One TEF concept does not fit all: The case for human risk assessment of polychlorinated biphenyls

Human risk assessment of dioxins and dioxin-like compounds relies heavily on toxic equivalency factors (TEFs) that are mainly based on in vivo rodent studies. However, especially for the PCBs there are many uncertainties with respect to the actual dioxin-like activities and subsequent health effects in humans. For example, the relative effect potencies (REPs) for PCB126 are consistently up to two orders of magnitude lower in human cell models than in rodents and rodent cell cultures. For other dioxin-like (DL) PCBs, REPs can often not be obtained in human models due to a lack of AHR-mediated responses. In addition, DL-PCB-related effects such as thyroid disruption are largely attributed to mechanisms that are not (directly) AHR-mediated. Consequently, the AHR-mediated risk in humans for DL-PCBs is likely overestimated in the current TEF concept. The increasing availability of in vitro models using human cells will provide great opportunities to determine human-specific REP/TEFs based on toxicologically relevant endpoints. A better understanding of human-specific responses should lead to more reliable potency estimates of human effects and ultimately improved human risk assessment for DL-PCBs.