Forecasting Chemical Toxicity with U.S. EPA’s Dr. Robert Kavlock

October 10, 2011

JOHN SHEGERIAN: Welcome back to Green is Good, and we’re so honored today to have on the line with us Dr. Robert Kavlock from the U.S. EPA. It’s the U.S. EPA’s first time today on Green is Good. Welcome to Green is Good, Dr. Kavlock. DR. ROBERT KAVLOCK: Thanks. It’s good to be here. JOHN SHEGERIAN: Hey, you know, Doc, you have such a fascinating journey, and you’re so well accomplished. Instead of me sitting here for the next 22 minutes reading your bio, why don’t you share with our listeners, before we get going to talk about your great program, ToxCast, over at the EPA, share your journey and how you even got involved with environmental issues and came to do this kind of great work at the U.S. EPA. DR. ROBERT KAVLOCK: Alright. I went to college to become a marine biologist because I thought Jacques Cousteau was an idol to follow. I went to the University of Miami, and I studied marine biology and ecology there for several years. Then one day, I had an opportunity to get a work study job with a government lab, which I didn’t know much about at the time, but it would actually wind up to be one of the labs of the Environmental Protection Agency. I liked the work I was doing so much as work-study student, that I actually switched careers. When I went to graduate school, I went to graduate school in a field called teratology, which is the study of birth defects. I was in graduate school studying the effects of air pollution on fetal development, and from there the saga went that the lab closed and it moved to North Carolina. When I graduated, they offered me a position as a postdoc to come up and work in their labs in Research Triangle Park, and I thought, “That sounds good. I’ll go up there for a few years. I’ll get some more training, and I’ll go somewhere else.” 33 years later, I’m still here. I’ve had a great career at EPA. I’ve done a lot of work in reproductive toxicology, that’s the effects of chemicals on embryonic development and male and female reproduction. Then a few years ago, I had the opportunity to get involved in a program that EPA was starting called Computational Toxicology, where we were going to try to bring in some of the new tools in molecular biology and in computer science, and try to apply them in new ways to look at what kind of effects chemicals might have on biological systems. Today I’m the head of the National Center for Computational Toxicology, and our flagship program is this ToxCast that we’ve been working on for the last five years. JOHN SHEGERIAN: Dr. Kavlock, you sit actually in North Carolina still, or you sit in Washington, D.C.? DR. ROBERT KAVLOCK: My official job is in North Carolina, although I get up to Washington about once a week now for talking or partners in the program offices who are going to use the kind of information that’s being generated in the research part of EPA for the regulatory purposes. JOHN SHEGERIAN: This is going to be a great show today because this is a fascinating topic. We’ve had other guests on before touching on this issue of green chemistry, and we’d like you to walk our listeners through and educate them to these critical issues that are important to all of us. Talk a little bit about definitions. We’re on your great website. For our listeners out there, if you’ve got your iPad or your laptop or your desktop in front of you, go to That’s what we’re on right now, Mike and I, and it’s a fascinating site. Talk a little bit about what is ToxCast? What does that mean definitionally, and what are some of your major goals? DR. ROBERT KAVLOCK: The word was actually coined to denote two things. One is we’re going to try to forecast toxicity, much like you try to forecast the weather by having all kind of sensors around the globe and getting that information and putting it in the computer models and making that forecast. The other part is we’re going to cast a real broad net as we look for those sensors out in the world that’s going to give us information about the chemical that we’re going to put together with the computer model. The reason we got started on this is if you go around to the program offices in EPA, and I mean the office in charge of clean air, clean water, pesticides, industrial chemicals, super fun sites, and you ask the question, “What chemicals are you concerned about from a regulatory viewpoint? Which ones really keep you up at night, worrying about whether they might cause damage to public health or the environment?” We actually created a spreadsheet that has about 9,900 chemicals on it. We got rid of all the duplicates and are left with about 9,000 chemicals that the program offices were concerned with. Then we went to the public information on the internet, and we created a site called Actor — you can link to that from our website — where we went out to what is now about 1,000 different sources of information, and we asked the question, “What do we know about those 9,900 chemicals?” For a large majority of those chemicals, there’s very little information in the public domain about whether they might cause harm to human health or the environment. So, there’s a huge information gap in terms of what would we like to know about chemicals versus what we do know about chemicals. When we started this ToxCast program, we had basically a blank wall, and we said, “What could we really do to start to change that situation? How could we get information on hundreds or thousands of chemicals, and try to help the people in Washington who have to