Archived Comments for:
Translational toxicology in setting occupational exposure limits for dusts and hazard classification – a critical evaluation of a recent approach to translate dust overload findings from rats to humans
Reply on behalf of the 'Permanent Senate Commission for the Investigation of Health Hazards of Chemical Compounds in the Work Area' (MAK Commission)
Andrea Hartwig, Karlsruhe Institute of Technology (KIT)
10 September 2015
Threshold value for Granular biopersistent particles (GBP)
Recently, Morfeld et al. [1] commented on the OEL for Granular biopersistent particles (GBP) without known specific toxicity proposed by the German MAK commission [2]. They concluded that the scientific reliability of the proposed GBP OEL and carcinogen classification is limited due to unjustified "translational toxicology". Before going into detail, I want to start with some general remarks.
The MAK documentation is based on the outcome of a working group consisting of members of the MAK commission as well as of the "Committee on Hazardous Substances" (AGS), which has been discussing the subject for several years. Thus, due to the great impact of a respective OEL, all arguments, including those of international renowned scientists in the field of particle toxicology outside the working group, have been considered within this process, deriving to a consensus as published in the MAK documentation [2]. Also Peter Morfeld and Joachim Bruch, two of the authors of the comments, have been involved in this process, the latter as a member of the working group.
Their criticism is based on the following arguments:
1. Using the methods proposed within the MAK documentation, they were unable to reproduce the OEL for GBP recommended by the Commission, and claimed to identify substantial errors in the models.
2. In their opinion, considerable shortcomings were:
Dimension of lung surface area
Lung clearance rates
Particle deposition fractions
Particle mass and volumetric metrics as opposed to the particle surface area metric
Particle density
The MAK commission has been well aware of this criticism because Morfeld and Bruch intervened several times, repeating mainly the same arguments. The working group, which has prepared the draft recommendation, appreciated these comments and carefully discussed all aspects including the input data mentioned above. However, most of them could not be accepted as outlined below.
To derive a MAK value, the MAK commission applied two different models [2]. Using the methods proposed within the MAK documentation, Morfeld et al. [1] claim that they were unable to reproduce the OEL for GBP recommended by the Commission, and stated that they have identified substantial errors in the models.
One approach (Model A) the MAK Commission used to derive a limit value for GBP is based on chronic inhalation study with toner dust and TiO2 in rats. The NOAEC with no increased numbers of polymorphonuclear neutrophilic leukocytes (PMN) in lung lavage determined in these experiments was extrapolated to the human equivalent concentration calculating the NOAE particle dose deposited in the lung employing the MPPD (multi-path particle dosimetry) model Version-V2.0 program. All the input data for this calculation are given in the appendix of the MAK documentation and have been justified in detail in the MAK documentation.
On the basis of these experimental data with toner and titanium dioxide and by means of the MPPD-V2.0 model, human equivalent concentrations (HEC) of 0.11 mg/m³ and 0.25 mg/m³ were derived from the toner and the titanium dioxide study for GBP with the density of 1 g/cm³.
To assure that the approach is correctly described, two experts from two different institutions have been asked to independently use the model and the respective input data to calculate the deposited particle dose. Using the MPPD model and the input data described in the MAK documentation, both experts derived at the same deposited dose values as described in the MAK documentation.
Thus, the statement by Morfeld et al. [1] that they were unable to reproduce the deposited dose is not justified. Obviously, Morfeld et al. used different input data and thus consequently obtained different deposited doses than the MAK Commission.
Although the rational for using the specific input data for the HEC approach are described in the MAK Documentation we briefly address them again.
Lung surface area
The MAK Commission used the value of alveolar lung surface area of 57.22 m² for humans; this is the lung volume at the end of a normal exhalation and can be determined by mechanical lung function measurements as the Functional Residual Capacity [3, 4]. The MAK Commission extensively discussed the study of Gehr et al. [5] and came to the conclusion that the value of Gehr of 143 m² represents the lung area after maximal inhalation and is not realistic for workplace conditions. The MAK Commission states that the value of 57.22 m² is in the lower range of the published values [2].
