Toxicogenomics in risk assessment: applications and needs.

Since its inception, there have been high expectations for the science of toxicogenomics to decrease the uncertainties associated with the risk assessment process by providing valuable insights into toxic mechanisms of action. However, the application of these data into risk assessment practices is still in the early stages of development, and proof of principle experiments have yet to emerge. The following discusses some potential applications as well as impediments that warrant a concerted investigation from all stakeholders in order to facilitate the acceptance and subsequent incorporation of toxicogenomics into regulatory decision making.

Workshop overview: use of genomic data in risk assessment.

The completion of the Human Genome Project has provided the foundation to analyze the expression of all genes transcribed in a specific cell, as well as a reference against which to assess genetic variability and its impact on susceptibility. Recent advances in genomic technologies have set the stage for better understanding and predicting individual adaptive and toxicological responses after toxicant exposure. Thus, it is now possible to simultaneously assess expression levels for thousands of different genes using DNA microarrays, as well as assess posttranscriptional and posttranslational events using high throughput proteomics. Similarly, the risk of toxicant exposure-induced disease used to be estimated across populations with widely varying responses. However, new high-throughput genomic technologies have the potential to greatly improve the accuracy of risk assessment, allowing identification of sensitive subpopulations at risk and ultimately leading to personalized risk profiles based on genetic composition. Recognizing the importance these technologies will have on the practice of risk assessment and the development of regulatory guidelines, the Society of Toxicology sponsored a workshop to evaluate the current status of genomic research; the ethical, legal, and social issues associated with these approaches; and, in this context, how data derived from these technologies would impact risk assessment. This article summarizes the evaluation by experts in genomic research and risk assessment in the first workshop to provide a forum for interaction between these scientific disciplines.

Assessment of interactions of diverse ternary mixtures in an estrogen receptor-alpha reporter assay.

This study used an MCF-7 cell based ER-alpha reporter gene assay to assess chemical interactions within the following ternary mixtures: (1) three synthetic pesticides, methoxychlor (MXC), o,p-DDT, and dieldrin; (2) three polyaromatic hydrocarbons, benzo[a]pyrene (BAP), 1,2-benzanthracene (BENZ), and chrysene (CHRY); and (3) an endogenous estrogen, [17beta-estradiol, (E(2))]; a phytoestrogen, genistein (GEN); and a synthetic estrogen, o,p-DDT. A full factorial design in which four concentrations of each chemical were assessed in all possible combinations (64 treatment groups) was utilized. In addition, mixtures were tested in both a low range (concentrations near the individual chemical response thresholds) and a high range ( approximately 2-10x higher) experiment. A response surface was estimated using a nonlinear mixed model, and the cumulative response in each mixture was evaluated for departure from additivity. The mixture of E(2), GEN, and DDT exhibited antagonistic interactions (p < 0.001) in both concentration ranges. However, specific interactions between E(2)/GEN and E(2)/DDT differed between the low and high range concentrations. The BAP/BENZ/CHRY mixture did not depart significantly from additivity (p = 0.66) in either concentration range, although response levels were generally low. The MXC/DDT/dieldrin mixture did not depart significantly from additivity in either the high (p = 0.065), or low dose range (p = 0.506), with generally minimal responses dominated by MXC and DDT. This methodology has allowed for a rigorous statistical evaluation of potential departures from additive interactions in endocrine active mixtures. In no case was a significantly greater-than-additive (synergistic) interaction observed.