Registration Dossier

Diss Factsheets

Administrative data

Description of key information

Skin sensitisation

The skin sensitization potential of HPMA is clearly documented by human data which however does not allow robust subcategorization beyond the exclusion of a high potency. Based on the non-human data that partially do not support classification at all (in-chemico, in-vitro, valid animal data) or partially support classification as low-to-moderate potent sensitizer (defined approaches acc. OECD 497), HPMA is considered as low-to-moderate potent sensitizer (Cat 1B).

Respiratory sensitisation

Insufficient evidence to consider HPMA as a cause of respiratory sensitisation or OA (Weight-of-Evidence assessment)

Key value for chemical safety assessment

Skin sensitisation

Link to relevant study records
skin sensitisation: in vivo (LLNA)
Type of information:
experimental study
Adequacy of study:
key study
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Basic data given
equivalent or similar to guideline
OECD Guideline 429 (Skin Sensitisation: Local Lymph Node Assay)
Principles of method if other than guideline:
Modified by Kimber, I., et al (1989, 1991a, 1991b)
GLP compliance:
Type of study:
mouse local lymph node assay (LLNA)
other: CBA/Ca
Details on test animals and environmental conditions:
- Source: Harlan Olac Ltd., Bicester, Oxon or Barriered Animal Breeding Unit, Alderley Park
- Age at study initiation: young adult, 8-12 weeks

other: Acetone olive oil or DMF
5.0, 10.0, 25.0, 50.0%
No. of animals per dose:
Details on study design:
- Name of test method: murine local lymph node assay
- Criteria used to consider a positive response: LLNA results are expressed as testlcontrol (TIC) ratios, where ratios equal to or greater than three indicate a positive proliferative response.

TREATMENT PREPARATION AND ADMINISTRATION: 25 uI of one of three concentrations of the test chemical on the dorsum of both ears daily for three consecutive days.
Positive control substance(s):
mercaptobenzothiazole (CAS No 149-30-4)
Test group / Remarks:
Remarks on result:
other: in the paper the SI values were expressed as T/C values were values < 3 indicate a positive result
Test group / Remarks:
Remarks on result:
other: in the paper the SI values were expressed as T/C values were values < 3 indicate a positive result
Test group / Remarks:
Remarks on result:
other: in the paper the SI values were expressed as T/C values were values < 3 indicate a positive result
other: disintegrations per minute (DPM)
Remarks on result:
other: see Remark
Concentration (%) Laboratory A (vehicle) Laboratory B (vehicle) Laboratory C (vehicle) Laboratory D (vehicle) 5 1.1 (AOO) - (AOO) - (AOO) - (DMF) 10 1.2 0.8 1.0 1.4 25 1.3 1.0 1.9 0.7 50 - 0.9 0.8 0.9

To document the consistency in results of collaborating laboratories, the local lymph node assay was conducted under field conditions by four laboratories.

Interpretation of results:
not sensitising
The test substance is considered to be non-sensitizing following conduct of the local lymph node assay.
Executive summary:

The sensitization potential of Hydroxypropyl methacrylate was tested in a local lymph node assay using a modified protocol. This inter-laboratory study used four laboratoires to demonstrate the consistency of results. Groups of mice received 25 uL of one of

four concentrations of the test chemical on the dorsum of both ears daily for three consecutive days. Control mice received an equal volume of the relevant vehicle alone. The proliferative activity lymph node cells (LNC) was expressed as the number of radioactive disintegrations per minute (dpm) per lymph node for each experimental group. The ratio of 3HTdR incorporation by LNC of test

lymph nodes relative to that recorded for control lymph nodes test/control (T/C) ratio was calculated for each test group. HPMA was considered not to be a sensitizer as the T/C ratio was not greater than 3.

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (sensitising)
Additional information:

  1. General introduction to sensitisation

The capacity of HPMA to induce sensitization is directly related to its metabolism and reactive chemistry. The general metabolic pathway for HPMA consists of ester hydrolysis to methacrylic acid and propylene glycol, a process catalyzed by carboxylesterase enzymes, which are further metabolized to carbon dioxide and water, in the case of methacrylic acid, and glucose or D-lactate, in the case of propylene glycol. Hydrolysis and further metabolism of HPMA occur rapidly following exposure as discussed in the chapter “Toxicokinetics”.   

