Distilling the Legal Implications of IARC's Hazard Classification for Automotive Gasoline

On 21 March 2025, the International Agency for Research on Cancer (IARC) announced that it would classify automotive gasoline as “carcinogenic to humans,” (Group 1) based on its interpretation that there is sufficient evidence for cancer in humans, and the combination of sufficient evidence for cancer in experimental animals and strong mechanistic evidence in exposed humans. Since 1988, gasoline, generally, has been classified by IARC as “possibly carcinogenic to humans.” IARC’s more definitive hazard classification of automotive gasoline will undoubtedly have a legal ripple effect, including resulting in increases in personal injury litigation. Affected industries include gasoline producers and distributors, whose customers that are chronically exposed may assert claims for cancer diagnosis based upon the new classification. Here is what you need to know.  

What Is Automotive Gasoline?

Automotive gasoline is largely comprised of a complex mixture of hydrocarbons derived from the fractional distillation of petroleum products. Often further fortified with additives to improve functionality, automotive gasoline and gasoline more generally are mixtures that are classified as “Substances of Unknown or Variable composition, Complex reaction products, and Biological materials” (UVCBs). The chemical composition of gasoline, which is closely linked with its corresponding physicochemical and toxicological properties, is dependent on the geographical origins of the crude oil from which it is derived, the specific boiling point fraction of the light naphtha used in the gasoline formulation, and the specific performance additives used to refine the properties of the resulting product, which vary regionally and seasonally. Thus, not all gasoline is created equal. 

Hazard Identification of UVCBs

Recognition that gasoline is a complicated mixture of variable composition is particularly significant for understanding the impacts of IARC’s hazard classification. What is not made clear from IARC’s announcement of their findings is that the hazard assessment may not be definitive for all automotive gasoline in perpetuity. This is because a single constituent in the gasoline as evaluated could be responsible for the results observed in occupational cohort, case-control, and animal studies, rather than the bulk of the mixture. Identification of the specific carcinogenic constituent(s) responsible for the results observed may provide greater insights into potential hazard/risk variabilities in automotive gasoline as a function of composition and allow for potential reformulation/refinement to reduce or remove the most egregiously hazardous substances. In the long term, this has the potential to significantly alter the recognized hazard profile of gasoline more broadly. 

The relevance of composition to hazard profile applicability will lead many to ask what, specifically, did IARC determine to be carcinogenic to humans and how does that relate to the gasoline I am using today? IARC’s conclusions about the carcinogenicity of gasoline were based on occupational cohort and case-control studies of service station attendants and gasoline distribution workers from 1976–1996 in six European countries, results of chronic carcinogenicity bioassays in experimental animals using gasoline samples certified to meet 1976 or 1990 industry average gasoline properties, and mechanistic evidence indicating that gasoline is genotoxic and induces oxidative stress and chronic inflammation.¹ Although IARC’s correlations of cancer with gasoline exposures across these multiple diverse studies does seemingly lend itself to a clear hazard profile for this mixture, the significance of any classification is complicated due to gasoline’s status as a UVCB. 

Hazard identification and risk assessment of UVCBs are notoriously complex. Not only do UVCBs generally lack standardized and universally representative reference materials for analysis, but many common toxicological investigations performed on UVCBs are incapable of elucidating whether an individual constituent of a UVCB is the putative toxin responsible for any effects observed or whether the collective makeup of the mixture, wherein the individual constituents act synergistically, is responsible for the biological results observed. 

Analogous hazard assessments of hydrocarbon UVCBs, e.g., lubricating base oils, by international regulatory authorities set a precedent for avoiding a “broad-brush” carcinogenicity classification. Hazard assessments of lubricating base oils do generally account for the fact that petroleum-derived hydrocarbon mixtures are not compositionally or toxicologically identical and that the concentration of certain constituent elements of the mixture can be significant for purposes of hazard classification, let alone risk assessment. Specifically, in Note L of Annex VI of the European Union’s regulation on the classification, labeling, and packaging of substances and mixtures,² it is noted that: 

The classification as a carcinogen need not apply if it can be shown that the substance contains less than 3% DMSO extract as measured by IP 346 “Determination of polycyclic aromatics in unused lubricating base oils and asphaltene free petroleum fractions – Dimethyl sulphoxide extraction refractive index method”, Institute of Petroleum, London. This note applies only to certain complex oil-derived substances in Part 3.

In other words, relevant authorities recognize that, for certain petroleum derived UVCBs, the concentration of a low-level constituent, such as polycyclic aromatic hydrocarbons in the case of lubricating base oils, can be significant and determinative for hazard classification. While some may view this as being irrelevant to the case of automotive gasoline due to a lack of clarity on whether the biological effects observed in the cohort, case-control, and animal studies should be attributed to the mixture itself, or a specific constituent therein, a deeper-dive into the circumstances surrounding automotive gasoline does appear to be necessary for truly understanding the applicability of IARC’s hazard assignment.

