animals were observed for clinical reactions, as well as the time of appearance and resolution of the response. Another 24 guinea pigs were divided into four groups (6 animals per group) and intra-dermally injected with antisera and 48 h later they were given i.v. injections of 1.5 or 4 mg/mL of Al(OH)3, physiological saline and 1% BSA + 1% Evans blue (v/v = 1:1), respectively, for excitation.The sizes of blue spots on inner layers of guinea pig skin were measured 30 min after excitation.The results showed that the i.v. administration of Al(OH)3 at a concentration of less than 4 mg/mL caused no anaphylactic reaction in guinea pigs, while treatment with 7-10 mg/mL caused strong anaphylactic shock and treatment with 13 mg/mL caused extreme reactions. However, in the passive cutaneous anaphylaxis (PCA) test, neither of Al(OH)3 at concentrations of 1.5 or 4 mg/mL caused anaphylaxis. The authors concluded that Al(OH)3at a concentration of less than 4 mg/mL failed to elicit any sign of anaphylactic reactions. The type of test compound used in the study is not clear. Currently, only the abstract is available for review thereby precluding an assessment of reliability. A Reliability Score is 4 (not assignable).

 

Overall,the weight of evidence supports a low sensitisation potential for aluminium oxide on dermal exposure.


Respiratory sensitisation

Link to relevant study records
Reference
Endpoint:
respiratory sensitisation: in vivo
Type of information:
experimental study
Adequacy of study:
supporting study
Study period:
2008
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Basic data given
Qualifier:
no guideline available
Principles of method if other than guideline:
Examination of the absolute and relative numbers of different cell types and specific biochemical parameters in fluid obtained from bronchoalveolar lavage of animal lungs (BALF) following either intratracheal or inhalation exposure to a substance can provide information on the nature of the reaction of the tissue, i.e the mechanism of action. This information is useful to support and interpret histopathological observations and/or external observations of respiratory symptoms or changes in lung function. In this study, histopathology and BALF analyses were done in the the presence and absence of ovalbumin, a known inducer of eosinophilic inflammation, in order to provide further information on the nature of the mechanism of action (allergenic versus non-specific irritative).
GLP compliance:
no
Species:
mouse
Strain:
ICR
Sex:
male
Details on test animals and environmental conditions:
TEST ANIMALS
- Source: Charles River Japan, Inc. (Kanagawa, Japan)
- Age at study initiation: 5 weeks at arrival; 6 weeks at testing
- Weight at study initiation: 307-315 g
- Housing: Plastic cages placed in a “conventional” room. Bedding with soft wood chips not otherwise specified.
- Diet: commercial diet CE-2 obtained from CLEA Japan Inc., Tokyo, Japan.
- Water: ad libitum
- Acclimation period: 1 week

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 23 °C
- Humidity (%): 55-70%
- Air changes (per hr): not reported
- Photoperiod (hrs dark / hrs light): Artificial light 12 hours daily from 6 a.m. to 6 p.m.

Justification of species and strain: ICR mice were chosen based on the results of an earlier study (Ichinose T et al., 2003; Toxicol Appl Pharmacol 187: 29-37) that showed that this species is “moderately” responsive to airway inflammation on exposure to OVA by intratracheal instillation (IT).
Route of induction exposure:
other: Intratracheal instillation
Route of challenge exposure:
other: not applicable
Vehicle:
other: Normal saline
Concentration:
Negative vehicle control (saline),
OVA alone,
Asian SD alone,
Arizona SD alone,
SiO2 alone,
Al2O3 alone,
Asian SD + OVA,
Arizona SD + OVA,
SiO2 + OVA,
Al2O3 + OVA.

Concentration: 0.1 mg per mouse

Deposition efficiencies were calculated using a tidal volume of 0.15 mL/mouse and a breathing rate of 200 breaths per minute. 0.1 mg/mouse was reported to be 111 times the weekly amount of particulate matter that would be deposited in the alveoli assuming 3% deposition (based on the ICRP model at 5.5 micron particle size) at an exposure level of 0.1 mg/m3.
No. of animals per dose:
16 animals/group
- 8 animals used for pathology and 8 animals for BALF analyses.
Details on study design:
Preparation of Test Solution:
In the absence of OVA, 5 mg of particulate was suspended in 2.5 or 5 mL of vehicle and sonicated for 5 minutes while being cooled (temperature not specified)
For the combined OVA and particulate exposure, the OVA (100µg) was dissolved in 10 mL or 5 mL of saline vehicle. 2.5 mL of the OVA solution was then mixed with 2.5 mL of the particle suspension.