Another potentially important metabolic pathway for HPMA and related esters involves reaction with tissue nucleophiles via Michael addition on Cβ of the α,β-unsaturated carboxyl group. However, the electrophilic activity of methacrylate esters, as reflected by GSH conjugation, is less than that of corresponding acrylate esters, as shown in Borak, et al. 2011. The electrophilic reactivity of low-molecular-weight molecules is an important aspect of their potential to act as sensitizers. In skin sensitization studies for example, a key early step in the process leading to sensitization is the formation of covalent adducts with a carrier protein, thereby forming an antigenic hapten-protein complex. Accordingly, electrophilic reactivity also serves as a predictor of sensitization potential. 

Hydrolysis of HPMA reduces its sensitization potential because, under physiological conditions, methacrylic acid is not electrophilic or protein reactive as shown in vivo (as discussed by Borak et al. 2011). Also, propylene glycol as other primary metabolite is not a skin sensitizer (e.g., Basketter et al. 1998). Metabolism data suggest that hydrolysis is the principal pathway of HPMA metabolism, that the hydrolysis is basically a detoxification process for the sensitisation potential of HPMA. Electrophilic reaction via Michael addition is expected to play only a minor role and thus may explain why in human patch tests, skin sensitization occurs apparently only at high tissue concentrations.


2. Skin sensitisation


2.1 Dermal absorption

Heylings (2013) used a QSPeR model for whole human skin based on that described by Potts and Guy (1992) to predict the dermal penetration rate of a large number of methacrylate esters with the help of molecule size and distribution coefficient, including the hydroxyalkyl methacrylates HEMA & HPMA. As relatively small, hydrophilic esters the hydroxyalkyl methacrylates are predicted to be rapidly absorbed (HPMA: “high” relative dermal absorption with a predicted dermal flux of 151 µg/cm2/h). However, as indicated by studies summarised in CSR section 5.2 metabolism (IUCLID 7.1.1), they will be subject to hydrolysis by local esterases in the stratum corneum.

2.2. Human data

The most relevant clinical studies for HPMA are presented in Table 1 that is available in the attached document “Endpoint summary skin sens HPMA”. In all these studies with either consecutive or unselected patients (where allergic contact dermatitis/ACD is generally suspected for a group of patients) or selected patients (ACD specifically suspected for acrylic compounds), the diagnostic method was patch testing (human diagnostic patch test, HDPT) according to international standards by dermatologists (Johansen et al. 2015). These typically retroperspective studies can allow an assessment of the prevalence of positive reactions in unselected or selected patient groups.

Details on Case Reports on numerous single cases have been compiled in a separate document attached to this endpoint summary.

In contrast to the numerous HDPT studies and case studies, data on dermal exposure levels to HEMA is scarce.

2.2.1 Adequacy of human data

In general, HDPT studies have several limitations for classification and labelling, as stated by an OECD expert group for the assessment of human data for skin sensitisation potential (OECD TG 497, Annex 4, 2021) that lead to a nonconsideration of this study type for potency considerations:

  • HDPT studies cover elicitation (response to challenge with allergen) and not induction of sensitisation.

  • In the case of HDPT data, previous exposure to the chemical often cannot be established with enough certainty to evaluate its potency for induction of sensitisation; even if it is established qualitatively, exposure can almost never be quantified in a satisfying manner (there might be exceptions at the workplace, but for consumers it is practically not possible). The situation is further complicated because patients might have had contact with other chemicals cross-reacting with the chemical in question.

For this assessment here, the OECD approach was partially followed in a way that HDPT data (and related cross-sectional studies) were considered only from a qualitative perspective, i.e. not used for considerations on potency/ subcategorisation.

Most of the HDPT studies available for HPMA have several additional limitations. For example, cross-reactivity between different (meth)acrylate monomers, concomitant sensitization, as well as impurities with undeclared mixtures of monomers in preparations used at work and in test substances, make accurate mapping of contact sensitization difficult (Kanerva et al. 1994, Rustermeyer et al. 1998, Kanerva et al 1988, Wrangsjö et al 2001). In most of the clinical reports, specific exposure to HPMA in patch-tested patients, or those who tested positive to the substance, is not verified. As another example, sorting irritant from allergic reactions can be challenging and was often not described in detail. Thus, the relevance of reactions was often not proven.


Prevalence in unselected patient groups

No studies in the general population or with unselected clinical patient groups are available.


Prevalence in selected patient groups

HPMA was usually tested as part of (meth)acrylate patch test series and its established test concentration is 2% in petrolatum. The patients were pre-selected strongly based suspected of having contact with (meth)acrylic compounds (women with artificial nails) or special occupational groups (dentists, dental workers, nail artists). A high number of diagnostic patch test studies on selected patients could be identified for the substance. The frequency of positive reactions was generally highly dependent on the selection of patient groups.

