This guest blog post from our Finnish colleagues summarizes the challenges of identifying OELs for new nanomaterials as part of the development of a WHO guideline for working safely with nanomaterials.

Engineered nanomaterials are fascinating. Just by making stuff smaller researchers have discovered forms of materials and even completely new materials that can be applied as diversely as better drugs, better paints or faster electronics. Using chemicals in a nanoscale version can completely alter their nature. We find nano-gold one of the most striking examples. The ‘big’ gold in its normal state is the yellow colored metal that does not interact with the body but nano-sized gold is purple or red, depending on its nanosize, and binds proteins very easily ¹. Therefore, nano-gold is interesting both as a potential cancer drug and as a potential toxic substance. For society, the question is how to find the right balance between the benefits of new products and the potential harms to health and the environment as new products are developed. One way of preventing harm to workers is to set and adhere to occupational exposure limits (OEL), concentrations of a chemical below which adverse health effects are not likely to occur. Therefore, we systematically searched for proposals for OELs for new nanomaterials as part of the development of a WHO guideline for working safely with nanomaterials ².

Occupational exposure limits fascinate us as well. They have been in use for over 100 years to protect the health of workers who are exposed to potential toxic chemicals during work. For many chemicals there is a concentration below which exposure to the chemical is considered safe and unlikely to have a negative impact on health. Other chemicals, especially those that cause cancer, are considered not to have a safe threshold. For those chemicals, concentrations can be estimated below which the risks are considered negligible. Some chemicals have been considered so potentially harmful that countries such as Finland, the UK and Korea have restricted or forbidden their use. Regulation using OELs to require employers to lower exposures has worked well. In the past 50 years workplace exposures to chemicals like solvents have decreased substantially in the US and part of Europe most probably due to regulated OELs ³. Also as part of a risk management strategy OELs are helpful. They help employers and workers to decrease exposure levels as a clear benchmark value that should not be exceeded. However, there are also many issues around OELs that are good for heated debates. Because there is no generally accepted way of deriving OELs, this has led to differences between OEL values. These variations are also reflected in the names of the OELs such as Threshold Limit Values (TLV®) in use by the occupational hygienists, Permissible Exposure Levels in use by OSHA in the US, Recommended Exposure Limits in use by NIOSH and Derived No-Effect Levels in use by the EU REACH agency. Basically, these are all concentrations of chemicals that should help to achieve a safe working environment but in different regulatory and cultural contexts ⁴.

Historically, with the introduction of then-new materials such as asbestos or PVC, a number of workers died before exposure to these chemicals in workplaces was regulated. No one wants this to happen again with nanomaterials. The European Union has explicitly stated that a precautionary approach should be taken. This means that we don’t wait with taking safety measures until we have full certainty about the adverse health effects of nanomaterials but that we act immediately. When more knowledge comes available these measures can be tightened or loosened based on the results of new research.

Even though there is so far not much evidence that simply nanosize leads to toxicological properties that are different from the bulk material or unique for the nanosize, many studies have shown that nanoscale particles can be more biologically active such as eliciting greater pulmonary inflammation at a given mass dose ^{5 6}. Thus, the mass-based OEL for the bulk material is not automatically a safe level for the nanomaterial. For example, some specific nanofibers are made of carbon but the nanofiber structure makes them similar to asbestos fibers that cause cancer. It would be unwise to assume that the same OEL of 2.5 mg/m³ for graphite dust would also be applicable to these carbon nanofibers. Therefore, experts in the UK, Germany, Finland and the Netherlands have stated that keeping exposure for these specific nanofibers somewhat below (0.01 fiber/cm³) levels considered acceptable for asbestos is the safest strategy ².

OELs are usually based on the available knowledge stemming from exposure of humans and/or animals. Because the nanomaterials are relatively new the knowledge about the toxicity comes mainly from animal and in vitro studies. For example NIOSH has proposed a Recommended Exposure Limit (REL) of 300 µg/m³ for ultrafine titanium dioxide, which is one of the most produced nanomaterials, based on chronic inhalation data in rats ⁵. But animal tests can take several years to develop which is very long compared to the potential and commercial value of new nanomaterials. The number of nanomaterials also increases rapidly which makes the traditional method of quantitative risk assessment for OEL derivation not always feasible. Therefore, scientists have proposed other methods.

The most pragmatic approach is what has been done for nanofibers ². Compare the properties of the new material to an existing one such as comparing specific nanofibers to asbestos and assume that the toxicity will be similar. This procedure is variously called ‘bridging’ or ‘read across’.  Another proposal is to make groups of nanomaterials with similar toxicity also called ‘grouping’. For example, for the nanosize variant of chemicals that already have established toxic effects such as causing cancer, the British Standard Institute (BSI) proposed to use an extra safety factor of 10 because of the nanosize. This means that the OEL for the bulk material is divided by ten to derive the OEL for the same but nanosized material. The assumption is that the toxic properties will be enhanced by the nanosize. Another proposed group is a group of insoluble or biopersistent nanomaterials, materials that, after inhalation, are not easily cleared away from the human body. The nanosize will lead to an increased biopersistence with accompanying toxicity and the OEL for the nanosized material proposed by BSI is therefore a factor 15 lower than the bulk OEL. For soluble nanomaterials, BSI applies a safety factor of two for the nanosize.

