Assessment of possible carcinogenicity of oxyfluorfen to humans using mode of action analysis of rodent liver effects.

Oxyfluorfen is a herbicide that is not genotoxic and produces liver toxicity in rodents, following repeated administration at high dose levels. Lifetime rodent feeding studies reported in 1977 with low-purity oxyfluorfen (85%) showed no increase in any tumor type in rats (800 ppm, high dose) and only a marginally increased incidence of hepatocellular tumors in male CD-1 mice at the highest dose (200 ppm). To evaluate the potential carcinogenicity of the currently registered oxyfluorfen (> 98% purity), we conducted a series of short-term liver mode of action (MOA) toxicology studies in male CD-1 mice administered dietary doses of 0, 40, 200, 800, and 1600 ppm for durations of 3, 7, 10, or 28 days. MOA endpoints examined included liver weight, histopathology, cell proliferation, nuclear receptor-mediated gene expression, and other peroxisome proliferator-specific endpoints and their reversibility. Minimal liver effects were observed in mice administered doses at or below 200 ppm for up to 28 days. Increased liver weight, single-cell necrosis, cell proliferation, and peroxisomal acyl-CoA oxidase (ACO) were observed at 800 ppm after 28 days, but there was no increase in peroxisomes. Expression of Cyp2b10 and Cyp4a10 transcripts, markers of constitutive androstane receptor and peroxisome proliferator activated receptor α nuclear receptor activation, respectively, were increased at 800 and 1600 ppm after 3 or 10 days. Collectively, these data along with the negative genotoxicity demonstrate that oxyfluorfen (> 98% purity) has the potential to induce mouse liver tumors through a nongenotoxic, mitogenic MOA with a clear threshold and is not predicted to be carcinogenic in humans at relevant exposure levels.