Registration Dossier

Administrative data

Workers - Hazard via inhalation route

Systemic effects

Long term exposure
Hazard assessment conclusion:
DNEL (Derived No Effect Level)
Value:
0.12 mg/m³
Acute/short term exposure
Hazard assessment conclusion:
high hazard (no threshold derived)
DNEL related information

Local effects

Long term exposure
Hazard assessment conclusion:
hazard unknown (no further information necessary)
Acute/short term exposure
Hazard assessment conclusion:
low hazard (no threshold derived)
DNEL related information

Workers - Hazard via dermal route

Systemic effects

Long term exposure
Hazard assessment conclusion:
DNEL (Derived No Effect Level)
Value:
16.73 mg/kg bw/day
Acute/short term exposure
Hazard assessment conclusion:
hazard unknown (no further information necessary)
DNEL related information

Local effects

Long term exposure
Hazard assessment conclusion:
low hazard (no threshold derived)
Acute/short term exposure
Hazard assessment conclusion:
low hazard (no threshold derived)

Workers - Hazard for the eyes

Local effects

Hazard assessment conclusion:
no hazard identified

Additional information - workers

Background information – essentiality of selenium

 

Selenium is an essential element that is part of a significant number of selenoproteins that have been identified in the human genome. Once absorbed by the human body, Se-compounds are transformed into an intermediate organic Se-compound (selenocysteine) which used in the synthesis of these selenoproteins. 

A lack of Se will have an adverse impact on both the expression and functioning of these selenoproteins. Several clinical manifestations have been attributed to Se-deficiency (Ge and Yang, 1993; Gramm et al, 1995; Fairweather-Tait et al, 2011; Loscalzo, 2014)

* Skeletal- and cardiomyopahy, muscle weakness

* Pseudoalbinism and red blood cell macrocytosis

* Degeneration of organs and tissues (Keshan and Kashin-Beck disease)

A chronic excess of selenium, on the other hand, will cause adverse effects, referred to as selenosis. Until now, the molecular mechanisms that cause Se-toxicity are not yet fully understood, but the typical symptoms of selenosis are headache, loss of hair, deformation and loss of nails, skin rash, malodorous (garlic) breath and skin, excessive tooth decay and discoloration, as well as numbness, paralysis and hemiplegia (Fairweather-Tait et al, 2011).

 

When setting a DNEL for selenium and selenium compounds, both aspects (deficiency, toxicity) need to be taken into account. A DNEL should be sufficiently low in order to protect humans against selenosis, but at the same time the DNEL should be sufficiently high in order to prevent Se-deficiency. Relationships/equations between concentrations levels of Se in blood (plasma/serum) and (dietary) Se-intake have been reported by various authors (Yang et al., 1989b; Yang and Xia, 1995; Longnecker et al, 1996; Burk et al., 2006; Combs et al., 2012.), and Fairweather-Tait et al (2010) concluded that there is no homeostatic regulation of plasma selenium concentration and it does not appear to reach a plateau. Elimination via urinary excretion is the main mechanism with regard to maintain the internal Se-levels at an acceptable levels. The amount of Se that is excreted depends on the amount that is taken up (dose-dependent response) and the chemical nature of Se that was ingested (Burke et al, 2006; Combs et al, 2011; Kokarnig et al, 2014). 

The observation that there is no homeostatic regulation mechanism that can keep the Se-concentration in blood plasma at a constant level (Fairweather-Tait et al., 2010) may explain why the ”therapeutic window” of Se in human is considered to be relatively narrow. Consequently, the selection and application of an assessment factor (AF) on the dose descriptor that is retained for setting a DNEL should be done with caution as an AF that is too high may result in a DNEL that could cause Se-deficiency.

 

In this context, a dose descriptor that is based on human health data (e.g. epidemiological studies) would be preferred over a dose-descriptor that is derived from animal test data. According to ECHA Guidance on the derivation of DNEL-values, a specific AF for taking inter-species differences into account has to be applied (e.g. AF of 10 from rat to human; allometric scaling factor of 4 + factor of 2.5 for remaining differences) on top of the AF of 10 that is typically require due to intra-species differences. It is not unreasonable to assume that a default AF of 100 (or higher) on an animal-based dose descriptor could result in a DNEL below the Se-deficiency threshold.

