Rationale and decision rules behind the ECETOC NanoApp to support registration of sets of similar nanoforms within REACH

Figures & data

Figure 1. Tiered approach for building and justification of sets of nanoforms.

Table 1. Properties taken into consideration for evaluation of similarity.

Properties considered in both Tier 1 and Tier 2	Additional data required in Tier 2	Additional data optional in Tier 3^a
Main constituents (CAS Number, relative content)	Reactivity, environmental dissolution, physiological dissolution AND in vitro toxicity
Crystallinity (crystalline forms, relative content)	Literature data on comparative toxicity OR reactivity, environmental dissolution and physiological dissolution.
Impurities/non-covalently bound additives (relative content)	Environmental dissolution, physiological dissolution, reactivity, in vitro toxicity, AND environmental dispersion stability
Covalently bound surface treatments (CAS Number, relative content, CLP)	Environmental dissolution, physiological dissolution, AND reactivity. AND in vitro toxicity if surface treatments are not of identical chemistry.
CLP based in chemical composition
Particle size distribution (d50 and d90)	Environmental dissolution, physiological dissolution, reactivity, in vitro toxicity, AND environmental dispersion stability	In vivo toxicity and biokinetics
Shape category, aspect ratio, assembly structure, and length (d90, only for elongated NFs)	In vitro toxicity
Mass specific surface area	Environmental dissolution, physiological dissolution, reactivity, in vitro toxicity, AND environmental dispersion stability
Production process (in relation to environmental dispersion stability)	Environmental dispersion stability and dustiness

^aA Tier 3 based on an in vivo study is only optionally presented, as a temporary solution, for special cases in which existing in vitro studies do not provide reliable predictions. As an alternative to performing such in vivo study, the sets can also be justified based on preselection of the worst-case nanoform in the set for further regulatory testing (cf. chapter 8).

Table 2. Similarity rules considered within Tier 1 and Tier 2.

Property (Descriptor)	Tier 1–Similarity rules and thresholds	Tier 2–Similarity rules, thresholds, and additional assays required for evaluation of similarity (Tier 3 in exceptional cases)
Composition (Relative content for all particle constituents, except for additives, impurities, and surface treatment agents)	Ratio^b of content for each constituent between the two nanoforms <1.5	Ratio of content for each constituent between the two nanoforms <3 Dissolution, reactivity and in vitro toxicity tests justify similarity (see descriptors and thresholds in Table 3).
Impurities/additives (identity^c, relative content)	Total impurities/additives <3% in both NFs OR ratio of each of the individual impurities/additives (above 2% in any of the two NFs) <2.	Total impurities/additives <10% in both NFs OR ratio of each of the individual impurities/additives (above 2% in any of the two NFs) <5. Dissolution, reactivity, in vitro toxicity, and environmental dispersion stability justify similarity (see descriptors and thresholds in Table 3).
Surface treatment (identity and relative content of the surface treatment agent)	Surface treatment agents are identical between the two nanoforms. The ratio of the surface treatment agents between the two nanoforms <1.5.	If surface treatment agents are identical between the two nanoforms: The ratio of the surface treatment agents between the two nanoforms <5. No differences in particle coverage are expected, considering the synthesis process. Dissolution, and reactivity justify similarity (see descriptors and thresholds in Table 3).
		If surface treatment agents are not chemically identical but similar between the two nanoforms: The ratio of the similar surface treatment agents between the two nanoforms <5. No differences in particle coverage are expected, considering the synthesis process. Dissolution, reactivity, and in vitro toxicity justify similarity (see descriptors and thresholds in Table 3).
CLP classification	Based on their chemical composition, the same CLP classification applies for the two NFs^a.	Same as Tier 1.
Size (d50 and d90 for the smallest and largest dimensions)	Ratio of each of these selected percentiles between the two nanoforms <1.5	If d50 for the two NFs ≥10 nm (or <10 nm when dissolution half-time in lung lining fluid is below 2 days): Ratio of each of these selected percentiles between the two nanoforms <3 (OR <5, only if the smallest nanoform will be used as a worst-case representative for testing to fill in data requirements in REACH). Dissolution, reactivity, in vitro toxicity tests, and environmental dispersion stability (see descriptors and thresholds in Table 3) justify similarity.
		If d50 for at least one of the NFs < 10 nm and at least one of the NFs has a dissolution half-time in lung lining fluid above 2 days: Ratio of each of these selected percentiles between the two nanoforms <3. Dissolution, reactivity, in vitro toxicity tests, and environmental dispersion stability justify similarity (see descriptors and thresholds in Table 3). In vivo short-term toxicity study justify similarity (see descriptors and thresholds in Table 4) OR the smallest nanoform will be used as a worst-case representative for testing to fill in data requirements in REACH.
Specific surface area (SSA)	Ratio of SSA between two nanoforms <1.5	Ratio of SSA between two nanoforms <3 (OR <5) with same additional data and conditions as described for size.
Morphology-shape (shape category and aspect ratio)	d90 value for the largest dimension of all NFs <5 µm. Ratio of aspect ratios between the two nanoforms <2. Platelets cannot be part of a set with other shape categories defined in the ECHA Guidance. For mixed-shape category NFs that contain platelets, the ratio of % of platelets between the two NFs <3.	d90 value for the largest dimension of all NFs <5 µm. Ratio of aspect ratios between the two nanoforms <5. Platelets cannot be part of a set with other shape categories defined in the ECHA Guidance. For mixed-shape category NFs that contain platelets, the ratio of % of platelets between the two NFs <5 (excluding from this condition mixed shape NFs that only contain up to 10% platelets). In vitro toxicity tests (see descriptors and thresholds in Table 3).
Morphology-crystalline form (relative content of each crystalline form^d, or amorphous from)	Ratio of each of the crystalline forms between the two nanoforms <1.5. If the user has data that shows that the toxicological profile of two crystalline forms present in the nanoforms is similar, these two crystalline forms will be considered as a single crystalline form for the rule above.	Ratio of each of the crystalline forms between the two nanoforms <4. Dissolution, and reactivity justify similarity (see descriptors and thresholds in Table 3). If the user has data that shows that the toxicological profile of two crystalline forms present in the nanoforms is similar, these two crystalline forms will be considered as a single crystalline form for the rules above.
Morphology-assembly structure (Yes/No; if Yes: selection from list: shell-like structures, multiwalled, nano-onions, multi-layers, hollow structures)	None of the nanoforms has an assembly structure OR Both of the nanoforms have the same assembly structure.	Same as for Tier 1
Aggregation state	Production processes for the two NFs under comparison are not expected to lead to different aggregation states.	Production processes for the two NFs under comparison may lead to different aggregation states. Environmental dispersion stability and dustiness justify similarity (see descriptors and thresholds in Table 3).

