2023 年 48 巻 9 号 p. 507-511
Allergic contact dermatitis is a common occupational and environmental health problem and setting of health-based exposure limits (HBELs) to prevent induction of skin sensitization is strongly desired. When manufacturing pharmaceuticals in a shared facility, cleaning validation using surface residue levels (SRLs) derived from permitted daily exposures (PDEs) is conducted to avoid cross-contamination from the perspective of protecting patients; however, it is unclear whether the SRLs are sufficient to prevent induction of skin sensitization for workers as well. In this study, we compared acceptable surface limits (ASLs) derived from acceptable exposure levels (AELs) based on EC1.6 obtained from local lymph node assay (LLNA): BrdU-ELISA for occupational risk management of skin sensitizers with PDE-based SRLs. ASLs for 1,4-phenylenediamine (GHS skin sensitization sub-category 1A), isoeugenol (sub-category 1A), and methyl methacrylate (sub-category 1B) were compared with SRLs based on the PDEs derived from their systemic effects. The results yielded an SRL for 1,4-phenylenediamine (PDE: 0.8 mg/day) of 30 mg/100 cm2, almost 1,000 times higher than ASL (0.031 mg/100 cm2) derived from its skin sensitization potency. SRL for isoeugenol (PDE: 3.1 mg/day) was 130 mg/100 cm2, over 500 times higher than ASL (0.18 mg/100 cm2). For methyl methacrylate (PDE: 5 mg/day) as well, SRL (200 mg/100 cm2) was higher, but it was within 20 times the ASL (10 mg/100 cm2). These results showed that ASL-based risk management is extremely important especially for strong sensitizers classified as GHS sub-category 1A for occupational skin sensitization risk management.
Allergic contact dermatitis, resulting from skin sensitization, is a common occupational and environmental health problem (Kimber et al., 2003). As allergic skin dermatitis can be induced by very low levels of exposure to skin sensitizers in sensitized workers and have long-lasting effects, setting of health-based exposure limits (HBELs) to prevent induction of skin sensitization is strongly desired. However, caution against skin sensitization risk in the work place is limited to the hazard communication by the Dermal Sensitization Notation (DSEN) by the American Conference of Governmental Industrial Hygienists (ACGIH) and quantitative HBELs based on the skin sensitizing potencies of chemicals are rarely established.
Regarding exposure control levels at shared facilities manufacturing different pharmaceutical products, swab limit level (SRL) is commonly used in cleaning validation to prevent cross-contamination from the perspective of protecting patients (EC, 2015; PIC/S, 2018; WHO, 2021). The SRLs are established based on permitted daily exposures (PDEs) as a HBEL, usually derived from the oral systemic effects. Although SRLs are considered to be sufficient to prevent adverse health effects in patients administrated drug products orally, it is unclear whether residue levels on equipment surfaces controlled by PDE-based cleaning validations is adequate from an occupational sensitization risk protection perspective.
In our previous study, a new approach for setting acceptable surface limits (ASLs) for chemicals on the surface of manufacturing equipment was proposed to reduce the risk of skin sensitization in workers (Fukushima et al., 2022). In the approach, the ASLs for the risk assessment are derived from the EC1.6 value obtained from an LLNA: BrdU-ELISA (OECD, 2018) with consideration of the sensitization assessment factors (SAFs) that reflect interspecies differences, individual differences, and the conditions of use. In this study, known skin sensitizers for which EC1.6 values of the LLNA: BrdU-ELISA are available were selected and ASLs derived from the skin-sensitizing potency were compared with the PDE-based SRLs calculated from the oral systemic effects, then the need for the exposure control based on the ASL to minimize skin sensitization risk for workers was examined.
Three chemicals, 1,4-phenylenediamine, isoeugenol and methyl methacrylate, were selected from the 15 chemicals listed in Kobayashi et al. (2020) with EC1.6 values of LLNA: BrdU-ELISA, as chemicals classified as having a specific target organ toxicity (repeated exposure) based on oral toxicity studies in the national GHS classification results published in the National Institute of Technology and Evaluation (NITE) website (NITE, accessed 14 April 2023), excluding chemicals classified as mutagenic or carcinogenic.