Patients with suspected exposure related to nail cosmetic products (occupational and general) had the highest incidences of positive reactions to HPMA. The respective publications rarely differentiated between consumer and/or professional use so it is nearly impossible to estimate exposure levels or a reliable reference size. Compared to the professional users of artificial nail systems, the positive reactions to HPMA seem to be less common among those who are only consumers.

No strict workplace studies could be identified for HPMA. However, some cross-sectional studies on dental technicians, who are at risk of developing a contact allergy due to exposure to acrylic compounds at work, share a similar design. Only the workers with skin symptoms were patch tested in these studies. Frequencies of positive reactions to the substance were between 12.7 and 28.5%. When evaluating the frequency of allergic reactions, it should be borne in mind that the targeted selection of patients tested with acrylates results in an artificial increase in rates (this is particularly clear for the sensitisation rates in tested dental technicians).

The numbers / incidences amongst dental technicians must be rated against a very high number of workers in this field. Because most of the reported cases for dental technicians were reported by IVDK database in Germany, a suitable reference is the total number of people working in this field. For Germany in 2018 total of around 64,700 dental technicians was published (Atlas Dentalstudie 2018). Between 1993 and 2015 a total number of 102 dental technicians tested positive for HPMA (Rustemeyer and Frosch 1996, Peiler and Rustemeyer 1996, Schnuch et al. 1998, Peiler et al. 2000, Heratizadeh et al. 2018). Related to the total number of dental technicians in Germany this is about 0.15%. This simple calculation does not take into account that some of these individual patients were probably included in more than one publication because the time spans overlap considerably (e.g. Peiler and Rustemeyer 1996, Schnuch et al. 1998).

Acrylates apparently do not play a substantial role as allergens in the general population as indicated by HDPT data on male painters with occupational dermatitis. In this subgroup 0.2% (1/415 patients) reacted positive to HPMA (Schubert et al. 2021).


Evaluation of human data

Taken together each of these studies is of limited power. Bias or confounders have not been ruled out fully with reasonable confidence. Nevertheless, there is evidence that HPMA has the potential to cause skin sensitisation in humans.


2.3 Non-human data

2.3.1 In-chemico/in-vitro:

There are in-chemico and in-vitro studies available covering all three key events (KE1-3).

Key Event



rating of potency in report (quantitive results)



Direct Peptide Reactivity Assay (DPRA)

2 x positive

moderate reactivity
weak positive
(mean peptide depletion 33.0 and 22)

Kolle S. (2013)
Nukada (2013)


Amino Acid Derivative Reactivity Assay (ADRA)

1 x positive
2 x negative


Fujita (2013)
Fujita (2019)
Yamamoto (2015)



Keratinocyte Activation Assay – LuSens


No assessment of potency

Kolle S. (2013)


U937 cell line activation test (U-SENS)*


No assessment of potency

Kolle S. (2013)


human Cell Line Activation Test (h-CLAT)



Ashikaga (2008, 2010)
Nukada (2011)


Interleukin-8 Reporter Gene Assay (IL-8 Luc Assay)



Takahashi (2011)

*Former MUSST: Dendritic Cell Line Activation Assay Myeloid U937 Skin Sensitization Test

All those studies are non-GLP studies and have been performed before the respective guidelines were fully validated.

There is no reliable explanation for the inconsistent results obtained in the available K1 studies apart from a possible low potency.

2.3.2 Integrated approaches

HPMA was evaluated as part of a comprehensive dataset of 196 chemicals attached as Annex 2 to the Supporting document to the Guideline (GL) on Defined Approaches (DAs) for Skin Sensitisation (Series on Testing and Assessment No. 336) on the basis of in chemico, in vitro and, for potency aspects, in silico data.

The outcome of the evaluation “Category 1B with high confidence” for the two DAs that were used for potency estimations (OECD TG 497, Annex 2, 2021).


2.3.3 Animal data

The sensitising potential of HPMA has been investigated in two murine local lymph node assays and in several guinea pig studies.


For HPMA two negative LLNA studies are available.

Negative results were obtained in the local lymph node assay in CBA/Ca mice with 10, 25 and 50% HPMA in acetone/olive oil (4:1) using a modified protocol in this inter-laboratory study (Basketter and Scholes 1992; Scholes et al. 1992). HPMA was considered not to be a sensitizer as the T/C ratio for 3HTdR incorporation was below 3.

In a validation study for the LLNA:BrdU-ELISA using a different mouse strain (BALB/c), a series of 32 substances was tested which were previously tested in CBA/JN mice. HPMA was tested in only one concentration (50%), which was the highest concentration tested in the ICCVAM formal validation study for the LLNA:BrdU-ELISA. HPMA was identified as non-sensitising in this study with a mean SI of 0.75; , no EC3 value was calculated (Hou 2015).


