 who is a Chair of Cell Biology and Neuroscience at UC Riverside, and he, as we've been discussing, took the straight science path to where he is now. He got his PhD at UC Berkeley in Environmental Health Sciences, and also working with the biomedical sciences program there. And he came to UC Riverside after a post-doc at Lawrence Livermore National Labs, always throughout his period, working consistently on managing risks from toxic chemicals. And then he became Chair at UC Riverside a few years ago. And in the path of publishing a lot on how to evaluate toxicity risks, he also took a year to be a prestigious Jefferson Science Fellow at the Department of State in Washington, D.C., and that was in 2004. But he's also guided the environmental toxicology program at UC Riverside for a while. Almost all the UC campuses have an environmental toxicology program, and we had a UC system-wide toxic substances recycling teaching program that I also participated in until the system-wide program was discontinued. But we hope that on each campus we can keep the flame of environmental toxicology going through training programs, through visits such as these to interact with faculty and students, and to engage incoming and new students into this really complex interaction between the basic science, the applied science, the policy issues that surround every day products that we use. Eighty thousand chemicals is a lot, and I'm sure he's going to tell us how many of those we know, something without, to make important decisions. But really, it's my pleasure, and I can't wait for conversation with us. Thank you so much. I should probably talk more about the Cannibal Legislation Regulation. You know, I think that we had this opportunity to work together on the CC toxics, some of the research and teaching program, and we were involved in a program involving green materials, unfortunately, that ended a little prematurely, and hopefully to keep able to do the work as we go on. I gave a little bit of an introduction, but for those who came in later, my research is really in the area of toxicology, and as I've gone into my career, I've become increasingly interested in risk assessment, and so what I'm trying to do is really talk more about bigger issues in risk assessment. So I won't be talking about any of the actual research that my lab is doing, per se, or bench research, but I will talk about some work at the end that I've been involved with, the EPA, and I will really try to understand how the science decisions are made at an agency's level. So, as I mentioned, I should probably do some background and some introduction to risk assessment, so I will go through this. So the first five or six slides will kind of risk assessment initially, and then the middle, there's another five or six that we're going to give you background information. Generally, there are a couple of definitions that are important to be aware of, so that if hazard is an ability of a substance to cause an adverse effect, kind of an innate ability of that substance, risk is the probability that hazard or the effect will occur. Risk assessment is really characterizing the likelihood or probability that a risk will occur under specific exposure conditions, usually to a specific population, and these are considered to be kind of a scientific aspect of things. Risk management is more of a policy decision. This is the process of evaluating alternative regulatory actions and selecting among them. So to illustrate this difference between hazard and risk, this was an idea I got from someone else I thought was actually fairly clever, but the idea of the hazard from the shark, if you're looking at an aquarium and people are there taking pictures and looking at this and observing, my shark is clearly a very dangerous animal species, but it poses essentially virtually no risk of a shark to these people themselves. So that would be hazard. The innate ability of a shark is a dangerous animal, but under these conditions, it poses very little risk. In contrast to this, where you have a surfer, it might relate, and I don't know if you can see it, but in this swell here, there's actually a shark underwater, and this is a very different, this is risk, and this is the real hazard, maybe an abstract, this is what I figure out, the real consequences. So generally, there's thought to be four steps in risk assessment. There's a really standard. This was developed by National Research Council. It came out of a book in 1983 called The Red Book, Risk Assessment of the Government, and it broke it down into four steps. Generally, it's considered to be used post-office. So hazard identification, determination of an ape, adverse toxic effects of an agent, then assessing the dose response. What's the relationship between dose and this health effect in humans? Then relief, what's assessing exposure, the measurement of the intensity, the frequency and duration of human exposure, and last, characterizing risk, estimating the incidence of health effects out of various conditions of exposure in a given population. I use this in my teaching class on the final, if you can read it, but I like to do it graphically, illustrates this sort of thing. So the three main steps of risk assessment, risk assessment is really locked, you have the hazard identification, does the agent cause an adverse effect? The dose response assessment, what is the relationship between those incidents? Closure assessment, what exposure is currently experienced for an antispiece in the future? Now the information you mentioned, you've stepped to come from research, a lot of people are doing, some people here at UCLA, and elsewhere in my lab, where you look at lab and field of observations, could come from information even from suicide and sexual exposure, dose response information, so funneling in, so the research funnels in these categories, and then these three, the hazard identification, dose response assessment, exposure assessment, they're combined into characterizing the risk, coming up with an estimated incidence in a given population. Now the combination of this is risk management, and this is really what someone in agency, an official, maybe someone's elected official or appointed, has to make a decision and say, what are the regulatory options? And you look and say, well, okay, evaluation public health, economic, social, political, you know, comparison of the risk, and finally you would make an agency decision. Now that's general framework that's used. So what I thought I'd do is talk about some ongoing developments that are happening currently in this area of hazard identification, and I'm going to focus on, really a focus on genotoxicity