Why so much uncertainty about nano risks?

Why is there so much uncertainty about the risks associated with nanos?

Although public budgets for studies on health and environmental risks dedicated to research in nanosciences and nanotechnologies remain low and are much more limited than for research on the expected benefits of nanomaterials, research on the potential risks associated with nanomaterials is the subject of several scientific publications. However, many uncertainties remain, as explained below.

Nevertheless, an absence of certainty about the risks should not mean absence of risks – nor should it lead to inaction. On the contrary, we must not repeat the mistakes of the past (leaded gasoline, asbestos, etc.).

Limitations encountered in the assessment of nanos risks

Insufficient characterization of the tested nanomaterials

The physicochemical characteristics of the tested nanomaterials have long been insufficiently described¹Improvements are already noticeable. Various working groups have defined the parameters that should be systematically specified in all papers (detailed description of experimental conditions is also essential). Following the August 19, 2012 editorial in Nature Nanotechnology, the scientific community has continued to define the information needed to stabilize the characterization foundation that all nanotoxicology papers should include. Cf. in particular The dialogue continues, Nature Nanotechnology, 8, 69, February 2013: The nanotoxicology community has numerous ideas and initiatives for improving the quality of published papers. or too heterogeneous to be able to reproduce the experiments and/or compare the results between the different studies. However, these characteristics play a very important role on the toxicity of these materials which undermines a key principle of toxicology that “All things are poison, and nothing is without poison; the dosage alone makes it so a thing is not a poison“²Sentence of the physician and alchemist Paracelsus who founded toxicology and is very often invoked to assess the risks associated with synthetic chemicals.

On the one hand, because the toxicity and ecotoxicity of nanoparticles vary according to their physicochemical characteristics (size, shape, structure, charge state, degree of agglomeration, composition, solubility, etc.), which themselves vary according to the conditions under which the nanoparticles are synthesized, stored, possibly coated, integrated into a product and then released into the environment. For example, in humans or animals, fibrous nanos are more likely to generate inflammatory effects than spherical nanoparticles³See for example Carbon nanotubes, but not spherical nanoparticles, block autophagy by a shape-related targeting of lysosomes in murine macrophages, Autophagy, 14:8, 1323-1334, DOI: 10.1080/15548627.2018.1474993.

On the other hand, it is also necessary to take into account what the nanomaterials under consideration – or their residues – will come into contact with: living plants, animals, micro-organisms, and other chemical substances. For example, in the biological environment, nanos can be rapidly coated with proteins, forming a protein corona that can also impact their toxicological effects⁴See for example:
– in French: Couronne protéique autour des nanoparticules: une affaire de taille, IRAMIS and Joliot Institutes, I2BC, March 2021
– in English: The bio-corona and its impact on nanomaterial toxicity, Westmeier D et al, European Journal of Nanomedicine, 7(3), 2015.

Any risk assessment of nanomaterials is therefore very complex but the scientific community has proposed a number of improvements⁵See for example:
– Environmental Risk Assessment of Nanomaterials in the light of new obligations under the REACH regulation – Which challenges remain and how to approach them, Integrated Environmental Assessment and Management, Schwirn K et al, March 2020 Experts call for updated guidance on nanomaterial risk assessment, Chemical Watch, 26 March 2020
– Harmonizing across environmental nanomaterial testing media for increased comparability of nanomaterial datasets, Geitner NK et al, Environ. Sci.: Nano, 7, 13-36, 2020.

Studies often difficult to extrapolate to humans

Nanomaterials are mainly tested in vitro, on different cell strains (human, animal, plant), on many micro-organisms (bacteria, viruses, fungi…). These studies mainly give indications in terms of carcinogenesis and cell viability. They cannot totally replace in vivo tests.

The studies in vivo also have limitations: animal models of toxicity pose ethical and financial problems as well as methodological ones: their extrapolation to humans is certainly more reliable than in vitro tests but is not guaranteed⁶See for example Comment le test sur les rats échoue à protéger les hommes, Stéphane Foucart, Le Monde, 22 October 2012.

There is a real need for data on the human body. Thanks to the progress of nano-metrology and respecting obviously indispensable ethical considerations, it is possible to advance knowledge at the human level⁷Nanotoxicology: the need for a human touch?, Miller M & Poland C, Small, July 2020.

Studies sometimes carried out under conditions that are not representative of actual exposure

As industrial applications are currently not well known or quantified, they can only be estimated. The “probable” exposure of the population and the environment can therefore only be estimated. On what criteria and with what reliability?

Non-representative exposure routes

For practical reasons, the nanomaterials tested are often introduced directly into certain parts of the body and organs (e.g. intracerebral, intraperitoneal injections), according to modalities that are sometimes very different from the actual conditions under which the environment or the population are exposed. This prevents a proper understanding of the important mechanisms that come into play “in real life” (processes involved in digestion / fermentation / detoxification for example).

