Speciality Chemicals Magazine JUL / AUG 2025 | Page 33

CONTRACT RESEARCH & TOXICOLOGY

The paper noted that, for most elemental impurities, ICH parenteral PDEs have been determined by dividing the oral PDE by a factor of 10, which assumes that estimating oral bioavailability at 10 % is sufficiently conservative. 7 However, if very extensive absorption is known or expected, a health risk assessor might feel able to justify smaller factors.

Other uncertainty factors

There are other uncertainty factors that are routinely considered in health risk assessment. For example, while a NOAEL is an ideal PoD, sometimes effects are seen even at the lowest tested dose – i. e. we have a lowest observed adverse effect level( LOAEL) as our PoD.

If this is the case, it is generally considered to be health precautionary to apply an uncertainty factor to generate an effective NOAEL. Medical device guidance suggests that a factor of 10 is sufficient for this, although a smaller factor might be justified in certain circumstances( e. g. very mild effects at the LOAEL). ICH pharmaceutical guidance suggests a factor of up to 10 for use of a LOAEL.

Other examples where an uncertainty factor of 10 can be appropriate include where a study lacks reliability or relevance, where the study was underpowered( for example, low numbers of animals, or animals of only one sex), or where the critical data was generated on a structurally

similar compound( the so-called‘ readacross’ approach).

Worked example

Let’ s consider a hypothetical example of a toxicity study used to derive tolerable daily( lifetime) exposures in humans under the guidance of different industries. In this example, Compound A might be present in medicines and medical devices and is subject to REACH registration. Both the medical scenarios involve application to the skin. The critical study is a sub-acute study by the dermal route in rabbits. The point of departure is a NOAEL of 250 mg / kg of body weight / day.

The differences in factors to be applied in the risk assessment of compound A vary considerably across these three industries( Table 1). Medical devices take the most precautionary approach, applying factors of 10 for inter-individual and inter-species variability and study duration. In common with the other industries, there is no need to consider route-toroute extrapolation for a dermal risk assessment based on dermal data.

Guidance in both the pharmaceutical and chemical industries suggest lower factors for inter-species differences, 2.5 and 6, respectively. For extrapolation for study duration, medicines align with medical devices, a factor of 10 for‘ shorter studies’, while REACH guidance offers 6( for subacute to chronic extrapolation). While for medicinal products or devices the standard factor of 10 applies for interindividual variations, a lower factor, of 5, is applied under REACH guidance when considering the health risk to workers rather than the population as a whole.

Table 2 shows how these differences in individual factors accumulate and lead to rather large differences in the overall uncertainty factor and thus

the tolerable dose that is used for risk assessment. There is a more than five-fold difference between the most precautionary dose, established using the medical device guidance, and the least, for workers under REACH. Even excluding the special case of worker risk assessment, there is still a fourfold difference( medical devices vs. pharmaceuticals) in the level that is considered to be without harm to the health of the general population.

Conclusion

Different chemical industries have their own approaches to conducting toxicological risk assessments, with long-established guidance including selection of uncertainty factors. These factors can vary considerably, often without obvious scientific support. This, in turn, leads to the head-scratching situation that a chemical substance can be considered safe by one industry at a level that would cause significant concern in another. Perhaps it is time for the various industries to come together and begin the move towards closer harmonisation. ●

JUL / AUG 2025 SPECCHEMONLINE. COM

33