enforce the laws to figure out which chemicals to really focus on?” So, that was the start of the program. JOHN SHEGERIAN: That was back when? Is that 2007? DR. ROBERT KAVLOCK: The CompTox center was formed in 2005. It was actually given a blank slate to work on, but was also given a five-year period to make a difference. So, could we do something completely different in five years? And see if we could find solutions. We spent about a year just thinking about what was available, what could we really do that was different. We really launched the program officially in 2007. JOHN SHEGERIAN: Now I understand the goals and I understand some of the history here and why you started. In laymen’s terms, because this is highly complicated but also unbelievably fascinating, is one of the major goals to then create visibility for the consumers at large, eventually, to understand what they should surround themselves with and what they should avoid? DR. ROBERT KAVLOCK: Ultimately, it could be used for that, but for now, the real sense is can we give some priority to this huge chemical information gap? Because if you wanted to do it the way we traditionally do it, you would use rodents. It’s almost like recipes you have to run if you want to find out does it cause cancer, does it cause birth defects, does it cause reproductive problems, does it cost nervous system problems, does it affect immune system? If you really wanted to do it the way traditional toxicology has been done, it would take you probably five years for chemicals, and it would take you $10-$15 million in chemicals to do that. And it would take EPA about $1 million to interpret that and figure out the information. If you do the 9,900 times about $10 million, you get to a real big number. You’re just not going to be able to solve the problem that way, so you need a whole new way of doing it. How the pharmaceutical industry thinks about making a new drug, they develop a target that’s related to the disease, and then they take lots and lots of chemicals and they develop a simple assay that tells you whether the chemicals interact with the process that causes that disease. If they wanted to replace a hormone that your body lost, they would develop an assay for that hormone, and they would test a lot of chemicals, and they would find something that worked like the hormone would do. Then they could develop that as a drug. MIKE BRADY: So, basically, is this done with computer models as opposed to live animal testing? They run a matrix, then, with the assay? DR. ROBERT KAVLOCK: What we use are proteins from animals or cells from animals or even cells that haven’t been derived directly from animals, but what are called cell lines, so we actually don’t use animals themselves, but we use products from the animals to do this. We use robots in a laboratory to actually test thousands of chemicals against the proteins or cells that we know might be targets. Firstly, it’s done in a laboratory with robots, and then we take that information, and then we feed it in the computer model, and we try to make sense of what those tests are telling us. MIKE BRADY: Does that shorten the length of time to get the results? Because we’re talking about time and money involved in testing all of these different chemicals. What would you say roughly? Is there an average? DR. ROBERT KAVLOCK: We can’t completely replace the way things are done traditionally right now, but in the space of literally a few weeks or a month, we can take 1,000 chemicals and we can test it against 600-800 different types of these potential disease targets. We need about $30,000 to do that. JOHN SHEGERIAN: So, really, what you’re saying is from the big number that you were explaining earlier to us, how it would take based on the traditional paradigm, you’ve now created an exponential methodology to do this testing, get accurate results, and be able to deliver information to your organization, and hopefully, then, down the road, be disseminated to others for use so we can better understand what’s in the chemicals. DR. ROBERT KAVLOCK: It’s cheaper, better, faster. We know we can do it cheaper and faster now, but the question is, can we do it better? That’s what the research program is all about. Can these technologies really replace those animals or tell us how to use those animals more effectively? JOHN SHEGERIAN: Do you have a definitive answer yet to that? DR. ROBERT KAVLOCK: We’re gaining confidence every day that we’re on the right path, but because the field has basically done toxicology as a field, it’s pretty much done what it’s done for 40 years the same way. There have been some changes, but it’s a fairly conservative approach. It takes a lot of evidence to convince people that a different way is a better way, and that’s what we’re trying to do now, from doing our research to making the publications, through giving talks at scientific societies and even doing interviews like this. JOHN SHEGERIAN: Got it. How long until you feel that you and your colleagues can prove the concept and actually become the new state-of-the-art methodology? DR. ROBERT KAVLOCK: I think in the next year or two, we’ll see the Environmental Protection Agency use this technology for the first time in the business side of the organization. They’ll be using it to help prioritize chemicals that might be considered to interact with the endocrine system. We’re very concerned that there are chemicals out in the environment that could potentially act like hormones, and they could cause birth defects, they could cause nervous system