Berninger Jason P, LaLone Carlie A, Villeneuve Daniel L, Ankley Gerald T
National Research Council, US Environmental Protection Agency, Duluth, Minnesota, USA.
Water Resources Center, College of Food, Agricultural and Natural Resource Sciences, University of Minnesota, St. Paul, Minnesota, USA.
Environ Toxicol Chem. 2016 Apr;35(4):1007-20. doi: 10.1002/etc.2965. Epub 2015 Jun 22.
The potential for pharmaceuticals in the environment to cause adverse ecological effects is of increasing concern. Given the thousands of active pharmaceutical ingredients (APIs) that can enter the aquatic environment through human and/or animal (e.g., livestock) waste, a current challenge in aquatic toxicology is identifying those that pose the greatest risk. Because empirical toxicity information for aquatic species is generally lacking for pharmaceuticals, an important data source for prioritization is that generated during the mammalian drug development process. Applying concepts of species read-across, mammalian pharmacokinetic data were used to systematically prioritize APIs by estimating their potential to cause adverse biological consequences to aquatic organisms, using fish as an example. Mammalian absorption, distribution, metabolism, and excretion (ADME) data (e.g., peak plasma concentration, apparent volume of distribution, clearance rate, and half-life) were collected and curated, creating the Mammalian Pharmacokinetic Prioritization For Aquatic Species Targeting (MaPPFAST) database representing 1070 APIs. From these data, a probabilistic model and scoring system were developed and evaluated. Individual APIs and therapeutic classes were ranked based on clearly defined read-across assumptions for translating mammalian-derived ADME parameters to estimate potential hazard in fish (i.e., greatest predicted hazard associated with lowest mammalian peak plasma concentrations, total clearance and highest volume of distribution, half-life). It is anticipated that the MaPPFAST database and the associated API prioritization approach will help guide research and/or inform ecological risk assessment.
环境中的药物引发不良生态效应的可能性日益受到关注。鉴于数千种活性药物成分（APIs）可通过人类和/或动物（如牲畜）粪便进入水生环境，水生毒理学当前面临的一项挑战是识别出那些风险最大的成分。由于通常缺乏针对水生生物的药物经验毒性信息，用于确定优先次序的一个重要数据源是在哺乳动物药物研发过程中生成的数据。应用物种类推的概念，以鱼类为例，利用哺乳动物的药代动力学数据，通过估计药物对水生生物造成不良生物学后果的可能性，系统地对活性药物成分进行优先排序。收集并整理了哺乳动物的吸收、分布、代谢和排泄（ADME）数据（如血浆峰值浓度、表观分布容积、清除率和半衰期），创建了代表1070种活性药物成分的水生生物靶向哺乳动物药代动力学优先排序（MaPPFAST）数据库。基于这些数据，开发并评估了一个概率模型和评分系统。根据明确界定的类推假设，将源自哺乳动物的ADME参数转化为对鱼类潜在危害的估计值（即与最低哺乳动物血浆峰值浓度、总清除率以及最高分布容积、半衰期相关的最大预测危害），对各个活性药物成分和治疗类别进行了排名。预计MaPPFAST数据库及相关的活性药物成分优先排序方法将有助于指导研究和/或为生态风险评估提供信息。