The LLNA: BrdU-ELISA EC1.6 data reported in Kobayashi et al. (2020) were used to set the ASL based on the skin sensitization potency. As a reference for the sensitizing potency of selected chemicals, GHS sub-categories according to the criteria for the LLNA: BrdU-ELISA (1A: EC1.6 value ≤ 6%; 1B: EC1.6 value > 6%), which was decided to be added to the GHS text by the UN sub-committee on its forty-third session (UN, 2022), were shown as well.
Toxicological information for setting the PDEs was obtained from the latest assessment report used for the GHS classification (NITE GHS Information, 2019; NITE GHS Information, 2011; NITE GHS Information, 2017), and the POD for systemic effects via the oral route was investigated.
Calculation of the ASL based on the skin sensitization potency
ASLs were calculated from acceptable exposure levels (AELs) derived from LLNA: BrdU-ELISA EC1.6 for the selected chemicals (Kobayashi et al., 2020) in accordance with the procedure proposed in the previous report (Fukushima et al., 2022). Briefly, the EC1.6 (%) values were converted to a load amount per unit skin surface area (μg/cm2) according to the following formula 1.
EC1.6 (mg/cm2) = [EC1.6 (%) x 10 mg/mL x 0.025 (mL/ear)] / applied area (1 cm2/ear) (formula 1)
To calculate the AELs, a safety assessment factor (SAF) interspecies value of 6, based on the EC1.6 values/human repeat insult patch test (HRIPT) NOEL ratios, a SAFinterindividual value of 10, and a SAFfrequency/duration value of 3 were applied, referring to previous literature on SAFs for skin sensitization QRA (Api et al., 2008; Basketter and Safford, 2016; Api et al., 2020) and the Guidance on IR&CSA (ECHA, 2012), then the composite SAF was calculated as 180. Consequently, the AELs were derived according to the following formula 2.
AEL (mg/cm2 skin surface) = EC1.6 (mg/cm2 skin surface) / composite SAF (formula 2)
Where, composite SAF = SAFinterspecies (6) x SAFinter-individuals (10) x SAFfrequency/durations (3) = 180
Consequently, ASLs for 100 cm2 equipment surface were derived according to the following formula 3. In the calculation of the ASLs, it was assumed that 100% of the residual chemical on the equipment surface may transfer to the skin as a conservative approach.
ASL (mg/100 cm2 equipment surface) =AEL (mg/cm2 skin surface) x transfer rate (%) x 100 (cm2) (formula 3)
Calculation of the PDE-based SRLSRLs for the cleaning validation were calculated using the PDEs for systemic effects.
First, the PDEs for the selected chemicals derived from the oral systemic effects were determined in accordance with the PIC/S guideline (PIC/S, 2018) (formula 4).
PDE (mg/day) = NO(A)EL (mg/kg/day) x weight adjustment / F1 x F2 x F3 x F4 x F5 (formula 4)
Where, weight adjustment: 50 kg
F1: A factor (values between 2 and 12) to account for extrapolation between species
F2: A factor of 10 to account for variability between individuals
F3: A factor 10 to account for repeat-dose toxicity studies of short duration
F4: A factor (1-10) that may be applied in cases of severe toxicity
F5: A variable factor that may be applied if the no-effect level was not established
Then, SRLs were derived following the derivation of maximum safe carryover (MSC) using the derived PDEs (formula 5), as well as maximum safe surface residue (MSSR) (formula 6) in line with the procedure described in WHO (2021). Then, SRLs were set as acceptable residue levels in 100 cm2 usually used in swab testing (formula 7). The example values of batch size (BS), maximum daily dose (MDD), and example shared surface area shown in Walsh et al. (2017) were used for the SRL calculations in this study.