disorders, they could cause some cancers. And so we have a screening program to try to find those chemicals. What ToxCast will be used for, initially, is to help shed light on which of the thousands of chemicals that we might look at in that program are we going to look at first? JOHN SHEGERIAN: Let’s talk a little bit about that, though. The program’s evolving. What was the term life of the program, originally when you set out? Was it five years or is it going to be extended now? DR. ROBERT KAVLOCK: We had five years. We were given a blank board, and we were also given a board of scientific counselors who came and looked at us about every 15 months to say are we doing something that’s on the right track? So, there was an external body who was always looking over our shoulder and giving feedback to us. That five-year period ended last year in February of 2010, and we’re still around, so the Agency likes us. I think that we continue to get good support from the Agency and from Congress and from a number of what we call external stakeholders, like the American Chemistry Council or the Humane Society or a number of bodies who are interested in doing a better job of chemical safety assessments. JOHN SHEGERIAN: So, you’ve created, really then, a consensus of many different groups to continue the support because they really do believe your work is very credible and creating some great new paradigms that could benefit everybody. DR. ROBERT KAVLOCK: It really is a transformation how things were done or have been done, so that’s why I think we had a lot of people who were interested. I’m optimistic this new vision might actually work. JOHN SHEGERIAN: If you’ve just joined us right now, we’re honored to have Dr. Robert Kavlock on from the U.S. EPA, the first time the U.S. EPA’s ever been on with us. We’re talking about ToxCast, this fascinating new program that they’ve evolved over the last almost six years now. Go to the website. If you’re in front of a computer or a laptop or an iPad, go and check what Dr. Kavlock and his colleagues are doing at Believe me, you will not be sorry. It’s fascinating information, stuff that you should know that’s going on that could benefit not only us as Americans, but the world at large. Doc, talk a little bit about the process, how ToxCast chemical screening works, and how is that data then available to others? When does it get shared, and when does it actually trickle down to a general consumer like Mike Brady and myself? DR. ROBERT KAVLOCK: We generate the data. We have contracts with a number of companies around the United States. We have about 10 different contracts that give us different kinds of technologies for looking at how proteins or cells are affected by chemicals. We chose those technologies because we think they’re shedding light on important disease processes. For example, because endocrine destruction is an important topic at the agency, we have a number of assays that look to see whether a chemical looks like it might be an estrogen, which would be the female sex hormone, or androgen, which is the male sex hormone, or interact with the thyroid, which would be another way that you can cause endocrine disruption. For each of those assays, we try to use multiple different technologies to tell us is there something to be concerned about there. For each one of the technologies, we’ll write up a description of the results of that experiment. We’ll publish it in a scientific journal, and then we post that data on our website, so all the data is available to you, to anyone else in the world, who can then download it and do their own analysis of it in a way that would give independent confirmation that what we’re doing is appropriate and good science. JOHN SHEGERIAN: And that’s on your ToxCast database? DR. ROBERT KAVLOCK: That’s right. If you go to the ToxCast DBO tab, it will actually take you into a little browser on your Internet, and you can put in the chemical, and you can see whether we tested it and whether we got any results. If you’re there right now, you can try. Bisphenol A is a plasticizer that has gotten a lot of attention in the press because it has some potential estrogenic activity. If you type in Bisphenol A, you’ll bring up our data on that chemical, and you’ll see that we would have found out that it does have estrogenic activities, and we would have found that out fairly quickly. JOHN SHEGERIAN: Doc, has any behavior changed since you started this process, and some of the discoveries you’ve made along the way or confirmations you’ve made along the way? Have any chemical interactions with the public at large, has any industry stopped using chemicals because of your research yet? Or have any chemicals been removed from any of the approved compounds that were historically out there because of some of this research? Or is that yet to come? DR. ROBERT KAVLOCK: The answer to your second question is no. The EPA is still evaluating the quality of the research that we’re doing before it would use it for making a public health decision. Because this is Green is Good, just an example of what we’re doing with green chemistry. JOHN SHEGERIAN: Sure. And Doc, do me a favor. Our listeners are fascinated by this new term of art, this new vernacular, green chemistry. Again, we’ve had a couple shows on this before, but please explain to our listeners what that really means to you and to all of us. What does the term green chemistry mean? DR. ROBERT KAVLOCK: Well, the term has been around probably for 10 or 15 years. In fact, my boss is the father of green chemistry, Paul