The Evolution of Gasoline Formulations

As noted above, automotive gasoline is a UVCB whose composition, both in terms of hydrocarbon distribution and performance additives can vary extensively as a function of region. Further complicating any broad-brush assessments of the compositional, physicochemical, and toxicological properties of gasoline is the fact that the composition of gasoline has intentionally changed over time in response to market changes and fuel quality standards imposed by the US Environmental Protection Agency (EPA) and other comparable authorities worldwide.³ Over the past 30-plus years, fuel quality standards have been particularly impactful in reducing the levels of hazardous substances in gasoline formulations, including benzene and methyl tert-butyl ether (MTBE). Benzene is not only considered to be genotoxic in vivo and carcinogenic to humans by numerous global authorities, but its exposures have been associated with the same types of cancers, i.e., acute myeloid leukemia,⁴ urinary bladder cancer,⁵ renal cell carcinoma,⁶ and hepatocellular adenoma or carcinoma,⁷ that IARC attributed to gasoline in its recent assessment. In contrast, many of the primary constituents of gasoline vapors, such as isopentane,⁸ n-butane,⁹ methylpentane isomers,¹⁰ and toluene,¹¹ have tested negative in genotoxicity screening tests or carcinogenicity bioassays performed in experimental animals. Such factors may suggest that a minor constituent of the gasoline samples tested may have been responsible for results observed in the various studies referenced by IARC. 

Considering the historical presence of benzene as a constituent in gasoline, as well as in gasoline samples investigated in the noted carcinogenicity bioassays performed in experimental animals, further understanding the role that benzene played in the outcomes of the gasoline studies is extremely significant to understanding the broad applicability of IARC’s classification for gasoline produced today. Whereas the gasoline samples used in the noted carcinogenicity bioassays contained 1.5% to 2.1% benzene, mandates have required significantly lower levels of benzene in gasoline for decades, and EPA’s “Fuel Trends Reports for Gasoline” show that levels in gasoline continue to be reduced.¹² Regulation of benzene as a constituent component of gasoline dates back to the 1990s, after the IARC first classified it as carcinogenic to humans in 1988.¹³ The EPA placed a limit of 1% benzene in reformulated gasoline in 1995.¹⁴ The EPA’s Mobile Source Air Toxics program further reduced the limit to 0.62% starting in 2011.¹⁵  These regulatory levels, notably, are lower than the samples used by IARC, but nonetheless, the new classification of gasoline as carcinogenic to humans carries with it the potential for regulators, including the EPA, to further reduce the percentage of benzene permissible in gasoline. 

MTBE, which IARC has also recently classified as “possibly carcinogenic to humans” was also known to be present several of the studies animal studies that the IARC Working Group considered in its evaluation, despite the fact that its use in gasoline has declined significantly in the United States and has been banned by several states. 

Further confounding the ultimate conclusion to IARC’s assessment are potential co-exposures to diesel and gasoline engine exhaust and other hydrocarbon mixtures among occupational cohort and case-control study participants, as well as certain limitations of the chronic animal studies.¹⁶

What to Know Moving Forward

Ultimately, it is important to recognize that IARC’s conclusions concerning automotive gasoline speak to its hazard profile and are not an assessment of risk. Notwithstanding any potential limitations of the hazard classification, recent studies have shown that consumer and worker risks resulting from potential exposures to automotive gasoline vapors, and benzene specifically, associated with commercial gasoline station fueling events do appear to be below relevant risk management limits. Nevertheless, the public-facing announcement of IARC’s reclassification of automotive gasoline should serve as a reminder about the significance of ongoing hazard and exposure assessments for chemical substances that consumers and workers may encounter every day. In the current case, it is also critically important that necessary steps be taken to more definitively elucidate the specific toxins responsible for the outcomes of the referenced animal studies and for impacted parties to ensure that worker and consumer exposures to, and thus, any potential human health risks associated with, automotive gasoline vapors are minimized to the extent possible.

The new IARC classification carries with it the potential for new regulatory action, and enterprising plaintiff lawyers will probably pursue litigation over cancer diagnoses for those who work in industries with heavy exposures to automotive gasoline under the familiar “no safe level of exposure” theory. Whereas, in the past, personal injury cases arising out of gasoline exposures primarily focused on constituent components, most notably benzene, the broad scope of the new IARC classification for gasoline may lead that litigation to expand beyond instances where elevated contents of benzene are detected. Gasoline manufacturers should consult with counsel regarding these increased regulatory and litigation risks.