Gehr et al. [5] themselves give an explanation why the determined alveolar surface area in their study does not represent the "true" value: "On the other hand it is known that the epithelial surface (defined in the publication as surface membrane of alveolar epithelial cells) does not correspond to the true alveolar surface in the living air-filled lung; this surface is rather formed by the air-fluid interface of the surface lining layer which smoothes the epithelial surface. We have shown on rat lungs that the "true" alveolar surface available for gas must be by 25-50% smaller than the epithelial surface depending on the level of air space inflation. If this is taken into consideration the "true" alveolar surface of the human lungs included in this study is reduced to 70-100 m²…" In their study the lungs were "fully inflated to near total lung capacity" and this inflation was achieved by instilling the fixative in aqueous solution into the airways. This does not represent the real alveolar surface area for a lightly working person and therefore this value was not taken into consideration by the MAK Commission.
Clearance rates
The average clearance half-time value for humans of 400 days was calculated by Bailey et al. and Kreyling and Scheuch [6, 7]. The same value was used also for the derivation of "General Threshold of Dust" [8].
The study of Gregoratto et al. [9] calculates a clearance half-time for humans of 300 days. "About 40% of an insoluble material deposited in the alveolar-interstitial region remains sequestered indefinitely and slowly clears only to the lymph nodes. The remaining material is cleared with half-time of about 300 days." The model is based on studies with long-lived radionuclides uranium-238, plutonium-239, americium-241 and cobalt-60. The authors do not state clearly whether the calculation based on this model is valid for other insoluble or poorly soluble particles.
For rats the clearance half-time used is 60 days [10] (see also figure 5 in "General Threshold of Dust" [8]). Borm et al. [11] refer to a normal clearance half-time of 60-80 days for rats.
Deposition fractions
The mean deposition fractions were calculated for rats and humans by applying the MPPD (multi-path particle dosimetry) Version-V2.0 program. All the input data for this calculation are given in the appendix of the MAK documentation; therefore the approach of the MAK Commission is transparent and the results should be easily reproducible.
Mass and volumetric metrics as opposed to the particle surface area metric
The MAK Commission based the derivation of the MAK value on the principle that inflammation and particle retention kinetics are largely driven by the volumetric particle dose. Since phagocytosis of GBP depends primarily on their size and not on their density, the most appropriate manner to compare various dusts appears to be the volume of the material [10].
The data available to the MAK Commission did not provide values for the surface area of the particles as metric. The BET surface area is generally used as a surrogate for particle surface area. However, the issue of surface area measurement in practice is complicated. Therefore and because it is not known what part of the BET surface may be the biological/toxicological relevant particle surface, this parameter was not chosen as a dose metric by the MAK Commission. Also, according to Borm et al. [11] the debate between "surface" and "volume" metric is still ongoing.
Density issue
According to Morrow and Mermelstein [12] "The volumetric amount of dust available for phagocytosis is the significant factor in dust overloading: consequently a correction should be made in comparing dust of high and low density."
In general, the retained particle dose is determined through the deposited particles minus the particles which are taken up by the macrophages and have been carried away. The highest inhaled particle mass below overload is 6% of the macrophage volume, which is considered as NOAEC. The deposited particle mass on the lung surface takes a certain volume, depending on the density. From this particle volume the macrophages can only phagocytose a portion which will fill up 6% of their volume.
Because the particle mass of TiO2 with a density of 4.3 represents a different particle volume at NOAEC than the particle volume of a particle with the density of 1, the 6 % of macrophage volume will contain different masses of particles depending on their densities. Therefore, at the same mass concentration (mg/m³) the particles with low density give rise to higher volume than the particles with higher density. As a consequence, the particle clearance and therefore the retained particle dose is not dependent on the particle density per se but on the particle volume (Density=mass/volume).
The exposure concentration is not identical with the effect concentration, which is the deposited/retained particle dose. The HEC calculation is based on the assumption that humans and rats react equally sensitive by the same particle dose in the lungs, as long as there are no quantitative data available that prove the opposite.
With the second approach (Model B) employed by the MAK Commissionan OEL for GBP was calculated based on NOAECs in rat inhalation studies and on volume loads of GBP not exceeding 6% of alveolar macrophage volume as described by Pauluhn [13] and further developed in Pauluhn [14]. The latter publication confirmed the value of 0.5 mg/m3, which is described in the MAK documentation [2].
