The May release of the National Academy of Sciences (NAS) review of the U.S. Environmental Protection Agency’s (EPA) progress in improving the Integrated Risk Information System (IRIS) was accompanied by much fanfare. The “kudos” directed to Ken Olden, who took over the IRIS program two years ago, are well-deserved. Dr. Olden has brought about a number of needed changes to the IRIS process, including a commitment to get the best science to support IRIS decisions, and for EPA-hosted, bimonthly scientific dialogues with stakeholders – whether public, in-person or via webcast.
In its report, NAS recommended that EPA continue to improve its systematic approach for analyzing results from human epidemiological studies, animal toxicology studies and mechanistic studies, and to incorporate them into future assessments. These recommendations echo a number of points in ACC’s Risk Principles, which call for more transparency and consistency in evaluating data.
While some improvements are indeed evident, the NAS report makes it clear that more changes to IRIS are needed to ensure EPA’s scientific approaches and analyses are objective and make full and appropriate use of up-to-date scientific research rather than continuing to rely upon default approaches to cancer risk assessments that are based on 1970’s understanding of the processes of carcinogenesis.
If EPA is serious about improving the science basis of IRIS assessments, then it must consistently address mode of action – how the human body works and the way chemicals interact with the body and the environment at different levels of exposure. By correctly designing IRIS assessments from the start – in the problem formulation step – the Agency can organize the available data to objectively evaluate alternative hypotheses. EPA has a real opportunity to implement some of the NAS suggestions related to mode of action, which is a key topic for the June 25-27 IRIS public science dialogue meeting where arsenic and hexavalent chromium will be discussed.
Why is mode of action so important? Mode of action information, when available, plays a critical role in interpreting results from animal toxicity studies and whether those results may (or may not) apply to humans who are exposed to substantially lower concentrations than animals. As described in EPA’s cancer risk assessment guidelines, extrapolating results of animal studies at high doses to humans at much lower doses ideally should be based upon an understanding of the mode(s) of action underlying the development of tumors in the animal studies.
By way of an example, a significant research program focused on characterizing the mode of action (following EPA’s guidelines) for hexavalent chromium was undertaken to replace default assumptions with actual data. With data, EPA can assess the potential risks to humans from environmental levels of exposure. This extensive research made use of cutting-edge science from over a dozen research organizations and culminated in more than a dozen scientific articles published in leading peer reviewed journals. Importantly, the research provides data to support physiologically-based pharmacokinetic (PBPK) models—EPA’s ‘gold standard’—and to clarify the dose dependent sequence of events that led to tumor formation in rodents from ingestion of hexavalent chromium.
Like arsenic, EPA’s handling of hexavalent chromium will be a litmus test of sorts. If EPA objectively evaluates and fully uses the mode of action data for each of these chemicals, then it will signal a commitment to implement both the NAS recommendations and its own guidelines.
We are confident that EPA’s full implementation of the NAS recommendations will result in timely, high quality assessments, and we are committed to working with the IRIS program toward that objective.