Registration Dossier

Toxicological information


Currently viewing:

Administrative data

Description of key information

Carcinogenicity - NOEL, 0.5 ml. Chronic dermal application of blended gasoline did not significantly change the incidence of liver hemangiomas, lung adenomas, or of malignant lymphomas in treated animals compared to negative and historical controls. 
Carcinogenicity - NOEL, 292 ppm (~1400 mg/m3). Male rat kidney tumors and female mouse liver tumors were observed following chronic inhalation exposure (OECD TG 453). If these effects are discounted as not being relevant to humans, NOAEL is 2056 (~10,000 mg/m3).

Key value for chemical safety assessment

Additional information

The animal data indicate that gasoline exposure by inhalation at high levels can produce kidney tumors in male rats and liver tumors in female mice. There is also evidence that some of the naphtha streams can produce skin tumors following repeated dermal application. The mechanistic data suggest that both the male rat kidney tumors and the female mouse liver tumors were the consequence of promotional processes. The male rat kidney tumors were the consequence of a process that does not occur in humans, and, therefore, are not relevant to human risk assessment. The mouse liver tumors may have been the consequence of a hormonal imbalance although there is no direct evidence that that was the case. Nevertheless, the absence of such tumors in female rats or of male rats or mice brings into question the direct relevance of these tumors to humans. The overall no adverse effect level was 292 ppm (or approximately 1400 mg/m3). However, if the kidney and liver tumors are discounted as not being relevant to humans, the overall NOAEL is 2056 ppm or approximately 10,000 mg/m3. 

The mechanism of skin tumor induction by "low boiling," petroleum-derived materials i.e., gasoline and distillate fuel-range materials has been under investigation for some time. It has been suggested that these skin tumors could be the result of a promotional process secondary to repeated skin injury (e.g., McKee et al., 1989). In support of that hypothesis, Nessel et al., (1998; 1999) have shown that when irritation is ameliorated by dilution in a non-irritating carrier, tumors do not develop. Further, as discussed in section 7.6 Genetic Toxicity, the genetic toxicity data for gasoline and naphtha blending stocks suggest that mutagenicity does not play a role in tumor development. Thus it seems most likely that the low levels of dermal tumors associated with repeated application of some of the naphtha blending stocks are the consequence of repeated irritation and/or skin injury acting via a promotional process on pre-existing, spontaneously initiated cells (Przygoda et al., 1994). Thus, skin tumors could be avoided by the use of good industrial hygiene practice to avoid excessive skin contact.

Justification for classification or non-classification

The chronic toxicity studies were conducted by standard procedures and under good laboratory practice guidelines. The majority of the mechanistic data have been published in the open literature. No further testing for carcinogenic potential is necessary. With respect to regulatory purposes, one important question relates to the relevance of data on “wholly vaporized gasoline” to humans who are exposed predominantly to the more volatile constituents. This is particularly important in assessing the significance of the female mouse liver tumors as they were apparently the consequence of increased metabolic activity in the liver. The substances that have the largest effects on metabolism are the higher boiling aromatic constituents that constitute a very small fraction of gasoline vapor but a much more substantial fraction of wholly vaporized gasoline. An analogous situation pertains to the kidney tumors in male rats. The constituents that are the most potent inducers of tumors of this type are high molecular weight isoparaffins which are present in wholly vaporized gasoline but not gasoline vapor. Considering the questions about the relevance of the various tumor types produced, as well as questions relating to the exposure conditions, it seems more appropriate to rely on data from gasoline vapor studies than the data from wholly vaporized gasoline for human health risk assessment. As the volatile fraction of gasoline has not been tested in chronic studies, the results of repeated dose and developmental and reproductive toxicity studies of gasoline "light ends" should be used as the basis for risk assessment. The data do not support classification of gasoline per se for carcinogenic potential, although there is a regulatory requirement to classify as carcinogenic gasoline and naphtha streams containing > 0.1% benzene.