Progress is nevertheless being made in the environmental field, with studies carried out in mesocosms, for example – huge aquariums reproducing a mini eco-system in which the behavior of different dosages of nanoparticles in contact with plants, fish, soil and water is studied⁸Cf. MESONNET: Utilisation de mésocosmes terrestres et aquatiques en réseau pour l’évaluation du risque associé à la dispersion de nanoparticules manufacturées, CEREGE project; andINERIS equipment: Les leçons des écosystèmes synthétiques, Le Monde, 20 Nov. 2013.

Studies conducted over too short a period

The studies are often conducted over durations that are far too short to reflect realistic conditions of exposure, which are actually largely chronic (accident cases should also be taken into account, but in very specific configurations). Research that study the impact of the ageing of nanomaterials on their toxicity are still very rare⁹Voir par exemple Ageing influences the toxicity of two innovative nanofertilizers to the soil invertebrates Enchytraeus crypticus and Folsomia candida, C. Maleihro, Environmental Pollution, 350, Juin 2024.

Epidemiological studies of nano risks in humans are almost non-existent. Only one epidemiological surveillance program for workers potentially exposed to nanomaterials has been set up in France (EpiNano), but it is facing many difficulties and its results will not be known or usable for many years.

Synthetic nanomaterials different from those to which ecosystems and human populations are exposed

The nanomaterials considered are often synthesized in the laboratory and therefore different from the nanomaterials (and nanomaterial residues) to which ecosystems and human populations are actually exposed¹⁰See for example Yang Y et al, Characterization of Food-Grade Titanium Dioxide: The Presence of Nanosized Particles, About. Sci. Technol., 2014, 48 (11), pp 6391-6400often more complex and mixed with elements from the living world. Currently, scientists have very limited knowledge of the types of nanomaterials that are incorporated into products currently on the market and a fortiori of the degradation residues of nanomaterials released into the environment throughout the “life cycle” of these products. They still lack knowledge about the mobility and transformations of nanomaterials in the environment.

Little consideration of the life cycle

Nanomaterials can change during their “life cycle“, whether in the environment or in the human body: many parameters come into play, such as the degree of acidity or salinity for example. The limits listed above apply specifically to the degradation residues of nanomaterials released into the environment from their design to their end of life.

Tested doses higher than real exposure

The doses of nanomaterials tested are often higher than the concentrations to which ecosystems and human populations are actually exposed (notably because of the limits of the detection and measurement devices used in the laboratory). However, the effects observed (or not) could also occur at lower concentrations; certain nanomaterials (silica in particular) are more genotoxic at low doses than at high doses¹¹Cf. Results of the European Nanogenotox program on the genotoxicity of nanomaterials, presented in French at ANSES, during the Restitution du programme national de recherche environnement santé travail: Substances chimiques et nanoparticules: modèles pour l’étude des expositions et des effets sanitaires: Dossier du participant et Diaporama, November 2013. And “Toxicological assessment of nanomaterials needs to evolve, says European research project,” APM International, November 14, 2013. More generally, we are beginning to better understand the effect of low doses and to realize that these effects can be just as harmful as high doses or have antagonistic effects depending on the dose. Dose-effects add considerable complexity to toxicology research. See for example The health problem of low doses, Elizabeth Grossman, July 2012; The second death of the alchemist Paracelsus, Stéphane Foucart, April 11, 2013. In addition, high concentrations make it possible to simulate situations of acute and specific contamination (for example an accidental spill on a production site, or during transport). In late 2019, a study further showed that a significant fraction of nanoparticles tested in nanotoxicity and nanomedicine studies can remain in the plastic syringes used to dose the nanoparticles. This undermines the reliability and reproducibility of the studies¹²Cf. Failure to launch: nano toxicity studies may be affected by nanoparticles staying behind in syringes European Union Observatory for Nanomaterials, 25 November 2019 and Unpredictable Nanoparticle Retention in Commonly Used Plastic Syringes Introduces Dosage Uncertainties That May Compromise the Accuracy of Nanomedicine and Nanotoxicology Studies, Holtzwarth U et al., Frontiers in pharmacology, November 2019..

Nanos are difficult to detect

Because of their small size and the transformations they undergo as they pass through the body or ecosystems, nanos are difficult to detect, quantify, characterize and follow in the human body and in ecosystems.

Data gaps

Finally, many aspects related to the toxicity and ecotoxicity of nanomaterials remain to be studied. For example, in April 2020, a scientific literature review highlighted the lack of data regarding the impact of nanomaterials on female fertility and the need to study their effects on the reproductive system¹³Cf. Female fertility data lacking for nanomaterials, European Observatory of Nanomaterials, 6 April 2020 and A critical review of studies on the reproductive and developmental toxicity of nanomaterials, ECHA / Danish National Research Centre for the Working Environment, April 2020.