Taking into account that all three values were determined by means of experimental data and model calculations, the MAK Commission decided to use the average of the values obtained with Model A (0.11 and 0.25 mg/m3) and and B (0.5 mg/m3) resulting in an OEL of 0.3 mg/m3 for the respirable dust fraction of GBP with a material density of 1 g/cm3.
It should be emphasized that the General Dust Limit for GBP was chosen to avoid particle induced chronic inflammatory response in the lung. By avoiding this GBP related adverse health effect, the risk of possible fibrogenic and neoplastic effects will be minimized as well. Epidemiological studies available on GBP do not allow a quantitative evaluation of an OEL for GBP induced chronic inflammation in the lung of exposed workers. Therefore, the Commission had to rely on data from inhalation studies with rats, the most sensitive species for GBP induced effects in the lung.
Regarding the carcinogen classification group 4, this is based on lung cancer induction on overload conditions in experimental animals. However, as long as lung overload is prevented and thus chronic inflammation, no excess cancer risk is expected.
Altogether, the MAK proposal of an OEL for GBP has been discussed comprehensively, taking into account all available epidemiological, experimental and mechanistic data, a scientific procedure always applied by the MAK Commission to substances under evaluation.
References:
1. Morfeld P, Bruch J, Levy L, Ngiewih Y, Chaudhuri I, Muranko HJ, Myerson R, McCunney RJ. Translational toxicology in setting occupational exposure limits for dusts and hazard classification – a critical evaluation of a recent approach to translate dust overload findings from rats to humans. Particle and Fibre Toxicology. 2015;12:3. doi:10.1186/s12989-015-0079-3.
2. Hartwig A. General threshold limit value for dust (R fraction) (biopersistent granular dusts) The MAK collection for occupational health and safety, part I, MAK value documentations, Wiley-VCH, Weinheim, 2012 (in German language, English version published in 2014).
5. Gehr P, Bachofen M, Weibel ER. The normal human lung: ultrastructure and morphometric estimation of diffusion capacity. Respir Physiol. 1978; 32:121-140.
6. Bailey MR, Fry RA, James AC. Long-term retention of particles in the human respiratory tract. J Aerosol Sci. 1985;16:295-305.
7. Kreyling WG, Scheuch G. Clearance of particles deposited in the lungs. In: Gehr P, Heyder J (edts) Particle-Lung Interactions. Marcel Dekker, New York, Basel, 2000; 232-376.
8. Greim H (ed) General threshold limit value for dust. The MAK collection for occupational health and safety, part I, MAK value documentations, Wiley-VCH, Weinheim, 1997 (in German language, English version published in 1999).
9. Gregoratto D, Bailey MR, Marsh JW. Modelling particle retention in the alveolar-interstitial region of the human lungs. J Radiol Prot. 2010; 30:491-512.
10. Muhle H, Creutzenberg O, Bellmann B, Heinrich U, Mermelstein R. Dust overload in lungs: investigations of various materials, species differences, and irreversibility of effects. J Aerosol Med. 1990; 3, Suppl 1: 111-128.
11. Borm P, Cassee FR, Oberdörster G. Lung particle overload: old school – new insights? Particle and Fibre Toxicology. 2015; 12:10. doi:10.1186/s12989-015-0086-4
12. Morrow PE and Mermelstein R. Inhalation toxicology: The design and interpretation of inhalation studies and their use in risk assessment. Springer Verlag, New York, 1988; 103-117.
13. Pauluhn J. Poorly soluble particulates: Searching for a unifying denominator of nanoparticles and fine particles for DNEL estimation. Toxicology. 2011; 279:176-188.
14. Pauluhn J. Derivation of occupational exposure levels (OELs) of low-toxicity isometric biopersistent particles: how can the kinetic lung overload paradigm be used for improved inhalation toxicity study design and OEL-derivation? Particle and Fibre Toxicology. 2014; 11:72.
