Dimethyl sulfoxide

Carcinogenicity

Description of key information

There are no standard carcinogenicity studies conducted with DMSO, but considering the lack of genotoxic potential, the absence of target organs (excepted the eye lens in some sensitive species) in many repeated dose toxicity studies performed with diverses animal species, routes of administration and exposure durations up to 2 years and the results of some initiation/promotion studies, it is not scientifically justified to perform a carcinogenicity study.
Initiation/promotion studies by oral administration DMSO had no effect on the latency or frequency of tumours induced by DMBA. Dermal administration gave contrasted results, enhancement, inhibition or no effects on the tumour frequency. It is difficult to find a common explanation for the range of effects of DMSO on experimental tumorigenesis. The likely improved skin penetration of the carcinogen when dissolved in DMSO could account for the enhancement observed in single-stage trials by dermal route. However, in two-stage trials, tumorigenesis was not enhanced with DMSO as the vehicle for the initiator and was actually inhibited when DMSO was used as the vehicle for the promoter.

Justification for classification or non-classification

No classification for carcinogenicity has to be applied for DMSO according to EU Directive 67/584/EEC and EU regulation (EC) No 1272/2008 (CLP).

Additional information

There are no standard carcinogenicity studies conducted with DMSO.Subcutaneous injections of 0.05 ml DMSO, once weekly for lifetime, to 30 ICR/Ha swiss mice did not induce tumour at the injection site (Van Duuren et al., 1971).Additional information is available from initiation/promotion studies.

Initiation/Promotion Studies

The effects of DMSO on the tumorigenic activity of 7,12-dimethylbenz[a]anthracene (DMBA) was investigated in rats. Three groups of 50 male Sprague-Dawley rats were given 20 mg DMBA by gavage. In the two first groups, DMBA treatment was followed by 50 ppm DMSO in their drinking water 3 days before or 3 days after DMBA administration, respectively. The third group received no DMSO and served as untreated controls. The animals were treated or observed for up to 18 months, DMSO had no effect on the latency or frequency of tumors induced by DMBA. Rats receiving DMSO weighed more and had fewer tumors than did the controls at the end of the 18-month study period. This was suggestive that DMSO decreased the total number of tumors, although the difference between treated and control rats did not reach statistical significance (Fletcher & Dennis, 1967).

Initiation-promotion studies based on the skin tumor model were carried out with DMSO in mice. The animals were treated twice weekly with 125 mg benzo[a]pyrene dissolved in 40 µl DMSO or acetone and applied dermally. The number of skin tumors was doubled when DMSO was used as the solvent. In another group, tumor initiation was carried out with DMBA dissolved in DMSO. Without promotion treatment the tumor frequency was unchanged relative to that in the controls. Promotion treatment with a solution of phorbol myristate in DMSO reduced the frequency of skin tumors (Jacoby & Weiss, 1986).

In 20 ICR/Ha Swiss mice, dermal application of 0.1 ml DMSO, 3 times weekly over a period of 400 days, after a primary treatment with DMBA (applied once only, 20µg in 0.1 ml acetone), induced no skin tumors (Loewengart & van Duuren, 1977).

The effect of DMSO as the solvent for 20-methylcholanthrene (MC) on induced carcinogenesis of the skin was investigated in male CBA x C57BL hybrid mice. MC was used as a 0.25 and 0.5% solution in benzene or DMSO and applied in a dose of 0.2 ml to the previously shaved skin of the interscapular region once a week until the end of the experiment. The following parameters were determined: the time of appearance of the first papilloma and the first carcinoma, the mean latent period of development of the two types of tumor. After application of a 0.25% solution of MC in DMSO only a tendency toward the acceleration of carcinogenesis was observed, but if the MC concentration was doubled, the acceleration of carcinogenesis became significant (Finogenova, 1974).

Two groups of 40 Sprague-Dawley rats were treated dermally, 3 times a week, for at least 26 weeks, with 0.02 ml of a 1% solution of 9,10-dimethylbenzanthracene (DMBA) dissolved either in DMSO or acetone. A third group (40 animals) received only DMSO (0.02 ml). The animals were observed weekly and the number of tumors counted. Specimens were taken from all tumours for histological examination at the end of the experiment. In the rats treated with DMBA dissolved in acetone, the time of appearance of the first tumor was shorter (26 weeks) compared to DMBA in DMSO (34 weeks) and the number of papillomas was greated (141 versus 32). No skin tumor was observed in the rats treated with DMSO alone (Stenbäch, 1970).

Hairless mice were given 5 topical applications of 470 nmol 20-methylcholanthrene (MC) at one-week intervals. In one experimental group the MC was dissolved in reagent grade acetone alone, in another group it was dissolved in a mixture consisting of 50% acetone and 50% DMSO. A control group received the mixed solvent alone. The animals were observed for development of papillomas and malignant tumors for 75 weeks. The group treated with MCA in acetone/DMSO had a higher mortality than the two other groups. The admixture of 50% DMSO to the solvent had a significant inhibitory effect on tumor and cancer rates and on cancer yield, whereas no effect could be observed on the tumor yield. Hence, 50% DMSO in the solvent has a moderate, but significant, inhibitory effect on MC-induced skin carcinogenesis (Iversen et al., 1981).