All of the properties included under Tier 1 should be considered similar in order to conclude that the NFs can be allocated into the same set of similar nanoforms based on their intrinsic similarity. When Tier 1 is not fulfilled for a given property, but Tier 2 conditions for such property are fulfilled, these are considered sufficient to justify the allocation of these NFs into the same set of similar nanoforms. When several conditions are included under a property and tier, all of them should be fulfilled unless otherwise indicated.

^aAssuming the CLP classification that would result from application of the rules in Regulation (EC) No 1272/2008 for classification of mixtures.

^bStrictly speaking, when we refer to the ratio of a descriptor between two nanoforms, we imply the following: $\frac{\max ({\mathit{DescriptorX}}_{i,{NF}_a},\mathrm{ }{\mathit{DescriptorX}}_{i,{NF}_b})}{\min ({\mathit{DescriptorX}}_{i,{NF}_a},\mathrm{ }{\mathit{DescriptorX}}_{i,{NF}_b})}<X$

where min and max denote minimum and maximum value, DescriptorXi,NFa refers, as an example in the first row, to the content in % of the constituent ‘i’ in the NFa, and DescriptorXi,NFa refers to the relative content (%) of that same constituent in the NFb. In this example, the condition should be fulfilled for each of the main constituents in NFa and NFb.

^cIf identity of impurities/additives is not reported, they will be assumed to be different in the two nanoforms, and only the thesholds on total impurities will apply.

^dThis criterion is waived if the user can provide evidence showing that no relevant differences in the toxicity of the two crystal forms can be expected (e.g. data for other NF or non-NF for those crystal forms).

Table 3. Descriptors and thresholds for justification of similarity between two nanoforms based on the Tier 2 additional assays.