Reply on behalf of the 'Permanent Senate Commission for the Investigation of Health Hazards of Chemical Compounds in the Work Area' (MAK Commission)
10 September 2015
Threshold value for Granular biopersistent particles (GBP)
Recently, Morfeld et al. [1] commented on the OEL for Granular biopersistent particles (GBP) without known specific toxicity proposed by the German MAK commission [2]. They concluded that the scientific reliability of the proposed GBP OEL and carcinogen classification is limited due to unjustified "translational toxicology". Before going into detail, I want to start with some general remarks.
The MAK documentation is based on the outcome of a working group consisting of members of the MAK commission as well as of the "Committee on Hazardous Substances" (AGS), which has been discussing the subject for several years. Thus, due to the great impact of a respective OEL, all arguments, including those of international renowned scientists in the field of particle toxicology outside the working group, have been considered within this process, deriving to a consensus as published in the MAK documentation [2]. Also Peter Morfeld and Joachim Bruch, two of the authors of the comments, have been involved in this process, the latter as a member of the working group.
Their criticism is based on the following arguments:
1. Using the methods proposed within the MAK documentation, they were unable to reproduce the OEL for GBP recommended by the Commission, and claimed to identify substantial errors in the models.
2. In their opinion, considerable shortcomings were:
The MAK commission has been well aware of this criticism because Morfeld and Bruch intervened several times, repeating mainly the same arguments. The working group, which has prepared the draft recommendation, appreciated these comments and carefully discussed all aspects including the input data mentioned above. However, most of them could not be accepted as outlined below.
To derive a MAK value, the MAK commission applied two different models [2]. Using the methods proposed within the MAK documentation, Morfeld et al. [1] claim that they were unable to reproduce the OEL for GBP recommended by the Commission, and stated that they have identified substantial errors in the models.
One approach (Model A) the MAK Commission used to derive a limit value for GBP is based on chronic inhalation study with toner dust and TiO2 in rats. The NOAEC with no increased numbers of polymorphonuclear neutrophilic leukocytes (PMN) in lung lavage determined in these experiments was extrapolated to the human equivalent concentration calculating the NOAE particle dose deposited in the lung employing the MPPD (multi-path particle dosimetry) model Version-V2.0 program. All the input data for this calculation are given in the appendix of the MAK documentation and have been justified in detail in the MAK documentation.
On the basis of these experimental data with toner and titanium dioxide and by means of the MPPD-V2.0 model, human equivalent concentrations (HEC) of 0.11 mg/m³ and 0.25 mg/m³ were derived from the toner and the titanium dioxide study for GBP with the density of 1 g/cm³.
To assure that the approach is correctly described, two experts from two different institutions have been asked to independently use the model and the respective input data to calculate the deposited particle dose. Using the MPPD model and the input data described in the MAK documentation, both experts derived at the same deposited dose values as described in the MAK documentation.
Thus, the statement by Morfeld et al. [1] that they were unable to reproduce the deposited dose is not justified. Obviously, Morfeld et al. used different input data and thus consequently obtained different deposited doses than the MAK Commission.
Although the rational for using the specific input data for the HEC approach are described in the MAK Documentation we briefly address them again.
Lung surface area
The MAK Commission used the value of alveolar lung surface area of 57.22 m² for humans; this is the lung volume at the end of a normal exhalation and can be determined by mechanical lung function measurements as the Functional Residual Capacity [3, 4]. The MAK Commission extensively discussed the study of Gehr et al. [5] and came to the conclusion that the value of Gehr of 143 m² represents the lung area after maximal inhalation and is not realistic for workplace conditions. The MAK Commission states that the value of 57.22 m² is in the lower range of the published values [2].
Gehr et al. [5] themselves give an explanation why the determined alveolar surface area in their study does not represent the "true" value: "On the other hand it is known that the epithelial surface (defined in the publication as surface membrane of alveolar epithelial cells) does not correspond to the true alveolar surface in the living air-filled lung; this surface is rather formed by the air-fluid interface of the surface lining layer which smoothes the epithelial surface. We have shown on rat lungs that the "true" alveolar surface available for gas must be by 25-50% smaller than the epithelial surface depending on the level of air space inflation. If this is taken into consideration the "true" alveolar surface of the human lungs included in this study is reduced to 70-100 m²…" In their study the lungs were "fully inflated to near total lung capacity" and this inflation was achieved by instilling the fixative in aqueous solution into the airways. This does not represent the real alveolar surface area for a lightly working person and therefore this value was not taken into consideration by the MAK Commission.