Anastos, so if you really want to get it from the horse’s mouth, I’d go to Paul and get him on your program. But basically it’s about when you look at a chemical, you think about its entire life cycle. Think about the materials that are going into make that chemical. Are they renewable? Are you using solvents that are toxic? Are you using a lot of energy in order to do that? There may be other ways to do that synthesis that reduces the waste that you’re making or you’re not using oil, which we’re going to wind up some day being out of, but you can use some materials which are renewable. Then when you get into the marketplace, you want to make sure that the chemical isn’t going to be very persistent in the environment. It’s not going to transport and wind up in the polar bears. We don’t want it in polar bears or even humans. We want to watch how it goes through the environment. At the end of its life cycle, we want to figure out what’s the best way to dispose of it so it doesn’t become a risk later, like we may find that ECDs are now. So, it’s really taking a holistic look at the making of a chemical, the synthesis of it, the transport of it, the use of it, and then finally the disposal of it. JOHN SHEGERIAN: So, the cradle to grave approach of chemicals is called green chemistry. DR. ROBERT KAVLOCK: In a nutshell, yes. So, what we’re trying to do, there have been some plastics which have been implicated in causing certain human diseases. These include Bisphenol A and I’m sure you’ve probably heard of some of the phthalates, which are basically in a lot of products as softening agents for the plastics. There have been a number of alternatives to these plasticizers that have been proposed, and they’ve been called greener because they don’t have whatever the emotive action or the toxicity that the initial phthalate or the Bisphenol A has. The question is, when you look holistically at those replacements, are they really greener? So, what we’re able to do in ToxCast is take some of those chemicals and actually compare them across hundreds of different assays to see they differ here, but is the one that’s being posed as an alternative, does it pop up somewhere else getting a red flag, or is it really greener across the board? That’s one of the things we’re working on now, is to take a number of these categories of green chemicals and actually try to see if they really get the green stamp when we look at it with our glasses on. JOHN SHEGERIAN: Got it. Do you have a re-up on another five years? Or now does the program have blue sky in front of it? Is it limitless in terms of how much can be done in, as you say, predicting toxicity faster and better? Give us a little visibility, Doc, on the next couple years. DR. ROBERT KAVLOCK: I mentioned before we’re still in business. We haven’t been made permanent, but we haven’t been sent away yet. We’re actually a small group of researchers. We’re about 24 who work here along with a number of post-docs that we have. It’s a pretty small group, and it’s a little bit like a startup company is out in the public sector. It’s a very energetic group. It’s a very dynamic group. It’s a very interactive group, and they’re all working really hard to carry their bucket of water to make sure this whole process works. For the next year or two, I think we’ll be coming up with some very critical experiments to see whether our first prediction models, and we’ve recently in the last month, published what we call our first generation model that predicts certain health outcomes. We have one on reproductive function, we have one on developmental toxicity, and we have one on how blood vessels form. So, those are our first three models. As we go forward in the program, the new chemicals that we’re testing, we’re making predictions before we test them, whether they’re going to have certain effects. This will be a real stringent proof that, for chemicals we haven’t looked at before, our models can still make a prediction. JOHN SHEGERIAN: Doc, what does the Tox21 partnership equal? DR. ROBERT KAVLOCK: Tox21 is a consortium of actually four government agencies that are working together on the same kind of technology. It has the Environmental Protection Agency, it has the National Toxicology Program, which is part of the National Institute of Environmental Health Sciences, and they live right across the lake from me. Our other partner in NIH is the NIH Chemical Genomic Center, up in Rockville. What they bring to us are the robots who actually do a lot of the hard work on this. Then our last partner that we were able to bring in last year to the consortium is the Food and Drug Administration. So, we’re all trying to, in this era of difficult resources, make sure that as we do our research using these new technologies, that we’re aware of what each group is doing, that we bring complementary expertise together, and that we try to synergize, rather than compete, for attention. And it really is a wonderful example of how the government can cooperate together. JOHN SHEGERIAN: That is wonderful. Dr. Robert Kavlock, we’re out of time today, but we’ve enjoyed this time with you. We’re so thankful that you took this time out of your busy schedule to share this amazing and literally visionary program, this ToxCast program. For our listeners out there that want to learn more about this important and critical work that Dr. Robert Kavlock and his colleagues are working on, please go to Dr. Robert Kavlock, you are a proven visionary and sustainability leader, and truly living proof that green is good.