Administration of Test Solution:
Volume: 0.1 mL
Method: The suspension was intratracheally administered under anaesthesia (4% halothane, Takeda Chemical, Osaka, Japan) using a polyethylene tube.
The animals were dosed 4 times in total. The interval between doses was 2 weeks.

Further Study details:
The animals were killed by exsanguination one day after the last intratracheal instillation when anaesthetised using an i.p. injection of pentobarbital. At this time, the animals were approximately 12 weeks of age.

Observations:
Pathology of Lung Tissue:
Pathologic examinations were done on the lung tissue from 8 out of 16 mice/group. The lungs were fixed with 10% neutral phosphate-buffered formalin, stained with haematoxylin and eosin (H&E) to assess infiltration of eosinophils and lymphocytes and also, to assess proliferation of goblet cells, periodic acid-shiff (PAS). The evaluation was done by two pathologists independently.

Bronchoalveolar fluid (BALF) analyses:
Cell counts:
The BALF of the remaining 8 mice was evaluated for cell counts. Slides were prepared using a Cytopsin (Sakura Co., Ltd., Tokyo, Japan) and were stained with Diff-Quik. 300 cells were counted.

Levels of cytokines:
The cytokines interleukin-5 (IL-5), IL-12, interferon-gamma (IFN-γ), tumour necrosis factor-alpha (TNF-α), macrophage inflammatory protein-1alpha (MIP-1α), IL-13, and eotaxin were measured using ELISA.

Levels of lactate dehydrogenase (LDH):
Measured using a lactate dehydrogenase C II-Test Wako from Wako Chemicals Ltd. (Osaka, Japan).

Antigen (OVA)-specific and total IgE and IgG1 antibodies:
The antibodies were measured using commercial ELISA kits.

Statistical Analyses:
Group differences were assessed using ANOVA and the Fisher’s projected least significant differences (P/SD) test.
Challenge controls:
Not applicable.
Positive control substance(s):
none
Negative control substance(s):
not specified
Results:
Pathology of Lung Tissue:
Al2O3:
The animals treated with aluminium oxide did not show a significantly increased proliferation of goblet cells or infiltration of lymphocytes when compared with the saline control.

OVA + Al2O3:
Lymphocyte infiltration was higher in the OVA+Al2O3 group compared with the control (p<0.001), and also when compared with OVA alone (p<0.01) and with Al2O3 alone (p<0.001).

Eosinophil infiltration was higher in the OVA+Al2O3 group compared with the control (p<0.01) and also OVA+Al2O3 compared with Al2O3 alone (p<0.01).

Goblet cell proliferation was higher in the OVA+Al2O3 group compared with the control (p<0.001), OVA+Al2O3 compared with OVA alone (p<0.05) and also when compared with Al2O3 alone (p<0.001).

Among the groups exposed to the particulates alone, the greatest changes were observed in the SiO2-treated group. Among the groups treated with OVA+particulates, the greatest changes were observed in the OVA+SiO2 group followed by the OVA+Arizona SD group, the OVA+Asian SD group and last the OVA+Al2O3 group.

Bronchoalveolar Lavage Fluid (BALF):
Cell counts
Total cells:
Among the groups exposed to particulates alone, only the SiO2-treated group had levels significantly greater than the control.
In the combined exposure groups, OVA+Asian SD, OVA+Arizona SD, and OVA+SiO2 had levels greater than the control (p<0.001). Total cells in the OVA + Al2O3 group were also significantly greater than the control but not to the same extent (p<0.05). No groups had cell counts greater than that observed in the group with OVA alone. The highest numbers were observed in the OVA+SiO2 group.

Macrophages:
Among the groups exposed to particulates alone, only Arizona SD and SiO2 caused significant increases in macrophages compared with the control (p<0.01 and p<0.001, respectively).
Among the combined exposure groups, OVA + Asian SD, OVA + Arizona SD, and OVA + SiO2 showed levels greater than the control. Numbers of macrophages in the OVA + Al2O3 group were greater than the control but not to the same extent (p<0.01). Only OVA+Arizona SD and OVA+SiO2 had greater numbers of macrophages than the OVA alone group (p<0.05 and p<0.001, respectively).

Eosinophils:
Particulates alone:
No effects observed.
Combined exposures:
OVA+Al2O3Numbers of eosinophils in the OVA+Asian SD (p<0.05), the OVA+Arizona SD (p<0.001) and the OVA+SiO2 (p<0.001) were greater than in the controls. None showed levels greater than OVA alone.

