Tobacco Documents Online

' EPA Moves to Reassess the Risk of Dioxin Urged on by the scientific community, EPA is developing a new model,for estimating dioxin's risk GALVANIZED BY THE REStiLTS OF A RECENT scientific meeting on dioxin's molecular ae - tions, Environmental Protection Agency (EPA) administrator William K. Reilly has launched a major new effort to reassess the toxicity of this ubi quitous'-and infamous- chemical. chemical. Responding to criticism that the model EPA now uses to assess dioxin's risk is obso- lete, Reilly has asked agency scientists to come up with a new "biologically based" model that will draw on an emerging under- standing of the fixst steps'that take place as dioxin enters a celI (for example, see pages 924 and 954). Reilly and others call the neww effort "precedenr setting" not only for how the agency regulates carcinogens but also for EPA's quick response to new scientific devel- opments-not i;s strong suit in the past. Until now, EPA has gauged the risk of dioxin exposure b,v using the same model it applies to most carcinogens: the linear mul- tistage model, which assumes that risk rises in, proportion to close. Agency officials have long viewed the caodel as'a "default"-one adopted for lack of a real understanding of how carcinogens work-and their intent was always to re place it with something more realistic ono.- mechanisms were under- stood. But so far, they say, such evidence has been lacking. Now it may at last be in hand, at least for dioxin and perhaps a handful of other chemicals tliat behave similarly. The turning point came in an 8 March briefing for Reilly and his top deputies given by three agency scientists: William Farland and Peter Preuss, both at EPA headquarters in Washington, D.C., and Linda Birnbaum of EPA's Health Effects Research Labora- tory tory in North Carolina. Part of the briefing was devoted to recent epidemiologic stud- ies, including the new one by Marilyn Fingerhut of the National Institute for Oc- cupational Safety and Health (NIOSH), which found perhaps the strongest link yet between high doses of dioxin and human cancer (see Science, 8 February, page 625). The EPA scientists. also discussed a reanalysis of data from a 1.976 study of cancer in dioxin-exposed rats that figured heavily in EPA's original risk assessment. After re- examining the original slides of liver tissue, ; investigators have concluded that the ani- mals developed fewer tumors than was origi- nally believed. But it was Birnbaum and Farland's de- scription of a meeting last November at the Banbury Center at Cold Spring Harbor Laboratory that Reilly says made the most compelling case for change. At that meeting a group of dioxin experts agreed that before dioxin can cause any of the ill effects it has been linked to-cancer, immune system suppression, chloracne, and birth defects-one "neces- sary but not sufficient" event must occur: the com- pound must bind to and activate a receptor, known as the aryl hydrocarbon or AH receptor (see Science, 8 February, p. 625). After that, the dioxin-receptor complex is transported to the nucleus, where it binds to specific sequences of meeting who is now working with EPA, is to pinpoint the threshold or "safe" dose below which none of dioxin's ill effects should occur. In building the model, Gallo and his EPA colleagues hope to draw on work on the dioxin receptor now under way in a number of labs around the countrv. In this issue of Science, for example, a group headed by Oliver Hankinson of the University of Cali- fornia at Los Angeles reports on the cloning of a protein that is necessary for the receptor to function. Various roles have been pro- posed for the new protein; one intriguing possibility is that it is part of the receptor itself. The dioxin receptor thus might contain at least two proteins, one that binds to dioxin (and presumably whatever natu- ral molecule dioxin mimics) and another that binds to DNA. "Boy, is that excit- ing," says Gallo, who adds that the new findings will feed directly into the model. Until the model is com- plete, no one can say for sure whether it will show dioxin to be more or less risky than EPA now calculates, though Gallo and others speculate that it will turn out to be less risky. One of the major ques- tions is how close the pre- sumed "safe" dose is to the background levels of dioxin to which the general popula- Key mover. Linda Birnbaum had been urging EPA to change how it does dioxin risk assessment. DNA and turns genes on and off, thereby causing its myriad effects. It had long been known that dioxin binds to a receptor, but before the Banbury meeting it had been unclear whether all of dioxin's effects or just some were mediated this way. The Banbury group also agreed that di- oxin has to occupy a certain number of AH receptors on a cell before any biological response can ensue. The result is a practical "threshold" for dioxin exposure, below which no toxic effects occur. That conclu- sion flies in the face of the linear model's underlying assumption: that the risk of harmful effects begins with exposure to a single molecule and increases from there. Faced with this new picture of dioxin's ac- tion, the Banbury participants urged EPA to develop a new, receptor-based model for dioxin risk assessment. Reilly bit. He has now asked scientists in EPA's Office of Research and Development, in collaboration with academic researchers around the country, to come up with just such a model. The goal, explains Michael Gallo of the Robert Wood Johnson Medical School, one of the organizers of the Banbury don is exposed. If background exposure is already near the "safe" dose, then there may not be much room for additional exposure. Those background levels are largely un- known, so Reilly has added that question to the EPA scientists' assignment. Over the next year Birnbaum and other EPA scien- tists, in collaboration with researchers from NIOSH, the Centers for Disease Control, and the Air Force, hope to get a fix on blood levels of dioxin and the handful of polychio- rinated biphenyls that behave similarly and thus could increase its risk. Meanwhile, other researchers will be studying the sources and routes of dioxin exposure- most of which are dietary-and how it is passed up the food chain. Reilly wants the new model and related work complete within a year, at which time the results will go on to EPA's Scientific Advisory Board (SAB ) for peer review. Three years ago, the SAB sent EPA scientists back to the drawing board when they tried to revise the dioxin standard, saving the sci- ence wasn't sound enough. Birnbaum and other EPA researchers predict a different outcome this time.  LESLIE ROBERTS 17,MLAY 1991 NEWS & COSiMENT 911