Clearance rates
The average clearance half-time value for humans of 400 days was calculated by Bailey et al. and Kreyling and Scheuch [6, 7]. The same value was used also for the derivation of "General Threshold of Dust" [8].
The study of Gregoratto et al. [9] calculates a clearance half-time for humans of 300 days. "About 40% of an insoluble material deposited in the alveolar-interstitial region remains sequestered indefinitely and slowly clears only to the lymph nodes. The remaining material is cleared with half-time of about 300 days." The model is based on studies with long-lived radionuclides uranium-238, plutonium-239, americium-241 and cobalt-60. The authors do not state clearly whether the calculation based on this model is valid for other insoluble or poorly soluble particles.
For rats the clearance half-time used is 60 days [10] (see also figure 5 in "General Threshold of Dust" [8]). Borm et al. [11] refer to a normal clearance half-time of 60-80 days for rats.
Deposition fractions
The mean deposition fractions were calculated for rats and humans by applying the MPPD (multi-path particle dosimetry) Version-V2.0 program. All the input data for this calculation are given in the appendix of the MAK documentation; therefore the approach of the MAK Commission is transparent and the results should be easily reproducible.
Mass and volumetric metrics as opposed to the particle surface area metric
The MAK Commission based the derivation of the MAK value on the principle that inflammation and particle retention kinetics are largely driven by the volumetric particle dose. Since phagocytosis of GBP depends primarily on their size and not on their density, the most appropriate manner to compare various dusts appears to be the volume of the material [10].
The data available to the MAK Commission did not provide values for the surface area of the particles as metric. The BET surface area is generally used as a surrogate for particle surface area. However, the issue of surface area measurement in practice is complicated. Therefore and because it is not known what part of the BET surface may be the biological/toxicological relevant particle surface, this parameter was not chosen as a dose metric by the MAK Commission. Also, according to Borm et al. [11] the debate between "surface" and "volume" metric is still ongoing.
Density issue
According to Morrow and Mermelstein [12] "The volumetric amount of dust available for phagocytosis is the significant factor in dust overloading: consequently a correction should be made in comparing dust of high and low density."
In general, the retained particle dose is determined through the deposited particles minus the particles which are taken up by the macrophages and have been carried away. The highest inhaled particle mass below overload is 6% of the macrophage volume, which is considered as NOAEC. The deposited particle mass on the lung surface takes a certain volume, depending on the density. From this particle volume the macrophages can only phagocytose a portion which will fill up 6% of their volume.
Because the particle mass of TiO2 with a density of 4.3 represents a different particle volume at NOAEC than the particle volume of a particle with the density of 1, the 6 % of macrophage volume will contain different masses of particles depending on their densities. Therefore, at the same mass concentration (mg/m³) the particles with low density give rise to higher volume than the particles with higher density. As a consequence, the particle clearance and therefore the retained particle dose is not dependent on the particle density per se but on the particle volume (Density=mass/volume).
The exposure concentration is not identical with the effect concentration, which is the deposited/retained particle dose. The HEC calculation is based on the assumption that humans and rats react equally sensitive by the same particle dose in the lungs, as long as there are no quantitative data available that prove the opposite.
With the second approach (Model B) employed by the MAK Commission an OEL for GBP was calculated based on NOAECs in rat inhalation studies and on volume loads of GBP not exceeding 6% of alveolar macrophage volume as described by Pauluhn [13] and further developed in Pauluhn [14]. The latter publication confirmed the value of 0.5 mg/m3, which is described in the MAK documentation [2].
Taking into account that all three values were determined by means of experimental data and model calculations, the MAK Commission decided to use the average of the values obtained with Model A (0.11 and 0.25 mg/m3) and and B (0.5 mg/m3) resulting in an OEL of 0.3 mg/m3 for the respirable dust fraction of GBP with a material density of 1 g/cm3.