Ongoing and future improvements

Ongoing research is leading to significant improvements, particularly in methodology¹⁴See for example: Scientific meeting on microplastics and nanomaterials: research in environment and health, ANSES / ANR, 20 May 2021 and Cahier de la recherche n°17: “Microplastics and nanomaterials”- Understanding where research stands, ANSES, May 2021.

For example, the Nanomics project developed at the CEA in partnership with the Lavoisier Institute (CNRS) of the University of Versailles, is a systematic screening approach to define the toxicity of about fifteen nanoparticles (already used in industry) on human lung cancer cell lines and on lung tissue grown in three dimensions. It is based on a high-throughput screening platform: a device that allows a large number of tests to be performed in parallel on cell cultures, allowing to quickly test different concentrations of nanoparticles and different types of cells¹⁵Cf. “Mesures des effets toxicologiques de nano-oxydes métalliques sur cellules humaines in vitro“, Chevillard S, in Nanomaterials and health – Understanding where the research stands, ANSES, Les cahiers de la recherche, October 2015. .

Progress is also being made in the environmental field, with studies carried out in mesocosms, for example – huge aquariums reproducing a mini eco-system in which the behavior of different dosages of nanoparticles in contact with plants, fish, soil and water is studied¹⁶Cf. MESONNET: Utilisation de mésocosmes terrestres et aquatiques en réseau pour l’évaluation du risque associé à la dispersion de nanoparticules manufacturées, CEREGE project; and INERIS equipment: Les leçons des écosystèmes synthétiques, Le Monde, 20 Nov. 2013.

Future progress in nano-metrology as well as improvements to the R-Nano register and the registration of nanomaterials under REACH should improve our knowledge, allowing us to work more precisely on nanomaterials produced or imported into France.

Adapting standardized tests used for toxicological testing of conventional chemicals (such as the OECD guidelines) to nanomaterials takes time¹⁷Work is underway to make this adaptation. See in particular:
– Adapting OECD Aquatic Toxicity Tests for Use with Manufactured Nanomaterials: Key Issues and Consensus Recommendations, Petersen EJ et al, About. Sci. Technol., 49 (16) : 9532-9547, 2015
– Nanotechnology Regulation and the OECD, CIEL, ECOS, Öko-Institute, January 2015
– Ecotoxicology and Environmental Fate of Manufactured Nanomaterials: Test Guidelines, the Working Party on Chemicals, Pesticides and Biotechnology, OECD, March 2014..

In the meantime, one should not “throw the baby out with the bathwater”: the numerous studies showing toxic effects and published before the development of these standardized tests¹⁸See for example our sheets Risks associated with carbon nanotubes; Risks associated with nanosilver; Risks associated with nano titanium dioxide; Risks associated with nanosilica should not be outrightly dismissed because they are not perfect in terms of characterization or other limitations listed above.

Better funding for risk studies is essential

In 2009, researchers estimated that fifty years of work and several hundred million dollars would be needed to study the risks of nanomaterials already on the market.