MSC (g) = PDE (µg/day) x BS (kg) / MDD (mg/day) (formula 5)
Where, batch size (BS): 100 kg
maximum daily dose (MDD): 10,000 mg
MSSR (mg/cm2) = MSC (g) x 1,000 / Shared surface area (cm2) (formula 6)
Where, shared surface area: 25,000 cm2
SRL (mg/100 cm2) = MSSR (mg/cm2) x 100 (cm2) (formula 7)
Comparison of the PDE-based SRL and ASLConsequently, the ASLs derived from AELs based on the skin sensitization potency and PDE-based SRLs derived from the systemic effects for the selected chemicals were compared.
The AELs derived from LLNA: BrdU-ELISA EC1.6 and the calculated ASLs for the selected chemicals are shown in Table 1.
| 1,4-Phenylenediamine | Isoeugenol | Methyl methacrylate | |
|---|---|---|---|
| GHS skin sensitization sub-category 1) | 1A | 1A | 1B |
| LLNA EC1.6 2) (%) | 0.22 | 1.3 | 75 |
| LLNA EC1.6 (mg/cm2) | 0.055 | 0.33 | 19 |
| AEL (mg/cm2) | 0.00031 | 0.0018 | 0.10 |
| Transfer rate (%) | 100 | 100 | 100 |
| ASL (mg/100 cm2) | 0.031 | 0.18 | 10 |
1) GHS skin sensitization sub-category classified in accordance with the criteria for LLNA: BrdU-ELISA 1.6 values (1A: EC1.6 ≤ 6%; 1B EC1.6 > 6) (UN, 2022)
The ASLs were determined as being 0.031 mg/100 cm2 (1,4-phenylenediamine), 0.18 mg/100 cm2 (isoeugenol), and 10 mg/100 cm2 (methyl methacrylate), based on the EC1.6 values of 0.22%, 1.3%, and 75%, respectively.
PDE-based SRL derived from the systemic effectsThe derivation of PDEs and PDE-based SRLs for the selected chemicals are shown in Table 2.
| 1,4-Phenylenediamine | Isoeugenol | Methyl methacrylate | ||
|---|---|---|---|---|
| PDE derivation | POD | 90-day repeated-dose toxicity study in rats NOEL = 4 mg/kg/day (based on the liver and kidney effects at 8 mg/kg/day) (SCCS, 2012) |
2-year repeated-dose toxicity study in mice LOAEL = 75 mg/kg/day (based on the histopathological changes in the olfactory epithelium) (NTP, 2010) |
2-year repeated-dose toxicity study in rats NOEL = 5 mg/kg/day (based on the increased kidney weights at 140 mg/kg/day) (MOE, 2013) |
| Adjustment factors1) | F1: 5 F2: 10 F3: 5 F4: 1 F5: 1 |
F1: 12 F2: 10 F3: 1 F4: 1 F5: 10 |
F1: 5 F2: 10 F3: 1 F4: 1 F5: 1 |
|
| PDE (mg/day) | 0.8 | 3.1 | 5 | |
| MSC (g) | 8 | 31 | 50 | |
| MSSR (mg/cm2) | 0.3 | 1.3 | 2 | |
| SRL (mg/100 cm2) | 30 | 130 | 200 | |
1) F1: A factor to account for extrapolation between species; F2: A factor to account for variability between individuals; F3: A factor to account for repeat-dose toxicity studies of short durations; F4: A factor that may be applied in cases of severe toxicity; F5: A variable factor that may be applied if the no-effect level was not established.
Taking into consideration the results of the toxicological study, the POD for PDE setting of 1,4-phenylenediamine was determined as the no observed effect level (NOEL) of 4 mg/kg/day based on kidney and liver effects observed at 8 mg/kg/day in a 90-day repeated-dose toxicity study conducted in rats (SCCS, 2012). The PDE of 1,4-phenylenediamine was calculated as 0.8 mg/day, and accordingly, the SRL was derived as 30 mg/100 cm2. As for isoeugenol, the POD for PDE setting was determined as the low observed adverse effect level (LOAEL) of 75 mg/kg/day based on the histopathological changes observed in the olfactory epithelium of the nasal cavity in a 2-year repeated-dose toxicity study conducted in mice (NTP, 2010). Then, the PDE and SRL were derived as 3.1 mg/day and 130 mg/100 cm2, respectively. In regard to methyl methacrylate, based on the increased kidney weights observed at 140 mg/kg/day in a 2-year oral repeated-dose toxicity study conducted in rats (MOE, 2013), the NOEL of 5 mg/kg/day was selected as the POD and the PDE and SRL were determined as 5 mg/day and 200 mg/100 cm2, respectively.