testing, this area which I've done a lot of research, and what's going on, kind of taking a bigger picture of discussion of what's happening at the moment, and I literally, some of this is going on very much at the moment, so, talk about these genotoxicity tests, these are short-term, in vitro and vivo tests designed to detect chemicals that can cause genetic damage, and agencies use this data in the absence of long-term animal studies to determine whether a chemical should be considered a possible carcinogen or not, or cause other types of terrible effects, and sometimes if you have long-term animal treatment studies available for the evaluation of carcinogenicity, this genetic toxicology data maybe assists in interpretation of these results. So the first part of what I'm going to talk about is really to focus on this, and the second part later on I'll put this in context, I mentioned beforehand before you would take a pesticide to market, in this case a pharmaceutical to market, you have to do a series of testing, so you might do the test in the United States and then you'd want to market that pharmaceutical in Japan, the Japanese would say, well, we didn't quite agree with how you did that test, therefore you can't market your pharmaceutical in Japan, so you'd have to redo the test in Japan, which is happening in Europe. So this ends up being what's called the trade barrier, where you can use scientific test requirements as trade barriers. This is very frustrating for industry. So they've gotten together and it's really called ICH is the International Conference on Harmonization to try and come to an agreement that these are like all the major sorts of countries accept or testing requirements and so that everyone will agree to accept them for the most part. So for looking at genetic damage, genotoxicity, screening for potential cancer risk, there are three tests that are required currently. One is a test of genutation of bacteria. It's a series of tests, but very similar. The second one is an in vitro test. This is tested in cells and culture, either if it comes on the damage in the male cells or screening for mutations affecting a particular gene of a primary kind of gene. So these are in vitro tests and the third one is an in vivo test where essentially we look for chromosomal damage in the bone marrow or blood of rodents and this is commonly not used in a test called the microcustacean. Dose the animals, period of time sacrifice them, harvest the blood or bone marrow, so that's the sort of standard test. Once you get a positive result on this it triggers usually additional tests. They're all negative, so it's probably fine. If it's positive, then you want to say well, under what conditions is it positive? And that can be illustrated. This is the test requirements, test besides, very similar, essentially the same three tests are required here on variations. What it shows here is once you get a positive result in these tests it triggers additional tests looking at the chemical interact with the DNA and the gonads. It causes bone legal effects it causes hurdle transitations specific low concentration. This is really focused on hurtable damage. But the idea here is when you have some vitro tests and a vivo test and if they're positive then you have a much more complicated, more involved in vivo tests. I'm kind of surprised that filling the first tier doesn't not have something from vivo market. Well, it may. It's kind of a risk reward. You know, these become much more expensive, but if you think you can make a large amount of money and that there are reasons that these will be negative and this is due to some artifact then you're willing to invest the money to do that. So, the key thing is you have vitro tests and some of the vivo followed up with the vivo tests that are much more helpful. As I mentioned earlier in roughly seven years ago Europeans passed the cosmetic directive the 7th amendment and this bans the marketing of cosmetic and personal care products that have been tested in the animals. So that whole scheme you walked off three fourths of it. And the ban on the testing of cosmetics went into effect in September of 2004 and a similar ban on testing ingredients took place in March 2009, almost two years ago. So, all these sort of strategies for follow up testing you can't do cosmetics. So it's been a scramble to figure out how do you determine safety of cosmetics and personal care products in your own. They went with six different in vitro tests and the problem is every single product failed. So that was not the same. So they're still really wrestling. It's a big issue how do you test for the safety of these to ensure that there's no problems when in fact you've lost the opportunity to do in-demo animal studies. So this is from one of the committees from the European community. It's the committee on consumer products. This is their position in 2009. So the current vitro tests are very sensitive. Based on the clearly negative vitro test battery mutagenic potential is excluded. No validated replacement methods are available that allow the follow up of positive results from standard vitro assays without further animal tests. So in many cases it will not be possible to evaluate the genetic potential of cosmetic ingredients on a sound scientific basis. Important part of the toxicological evaluation of cosmetic ingredients cannot be accomplished. That's one big thing that's going on in Europe. In terms of animal welfare not allowing vitro tests how do they go forward. They're really wrestling with this. Now the other right one I mentioned is called REACH. It's the restriction evaluation authorization sorry, registration evaluation authorization restriction of chemicals. And this went into law about two years ago. And it introduced registration and mandatory data requirements for new one. And it was estimated at a time of approximately 30,000 chemicals manufactured or imported greater than one metric ton per year in Europe. And the real key thing on this is they flipped the burden of proof and transferred responsibility to risk assessment from the government to the manufacturers and importers. And this is having a huge impact on industries. Not only in Europe because if you want to market or import chemical a product to Europe you have to apply by REACH. 