The grouping of nanomaterials with similar toxicity potentials (read-accross) is a strategy advocated by some industrial players, but it is also contested because of the numerous methodological pitfalls¹⁹See about grouping of nanomaterials (“grouping” and “read-across”):
– 1st Innovative and complex materials: Towards grouping to support hazard and risk assessment, Stakeholder Workshop, BMBF-project InnoMat.Life, 15 June 2021
– NanoApp, an ECETOC project, launched in December 2020 (“this tool is used to establish and justify sets of nanoforms and identify poorly soluble – low toxicity (PSLT) nanoforms”).
– A framework for grouping and read-across of nanomaterials- supporting innovation and risk assessment, Stone V et al, Nano Today, 35, December 2020
– Grouping all carbon nanotubes into a single substance category is scientifically unjustified, Bengt Fadeel & Kostas Kostarelos, Nature Nanotechnology, March 2020
– Categorizing nanomaterials to achieve effective hazard and risk assessment, Cordis, NanoREG II, December 17, 2019
– Material-specific properties applied to an environmental risk assessment of engineered nanomaterials – implications on grouping and read-across concepts, Wigger H and Nowack B, Nanotoxicology, 13(5): 623-643, February 2019
– Understanding the legal term Nanoform in REACH (and ‘set of similar nanoforms’) – A discussion Workshop between ECHA and Industry Experts, CEFIC & NIA, 16 October 2018
– Nanotechnology experts from across the globe join forces to advance nanomaterials safety testing through Grouping and Read Across, NanoReg2 and Gracious, September 2018
– Criteria for grouping of manufactured nanomaterials to facilitate hazard and risk assessment, a systematic review of expert opinions, Landvik NE et al, Regulatory Toxicology and Pharmacology, 95: 270-279, June 2018
– GRACIOUS: Grouping, Read-Across, CharacterIsation and classificatiOn framework for regUlatory risk assessment of manufactured nanomaterials and Safer design of nano-enabled products, H2020 research project, 2018-2021
– Grouping and read-across for nanoforms, ECHA, November 30, 2017
– Conference on Categorization of Next Generation Nanomaterials, FutureNanoNeeds, November 30 and December 1, 2017.
– Considerations about the relationship of nanomaterial’s physicalchemical properties and aquatic toxicity for the purpose of grouping, UBA, November 2017
– Effects of nanoparticles on human immune cells, Denis Girard, IRSST, November 2017: “The results as a whole clearly demonstrate that it is difficult to classify NPs strictly according to their potential to modify one or another of the functions studied. A more nuanced picture is preferred in which the effects caused by a given NP on human OE biology in vitro must be considered to clarify its mode of action. The effects of NPs are therefore extremely varied and this study aims to demonstrate that they do not all act in the same way.
– Approaches on nano-grouping/equivalence/read-across concepts based on physical-chemical properties (Gera-PC) for regulatory regimes, OECD, January 2016
– NanoToxClass – Assessment of the health effects of industrially used nanomaterials to be made more efficient, BfR, 18 January 2016
– ECETOC concept allows assessing the safety of nanomaterials undertaking animal testing only as a very last resort, ECETOC, 16 December 2015
– the work of the European research project NanoSolutions (2013-2017), which seeks to identify the characteristics of manufactured nanomaterials that determine their biological risk potential. It will develop a safety classification model for these nanomaterials, based on an understanding of their interactions with living organisms
– Clustering of nanomaterials: a tool for risk assessment, FOPH, August 2015: The clustering concept proposed by the FOPH first provides for the classification of very similarly designed nanomaterials as entities based on a set of criteria. In a second phase, the entities are attached to “clouds”. Within the same cloud, entities can be evaluated with the same test strategy. Walser & Studer, Sameness: The regulatory crux with nanomaterial identity and grouping schemes for hazard assessment, Regulatory Toxicology and Pharmacology, 72(3): 569-571, August 2015
– Grouping nanomaterials – A strategy towards grouping and read-across, RIVM, June 2015
– EU toxicology body publishes grouping framework for nanomaterials – Risk assessment tool contributes to sustainable development of nano products, Chemical Watch, 2 April 20115 and A decision-making framework for the grouping and testing of nanomaterials (DF4nanoGrouping), Arts JHE et al, Regulatory Toxicology and Pharmacology, 71(2): S1-S27, March 2015
– A critical appraisal of existing concepts for the grouping of nanomaterials, Regulatory Toxicology and Pharmacology, 70(2): 492-506, November 2014
– Grouping of nanomaterials for risk assessment, Bolt HM, Archives of Toxicology, November 2014
– A strategy for grouping of nanomaterials based on key physico-chemical descriptors as a basis for safer-by-design NMs, Nano Today, 9(3): 266-270, June 2014.

Some scientists advocate working on “model nanoparticles”²⁰See in particular:
– “Imogolite nanotubes: a new model material in nanotoxicology?” by Rose J et al, in Participant’s file prepared for the Restitution of the National Environment Health and Work Research Program (PNREST), October 2015
– “Towards a model material in nanotoxicology?”, Rose J and “Mesures des effets toxicologiques de nano-oxydes métalliques sur cellules humaines in vitro“, Chevillard S in Nanomatériaux et santé – Comprendre où en est la recherche, ANSES, Les cahiers de la recherche, October 2015. We are still far from obtaining knowledge on the toxicity and eco-toxicity of nanoparticles used in industry…

The OECD Working Party on Manufactured Nanomaterials is addressing hazard and exposure assessment and guidance for different types of manufactured nanomaterials. Late 2015, it suggested to focus on manufactured nanomaterials contained in gases or liquids, for which the risk of exposure is higher than for solids, insofar as gases and liquids propagate more rapidly and penetrate the human body more easily through inhalation or ingestion²¹Cf. OECD, Nanomaterials in Waste Streams (Chapter 1, Overview), November 2015.. Further publications are expected.

The existing uncertainties give rise to divergent interpretations

These uncertainties and difficulties explain why the results cannot be generalizable and should be treated with caution.

While some minimize the risks by arguing that the experiments were carried out on the basis of a “worst case scenario” (involving, for example, nanoparticles used in dispersed form and in very high doses), others emphasize that the conclusions lead to serious concerns.