Distillates (petroleum), hydrotreated light paraffinic

Carcinogenicity

Description of key information

Insufficiently refined other lubricant base oils with IP 346 ≥ 3% produce tumours when tested in mouse dermal application tests.  Accordingly they should be considered carcinogenic.  Sufficiently refined other lubricant base oils with IP 346 < 3% are not carcinogenic in dermal application studies.  Similarly, no tumours were noted in chronic repeat dose inhalation toxicity studies of lubricant base oils conducted in a number of species of laboratory animals (Section 5.6.3 of CSR).

Key value for chemical safety assessment

Carcinogenicity: via oral route

Endpoint conclusion

Endpoint conclusion:

no study available

Carcinogenicity: via inhalation route

Endpoint conclusion

Endpoint conclusion:

no study available

Carcinogenicity: via dermal route

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Study duration:

chronic

Species:

mouse

Justification for classification or non-classification

Some mildly or poorly refined other lubricant base oils are Carcinogenic Cat 1B (H350: May cause cancer) according to the EU CLP Regulation EC No. 1272/2008. Lubricant base oils are non-carcinogenic if DMSO extract as measured by IP346 is less than 3% and are considered sufficiently refined. Results are based on 37 negative cancer (dermal) studies available where the negative results have been compared to IP346 (DMSO extract <3%, sufficiently refined) and 4 positive cancer (dermal) studies available where the positive results have been compared to IP346 (DMSO extract >3%, insufficiently refined).

Additional information

Note that the CAS numbers are normally assigned on the basis of the last refining step employed and may not reflect the entire refining history. Note also that differences in refining severity may lead to differences in polycyclic aromatic constituent content which may in turn lead to differences in carcinogenic activity. Accordingly, other lubricant base oils with the same CAS numbers may have differing carcinogenic properties.  

No oral carcinogenicity studies have been conducted on other lubricant base oils.

Numerous lifetime dermal carcinogenicity studies have been carried out on lubricant base oils. Data from these studies have been reported and published in a number of comprehensive reviews (Bingham et al., 1980; Blackburn et al., 1984, 1996; CONCAWE, 1994; 1997; IARC, 1984; Roy et al., 1988; Shoda et al., 1997; Mackerer et al. 2003). Overall, these studies have shown that lubricant base oils that have been refined to a sufficient degree of severity, i.e., by solvent-extraction and / or severe hydrotreatment do not normally induce skin cancer in mice. However, lubricant base oils that have not been sufficiently refined may be carcinogenic to the skin. The IP 346 test is used to determine whether the lubricant base oils have been sufficiently refined to avoid dermal carcinogenic hazard. The method measures the quantity of DMSO extract which has been proved to correlate to the carcinogenic properties of the other lubricant base oils. Other lubricant base oils are classified for dermal carcinogenic hazard (Category 1B, H350 ) unless they have been shown to contain less than 3 wt% DMSO extractable material according to the IP346 method according to the IP346 method (also described in the current legislation 1272/2008/EU in Note L).

Doak et al. (1983) reported on an 18-month dermal carcinogenesis carcinogenicity study of twelve mineral oils originating from naphthenic and paraffinic feed stocks and refined by a number of processes. All of these oils are considered “sufficiently” refined. The oils were evaluated for their potential to induce cutaneous neoplasia in female CF1 mice. The oils were applied to the shorn dorsal skin for up to 78 weeks using the following treatment regimes:

a(i) -     undiluted oil applied twice weekly or 78 weeks

a(ii) -    undiluted oil applied twice weekly for 22 weeks with no further treatment

b(i) -    undiluted oil applied once weekly for 78 weeks

b(ii) -    undiluted oil applied once weekly for 22 weeks and then once fortnightly

c     -  diluted oil (1:1 v/v with medicinal grade liquid paraffin) applied twice weekly for 22 weeks and then once weekly

d     -  diluted oil (1:1 v/v with medicinal grade liquid paraffin) applied once weekly for 78 weeks 

This study lends further support to the concept that the degree of refining influences the carcinogenic potential of the oils. Whereas mild acid / earth refining processes are inadequate to substantially reduce the carcinogenic potential of other lubricant base oils, hydrotreatment and / or solvent extraction methods can yield oils with no carcinogenic potential. Based on the results of the skin painting studies, all of the solvent extracted oils are non-carcinogenic.  