Guinea pigs

Several Guinea Pig Maximisation tests (Basketter, 1992; Clemmensen, 1984; Bjoerkner, 1984) and a split adjuvant test (Rao, 1981) are available for HPMA. In these studies, none or few animals only (< 30%) were sensitized. Cross-reactions in animals

Cross-reactions have been described in various animal studies and are described in the attached document “Endpoint summary skin sens HPMA”.

Evaluation of hazard

Data from animal tests, performed according to existing guidelines, do not meet the criteria for classification.


2.4 Comparison against the CLP criteria

Integrated in-chemico/in-vitro/in-silico approaches

The two defined approaches of OECD 497 used data from in-chemico, in-vitro and in-silico and calculated a potency of Cat 1B with high confidence. Based on these approaches, a classification for skin sensitisation with Cat 1B would be warranted.

 Animal data

In the key animal studies (two LLNA and several GPMT) HPMA indicated a very low skin sensitisation potential below classification criteria. Based on these studies a classification for skin sensitisation would not be warranted.

 Human data


Frequencies of positive patch tests in selected dermatitis patients (≥ 2.0 %) indicate high frequency. Accordingly, in the available cross-sectional studies on a risk occupation (mimicking a workplace study), the frequency of positive patch tests was well above the cut-off value of 1.0% (>> 1.0 % for nail care workers; >> 1.0 % for dentists/ dental workers). For more details, see attached document “Endpoint summary skin sens HPMA”.

Exposure aspects

While cross reactivity to other methacrylates is expected and methacrylates are common substances in occupational uses in dental materials, glues and several other products with skin contact, relatively high exposure is expected at concentrations > 1%, repeated exposures and numbers of exposures can be assumed. For more details, see attached document “Endpoint summary skin sens HPMA”.


For the elicitation of skin sensitizing effects in humans by HPMA, relatively high local concentrations are required. Following CLP criteria it can be concluded that HPMA is not a high potency skin sensitizer, in spite of the high number of positive patch test results reported. While the quality of the human data is considered as limited in the context of classification (e.g., studies with unselected persons are missing), a combination of a relatively high frequency of occurrence with a relatively high exposure requirement would warrant a classification with Cat 1 without subcategorization, according to ECHA’s Guidance on the Application of CLP Criteria 2017b.


2.5 Integration of all available data

Two defined Approaches according to OECD 497 using the aforementioned in-chemico and in-vitro data plus QSAR/ in-silico data conclude on a low-to-moderate potency of HPMA for skin sensitation with a high confidence (Cat 1B), however did not take into account some uncertainty arising from the observed inconsistence of the in-chemico and in-vitro data.

HPMA has been evaluated for skin sensitization potential in reliable tests in experimental animals. Studies in which HPMA was applied to intact skin elicited consistently non-sensitizing responses. Only at very high concentrations for induction and elicitation – an aspect that lowers the adequacy of these studies - few animals were sensitized using the Magnusson-Kligman protocol (<30%, as CLP criteria). 

Only human diagnostic patch test (HDPT) studies in selected patients are available for evaluation of skin sensitization potential in humans. These studies have several limitations for classification and labelling, and especially for potency assessment.

The frequency of positive reactions to HPMA in diagnostic patch tests can be considered substantial in a defined population. Human data supports the classification of HPMA as a skin sensitiser. However, there is adequate information that this occurs only in relation to relatively high exposures. Overall and seen for its own, the human data alone justify classification of HPMA as skin sensitizer. A high potency can be excluded but further robust subcategorization is not possible due to inadequacy of the data.

In conclusion, the skin sensitization potential of HPMA is clearly documented by human data which however does not allow robust subcategorization beyond the exclusion of a high potency. Based on the non-human data that partially do not support classification at all (in-chemico, in-vitro, valid animal data) and that partially support classification as low-to-moderate potent sensitizer (defined approaches acc. OECD 497), a subcategorization towards Cat 1B for HPMA is warranted.


Additional literature (without robust study summaries in IUCLID)

Atlas Dentalstudie (2018) published by GFDI Gesellschaft zur Förderung der Dental-Industrie mbH.

Basketter DA, Gerberick GF, Kimber I. Strategies for identifying false positive responses in predictive skin sensitization tests. Food Chem Toxicol. 1998 Apr;36(4):327-33. doi: 10.1016/s0278-6915(97)00158-0. PMID: 9651050.

Borak J, Fields C, Andrews LS, Pemberton MA. Methyl methacrylate and respiratory sensitization: a critical review. Crit Rev Toxicol. 2011 Mar;41(3):230-68. doi: 10.3109/10408444.2010.532768. PMID: 21401327; PMCID: PMC3072694.