Comparison of the ASLs and PDE-based SRLsResults of comparison of the ASLs derived from the skin sensitization potency and PDE-based SRLs derived from the oral systemic effects for the selected chemicals are shown in Table 3.
| 1,4-Phenylenediamine | Isoeugenol | Methyl methacrylate | |
|---|---|---|---|
| ASL (mg/100 cm2) | 0.031 | 0.18 | 10 |
| SRL (mg/100cm2) | 30 | 130 | 200 |
| SRL/ASL | 967 | 722 | 20 |
SRLs of 1,4-phenylenediamine and isoeugenol, both classified as sub-category 1A according to the sub-categorization criteria (1A: EC1.6 value ≤ 6%; 1B: EC1.6 value > 6% (UN, 2022)), far exceeded the ASL for each chemical (i.e., almost 1,000 times higher than the ASL for 1,4-phenylenediamine and over 700 times higher than the ASL for isoeugenol, respectively). The SRL of methyl methacrylate, which was classified as sub-category 1B, was also higher, but within 20 times higher than the ASL of the chemical. These results indicate that exposure control using SRLs derived from the systemic effects may be insufficient to manage the skin sensitization risk in workers, especially for strong skin sensitizers corresponding to GHS sub-category 1A.
In this study, the ASLs derived from AELs based on the skin-sensitizing potency of known sensitizers were compared with the PDE-based SRLs calculated from the oral systemic effects in shared facilities. Although it is obvious that HBELs need to be set for each exposure route and endpoint concerned and setting HBELs based on sensitization is necessary for sensitization risk management for workers, it is assumed that in most cases the control of residue levels on the equipment surface is primary considered based on the cleaning validation aiming at the protection of the patients. Therefore, it is meaningful that this study specifically demonstrated the necessity of exposure control for worker focusing on sensitization by presenting the comparison results of ASLs derived from AELs based on the skin-sensitizing potency and PDE-based SRLs for some skin sensitizers. EC1.6 used for the derivation of AEL is defined as the concentration of chemicals which gives positive response in LLNA: BrdU-ELISA. Skin sensitization potency data such as EC1.6 is known to be correlated with human experimental induction threshold data (Basketter and Safford, 2016). In this study, the AEL of each chemical at which skin sensitization is not expected to be induced in humans was derived from the EC1.6 value of each chemical, then the ASL was calculated. Therefore, acceptable residue levels for workers need to be set below the ASL calculated for each chemical. The rate of transfer of the chemicals from the equipment to the skin was assumed 100% as a conservative approach in the present study, however, studies on the transfer rates of the chemicals are needed to clarify this point.
Induction levels are known to be lower than sensitizing levels (Basketter et al., 2016); therefore, the ASL-based risk management is considered not to be sufficient to prevent induction in individuals already sensitized with similar substances. Nevertheless, the approach to prevent allergic contact dermatitis by avoiding sensitization caused by contaminated chemicals is extremely important under the sensitization risk management for workers.
In conclusions, in this study, ASLs derived from AELs based on the skin-sensitizing potency and SRLs derived from the PDEs based on the oral systemic effects and were compared. For the three skin sensitizers examined in this study, PDE-based SRLs exceeded ASLs derived from skin sensitization potency for workers. HBELs need to be set for each exposure route and endpoint concerned and ASL-based risk management for worker is extremely important, especially for strong sensitizers classified as GHS sub-category 1A for occupational skin sensitization risk management.
Conflict of interestThe authors declare that there is no conflict of interest.
