Vaccari L, Ranzi A, Canova C, Ghermandi G, Giannini S, Pitter G, Russo F, Stefanelli J, Teggi S, Vantini A, Jeddi M Z, Fletcher T, Colacci A
Center for Environment, Prevention and Health, Emilia-Romagna Regional Agency for Prevention, Environment and Energy (Arpae), 40139, Bologna, Italy.
Unit of Biostatistics, Epidemiology and Public Health-University of Padua, Padua, Italy.
Heliyon. 2024 Jul 31;10(15):e35288. doi: 10.1016/j.heliyon.2024.e35288. eCollection 2024 Aug 15.
Long-term contamination of tap water and groundwater by perfluoroalkyl and polyfluoroalkyl substances (PFASs) has been documented in the Veneto region of northern Italy. This study aimed to assess the exposure of individuals residing in the contaminated area and to test several toxicokinetic (TK) models of varying complexities to identify an efficient method for predicting perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) concentrations in human serum using observed data.The ultimate goal is to provide public health officials with guidance on selecting the appropriate TK model for specific contexts, a reliable and rapid tool to support human bio-monitoring (HBM) studies.
Two simpler empirical TK models and a more complex multi-compartment physiologically based toxicokinetic (PBTK) model were compared with individual and aggregate data from an HBM study. In addition, the PBPK model was modified by adjusting input parameters and introducing new terms into the equations within the original model code. These modifications aimed to optimize the results compared to the original model, with some versions incorporating adjustments to account for the influence of menstruation in women. All models were evaluated to understand their strengths and weaknesses, providing guidance on the appropriate model to use according to specific scenarios.
The results obtained from the tested models were quite similar, with significant improvements observed only in the modified models. Simpler models also provided satisfactory results in scenarios involving low PFOS serum concentrations and recent exposure cessation. In many cases, predictions demonstrated high accuracy, particularly at the aggregate level and for women.
These findings suggest that environmental protection agencies and health authorities may benefit from employing the tested models at the aggregate level as an initial step in HBM studies, rather than conducting more invasive and expensive screening campaigns.
意大利北部威尼托地区已记录到全氟烷基和多氟烷基物质（PFASs）对自来水和地下水的长期污染。本研究旨在评估居住在污染地区的个体的暴露情况，并测试几种不同复杂程度的毒代动力学（TK）模型，以确定一种利用观测数据预测人血清中全氟辛酸（PFOA）和全氟辛烷磺酸（PFOS）浓度的有效方法。最终目标是为公共卫生官员提供在特定情况下选择合适的TK模型的指导，这是一种支持人体生物监测（HBM）研究的可靠且快速的工具。
将两个较简单的经验TK模型和一个更复杂的基于生理的多室毒代动力学（PBTK）模型与一项HBM研究中的个体和汇总数据进行比较。此外，通过调整输入参数并在原始模型代码中的方程中引入新项，对PBPK模型进行了修改。这些修改旨在与原始模型相比优化结果，一些版本纳入了调整以考虑女性月经的影响。对所有模型进行评估以了解其优缺点，为根据特定场景使用合适的模型提供指导。
测试模型得到的结果非常相似，仅在修改后的模型中观察到显著改进。较简单的模型在涉及低PFOS血清浓度和近期接触停止的情况下也提供了令人满意的结果。在许多情况下，预测显示出高准确性，特别是在汇总水平和对女性而言。
这些发现表明，环境保护机构和卫生当局可能会受益于在汇总水平上采用测试模型作为HBM研究的第一步，而不是开展更具侵入性和成本更高的筛查活动。