80 million citizens. Huge market. So virtually many, many countries, companies worldwide are trying to be compliant with REACH because of opportunity for certainly assorted products in Europe. Now this is a testing data requirements brought by REACH. And they vary depending upon amount that's being manufactured or imported in Europe. If you have greater than one ton there are certain test requirements for either one, 10 tons, others 100 tons, 1000 tons per year. And so if you look here the lowest test requirement, you'll need physical chemical data boiling point you know, volatility water, park, fish, things like this. There's some on environmental fate one as a ecotox data, maybe Daphne can remember that it's a good chance. Probably algae actually, but and then there are about 6 that are related to mania ecotoxicology. And as you go up with higher production volume, you actually get pretty high test requirements. Essentially over 50 different tests that are required in order to market that. In a lot of cases these have never been required before. So very costly proposition. So here these are the tests related to toxicology. So greater than one ton per year there's a test for skin irritation two in vitro tests are required one test for eye irritation in vitro skin sensitization one test eugenicity is bacterial test required and then there's acute oral toxicity tests. When you go up to 10 times it requires everything before it and then these are going to add it to it. So now you get in vivo skin irritation in vivo eye irritation sort of thing. Two more in vitro test eugenicity additional acute toxicity tests, repeated dose tests reproductive developmental tests, etc. That goes on, now you go up to 100 times now you get additional the first in vivo test eugenicity, but you're having 90 repeated dose tests. Developmental studies, one in vivo a two generation reproductive study etc. So this ramps up. So initially when REACH came out they thought it would affect about 30,000 chemicals. That was the number they were estimated. There were 30,000 chemicals that were manufactured or produced greater than one metric ton per year. But when the last stages of REACH legislation they thought we really don't know how many are coming in. Let's do a pre-registration phase so we can force people to work together from this animal usage. Don't work together to find out how many chemicals are coming. Well in that pre-registration phase there were 144,000 registrations that have been submitted. Now some of these might be duplicates but this was an article by Thomas Hartman published in Nature about two years ago. In essence because of duplications their estimates are that you think you're 30,000 too low. The worst case scenario is about 100,000 when you test a test case about 68,000 chemicals. Individual chemicals need testing. So you saw that list of tests you're looking at somewhere 60 to 100,000 chemicals are going to have to go through as a minimum of these tests but a lot of of these other tests. Very costly, very timely I see. Now the other thing that they thought of the animals used in the EU in the original REACH legislation they estimated it would require about 3 million animals to do the testing. And Hartman estimated that actually in the worst case scenario it would require 141 million animals to do the test and the best case scenario is 54 million animals to do the testing that's required. The European Chemical Agency which was established by REACH their more recent estimate from last year is actually 9 million animals kind of what they're thinking. So in Europe you've got this big push for animal welfare to reduce testing on animals that just passed the law that has very very large animal usage which is creating lots of challenges in Europe. So as I said the vast majority of these costs really are associated with reproductive toxicity testing. Repeated dose testing has about 20% of the cost and the cancer studies and skin sensitization etc. Because the cancer studies are only done for the most extreme cases that other animals use for costs are certainly much less than these. The animal usage the vast majority is reproductive toxicity testing but even if you're talking about a 10% here let's say these genotoxicity tests the follow-up tests represent a fraction of that you're still looking at hundreds of thousands of animals. So there's been a big focus in Europe with Sam Halens let's refocus on these short term tests how good are they how sensitive are they and this is really a figure based on David Kirkland's group. These are different types of short term and detrope tests. Ames test which is done in cell and olivacteria standard one and it looks at two different aspects of this. One is called sensitivity and one is called spasticity. So sensitivity is a situation when you have 100 chemicals that are known to cause cancer and neurons what percentage of those will be detected by your asset? So the Ames test is actually a little high but close enough 58, 59% so the Ames test is a spectral test that will detect 58% of root percentages. So that's a sensitivity issue. Spasticity is the opposite. When something has been tested and doesn't cause cancer it will cause mutations. So essentially it's saying this is spasticity. So it's negative for causing cancer and is it negative in your short term test? The Ames test does quite well in this way it's almost 75%. So this is kind of an interesting combination. It's not as sensitive as others but it's very basically. The other short term test is the mass diploma assay so it's mainly in vitro. They tend to be more sensitive they detect more root carcinogens but their spasticity drops off. They detect they get a positive result when things doesn't cause cancer. And currently what you're doing is adding combinations of test batteries shown before and this shows combinations. So when you start doing combinations of test batteries you pick up about 85% or 90% of the carcinogens almost 70% of your non-carcinogens are positive as well. So this is a problem because now if you have a positive result in a short term test it requires by reach to follow up studies in animals. So what they're saying is this is a huge amount this is a real problem. This lack of really spasticity the false positives and there's been a huge focus in Europe on what can be done to try and reduce this. So if you look at a reach these sorts of faults or misleading or irrelevant positives can be done under any necessary testing of really 10 to 100s of thousands of animals which is a very high priority. Also under reach in the cause of medical loss these can lead to the withdrawal or end of development of many beneficial or potentially beneficial agents these come with consequences so there's going to be major economic consequences one or the other. So just to kind of overview what they've been doing and looking at this is going on actively. So they're exploring the use of more represented cell lines we work with better cell lines that aren't as sensitive. They went lower the maximum cytotoxicity limits for these in