 

In summary, the rationale for the direct use of human (epidemiological) data for DNEL-derivation can be summarized as follows:

  • Selenium is an essential element for animals and humans, which is needed for the proper function of a number of enzymes and proteins, like for example for glutathione peroxidases, thioredoxin reductase and deiodase and also binding and transport proteins.
  • Its “therapeutic” window is narrow, such that Selenium deficiency can cause severe adverse health effects, but also overdosing may lead to the clinical manifestation of Selenosis.
  • Human dietary intake of Selenium (from foodstuff and drinking water) is necessary (and also “unavoidable”) and has been documented and reviewed with regard to any potential health concerns which may derive from that fact (see below for derivation of DNEL).
  • Nutritional supplemental Selenium and Selenium added fertilisers are required in areas with low Selenium concentrations in the soil. Soil which contains 0.02 to 2 mg Selenium/kg is the most important Selenium source for humans (and animals). Since also the amount of Selenium in plants depends on the concentration in soil, diseases of grazing animals have been observed where such levels were low enough.
  • Due to the facts of essentiality, the “unavoidable” human uptake, the need for nutritional supplementation (in some cases) and the relatively narrow “therapeutic” window accompanied with reliably human effect data national and international scientific bodies have thoroughly evaluated the benefits of Selenium for human health as well as its possible adverse effects.

 

 

Basic approach

The typical approach for the derivation of a DNEL uses effects data from appropriate studies with experimental animals (e.g. NOAEL: No Observed Adverse Effect Level), followed by a theoretical calculation for the human situation by application of standard assessment factors. The purpose of these AFs is to take into account inter/intraspecies differences and variability.

 

As explained in the introduction, such an approach can be questioned for essential elements (like selenium) that have a narrow therapeutic window as this method could result in a DNEL below required intake levels. However, when adequate human data exist, in particular when certain toxicological endpoints may have been observed only in humans but not in experimental animals such human data should be used instead of the animal data. This approach can be followed for Selenium and its inorganic compounds, where abundant human data exist and where critical human endpoints have never been observed in the experimental animal. It also allows to adequately consider the property of Se being an essential micronutrient for mammalian cells.

 

The starting point for the derivation of DNELs (workers/general population) is the NOAEL of approximately 850 µg/person/day that has been determined by Yang et al (1989) and which is considered to be reliable and relevant (see Section 5.10.2 – other effects, human information). This value also corresponds to a concentration of 12 µg/kg bw/day for clinical selenosis. Is summary, the NOAEL results from an increasing effect on prothrombine time by higher Se-levels.

 

The quality of epidemiological studies are assessed against a number of criteria:

#1: the proper selection and characterisation of the exposed and control groups

#2: adequate characterisation of exposure;

#3: sufficient length of follow-up for disease occurrence;

#4: valid method for observing an effect;

#5: proper consideration of bias and confounding factors; and

#6: a reasonable statistical reliability to justify the conclusion.

 

Yang et al (1989) assessed sufficiently large sub-populations (n=200-300) that were living in area’s with different natural Se-levels (low, medium, high) and the groups represented different sexes and ages (Criterion #1). Intake of Se was determined via a 3-day survey and analysis of the food that was consumed by the subjects of each group (Criterion #2). Subjects were life-long residents of the different areas, and cases of selenosis have been observed in the group of subjects that were living at the site with high natural levels (Criterion #3). Diagnosis of selenosis was based mainly upon morphological alterations of the fingernails (Criterion #4). This method is considered to be relevant as selenium concentration in (toe)nails have been related to chronic/long-term selenium intake and to selenium status (Rajpathak et al., 2005; Geybels et al., 2013). Blood, 24h-urine, breast milk and tissue samples were collected and analysed for Se, Hg, Cd, Zn, Co and As (Criterion #4, #5). In order to determine the expected level of Se-intake, individual Se intakes or tissue-Se levels were correlated with either biochemical parameters or clinical signs (Criterion #6).  

