Source:  Risk Policy Report, 1/14/11

Vincent Cogliano, the new director of EPA's Integrated Risk Information System (IRIS), says the agency will have to begin adopting and learning how to incorporate novel toxicological testing methods such as in vitro and computational testing in its risk assessments or it will "run out" of traditional animal data for use in assessments.

"If we're going to be able to evaluate chemicals, we need to use the new data," Cogliano said during a panel discussion at the Health and Environmental Sciences Institute's (HESI's) Jan. 11 conference on its Risk21 project, which seeks to provide recommendations on how to incorporate the new techniques into risk assessment.

But he also cautioned that because toxicologists are not conducting as many bioassays to derive traditional animal data as they once did, there may not be a smooth transition to the new approach. This may have an impact on agency efforts to assess the risks of some newer, high-profile chemicals. An agency source gave perfluorooctanoic acid (PFOA), which is used to make water- and stain-resistant products, and the ubiquitous plastic ingredient bisphenol-A (BPA), as examples of newer chemicals with limited traditional toxicological data.

Cogliano assumed the leadership of the IRIS program -- EPA's premiere hazard assessment program whose assessments are often the basis of environmental regulations -- late last month, following a seven-year detail with the International Agency for Research on Cancer. Prior to his work in Lyon, France, with that agency, he was one of EPA's top experts on the ubiquitous groundwater contaminant trichloroethylene and helped to write the agency's 2005 guidelines on how to perform cancer risk assessments.

Cogliano's perspective comes as EPA's research office is still grappling with how to interpret the new data and incorporate it into its risk assessments and IRIS assessments -- key hazard assessments of chemicals that are often the basis for the agency's air, water, waste and other regulations.

The issue of using alternative toxicology test methods has received increased prominence since the National Academy of Sciences issued a report in 2007, Toxicity Testing in the 21st Century: A Vision and a Strategy, that "presents . . . a vision for mobilizing the scientific community and marshaling scientific resources to initiate and sustain new approaches, some available and others yet to be developed, to toxicity testing."

But stakeholders have raised varying concerns about how the regulatory system will incorporate the new tests, or how to compare in vitro tests in single cells with those of whole animals with metabolic systems that process chemicals.

Agency staff in the National Center for Environmental Assessment -- which also houses the IRIS program -- have been working for more than a year on a "NexGen" risk assessment project that centers around a handful of pilot assessments that will use new and traditional data to compare the two approaches and determine how best to use the new types of information. The project will be laid out for the public at an initial meeting next month.

But in spite of these challenges, Cogliano said that adopting and learning how to incorporate the latest science and technology in toxicology is necessary for EPA to continue assessing the risks of environmental contaminants in the future. During a final roundtable discussion at the Risk 21 conference, Cogliano noted that, "For other chemicals in the environment, there is no requirement to test in bioassays. We're going to run out of data if we want to keep doing this the same way. Tests are not required for 98 percent of chemicals."

Bioassays are traditional toxicology tests, usually performed on lab rodents, that are the basis of most of EPA's regulations. Cogliano's concern echoes many stakeholders seeking reform of the Toxic Substances Control Act (TSCA), which regulates the manufacture and use of industrial chemicals in the U.S. Environmentalists and public health activists frequently raise the concern that the chemical law, unlike pesticides law, does not require hazard testing before chemicals go on the market. Meanwhile, industry and animal welfare activists protest the high cost, in money, time and animal lives, consumed by traditional testing.

During a final roundtable discussion at the HESI Risk 21 conference, one of its organizers, British consultant John Doe, asked panelists about the importance of incorporating the new toxicological methodologies into risk assessment practice. Cogliano described doing so as "very important. Most of our methods were developed in the 1980s. It's the same model for a lot of regulatory decisions."

Cogliano also raised concern that traditional methods of toxicity testing using laboratory animals may be phased out before the new computational and in vitro testing methods are fully developed. "I think we are rapidly abandoning all the bioassays we used to do," he said.

An agency source explains that while there is bioassay data for most of the chemicals currently on the IRIS program's agency for assessment, there is not for some of the newer chemicals. For new chemicals that IRIS will need to assess in the future, the agency will need to be able to use the newer data, the source says. IRIS assessments are often based on bioassays conducted by the National Institutes of Environmental Health's National Toxicology Program, but the source notes that NTP only conducts about ten bioassays a year, while there are thousands of chemicals in commerce that need assessing.

The source named Bisphenol-A and PFOA as examples of newer chemicals with limited traditional bioassay data, making it difficult to assess them without using the newer types of data. EPA has indicated in the past that it would perform IRIS assessments of both chemicals, but both have stalled while other offices sought to assess their risks. EPA delayed its BPA assessment in 2008, when the Food and Drug Administration began assessing its use in food-contact applications. And IRIS passed assessment of PFOA to the EPA's pesticides office, where it stalled last year, and the water office is now considering whether to regulate the chemical. BPA remains on the IRIS agenda, according to an October Federal Register notice, but PFOA does not appear.

"We all agree this is the way we have to go. We're never going to get there otherwise. I think the actual larger job is to translate it into [regulatory decisions,]" Cogliano said at the conference.

His thoughts echoed those of George Washington University professor Lynn Goldman, a former EPA toxics and pesticides chief during the Clinton administration. Goldman decried the ingrained nature of regulatory chemical testing at EPA and other federal agencies in prepared remarks at the beginning of the conference. She described the difference between what she called regulatory science and research science. While the former assesses chemicals through a prescribed battery of tests, the latter is more hypothesis driven, Goldman said.

She argued that agencies' scientists need to collaborate with scientists in academia and industry in order for the two groups to learn from each other. She noted that there are some limitations to research science, though it may have advantages when assessing risks. But this research often does not include some of the basics of regulatory science needed to perform dose-response analysis, such as statistically significant numbers of animals per group per dose, she said.

Cogliano also urged those at the conference to think more broadly about the public health issues in the environment. "Think about changing the problem formulation. The problem isn't to assess [risk] chemical by chemical. We have to think about mixtures, complex exposures, other exposures," Cogliano said. "We can do better thinking about the problem than [thinking about the risk assessment.]"
The HESI project, Risk Assessment for the 21st Century: A Vision and a Plan, or Risk 21, seeks to "initiate and stimulate a proactive and constructive dialog" between industry, government, academics and others to identify key new advancements in toxicology and incorporate them in risk assessment, a HESI director said last summer.

The Risk21 project is driven by two NAS reports, including Toxicity Testing in the 21st Century and a second NAS report, Science and Decisions: Advancing Risk Assessment, which encouraged consideration of possible solutions to environmental problems in the early scoping phases of a risk assessment -- an effort intended to make assessments more relevant.

ILSI/HESI, a group funded by major corporations like Coca Cola and Dow Chemical Co. and Procter & Gamble, aims to bring together government, industry and academic scientists. It's Risk21 project intends to develop guidance on how risk assessors can adopt the recommendations of the National Academy of Sciences' recommendations in Toxicity Testing in the 21st Century as well as Science and Decisions.
The project is divided into four subgroups, each consisting of academic, industry and government scientists studying the topics of dose-response, exposure, cumulative risk and integrated evaluation strategies, or testing.