Neutrophils:
Particulates alone:
The SiO2 group was significantly greater than the controls (p<0.001)

Combined exposures:
OVA+Al2O3
Lymphocytes:
Levels in the Arizona SD group were significantly greater than the controls (p<0.01).
OVA+Asian SD (p<0.05) and OVA+Arizona SD (p<0.001) were significantly greater than the OVA only group.

Cytokines
IL-5
Significant elevations were observed for OVA+Arizona SD and OVA+SiO2 compared to the control and compared to OVA alone.

IL-6
A significant elevation was observed for OVA+SiO2 compared to OVA alone.

IL-12
The SiO2 group had significantly higher levels than the control (p<0.001). The OVA+SiO2 group had significantly higher levels than the control (p<0.001), the OVA alone group (p<0.001), and the SiO2 alone group (p<0.001). Levels in the OVA+Arizona SD group were marginally significantly greater than the control group (p<0.05).

IL-13
The OVA+SiO2 group had significantly higher levels than the control (p<0.001), OVA alone (p<0.001), and SiO2 alone groups.

IFN-γ
Asian SD alone (p<0.01), Arizona SD alone (p<0.01), SiO2 alone (p<0.01) and Al2O3 alone (p<0.05) had significantly higher levels than the control.

TNF-α
The Asian SD group (p<0.01), the SiO2 group (p<0.01) and the Al2O3 group (p<0.05) had significantly higher levels than the control. In the combined exposures, OVA+Asian SD and OVA+Arizona SD had higher levels than the controls. None of the particulate-treated groups had levels significantly higher than the OVA alone group.

Levels of lactate dehydrogenase (LDH):
Control: 8.48±1.38 IU/L
Asian SD: 5.56±0.26 IU/L
Arizona SD: 8.19±1.05 IU/L
SiO2: 9.99±0.75 IU/L
Al2O3: 6.17±0.77 IU/L
OVA: 5.09±0.67 IU/L
OVA+Asian SD: 5.98±0.65 IU/L
OVA+Arizona SD: 7.35±1.11 IU/L
OVA+SiO2: 31.18±8.97 IU/L
OVA+Al2O3: 6.37±0.42 IU/L

For the chemokine eotaxin, significant elevations were observed only in the OVA+SiO2 group. KC (keratinocyte chemoattractant) was significantly elevated relative to the control in the Arizona SD (p<0.01) and SiO2 (p<0.001) groups. In the combined exposure groups, the OVA+Arizona SD and OVA+SiO2 groups were significantly elevated relative to both the controls and the OVA alone groups. Levels of monocyte chemotactic protein-3 (MCP-3) in the OVA+Arizona SD and OVA+SiO2 groups were also significantly elevated relative to the control and the OVA alone group. Macrophage inflammatory protein-1α was significantly elevated in the Arizona SD and SiO2 groups relative to the controls and in the OVA+SiO2 group relative to the OVA alone group. OVA+Asian SD, OVA+Arizona SD and OVA+SiO2 had levels significantly greater than the control.

Antigen (OVA)-specific and total IgE and IgG1 antibodies:
Serum levels of OVA specific IgG1 antibodies were significantly elevated relative to levels in the OVA group alone in animals treated with OVA+Arizona SD and OVA+SiO2 (p<0.001). No OVA-specific IgE antibodies were detected. Levels of total IgE in serum in the particulate-treated groups did not differ significantly from levels in the group treated with OVA alone.

Positive control results:
Not applicable.
Negative control results:
Not applicable.
Interpretation of results:
other: CLP/EU GHS criteria not met, no classification required according to Regulation (EC) No 1272/2008
Conclusions:
When co-administered with OVA, all the tested materials showed an increase in eosinophils on H&E stained slides (OVA+SiO2>OVA+Arizona SD>OVA+Asian SD>OVA+Al2O3). Considering the results for Al2O3, specifically, in BALF, Al2O3 alone did not lead to significant increases in total cells, macrophages, eosinophils, neutrophils or lymphocytes. When Al2O3 was administered with OVA, small but statistically significant increases in total cells (p<0.05), macrophages (p<0.01) and lymphocytes (p<0.05) were observed relative to the control but not relative to the OVA alone group. Significant (p<0.05) increases in IFN-γ and TNF-α relative to the saline control were observed in BALF of Al2O3 treated animals. IL-5, IL-6, IL-12 and IL-13 did not show significant increases in the Al2O3-treated animals. In exposures combined with OVA, the only significantly elevated cytokine was IFN-γ.

