Nonlinear cancer response at ultralow dose: a 40800-animal ED(001) tumor and biomarker study.

摘要: Assessment of human cancer risk from animal carcinogen studies is severely limited by inadequate experimental data at environmentally relevant exposures and procedures requiring modeled extrapolations many orders magnitude below observable data. We used rainbow trout, an model well-suited to ultralow-dose carcinogenesis research, explore dose-response down a targeted 10 excess liver tumors per 10000 animals (ED(001)). A total 40800 trout were fed 0-225 ppm dibenzo[a,l]pyrene (DBP) for 4 weeks, sampled biomarker analyses, returned control diet 9 months prior gross histologic examination. Suspect confirmed pathology, resulting incidences compared the default EPA LED(10) linear extrapolation method. The study provided observed incidence two above-background lowest dose (that is, unmodeled ED(0002) measurement). Among nine statistical models explored, three determined fit well-linear probit, quadratic logit, Ryzin-Rai. None these fitted compatible with assumption, all fell increasingly decreasing DBP dose. Low-dose tumor response was also not predictable hepatic DBP-DNA adduct biomarkers, which accumulated as power function (adducts = 100 x DBP(1.31)). Two-order predicted doses producing one million individuals (ED(10)(-6)) that 500-1500-fold higher than five-order extrapolation. These results are considered specific model, carcinogen, protocol used. They provide first estimation in any degree conservatism may exist assumption genotoxic carcinogen.