vitro mammalian cell tests. The in vitro mammalian cell ass is the ones most problematic. Can you reduce the maximum test concentrations for vitro mammalian cell tests? I'll tell you how tiny this is. Tomorrow there is an expert meeting in Paris called for OECD which stands for the Organization for Economic Cooperation Development. Again this is 45 countries or something that work together trying to agree upon test guidelines that they'll all accept. This is currently a matter of debate. Before I came here this morning I sent out an email to one of the people who's going to be there. So you're asking for my advice. Sorry. I don't really understand lowering the toxicity limits meaning to improve specificity. Yeah, essentially what happens is there's a belief that you get to really highly toxic, excited toxic levels you get to lots of damage that occurs. If I'm going to let you think of this as a cell dies as a cell starts dying starts releasing nucleases start degrading DNA. So if you catch it that period you're just looking for DNA, integrity DNA, you're going to sell causing DNA damage. The damage may not be a direct sort of due to reactivity of the chemical it may simply be to a secondary effect in the toxicity. There's really high levels of excited toxic if you bring those down it's believed you'll eliminate some of these sort of artifactual results. And you may not know this but there have been long debates about what that level should be. I mean, literally put on for a couple of years on some of the assets I've been involved in. I mean I was involved in this and right now this is one of the big focus is what's the maximum test concentration. What do you say I've tested too much. We're done. Another point is to modify the standard test battery to replace in vitro mammalian cell assays within the second and vivo test or there's a whole big thing in exploring your assays and approaches. EPA, FDA, National Toxicology program are trying to come up with what's called TOCS 21 TOCS College in the 21st century can we replace the existing tests with better tests using genomics, proteomics, screening acid, etc. And then there's a refining interpretation of test results which we're going to help out. There are a lot of people who spend quite a bit of time on this. So that I talked about those ICH TOCS test guidelines what they were trying to do they finalize pharmaceutical industry all got together, some FDA and essentially agreed so they talked about this and they agreed to what they thought was this maximum test concentration and they could eliminate the vitro mammalian cell assays. We all got together and they were going to have the final signatures. There was a meeting in Japan the representative from the FDA was there who was going to sign it and he got a phone call from his boss ordering him not to sign that document. But there had been enough push back from other scientists in the FDA that didn't agree with that that they wanted more time to discuss it. And so it was accepted by all the other regulatory bodies authoritative bodies around the world but the US FDA did not agree for it and they convened an expert panel in January 2010 to advise the FDA to actually chair that panel but the final decision is still not to be made whether they will accept it or not. So that's the kind of the status on this and what they're proposing to do is to keep the status quo in this option for the use of genitalia and they would turn around and have the imaging animals and then take a second measurement of genotoxicity as you know is the proposal replacement. So whether FDA will buy into that or not is still a question. You have anybody to say what core of the FDA scientist objection is to the Well well it's kind of interesting yeah basically the concern with the objections to the FDA scientist are a couple of two-fold really. One is that well let me put in context and I'll explain the objections the center for drugs is really the key one of all it's called cedar. For the center of drugs they do these short-term tests in vitro and these short-term genotoxiciasis and then they in 90% of the cases require a full-up animal to your animal bioassay for cancer so they require a cancer bioassay as a follow-up so they will get eventually the cancer bioassay results. So from an expert point of view that we were on under these conditions it makes sense to skip this one test because you're going to get the cancer data anyway. Well the FDA scientists there's two groups of them cancer results are not available until they've already gone into phase one and phase two clinical trials so you'll be testing things in humans without knowledge of potential genotoxicity and certainly your most sensitive genotoxicity test you're going to do. The flip side of this is that there are other centers within the FDA those that deal with food additives, those that deal with veterinary, medical products, etc. and they can't, well, cedars the drugs can require a test to be done as part of the package. Those other groups can't require things they only can get what's given to them so for food additives what's happening is you're eliminating what they think is one of the most powerful clues is that there's a problem they should pursue it more so that's why they approach them so it's a big actually it's a real it's caused quite a split within the FDA by acrimonious and you've got a bunch of policy makers that really don't understand the issues they're hearing from one group totally supports this and the rest of the world tends to support it but then they have this other group with their scientific sciences being injected so they have to make a decision and really it's not position what it is it makes it cause a huge bit but that helps that's what's going on with the FDA it's all about dose response assessment and harmonization of cancer and non-cancer approaches so and it's a risk assessment what you can look at what's called in the 1970s is called essentially a parallelogram approach by Fritz Stoltz but essentially if you're looking at what happens in roads going from high dose to low dose you can figure out what your dose response curve is you might assume that the same thing happens in humans going from high dose to low dose if these are comparable and their responses are comparable that's sort of the basis that's used and there are two approaches but what you're really interested in is what's happening in this really very low dose region down here how to make determinations and there are two approaches that are used one is you assume as shown here you've got a dose response you draw a line it's a zero point you have actually a threshold and so if you can identify what that threshold