Overall, the results from the study showed that allergic inflammatory effects of atmospheric dusts are likely due to SiO2. Al2O3 was the least inflammatory material tested and led to only weak effects on the mouse lung.
Executive summary:

Ichinose et al. (2008) studied allergic inflammation after intratracheal instillation of Asian sand dust,sand dust, amorphous silica and Al2O3in 6-week old male ICR mice. Four instillations were performed at 2-week intervals. There were ten groups of animals (n=16 in each). One of these groups received Al2O3(particle size 1~5 µm), a dose of 0.1 mg suspended in saline. The control group received saline only (0.1 mL). The animals were killed one day after the last instillation. Eight out of 16 animals in each group were used for pathologic examination. The lung samples were stained with haematoxylin and eosin to evaluate the degree of infiltration of eosinophils or lymphocytes in the airways, and with periodic acid-shiff to evaluate the degree of proliferation of goblet cells in the bronchial epithelium. The other 8 mice were used for examination of free cell counts (total and differential), determination of levels of lactate dehydrogenase (LDH), cytokines (Interleukins – IL-5, IL-6, IL-12, IL-13, interferon-IFN-gand tumor necrosis factor- TNF-a) and chemokines in bronchoalveolar lavage fluids (BALF), and also total IgE in serum using enzyme-linked immunosorbent assays (ELISA). In the group of mice exposed to Al2O3, the levels of eosinophil and lymphocyte infiltration in the submucosa and proliferation of goblet cells in the airways, the level of LDH, chemokines and interleukins, number of cells in BALF and the level of IgE in serum were not significantly different from those in the control mice. The results suggest that intratracheal administration of Al2O3does not produce allergic inflammatory effects in the lungs of mice.

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:

Based on an assessment of the available data, the weight of evidence does not support an important role for alumina in the development of potroom asthma. The weight of evidence supports an irritative effect of the fluoride component of pot emissions (Taiwo et al., 2006; Søyseth et al., 1992, 1994, 1997; Sorgdrager et al., 1998; Kongerud and Samuelsen, 1991). Work-related asthmatic symptoms in potroom workers may result, in part, from non-specific “irritant” and also from “allergic inflammation” reactions (Sariҫ et al., 1986; Mackay et al., 1990; Eklund et al., 1989). The relative contributions of these mechanisms are unclear and individuals with pre-existing bronchial hyper-responsiveness or a history of respiratory allergy may be at increased risk (Barnard et al., 2004; Kongerud and Samuelsen, 1991).

Some studies among aluminium welders have shown decrements in lung function (Fishwick et al., 2004; Abbate et al., 2003; Nielsen et al., 1993) and others have not (Sjogren and Ulfvarson, 1985; Kilburn et al., 1989). Sjogren and Ulfvarson (1985) observed an association between respiratory symptoms and exposures to ozone among welders. The small sample sizes, questionable comparability of referent groups (e.g. Kilburn et al., 1989) and possible (residual) confounding by smoking and irritative co-exposures such as ozone and other fume constituents are limitations of the available studies.

Results from a few case reports, notably Vandenplas et al. (1998) (also Park et al.,1996 and Burge et al., 2000) indicate that respiratory sensitization on exposure to poorly soluble aluminium substances can occur, but results from larger studies (Fritschi et al., 2001; Beach et al., 2001; Taiwo et al., 2006) suggest that it is rare. 

The only animal study identified that employed a guideline assay with results relevant to the respiratory sensitization endpoint and the target substances (the LLNA; Basketter et al., 1999) found that aluminium chloride, a more soluble aluminium substance, did not possess “significant ability to sensitize the skin”. As discussed by ECHA (2008):“In combination with other data it might be possible to conclude in a WoE assessment that chemicals that (at an appropriate test concentration and test conditions, i.e. skin penetration should have occurred) are negative in the LLNA, as well as being considered as not being skin sensitizers, can also be regarded as lacking the potential to cause allergic sensitisation of the respiratory tract.” (ECHA, 2008). The mouse study of Ichinose et al. (2008) supports a weak allergic inflammatory potential for Al2O3. In addition, the guideline compliant study by Lab Research Ltd. (2010) that tested the skin sensitisation potential of Al(OH)3reported negative results, thereby contributing to the weight of evidence supporting a low sensitizing potential for the poorly soluble target substances (See the companion RSI hazard assessment report on skin sensitisation, HAR_skin_sensit_100831_FINAL.doc).


Justification for classification or non-classification

The current weight of evidence supports a low skin and respiratory sensitizing potential for the poorly soluble substance aluminium oxide.

Overall, according to DSD (67/548/EEC) or CLP (1272/2008/EC) classification criteria for sensitisation, no classification is required.