Genomics and Environmental Regulation: Science, Ethics, and Law

5 Challenges in Applying Toxicogenomic Data in Federal Regulatory Settings

C H A P T E R 5 Challenges in Applying Toxicogenomic Data in Federal Regulatory Settings LYNN L. BERGESON Both the legal and the scientific literatures are replete with assessments of the implications of toxicogenomics for many aspects of human and ecological health. Based on a review of the research and writings in this area, there is consensus that toxicogenomic data will find their greatest application in at least three areas: susceptibility—identifying subpopulations and life stages uniquely susceptible to environmental factors; predictive toxicology—facilitating high-throughput chemical screening for potential adverse impacts and hence assisting in prioritizing chemicals believed to be most potentially toxic; and understanding disease mechanisms—helping to elucidate a substance’s mode of action to define better its dose-response continuum.1 Similarly, the application of high-throughput genomic data to toxicology is progressing, and the generation of large quantities of toxicogenomic data seems to be well along. The good news is that some of the important technical issues associated with collecting toxicogenomic data, such as how to manufacture microarrays, have been addressed. Database con- struction and maintenance are mature for genetic and genomic data, and microarray databases are developing. Many challenges remain, however. For example, these databases do not consistently incorporate data relevant to toxicology, including exposure , dose, or time parameters. Similarly, statistical models applicable to large data sets are only now developing and do not yet include in all cases dose and time effects at toxic and nontoxic exposures. The next scientific frontier in genomics and bioinformatics will be conquering key challenges such as standardizing methods, nomenclature, and database construction; automating data collection; integrating toxicological with gene expression data; developing statistical models applicable to gene expression data that consider dose and time; and, perhaps most important, validating all of the above. In short, additional scientific, policy, ethical, and legal concerns remain to be addressed. This brief summary of some of these issues is not intended to suggest that these challenges are readily amenable to solution, quite the contrary. Until these challenges are addressed, the true utility and promise of toxicogenomic data will remain unfulfilled in four areas the U.S. Environmental Protection Agency (EPA) identified as the most likely to be influenced by the generation of genomic information: prioritization of contaminants and contaminated sites, monitoring, reporting requirements , and risk assessment. EPA acknowledged this in its 2004 report Potential Implications of Genomics for Regulatory and Risk Assessment Applications at EPA, prepared by the Genomics Task Force Workgroup of EPA’s Science Policy Council (SPC; U.S. Environmental Protection Agency, 2004c). As EPA notes, “understanding genomic responses with respect to adverse ecological and/or human health outcomes is far from established.”2 Accordingly, the application of genomic data any time soon for any regulatory standard-setting purpose, and the use of gene expression data to classify toxic and nontoxic effects, would seem unwise, if not entirely premature. Nonetheless, despite their nascent stage of development, the temptation to use these data early will be hard to resist, and undoubtedly some will succumb. To benefit from the full potential of genomic data, many scientific challenges need to be overcome. Outlined below are selected illustrations of the types of challenges that agencies and stakeholders in human and ecological health areas can expect to confront when using genomic data in regulatory settings. No at68 L E G A L P E R S P E C T I V E S tempt is made here to put too fine a point on whether these challenges are fundamentally “legal,” “regulatory,” or “attitudinal” in nature. Information Quality and Peer Review A threshold issue for all federal agencies is the weight toxicogenomic data and related information should be given in regulatory and federal decision-making contexts. The EPA SPC’s Genomics Task Force was formed expressly to explore the broader implications genomics is expected to have for EPA programs and policies and to attempt to gain “further understanding of the appropriate usage of these data and the potential consequences of their use, as well as to identify possible infrastructure needs.”3 EPA’s consideration of the use of genomic information generally must be guided by and consistent with a broader federal mandate required under the Information Quality Act (IQA), a federal law directing all federal agencies to ensure and maximize the quality, objectivity, utility, and integrity of the information they disseminate.4 How...