Supporting data from dermal carcinogenicity studies conducted in mice (API, 1983; Horton and Denman, 1955; Kane et al., 1984; Halder et al., 1984; McKee et al., 1989) demonstrate that sufficiently refined other lubricant base oils are not carcinogenic.

In order to understand the effects of various types of refining processes (i.e., hydrotreatment, solvent extraction, combined solvent extraction and hydrotreatment) on carcinogenic potential, 94 lubricant base oils and related materials were evaluated in the mouse epidermal cancer bioassay (Chasey and McKee, 1993). In these studies, male C3H mice, ca. 6-10 weeks of age, were obtained from Jackson Laboratories,,(C3H / HeJ) or Charles River,,(C3H / HeNCrlBR). These mice were randomly distributed into test groups of 40 or 50 animals. In early studies, mice were housed five per cage in suspended wire-mesh cages. In later studies, they were housed singly, in the same type of cages. The hair in the interscapular area was clipped once weekly to facilitate test material application. The test materials were applied by automatic pipette in either 37.5 microlitres aliquots twice a week or 25 microlitres aliquots three times a week. In early studies, the treatment continued until the animals died spontaneously or were sacrificed in a moribund state. In later studies, surviving mice were sacrificed after either 24 months of treatment or at the time at which grossly diagnosed squamous cell carcinomas were recorded. Animals were examined twice weekly for the appearance of dermal tumours. Each tumour in the treatment area was examined carefully and classified grossly. All grossly diagnosed tumours were examined microscopically after study termination.

The two largest subsets of the database are the solvent extracted oils (i.e., 21 oils) and the solvent extracted and hydrotreated oils (i.e., 37 oils). The data for these subsets were highlighted due to the importance of solvent extraction in many lubricant base oil manufacturing schemes. All commercial products refined in this way produced negative bioassays. The solvent extracted oils that produced tumours were refined by experimental conditions that are not representative of those used in commercial refineries.

Lubricant base oils refined solely by hydrogen treatment produced a variety of responses in the bioassays. Solvent extraction removes polycyclic aromatic constituents. Hydrogen treatment at relatively low temperatures and hydrogen pressures is often used as a finishing step to reduce sulphur and nitrogen content, but may have minimal effects on polycyclic aromatic constituent content. At higher temperatures and hydrogen pressures, hydrotreatment may modify the molecular structure of polycyclic aromatic constituents via reactions such as hydrogenation or ring-opening. As a result, the dermal carcinogenicity of an oil which has been refined by hydrogen treatment only is highly dependent on the hydrotreating conditions. The International Agency for Research on Cancer (IARC) has concluded that mildly hydrotreated oils are carcinogenic. All of the samples listed in the database provided by Chasey and McKee (1993) processed by "severe" hydrotreatment, were non-carcinogenic.

Of the 94 samples tested for carcinogenic activity, 57 produced no tumour-bearing animals and the remaining 37 produced one or more. Among the groups containing tumour-bearing animals, seven had one, six had two, two had four and the remaining 22 had five or more. In order to differentiate statistically one of the treatment group responses from that of an equally sized negative control group (containing no tumour-bearing animals), at least five tumour-bearing animals are required. Thus, responses are statistically significant in 22 of the 37 groups containing tumour-bearing animals. When the results are examined by processing history classification, all raw vacuum distillates (UATO) (6/6), raw extracts (UDAE) (7/7) and hydrotreated extracts (TDAE) (6/6) produced tumours. Among the remaining classifications, most extracted distillates (19/21), extracted and hydrotreated distillates (33/37) and refined vacuum residua (2/2) produced fewer than two tumour-bearing animals per group. The hydrotreated distillates produced a mixed response.

Justification for selection of carcinogenicity via dermal route endpoint:

One of five carcinogenicity studies

Carcinogenicity: via dermal route (target organ): other: skin