It should be emphasized that the General Dust Limit for GBP was chosen to avoid particle induced chronic inflammatory response in the lung. By avoiding this GBP related adverse health effect, the risk of possible fibrogenic and neoplastic effects will be minimized as well. Epidemiological studies available on GBP do not allow a quantitative evaluation of an OEL for GBP induced chronic inflammation in the lung of exposed workers. Therefore, the Commission had to rely on data from inhalation studies with rats, the most sensitive species for GBP induced effects in the lung.
Regarding the carcinogen classification group 4, this is based on lung cancer induction on overload conditions in experimental animals. However, as long as lung overload is prevented and thus chronic inflammation, no excess cancer risk is expected.
Altogether, the MAK proposal of an OEL for GBP has been discussed comprehensively, taking into account all available epidemiological, experimental and mechanistic data, a scientific procedure always applied by the MAK Commission to substances under evaluation.
References:
1. Morfeld P, Bruch J, Levy L, Ngiewih Y, Chaudhuri I, Muranko HJ, Myerson R, McCunney RJ. Translational toxicology in setting occupational exposure limits for dusts and hazard classification – a critical evaluation of a recent approach to translate dust overload findings from rats to humans. Particle and Fibre Toxicology. 2015;12:3. doi:10.1186/s12989-015-0079-3.
2. Hartwig A. General threshold limit value for dust (R fraction) (biopersistent granular dusts) The MAK collection for occupational health and safety, part I, MAK value documentations, Wiley-VCH, Weinheim, 2012 (in German language, English version published in 2014).
http://onlinelibrary.wiley.com/doi/10.1002/3527600418.mb0230stwe5314/pdf
3. Yeh HC and Schum GM. Models of human lung airways and their application to inhaled particle deposition. Bull Math Biol. 1980; 42:461-480.
4. US-EPA (US Environmental Protection Agency) Air Quality Criteria for Particulate Matter (Final Report, Oct 2004) US Environmental Protection Agency, EPA 600/P-99/002aF-bF, Washington, DC, USA, 2004.
http://www.cfpub.epa.gov/ncea/cfm/recordisplay.cfmßdeid=87903
5. Gehr P, Bachofen M, Weibel ER. The normal human lung: ultrastructure and morphometric estimation of diffusion capacity. Respir Physiol. 1978; 32:121-140.
6. Bailey MR, Fry RA, James AC. Long-term retention of particles in the human respiratory tract. J Aerosol Sci. 1985;16:295-305.
7. Kreyling WG, Scheuch G. Clearance of particles deposited in the lungs. In: Gehr P, Heyder J (edts) Particle-Lung Interactions. Marcel Dekker, New York, Basel, 2000; 232-376.
8. Greim H (ed) General threshold limit value for dust. The MAK collection for occupational health and safety, part I, MAK value documentations, Wiley-VCH, Weinheim, 1997 (in German language, English version published in 1999).
http://onlinelibrary.wiley.com/doi/10.1002/3527600418.mb0230stwe0012/pdf
9. Gregoratto D, Bailey MR, Marsh JW. Modelling particle retention in the alveolar-interstitial region of the human lungs. J Radiol Prot. 2010; 30:491-512.
10. Muhle H, Creutzenberg O, Bellmann B, Heinrich U, Mermelstein R. Dust overload in lungs: investigations of various materials, species differences, and irreversibility of effects. J Aerosol Med. 1990; 3, Suppl 1: 111-128.
11. Borm P, Cassee FR, Oberdörster G. Lung particle overload: old school – new insights? Particle and Fibre Toxicology. 2015; 12:10. doi:10.1186/s12989-015-0086-4
12. Morrow PE and Mermelstein R. Inhalation toxicology: The design and interpretation of inhalation studies and their use in risk assessment. Springer Verlag, New York, 1988; 103-117.
13. Pauluhn J. Poorly soluble particulates: Searching for a unifying denominator of nanoparticles and fine particles for DNEL estimation. Toxicology. 2011; 279:176-188.
14. Pauluhn J. Derivation of occupational exposure levels (OELs) of low-toxicity isometric biopersistent particles: how can the kinetic lung overload paradigm be used for improved inhalation toxicity study design and OEL-derivation? Particle and Fibre Toxicology. 2014; 11:72.
Competing interests
No competing interests to declare