It should be noted that selenium in inorganic compounds (e.g. selenate, selenite) also appears to be well absorbed, but less is retained than with organic compounds (Thomson and Robinson, 1986; Fairweather-Tait et al., 2010). Therefore the results of this dietary study can be considered as a worst-case exposure (maximized uptake/absorption of Se after oral intake) (Criterion #5)

 

The assessment factor that is applied on the dose descriptor that is taken from a relevant study depends on the considerations with regard to the following aspects:

  • intraspecies differences;
  • differences in duration of exposure;
  • issues related to dose-response;
  • quality of human data

 

No significant differences between different sexes and/or age groups were noted, and all subjects had been long-life exposed to similar Se-levels in food. There was a clear relationship between elevated Se-levels in blood and the occurrence of selenosis. Overall; this study is considered to be reliable and relevant, and it should be emphasised that this study has been used by various organisations and competent authorities for setting safe levels of Se (e.g. food-levels, exposure-levels for workers) who never disputed this particular study with respect to its reliability. Examples where Yang et al (1989) were used as a starting point for the assessment of Se are given below:

·      available at:

http://www.efsa.europa.eu/en/scdocs/doc/ans_ej1082_L-Selenomethionine_op_en.pdf

·      SCIENTIFIC OPINION - Scientific Opinion on Dietary Reference Values for selenium, EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). EFSA Journal, 2014;12(10):3846.

·     German MAK Commission: Toxikologisch-arbeitsmedizinische Begründungen von MAK-Werten “Selen und seine anorganischen Verbindungen”, 1999, 2001 and 2010. available at:

http://onlinelibrary.wiley.com/book/10.1002/3527600418

   

 

DNEL derivation

 

The calculated DNELs are referring to selenium; the corresponding DNEL for the substance takes into account the selenium moiety based in the respective molecular weight (MW-translation).

 

Workers

In 1999 (and confirmed in 2001) the German MAK Commission derived its Maximum Workplace Concentration (MAK: Maximale Arbeitsplatz-Konzentration) on the basis of the Yang et al (1989) study, but also taking into account the relevant NOAEL of 0.15 mg/kg bw/day from a two-generation study on rats and a LOAEL of <0;1 mg/kg bw/day from a 4-generation study on mice (Rosenfeld and Beath, 1954; Schroeder and Mitchener, 1971; Schroeder and Mitchener, 1972).

 

Inhalation:

The MAK that was put forward in in 1999/2001 was 0.05 mg Se/m3(Göen et al, 2019), equivalent to 7 µg Se/kg bw/day (for an 8-hour shift). This MAK value covers all relevant systemic toxicological endpoints (acute toxicity, toxicity after repeated dosing, for reproduction, genotoxicity and carcinogenicity). 

 

Dermal:

For the dermal route the low absorption of 0.1 % has to be taken into account, that means, that the dermal DNEL is set to 7 mg Se/kg bw/day.

 

For disodium selenate a factor of 2.39 is used to account for molecular weight differences, resulting in the following DNELs for workers:

 

DNELdermal: 16.73 mg Na2SeO4/kg bw/day

DNELinhalation: 0.12 Na2SeO4/m3 (for an 8 -hour shift)

 

It is worth noting that the TLV in the United Kingdom for workplace exposure and its compounds (excl. hydrogen selenide) for an 8-hour shift is 0.1 mg/m3 (Health and Safety Executive: EH40/2005 Workplace exposure limits, Merseyside UK, 2011. Available at: www.hse.gov.uk)

General Population - Hazard via inhalation route

Systemic effects

Long term exposure
Hazard assessment conclusion:
DNEL (Derived No Effect Level)
Value:
0.036 mg/m³
Acute/short term exposure
Hazard assessment conclusion:
high hazard (no threshold derived)
DNEL related information

Local effects

Long term exposure
Hazard assessment conclusion:
hazard unknown (no further information necessary)
Acute/short term exposure
Hazard assessment conclusion:
low hazard (no threshold derived)
DNEL related information

General Population - Hazard via dermal route

Systemic effects

Long term exposure
Hazard assessment conclusion:
DNEL (Derived No Effect Level)
Value:
10.28 mg/kg bw/day
Acute/short term exposure
Hazard assessment conclusion:
hazard unknown (no further information necessary)
DNEL related information

Local effects

Long term exposure
Hazard assessment conclusion:
low hazard (no threshold derived)
Acute/short term exposure
Hazard assessment conclusion:
low hazard (no threshold derived)

General Population - Hazard via oral route

Systemic effects

Long term exposure
Hazard assessment conclusion:
DNEL (Derived No Effect Level)
Value:
10.28 µg/kg bw/day
Acute/short term exposure
Hazard assessment conclusion:
high hazard (no threshold derived)
DNEL related information

General Population - Hazard for the eyes

Local effects

Hazard assessment conclusion:
no hazard identified

Additional information - General Population

DNEL derivation

 

The calculated DNELs are referring to selenium; the corresponding DNEL for the substance takes into account the selenium moiety based in the respective molecular weight (MW-translation).

