Case Study Challenges in Human Health Hazard and Risk Assessment of Nanoscale Silver - Safety of Nanomaterials along Their Lifecycle - page 426

potential, an in vitro micronucleus assay, a screening test for reproductive and develop-
mental toxicity in rats, and a dermal toxicity study.
Recently, the nanosilver toxicity database was revisited in a conference report
by Schäfer et al. (2013) and a provisional margin of safety calculation by Hadrup
and Lam (2013). With regard to human health risk assessment it was concluded at a
conference on state of the art in risk assessment of silver compounds in consumer
products held at the German Federal Institute for Risk Assessment (BfR) in 2012
that development of consumer exposure scenarios for nanosilver products is still at
an early stage and reliable exposure measurement data are lacking. Major gaps in the
toxicological database were identified with regard to toxicokinetic behavior of dif-
ferent types of nanosilver, genotoxicity, reproductive toxicity, and further repeated
dose toxicity studies designed according to relevant exposure settings. It was pointed
out that a better mechanistic understanding is required to allow linking differences
in physicochemical properties between nanoforms of silver to their toxic potential.
Participants discussed that in such a data-deficient situation, approaches to risk
assessment and management can include law enforcement and/or communication
measures to limit nanosilver use to essential applications, or mandatory provisions
for labeling of nanosilver containing products and the obligation for manufacturers
to generate additional safety data. It was pointed out that in any case uncertainties
in current safety assessment would need to be clearly pointed out and adequately
accounted for (Schäfer et al. 2013). This is contrasted by considerations of Hadrup
and Lam (2013), who proposed a tolerable daily intake (TDI) of 2.5 µg/kg bw/d
based on the subacute study in mice (Park et al. 2011) and the default safety factor of
100. When compared to estimates for dietary oral exposure, the authors concluded
that “ a Margin of Safety calculation indicates at least a factor of five before a level
of concern to the general population is reached.”
Different approaches have been discussed to compensate for the current lack of
chronic toxicity data on nanoforms of silver. In its assessment of AGS-20, the EPA
has chosen to apply an additional uncertainty factor of 3 to the assessment of long-
term scenarios to account for extrapolations from a 28-day study in the case of oral
and dermal exposure, and a 90-day study in the case of inhalation exposure (US
EPA 2011). In contrast, Hadrup and Lam derived their TDI value from a no observed
adverse effect level (NOAEL) of a 28-day study using the default uncertainty fac-
tor of 100 without consideration of time extrapolation (Hadrup and Lam 2013). For
comparison, European Chemicals Agency's (ECHA) guidance on chemical safety
assessment chapter on the characterization of dose-response for human (R.8) recom-
mends an assessment factor of 6 for extrapolation from subacute to chronic exposure
to chemicals (ECHA 2013).
A more generic concept for derivation of chronic reference values was developed
in the NanoGEM research project: Experimental observations in the rat led to the
concept of dust overloading of the lung as an underlying mechanism for development
of an adverse effect. It was later formulated as the overload hypothesis by Morrow
(1988). The overload hypothesis postulates that loading of an alveolar macrophage
with inert biopersistent material occupying 6% of the macrophages volume causes
a reduced mobility of the cell. Chronic exposure at this level would lead to lung
inflammation and might ultimately result in tumorigenesis (Morrow 1988). Although