Property (Descriptor)	Recommended assay protocol	Similarity rules and thresholds required for justification of similarity
Dissolution in two media: one physiological fluid, and one environmental medium (dissolution half-time per mass)	Static or dynamic conditions. ISO_19057:2017 and OECD TG draft on dissolution of NMs in aquatic media, and OECD TG development ongoing.	Ratio of each of the dissolution half-times between the two nanoforms <3. Except in the following two scenarios: A worst-case nanoform is selected for testing, see Table 2, in which the threshold for dissolution is <5. The two NFs have dissolution half-times higher than 1 year, in which their dissolution is considered similar regardless of the actual value.
Reactivity (% reactivity per mass versus the positive control; absolute reactivity values in the case of assays with very wide dynamic range -over 100- such as FRAS)	Standardized FRAS, EPR, or DCFH assay. ISO_18827:2017, Gandon et al. 2017, Boyles et al., in preparation	Reactivity (as % of positive control) of the two nanoforms differs in <10%. Only in the case of assays with wide dynamic ranges (>100): ratio of each of the crystalline forms between the two nanoforms <5.
In vitro toxicity (LOAEC or BMD in two in vitro studies, one in macrophages, one in another cell line).	Macrophage function assay (with inflammation markers) and cytotoxicity in another mammalian cell line. Wiemann et al. 2016; Kroll et al. 2011; ISO 19007:2018.	Ratio of the selected dose descriptor for each of the in vitro toxicity assays between the two nanoforms <3.
Dispersion stability in an environmental media (% of nanoform stable in suspension)	OECD TG 318 in only one environmental medium. Based on Abdolahpur Monick et al. 2018, medium with NOM, 1 mM Ca, pH 7 (i.e. higher water stability) is recommended.	Percentage (of initially dispersed particles) stable in suspension of the two nanoforms differs in <40%
Dustiness (respirable dustiness mass fraction in mg/kg; number based dustiness index)	Standardized method from EN 15051; EN_17199:2019; OECD TG development ongoing.	Ratio of the selected dustiness descriptor between the two nanoforms <5.

Table 4. Descriptors and thresholds for justification of similarity between two nanoforms based on the Tier 3 in vivo study.

Property (Descriptor)	Recommended assay protocol	Similarity rules and thresholds required for justification of similarity
In vivo short-term toxicity study with evaluation of organ burdens (LOAEC or BMD; % internal dose in at least lungs, spleen, liver, kidney, and blood; lung t1/2 at nontoxic lung burdens)	A short-term inhalation toxicity study, such as STIS, is suggested (Wiench and Ma-Hock 2019).	Ratio of LOAEC (or BMD) between the two nanoforms <3 (OR <5, only if the most toxic nanoform in this assay will be used as a worst-case representative for testing to fill in data requirements in REACH). Ratio of organ burden between the two nanoforms for lung, spleen, liver, kidney, and blood (except when their burden represents <1% of total body burden) <3. Ratio of lung clearance half life between the two nanoforms <2.

Gandon, A., Werle, K., Neubauer N., and Wohlleben W. Surface Reactivity Measurements as Required for Grouping and Read-across: An Advanced FRAS Protocol. In: Journal of Physics: Conference Series 2017. IOP Publishing: 012033. doi:10.1088/1742-6596/838/1/012033.

 Google Scholar

Wiemann, Martin, Antje Vennemann, Ursula G. Sauer, Karin Wiench, Lan Ma-Hock, and Robert Landsiedel. 2016. “An In Vitro Alveolar Macrophage Assay for Predicting the Short-Term Inhalation Toxicity of Nanomaterials.” Journal of Nanobiotechnology 14.1 (1): 16.

 Google Scholar

Kroll, Alexandra, Christian Dierker, Christina Rommel, Daniela Hahn, Wendel Wohlleben, Christian Schulze-Isfort, Christian Göbbert, et al. 2011. “Cytotoxicity Screening of 23 Engineered Nanomaterials Using a Test Matrix of Ten Cell Lines and Three Different Assays.” Particle and Fibre Toxicology 8: 9. doi:10.1186/1743-8977-8-9.

 PubMed Web of Science ®Google Scholar

Abdolahpur Monikh, F., M. A. Praetorius, A. Schmid, P. Kozin, B. Meisterjahn, E. Makarov, T. Hofmann, and F. Von der Kammer. 2018. “Scientific Rationale for the Development of an OECD Test Guideline on Engineered Nanomaterial Stability.” NanoImpact 11: 42–50. doi:10.1016/j.impact.2018.01.003.

 Web of Science ®Google Scholar

Wiench, K., and L. Ma-Hock. 2019. “The Short-Term Inhalation Study (STIS) as a Range Finder and Screening Tool in a Tiered Grouping Strategy.” In: In Vivo Inhalation Toxicity Screening Methods for Manufactured Nanomaterials. Current Topics in Environmental Health and Preventive Medicine, edited by T. Takebayashi, R. Landsiedel, M. Gamo.Singapore: Springer.

 Google Scholar