is by definition anything below that would be safe because it's below a threshold and above that there's a risk the problem is we're working with animals most of the time and we don't know how humans compare with those animals for most chemicals so you want to build in some sort of uncertainty factor to deal with uncertainty usually you deal not only uncertainty from animals to humans but uncertainty within humans the standard way is to use uncertainty factors tenfold for human variability tenfold from extrapolation from animals to humans and depending upon data quality and some other things you might have other uncertainty factors it's not as soon as there's a threshold very commonly done this is the standard thing that's done for non-cancer agents things that don't cause cancer approaches now for things that cause cancer usually different approaches used you'll have a series of studies maybe you've got three bioassays that have been done mice and rats sex and studying rats and you look for which one caused cancer and the most sensitive sex species an endpoint and that may give you the most response curve here now the interval range usually you can't see anything below about 10% statistical reasons so at this point extrapolation is made you have to extrapolate to low dose region you try to figure out what's safe and generally what's considered a safe dose is one that affects one in a million individuals or the virtually safe dose and that's shown here so you have to extrapolate and there's different extrapolations you can assume it has a threshold it drops up very rapidly so there's no risk at low dose or a linear effect you can actually assume it has a super linear effect as well less common so this extrapolates to low dose region so the current EPA approach is to you look at your dose response and here would be modeling at dose response but in reality if you start here at a 10% effect level you think half of your individuals have a higher risk and half of the individuals have a lower risk right so you don't want to be conservative and protect most of the population so rather than working from this 10% line you actually work off of the lower confidence limit on the dose which it crosses and that's called this LED 10 the lower confidence limit the effective dose that affects 10% of the animals there they do a linear extrapolation down to zero extrapolating to very low dose it's kind of the crime approach that's used so again this is usually used for carcinogens for refinement so it's not all carcinogens it's really mutagenic carcinogens more of those used so as I said there's two approaches that are used for non-cancer agents it doesn't have a threshold if there is a variety of reference dose it's safe dose EPA calls this a reference dose World Health Organization calls this an acceptable daily intake ATSTR calls this a minimal risk level for carcinogens for the same thing mutagenic cancer carcinogens it's assumed that there's no threshold the risk is present very low doses, mathematical models extrapolate and usually 1 million risk is kind of what is the bench point and so you put this together to try and test the main risk although the same thing the breakdown initially was carcinogens with the extrapolation non-carcinogens and nonlinear extrapolation more recently had mutagenic carcinogens non-mutagenic carcinogens threshold industry wanted the threshold mixed life bodies, much much cheaper if you say there's no risk below a certain thing you can forget about it so they pushed over the last 10 years to try and harmonize they didn't like this dichotomy between carcinogenic and non-carcinogenic they proposed to harmonize based on a threshold should be used for everything while the National Academy of Sciences got the National Research Council they convened an expert group on risk assessment and the nature of the people in this group they basically said yeah, we should harmonize we should make these the same but we should harmonize on a linear rather than a threshold this has a lot of linear this came out in 2009 industry is very upset about this I mean there are many different working groups trying to figure out how to change change this recommendation came down to Morgan Zeiss one of the key people Gary Ginsburg, another key person and there's some pretty good arguments you can make for linear using linear extrapolation and the other thing on this is usually working animal data and it doesn't take an account heterogeneous populations so you work with rats you put a particular strain almost always white rat so these are really inbred and are very homogenous human populations aren't homogenous at all so even when you see and this is kind of theoretical but it makes sense and it's probably correct so animal models see this sort of sigmoidal nonlinear sort of sublinear response curve frequently however you don't see that very often in human studies the reason it's thought to be the following in a genetically homogenous population you may have a sort of nonlinear type of response seen however when you start working with heterogeneous populations where there may be genetics, susceptibility what's going to happen is you'll have your regular population be but there'll be a sensitive subpopulation when you start getting a heterogeneous population with all sorts of multiple genes involved and environmental lifestyle factors what happens is you get lots of bumps and this appears to be more misleading so that's why it's thought that in human studies usually you see linear types of responses but in the animal studies that would be very offensive now the other factors so we've got the sort of epidemiology where you can get into this the third approach on this is looking at mode of action information so if you understand how these chemicals work it ought to allow you to make some decisions and inform this decision making process so I'll talk a little bit about these mode of action and I'll get you guys thinking and giving you feedback for it so in the cancerous success guidelines that came out in 2005 essentially said if an agent acts through a mutagenic mode of action this will affect how one interpret makes decisions about the shape of the dose response terminal dose region at the same time there are supplemental guidance for assessing susceptibly early life exposure that came out with carcinogens and this determines if it's mutagenic it determines whether you add additional essentially safety factors an agent dependent adjustment factor to protect against early life susceptibility so whether something is mutagenic and not actually becomes very important from a regulatory point of view very simplistically you can say if it has a mutagenic