Ultra-Low Dose Carcinogen Testing with the Trout Model

描述(由申请人提供)：外推动物的剂量反应致癌数据以确定人类暴露的“安全”水平(百万分之一)是具有挑战性的，因为它需要对至少4个数量级的剂量反应进行建模。在许多情况下，保守方法假定为线性。在啮齿动物模型中，最大的肿瘤研究使用了24,192只小鼠，每100只小鼠中就有1只发现癌症。鲑鱼肿瘤模型已被证明在识别致癌物及其作用机制方面具有很高的价值。鲑鱼对黄曲霉毒素B1(AFB1)的肝癌致癌性非常敏感，黄曲霉毒素B1是IARC唯一通过食物供应接触的人类致癌物质。肝细胞癌每年造成全球60多万人死亡，占某些地区全部死亡人数的10-15%。肝细胞癌是美国增长最快的实体肿瘤。对于AFB1，鲑鱼和人类的靶器官、代谢、DNA加合和基因靶点相似，在这个例子中，鲑鱼模型优于小鼠。鲑鱼模型可以利用大量动物来评估广泛剂量范围内的癌症。这种方法是可能的，因为这种模式有许多优点，包括低自发肿瘤发生率和低每日费用。我们最近完成了所有动物模型中最大的癌症研究，利用42,000条鲑鱼来评估超低剂量的二苯并[a，L]芘(DBP)的致癌性。这个目标比小鼠ED01研究低一个数量级，即1000例癌症中就有一例。我们的数据确定了DBP的剂量，该剂量导致每5,000条鲑鱼中有1条癌症，并且剂量-反应是非线性的。我们现在建议利用这个超低剂量模型来检验这一假设，即在超低剂量下癌症发病率的非线性也适用于AFB1。肿瘤发病率数据将与使用定制微阵列的分子剂量学、细胞增殖、细胞凋亡和基因表达的测量相结合，以检验第二个假设，即与肿瘤发病率相反，这些致癌生物标志物在整个肿瘤剂量-反应范围内呈线性。

DESCRIPTION (provided by applicant): Extrapolation of dose-response carcinogen data from animals to establish "safe" levels for human exposures (1 in a million) is challenging as it requires modeling of dose-response over at least 4 orders of magnitude. In many cases, the conservative approach is assumed linearity. The largest tumor study in a rodent model used 24,192 mice to detect 1 cancer in 100. The trout tumor model has proven valuable in identification of carcinogens and their mechanism of action. Trout are very sensitive to the hepatocarcinogenicity of aflatoxin B1 (AFB1), the only IARC human carcinogen where exposure is through the food supply. Hepatocellular carcinoma (HCC) is responsible for over 600,000 deaths a year worldwide and accounts for 10-15% of all deaths in certain regions. HCC is the most rapidly increasing solid tumor in the U.S. For AFB1, the target organ, metabolism, DNA adduction and gene targets are similar in trout and human and, in this example, the trout model is superior to mouse. The trout model can utilize large numbers of animals to evaluate cancer across a wide range of doses. This approach is possible due to a number of advantages of this model including low spontaneous tumor incidence and low per diem costs. We have recently completed the largest cancer study in any animal model, utilizing 42,000 trout to assess carcinogenicity of dibenzo[a,l]pyrene (DBP) at ultra-low doses. The target was an order of magnitude lower than the mouse ED01 study, to one cancer in 1000. Our data established a dose of DBP that resulted in 1 cancer in 5,000 trout and the dose-response was non-linear. We now propose to utilize this ultra-low dose model to test the hypothesis that the non-linearity of cancer incidence at ultra-low dose also applies to AFB1. Tumor incidence data will be coupled with measurements of molecular dosimetry, cell proliferation, apoptosis and gene expression utilizing a custom microarray in order to test the second hypothesis, that in contrast to tumor incidence, these biomarkers of carcinogenicity exhibit linearity across the entire tumor dose-response range.