Kolle SN, Natsch A, Gerberick GF and Landsiedel R; A review of substances found positive in 1 of 3 in vitro tests for skin sensitization. Regulatory Toxicology and Pharmacology 106 (2019) 352–368

Schnuch A. (2017) Critical Comment on Evaluation of selected sensitizing fragrance substances. A LOUS follow-up project The Danish EPA, Copenhagen (2016)

Respiratory sensitisation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:

Weight-of-Evidence assessment on the potential of HEMA and HPMA for respiratory sensitisation - summary

All available information has been systematically assessed for its evidence that supports, or contradicts, the hypothesis that HEMA and/or HPMA can cause the development of respiratory sensitisation, or occupational asthma (OA), in subjects that were not previously asthmatic. This has been done in a structured manner following CLP requirements for the classification of potential respiratory sensitizers (section and on the basis of the respective ECHA template for Weight-of-Evidence (WoE) assessments. The complete assessment is attached to this endpoint summary.

As first conclusion it is noted that for none of the investigated Lines of Evidence (LoEs), there is high quality information that at the same time fullfills relevant causality aspects (in terms of consistency, specificity, biologically plausibility and temporality) and, moreover, has a proper level of certainty. As consequence, none of these LoEs provides a high level of confidence for, or against, the tested hypothesis.

According to CLP criteria, human data have the highest predefined weigh in a WoE assessment and have therefore to be evaluated with special attention. The 13 potential clinical cases from the dental and cosmetic sector related to HEMA, respectively, do not fullfill CLP requirements “of reliable and good quality evidence”. For example, none of the known SIC tests (specific inhalation challenge) have been performed with the singe substance so that none of these tests can be seen as conform to guidelines. It is noted that none of these cases is directly related to respiratory exposure to HPMA, and this is the only relevant differentiation that can be made between HEMA and HPMA within the total WoE assessment. One eventual reason for that observation is that HPMA appears to be used less frequent in the relevant sectors, especially in the dental sector. Other branches of the human LoE (i.e. national health surveillance databases – no cases - & human exposure – implausible) do provide weak evidence against the tested hypothesis. Main sources of uncertainties in this LoE are A) the significant co-exposure in the occupational setting as well in performed SIC tests, including co-exposure to the acknowledged asthmagen Glutaraldehyde in the dental sector; B) the confounding clinical history of all but one patients and C) the inconsistent immunological pattern. In summary, no relevant evidence for the hypothesis is derived from human data that these substances can cause the development of respiratory sensitisation. This is especially true for HPMA in absence of relevant clinical cases.

An earlier assessment, that are also based on human data and that thus have also a relatively high predefined weigh in a WoE assessment, came to the same conclusion, on a however weaker data base.

Other lines of evidences provide either weak evidence for[1] or against[2] the hypothesis that both substances can cause the development of respiratory sensitisation. In addition, all those LoEs are generally afflicted with significant uncertainties so that they are considered as of minor relevance for the WoE assessment.

The conclusion drawn from these analyses, with focus on human data, indicate that there is insufficient evidence to implicate HEMA or HPMA as a cause of respiratory sensitisation or OA. Therefore, there is no basis for classification as respiratory sensitisers under EU CLP, which is widely consistent UN GHS.


[1] Such LoEs were structural alerts, QSAR, ADME, other relevant hazards, in vitro/ ex vivo and in vivo data

[2] Such LoEs were phys-chem properties and analogous substances

Justification for classification or non-classification

Skin Sensitisation

The main isomer in HPMA, propenoic acid, 2-methyl-, 2-hydroxypropylester (CAS 923-26-2, 70-80%) is classified as Skin Sensitiser Cat 1 according to CLP Annex VI (Index No. 607-125-00-5). As consequence, HPMA, as product of two isomers, has  to be classified as well as Skin Sens. Cat 1 in the EU for the time being. The corresponding hazard statement is H317: May cause an allergic skin reaction.

Based on human data HPMA should be considered to have skin sensitization potential. The weighed sum of all available data justifies the subcategorization with Skin Sens. Cat 1B as a weak-to-moderate sensitizer. Such classification will be proposed in the upcoming classification process in the EU.

Outside the EU, HPMA should be classified as Skin Sens. Cat 1B according to UN-GHS.

Respiratory Sensitisation

The conclusion drawn from comprehensive weight-of-evidence analyses, with focus on human data, indicate that there is insufficient evidence to implicate HEMA as a cause of respiratory sensitisation or occupational asthma. Therefore, there is no basis for classification as respiratory sensitiser under EU CLP or UN GHS.