mode of action you assume it will put in extrapolation but it doesn't and it's a nonlinear threshold approach it's actually not entirely correct but it's close enough for illustration so the problem right now is what I've been involved with the EPA is the basis for determining whether a carcinogenic mutagenic mode of action is not firmly established what weight should be given to various test results what kinds and how much evidence is necessary to make that exclusion it's currently the focus of the EPA agency like task force they've been working on it for about four years they can't come to an agreement they came up with a proposal in an expert view, they got shot down revising it and this is what I've been involved in with the EPA is rather than we get theoretically what happens what's happening in practice so I'm going to give you a couple of examples here and this is kind of an idea of scientific positioning this is a small molecule called ethylinoxide it's commonly used as industrial intermediates, used to sterilizing agents but this has been a field for a long time I've seen it anyway this is an epoxide that's basically a reaction with DNA so it binds DNA both in vitro and vivo it's mutagenic and bacteria because it comes home damaged mutagenic and vivo both in non-target and target organs you can actually pick up and animals have been treated with ethylinoxide they develop cancer you can go to those cancers the tumors, you can sequence genes they're involved of course in jazz it's P53 and RAS you can actually see increased mutations in those genes sort of patterns it causes all sorts of effects in humans similar things are seen in basically similar agents mutations so one of the authoritative bodies in terms of other substance carcinogen is called international agency for research on cancer I'm going to give you a little trivia this is an international body organization you know which body does this within the state of California determines whether a chemical is known to the state of California is causing cancer a body called the carcinogen identification committee it's formed under Props of 65 so you go to gas pumps and you see that little sign that says this establishment is known to use patients that cause cancer anyway that's the one that does it for California the federal government level the international agency or the national toxicology program that makes a determination on that as well so the different authoritative bodies but the international agency for research on cancer is going to be very well respected for the body so they consider this as sort of a genotoxic for ketogenic mode of action I think everyone agreed it's positive in all these different texts let me pick another one something different chloroprene this is just in making synthetic rubber sorts of things like the hoses in your car that connect between the radiator engine that's the sort of thing, hard rubber the question is has been tested in animals and it is really positive it causes tumors and something like 8 tissues in rats 6 or 7 in the mouse borderline carcinogenesis is a screaming positive so now you're regulatory regulating you have to make decision to sysact for ketogenic mode of action when you assume it's threshold then we have to extrapolate so here's your data it does bind to DNA in vitro with metabolic activation but you have no information in vitro it's positive mutation in some bacterial studies but it's pretty iffy a lot of them are negative it's kind of a strange mix it's been tested in one study of mammalian cells in vitro for hemogenesis, it's negative it's largely negative it causes chromosomal damage DNA damage in vitro and in vivo it's negative for mutations in one study in vivo it has mixed results it goes off with studies however if you look at cancer-related genes in these tumor, multiple tumor sites there's kind of unique pattern of mutations a fingerprint can see and that fingerprint in types of tumors are actually quite similar to what's seen with 1,3D denying isoprime in vinyl fluoride so I would consider this a 2B essentially probable even in person actually it's possible in person so you're a regulator you've now graduated you've got a job with the EPA you're working the IRS for you you've got a risk assessment program so what do you call this? it's actually a mutagenical reaction I'm venturing the essence so what are the problems? well, you've got a little bit of data here it doesn't bind to DNA in vitro but you can force almost anything to bind to DNA in vitro if you play your conditions right almost totally negative in vivo you've got some in vitro stuff but in vivo stuff pretty much negative so how strongly has the mutation seen in cancer-related genes how strongly correlated is that with something being cancer is that like a very important... well it's actually a very good point like you said it had a very specific tumor for it well there's two arguments on this okay since cancer is a genetic disease even when something doesn't act for an engineering mode of action because a cancer tumor eventually you're going to have gene mutations occurring in that tumor so the fact that you have mutations by themselves may not be convincing but what is convincing in this case is the frequency and the type of mutations the type of mutations are very similar to what's seen with these other agents that are known so you see things sort of based on substitutions or preemption mutations and so it gives you kind of what's kind of like you can think of this as a fingerprint code so that is actually for me is a very influential this is very heavily influential any other thoughts? now if you're an industry you're going to want this to be non-engineering mode of action because it could save you a ton of work a lot of money and you're working in this sort of range that I'm going to do but EPA didn't weigh in on this one and they determined that the mode of action supported the evidence is part of the DNA mode of action and I'll see why they didn't come up with that let me have another one one for dioxy common solvent so if I'm not coming go back to that before I will forget so if I'm someone in the industry why can't I request for additional information additional stories meanwhile we would keep the product on the market so for example there's a lot of questions in one study and that makes results out of social life you say why don't we clarify those actually they try to do this all the time I know but this judgment by iris would be called premature by somebody who is a strong advocate well I should say it's not always industry yeah so naphthalene naphthalene is a simple sort of aromatic benzene rings used mothballs mothballs naphthalene was