 

General population

According to ECHA guidance, the traditional default AF of 10 should be considered on a human-based NOAEL in order to take intra-species variability into account. This AF represents two different components: a toxicokinetic parameter and a toxicodynamic parameter. However, when the observed effect in humans is associated with biomonitoring data such as urinary or blood level – as is the case For Se in than Yang et al (1989)-study, the toxicokinetic factor of intraspecies variation is accounted for and the AF could be reduced as it would only have to cover for toxicodynamic variability. Biomonitoring data reflect the internal exposure and thus, toxicokinetic parameters influencing the internal/systemic bioavailability do not play a role.

 

Taking the information of this data quality assessment information into account, an AF of 3 (instead of the default AF of 10) is proposed for the derivation of a DNEL that is based on Yang et al (1989).

 

It should be noted that the Scientific Committee on Food (SCF, 2000) for inorganic selenium compounds determined a Tolerable Upper Intake Level (UL) of 300 μg/day for adults (including pregnant and lactating women), on the basis of the No Observed Adverse Effect Level (NOAEL) of 850 μg/day for clinical selenosis and applying a comparable uncertainty factor of 3. They also referred three studies that reporting no adverse effects for selenium intake between about 200 and 500 μg/day. This UL was recently fully acknowledged by the Office of Dietary Supplements of the US National Institutes of Health (available at:http://ods.od.nih.gov/factsheets/Selenium-HealthProfessional/). Similarly, in a document from the World Health Organization (WHO) for drinking water quality, the human NOAEL for Selenium uptake was stated to be 4 µg/kg bw per day.

Finally the US Environmental Protection Agency has defined a Reference Dose (RfD) for Selenium to be 5 µg/kg bw per day (available at:http://www.epa.gov/iris/subst/0472.htm).

 

 

Normalising the DNEL of 300 µg/day for body weight (average of 70 kg), a DNEL of 4.3 µg/kg/day is determined.

It is worth mentioning that the EFSA-panel (2009; Scientific opinion on L-selenomethionine as a source for selenium added for nutritional purposes to food supplements) concluded thatthe toxicity of L-selenomethionine is comparable to other forms of selenium, in terms of equivalent amounts of bioavailable selenium. As with other selenium compounds, the results of toxicological studies with L-selenomethionine in animals are indicative of a steep dose-response curve, with a threshold for onset of toxicity in the range of 100–400 μg selenium/kg bw/day, dependent on the species. These thresholds are approximately a factor of 20-100 higher than the DNEL of 4.3 µg/kg bw/day that is derived from Yang et al (1989).

   

The derived human NOAEL (equivalent to the DNEL) seems to cover the general population for all routes of exposure.

 

Oral:

The DNEL of 300 µg Se/person/day or 4.3 µg/kg bw/day is directly used for the oral route.

 

Dermal:

For the dermal route the low absorption of 0.1 % has to be taken into account, that means, that the dermal DNEL is set to 4.3 mg/kg per day.

 

Inhalation:

The corresponding DNEL for inhalation expressed as concentration in the air is calculated by:

Daily respiratory volume = 20 m3according to ECHA guidance R.8 for 24 hours of exposure.

DNEL = 300 µg per person per day,

DNELinh= 0.3 mg Se/20 m3= 0.015 mg Se/m3or 15 µg Se/m3

 

For disodium selenate a factor of 2.39 is used to account for molecular weight differences, resulting in the following DNELs:

 

DNELoral: 10.28 µg Na2SeO4/kg bw/day

DNELdermal: 10.28 mg Na2SeO4/kg bw/day

DNELinhalation: 0.039 mg Na2SeO4/m³