Murdoch D J, Krewski D
Health Protection Branch, Health and Welfare Canada, Ottawa, Ontario
Risk Anal. 1988 Dec;8(4):521-30. doi: 10.1111/j.1539-6924.1988.tb01192.x.
Previous applications of carcinogenic risk assessment using mathematical models of carcinogenesis have focused largely on the case where the level of exposure remains constant over time. In many situations, however, the dose of the carcinogen varies with time. In this paper, we discuss both the classical Armitage-Doll multistage model and the Moolgavkar-Venzon-Knudson two-stage birth-death-mutation model with time-dependent dosing regimens. Bounds on the degree of underestimation of risk that can occur through the use of a simple time-weighted average dose are derived by means of comparison with an equivalent constant dose corresponding to the actual risk under the time-dependent dosing regimen.
以往使用癌症发生数学模型进行致癌风险评估，主要集中在暴露水平随时间保持恒定的情况。然而，在许多情况下，致癌物的剂量会随时间变化。在本文中，我们讨论了经典的阿米蒂奇 - 多尔多阶段模型以及莫尔加夫卡尔 - 文佐恩 - 克努森两阶段出生 - 死亡 - 突变模型，这些模型考虑了随时间变化的给药方案。通过与对应于随时间变化给药方案下实际风险的等效恒定剂量进行比较，得出了因使用简单时间加权平均剂量可能导致的风险低估程度的界限。