there was it was known as tautoxicity in a mass lung wasn't really considered by cancer until there was an NTP bio-acid that was done and it was basically suggested for cancer in a mass lung but very strong for cancer's nasal cavity of rats so this is moving this way it wants to regulate it help shut down the administration generally the white things go forward they're opposed to regulations well who's the big problem well department of defense has all sorts of contaminated groundwork sites they have naphthalene act they're on the hook for billions of dollars I mean we're talking billions of dollars for safety so they stop and say what additional studies do we need we better wait for those studies it goes into a delaying process because they don't have to spend the money while it's being delayed and you can debate this linear nonlinear nonlinear genetic to your advantage so if it's to your advantage economically to keep something on the market without having to pay you'll be arguing that point with you and that's part of the big courses that go on so anyway one for dioxane this is pretty well making maybe it can be a binding non-linear genetic it makes results largely negative for chromosome damage in vitro and in vivo no information on vitro and genus in vivo however it's been positive in several studies for chromosome damage in the target or in this case the liver one of those studies is that my laboratory so I have some interest no information mutations cancer-related genes no information on genotoxins and chemins no information on similar chemicals not evaluated by IR what decision is made what's the chemical used for it's a general solvent so there would be possible exposure to it yeah there's actually pretty high exposure to it it's a common need to solve it and so I should tell you a little trivia I'm doing a poster with colleagues in EPA on some of this stuff it's metagenic mode of action as I left my call car I got a voice message I called up one of my colleagues and said the poster's been approved but they need you to change the conclusion unless it's chemical so this is the way I interpreted it they said they think it's life and percentage on non-mode of action the evidence indicated is either non-genotoxic or ketogenic toxic they didn't like that conclusion so I'll have to look at it I can't say I'll have to change it no matter what I mean I mean you know you get into this how you want to work things on the action when you say genotoxic are you also looking at epigenetic differences in the chromosome so acetylation, mesylation are all their key genes that's actually a very good point right now the regulatory strategies don't know how to deal with epigenetic changes there are almost no good short term tests of epigenetic changes and even how to interpret those so it's clearly going to happen on the horizon it's an area for promising research but it's not nothing's done at this point people recognize they must be happening clearly happening but what chemicals do they probably the poster child for this is why they decided which I are considered a probable human carcinogen let's see anyway let me I don't have a series of these but I think I'll go to questions let me just summarize here well these are some of the other factors this is something I've been working on I was trying to identify you get the short term test results but what are the other factors that weigh in and what are the considerations what is understanding differences between in vitro and in vivo metabolism I think it's really critical in some cases you can have two nitritalia all your in vitro data is most your in vivo data is negative but this is positive consistently for what's called you can't schedule DNA synthesis although it doesn't confine to DNA as well so but realizing that you can explain away the in vitro data because the metabolism in vivo is really quite complicated part of the metabolism takes place in the intestine by microorganisms bacterial microflora in the intestines and that's critical you won't find that in vitro importance of uptake and its impact in vivo tox, genotoxicity results the difference between chromium 3 and chromium 6 is very likely uptake related structure similarity in genogenic carcinogens distinct mutations cancer related genes play a role in the impact of toxicity or high-dose effects one for dioxin I should told you the reason they went with it is the positive results were seen at really high doses and so they kind of said we think this is secondary to toxicity so we're going to ignore this it's not my day that got swept away it's not my day they have to do that origin oxidated damage, evidence for altering bonds of action let me just include quickly we'll skip that one generally new regulations are increasing these genotoxicity tests bringing more scrutiny the accuracy of the current methods there are no issues such as Pox 21 collaborations under light but they're likely to have the same problems in many respects with the current assays so given their extensive established use regulatory acceptance many of the current assays will likely be used in the foreseeable future so we're going to have to figure out what to do with it refining our interpretation what the Europeans are doing is actually quite I think valuable let me make these tests better so in mode of action termination where the chemical causes of cancer and immunogenic mode of action should be based on a thorough understanding of the chemicals metabolism and toxic effects and those of similar acting agents decisions should not be simply based on superficial counting of the number of positive and negative short contest results and the last one is this I'll start off with this this is a sculpture by sort of Johnson called the awakening reaching up from this hopefully it reaches coming out and it will actually lead to a better future rather than a more future anyway thanks for coming I think we have time for some questions so I'll ask so one of the concepts you talked about balance of possibilities in causality that is so when I think one of the chemicals that you showed no evidence of DNA binding but then from some other and so the uninitiated these are problems which are exclusive but apparently not so how would some results be interpreted in this context but maybe there is a net reason that we're just not that's a very good point most the vast majority of chemical carcinogens bind covalent with DNA not all of them there are things like interpolating agents that combine covalently with concentrations you can also get something that will cause oxidative damage so it doesn't bind covalently but cause damage and also get chemicals that interfere with enzymes such