Again a very different approach is to take the knowledge that is available from air pollution research. At least part of the polluting particles are nanosize and especially these nanosized particles are blamed for adverse health effects. These adverse health effects have been studied well in humans and could provide a model for nanomaterials.  One step further is to predict the toxicity based on the chemical structure of the material which is called Quantitative Structure-Activity Relationship (QSAR) modelling. As far as we are aware, this approach has not yet been used for nanomaterial OEL derivation. An ISO group of experts recently gave an overview of similarities and differences in the methods used to develop OELs for engineered nanomaterials ⁷.

Most OELs, including those for engineered nanomaterials, are expressed as a mass concentration (mg/m³) which implies that this would be the best way to measure them.  However, that is still a major problem with nanomaterials. In the nanoworld, measurement becomes very different. Where almost all OELs are provided as a mass concentration, such as the amount of milligrams of the chemical in a cubic meter of air, it is unclear what the best metric is for nanomaterials. One alternative is the number concentration: the number of particles in a cubic meter of air. Another one is the surface concentration: the total surface of the particles (nm²) in a cubic meter of air. It all depends on what, we think, causes an increase of toxicity. It could be that the best metric varies per group of materials: particles number for biopersistent materials but surface concentration for soluble materials. The challenge still is to find the best metric for most OELs that are only expressed as mass concentrations.

In spite of constantly increasing information on nanomaterials, there are few systematic reviews of nanomaterials available. For many interesting reasons, systematically summarizing the available evidence in toxicology seems more difficult than elsewhere ⁸. OEL derivation is located at the edge of science and regulation/practice which makes it also difficult to locate all available proposals. When we searched for proposed OELs, only about half were published in peer-reviewed scientific journals indexed in scientific databases like PubMed ². The other half was scattered across project reports published on the Internet only, conference abstracts or government reports. Because they are so difficult to locate, we believe that having a list of proposed values is already very helpful if one wants to use a proposed OEL as a benchmark value for a workplace.

Given the uncertainty surrounding adverse health effects of nanomaterials, it is no wonder that the OELs that we found varied. Some were rather similar such as those for nanofibers but for carbon tubes the OELs varied with a factor 50 and for metallic nanomaterials they varied even with a factor 100 to 300. The variation is due to variation in objectives and derivation methods as has also been shown for bulk material OELs. Deveau et al ⁴ provide a practical framework for how to find the best applicable OEL for a particular problem from a list of varying OELs for one chemical. The framework can also be applied to the list of nanomaterial OELs ².

The challenge is to find consensus on how to derive the OELs for nanomaterials and next to underpin the proposed values with more empirical research. For the time being, the OELs for nanomaterials are bench mark levels to which the exposure levels measured in a specific workplace can be compared. Compared to the concentrations that have been measured in workplaces ⁹, these OEL benchmark levels are often still substantially lower and should be an incentive to reduce exposure.

Jos Verbeek is the Coordinating Editor of the Cochrane Work Review Group, based at the Finnish Institute of Occupational Health, in Kuopio, Finland.

Raluca Mihalache is a Public Health expert interested in evidence synthesis and PhD student at the University of Eastern Finland.

References

1. Kulinowski K, B. L. Training workers on the risks of nanotechnology: National Institute of Environmental Health Sciences, 2011.
2. Mihalache R, Verbeek J, Graczyk H, et al. Occupational exposure limits for manufactured nanomaterials, a systematic review. Nanotoxicology 2016;10:1-35. doi: 10.1080/17435390.2016.1262920
3. Creely KS, Cowie H, Van Tongeren M, et al. Trends in inhalation exposure–a review of the data in the published scientific literature. Ann Occup Hyg 2007;51(8):665-78. doi: 10.1093/annhyg/mem050
4. Deveau M, Chen CP, Johanson G, et al. The Global Landscape of Occupational Exposure Limits–Implementation of Harmonization Principles to Guide Limit Selection. J Occup Environ Hyg 2015;12 Suppl 1:S127-44. doi: 10.1080/15459624.2015.1060327
5. N I O S H Occupational Exposure to Titanium Dioxide. Current Inteligence Bulletin. Cincinatti, OH, 2011.
6. Gebel T, Foth H, Damm G, et al. Manufactured nanomaterials: categorization and approaches to hazard assessment. Archives of toxicology 2014;88(12):2191-211. doi: 10.1007/s00204-014-1383-7
7. I S O IOfS. ISO/TR 18637:2016. Overview of available frameworks for the development of occupational exposure limits and bands for nano-objects and their aggregates and agglomerates (NOAAs). Geneva: ISO, 2016.
8. Stephens ML, Betts K, Beck NB, et al. The Emergence of Systematic Review in Toxicology. Toxicol Sci 2016;152(1):10-6. doi: 10.1093/toxsci/kfw059
9. Debia M, Bakhiyi B, Ostiguy C, et al. A Systematic Review of Reported Exposure to Engineered Nanomaterials. Ann Occup Hyg 2016;60(8):916-35. doi: 10.1093/annhyg/mew041