WRT2 - Round Table 2: Estimating Response at Low DoseSalon D Noon-1:30 pm |
| Chair(s): Jim Wilson |
Knowing the response at low exposure is crucial to being able to estimate risk, given the very small exposures to chemicals usually of interest to regulators and the public. For fifty years regulatory risk assessment has been dominated by two assumptions: that the risk to humans is negligible at one-one hundredth the "no observed adverse effect" dose rate in small rodents, or that response is directly proportional to exposure. These assumptions lead to identification of exposures that are considered to pose no or a negligible risk. Such policy-based assumptions are not satisfactory if estimates of low-dose response are needed. While a few instances of true low dose risk estimation exist (e.g., of inhaled lead, ingested aflatoxin), these come from epidemiologic observations on exposures near those of regulatory interest. In the absence of measured responses, we must develop an understanding of the processes involved in producing a response and let that knowlege inform us about the nature of the dose-effect relationship. This symposium will present reviews of current knowledge related to estimation of low-dose responses. |
|
WRT2.1 Low-dose response: Inferring Functional Form from First Principles. James D. Wilson, Daniel M. Byrd III; Resources for the Future (emeritus) and LSRO wilson.jimjudy@att.net Abstract: For decades risk assessors have argued over the functional form taken by exposure-response functions at low dose. Experiments cannot settle this question. However, the fundamentals of chemistry and physics place some limits on the functional form. For instance, the incidence of injuries due to cell-membrane disruption from acid must be zero at some exposure, because for each cell more sites than one must be protonated before the membrane loses integrity. A few kinds of toxic response are responsible for the majority of possible injuries, e.g., membrane disruption, mutation, effects on energy metabolism, and interference with inter- and intra-cellular signalling. Chemistry-based limits on the low-dose response for the common modes of toxicity will be identified and described. |
|
WRT2.2 Advances in Biologically-Motivated Dose Response Assessment: Physiologically Based Pharmacokinetic Models as a Tool for Gaining Insights into Response at Low Doses. Justin Teeguarden; Environ Institute for Health Risk Sciences jteeguarden@environcorp.com Abstract: Risk assessment for chronic exposures fundamentally depends on predicting or estimating risk (response) for exposures or doses significantly below those tested experimentally. Regulatory agencies and the regulated community share an interest in developing more informed, biologically motivated approaches to conducting dose response assessments in the low dose range (lower than in experimentally tested). The general belief is that representing the processes which, identified through experimental research, are believed to be involved in controlling the response, will improve confidence in estimating dose-response relationships in the low dose region. Physiologically Based Pharmacokinetic/Pharmacodynamic models are effective tools for integrating information on these various biochemical and physiological processes and making predictions of dose-response relationships. Examples are used to demonstrate the importance of biologically motivated quantitative approaches to dose-response assessment, and the results will be compared to default (linear, no threshold) approaches. The case studies will include an essential element, a weakly estrogenic compound and a poorly soluble well metabolized gas. |
|
WRT2.3 A Computer Simulation Model of Low-Dose Interactions Among VOC Hematotoxins In Bone Marrow. Cox, Tony; Cox Associates tony@cox-associates.com Abstract: Epidemiological investigations of human health responses (especially, acute myeloid leukemias (s-AML)and non-Hodgkins lymphomas) to low-dose occupational exposures to benzene and/or mixtures of volatile organic compounds (VOCs)have produced inconsistent results. Some (e.g., Paxton, 1996; Schnatter et al., 1996) have reported apparent exposure concentration thresholds for increased risks; others find sub-linear or zero dose-response relations (Crump, 1996; Albin, 2000), while still others descry positive risks from low estimated exposures (Qu et al., 2002; Hayes et al., 2001). To clarify the shapes of low-dose dose-response relations for benzene and VOC mixtures, we developed an integrated pharmacokinetic-pharmacodynamic (PBPK/PD) model of effects of benzene and VOC mixtures on hematopoietic stem cell populations (e.g., CFU-GM) involved in chemically-induced s-AML. The model, including a partly experimentally validated feedback-control sub-model of suppression and compensating proliferation of hematopoietic stem cells in response to hematotoxins, identifies combinations of PBPK and PD processes that produce chemical-to-chemical, chemical-to-enzyme system, and chemical-to-tissue interactions affecting toxicity and risk at physiologically relevant levels. A striking and unexpected prdiction is that sufficiently low concentrations of benzene or benzene-containing VOC mixtures can reduce stem cell proliferation, premature recruitment of early stem cells, and resulting risk of s-MDS and s-AML. The prediction of a U-shaped dose-response relation is robust to model uncertainties, but occurs only below the concentrations used in most experiments. The model successfully explains some past puzzles in published data (e.g., how and why smaller total doses of inhaled benzene can have larger hematotoxic effects if administered as a relatively short, concentrated dose) and may help to interpret the epidemiology of low exposures to benzene and hematotoxic VOC mixtures. |