as topoisomerase 2 which will inhibit topoisomerase 2 cause DNA breakage that will not result in being binding but for the most part that particular example is used as one in which that was part of the argument for this is probably not working for it for there are others or so I mean I was asked specifically could you come up with an idea they think this is working through causing mutations but they can't protect any of the DNA addicts why might that be the case well some of these addicts would be unstable and could result in a basic site sensitivity Mike, is there a question? yeah there are also compounds that are I don't know what you call co-carcinogens secondary carcinogens through more on the promotion side of cancer as opposed to the initiation side of cancer and is there any plan to integrate that into the risk assessment process not that I'm aware of the risk assessment process and one of the real weaknesses in the current risk assessment process is that it works one chemical at a time with multiple chemicals and they may cooperate in their interactions so this is kind of a synergistic differentiation and people talk about it but it is so difficult to try and figure out how to actually manage that because you can think of all these and you've got a dose response and you can think of a second chemical and it may have effects of a totally different series of doses so you have to test every dose with this with every possible combination of that one it gets in a matrix array and the numbers can possibly get really large I'm sure you've dealt with this I mean from an experimental point of view I've dealt with it as well and usually try to take the easiest way possible one chemical at a time so there are many is IRIS it's a database that EPS doing a lot of work to put together and support with strong evidence is it also supposed to be consulted by the general public you're second to the last slide about society and science because I know there are databases you can go to them online there are I don't know it's non-profit organizations type in your chemical you know should I buy this product I know that product so I don't know how good the science is and if EPS is not going to be the place to go for everybody then it's a really difficult situation for consumers to know and believe well the IRIS database which is Integrated Risk Information System database just by EPA is really considered to be kind of the agency-wide determination sometimes they'll put up things with pesticides or other offices may weigh in specifically it's supposed to be the general overview of what the EPA and how they evaluate and they're they're putting a lot of work into it I mean and they're it's in the past but the IRIS database really just had kind of the if you think of this the abstract or the summary that's all they would post online is the summary and you wouldn't see all the stuff behind the scenes now they're putting toxicological reviews that support the conclusions being made I mean some of these are you know the usual one is 100 pages long but typically several hundred pages long I mean you get things like TCDD there are thousands of pages long documentation and usually the more documentation the more controversial the longer it takes and the process is delayed oftentimes but that's considered a kind of the agency-wide decision but then again you know APSCR has their opinions FDA has their opinions differently Europeans will do things differently try and look at things make your own determination How do you propose the non-genotoxic carcinogen because in the U.S. non-genotoxic carcinogen um my approach on this is I'm fairly pragmatic how I approach this I mean I think you look at the same thing if you have any response information if you understand the mechanism but I'm open to sort of using the non-linear sublinear approach for non-genotoxic carcinogens it just needs to be established sufficiently if you have confidence that's how they're actually working one of the interesting things with the the class told proxonal proliferating PPR alpha this is a mechanism liver carcinogens established well unfortunately the knockout mouse still gets the mice have knockouts in this receptor still get cancer so what maybe argued is this mechanism when it's been jumping on the bandwagon may actually not be correct so you want to have at least enough assurity that you're making a decision that it's actually firmly based on scientific knowledge but I'm open I'm in the pain that if you take a center approach and use uncertainty factors appropriately it's probably adequately protected with human health but you can also make very good arguments for linear extrapolation approaches for many of these agents as well and how about very low dose radiation where we actually see an adaptive response so it actually goes below the zero point well there is a sense of getting cancer according to the models it is there is a movement that believes that some argue that very low dose may be beneficial because there's an adaptive response in vitro studies in radiation we can see this in vitro studies in humans I don't believe has the major the Bureau of Commission on this kind of concluded that from a public health point of view it's probably safer to use a linear extrapolation and to go with some of the hormesis and other approaches unless you have really convincing evidence that's the case you may have some data that shows this adaptive response also in terms of the tests they're trying to modify to to get some of the specificity issues have they tried licking in human cells more resistant to chromosome damage more so than rodent cells most of the mouse and other rodent cells tend to be highly sensitive to chromosome aberrations and chromosome damage compared to humans yeah you're right on the point one of the problems with the rodent cell lines for those of you I don't know there's a critical cancer related gene called p53 it's a retomersopressor gene usually it stops the cell cycle to allow time for DNA repair to take place and it's too much damage in use of apoptosis program cell done almost all of the rodent cell lines lack functional p53 activity and so that those cells that ordinarily would die off don't die and they go on and you pick up the cross-laborations whereas the human cell lines are almost always p53 competent to shift to the human cell lines because mainly because of p53 status so that's one of the things that's happening right now is trying to shift away from the B79 or the cho-cells to go to these other human cell lines like 90 no of TK6 in which they appear to be more resistant to certain chromosome damage that's thought to be the matter thank you and we can