 Well, thank you very much for this great invitation to be here today. I want to thank Eric and Barb and Colleen also for this wonderful invitation. I also want to thank Karen, who long ago and far away when we were both at Georgetown got me into the genetics of smoking and thinking about clinical translational issues. And I feel like this work really has the potential to affect health outcomes on the ground. It's very exciting to be working in an area where you feel like you can make a difference in people's lives, unlike some of the things that we do that have less applicability in the immediate. So challenges to successful translation. There are multiple challenges to translating exciting, novel, efficacious applications of genomics research into clinical practice. Tailored smoking cessation treatment is exciting because it would be potentially the first PGX treatment in primary care practice, a group of physicians that historically have not had much experience with genetics outside of the OBGYN context. So providers' willingness and capacity to incorporate new pharmacogenomic treatment strategies into practice is a major potential hurdle. Secondly, patients' willingness. And here there's two points of willingness. I think most of us have focused on consumers' willingness to undergo genetic testing in order to receive tailored treatment. But as you'll see in what I'll present later, a critical piece of this is patients' willingness to follow the treatment recommendations indicated by the test. If we don't have that second piece, we're dead in the water. There are also a host of coverage and financing issues. I also have a background in public health and in healthcare financing in the public sector of Massachusetts. And I'm happy to take questions on that. Right now, for example, pharmacotherapy is covered in about 66 percent, for 66 percent of folks who have private insurance, but only 16 percent of Medicaid, of this as a means tested low-income segment of the population, have access to coverage. So those issues are also important. And then lastly, there's the logistical hurdles. I think that we forget the primary care physicians, more than half of patients are seen by primary care physician practices that are either solo physicians or a partner, just two doctors. And many of them throughout the country are still the traditional model. You have the physician with perhaps his wife as the secretary, and that's it. So when we think about clinical integration, half the population is served by these very small physician practices who probably are going to be the last adopters of EHRs, right, and who don't have the infrastructure to think about delivering this kind of complicated service, then we really need to think about them also. So in the interest of time, I'm going to focus heavily on providers' willingness and capacity and patients' willingness and capacity. I also am a health disparities researcher, and I know that NIH has a commitment to ensuring that novel treatment strategies reach those most in need. As Eric has said with Mark, genomics will only achieve its full potential to improve human health when the advances in engenders become accessible to all. And we'll also face the greatest challenges dealing with underserved communities, and we don't want to leave them behind. Otherwise, the trajectory in terms of smoking-related morbidity and death will really become a much larger problem proportionally. So let's start with providers. A few years ago, this is work that I did with the T-Turk funding from Robert Johnson Foundation that partnered with NCI. We did a random sample of 2,000 physicians in the U.S. We had a great response rate, and we were trying to identify challenges anticipating. One of the things that was very forward-thinking about the T-Turk program is that they had Robert Johnson Foundation partner, and the idea then was, let's not just do the science first, but let's simultaneously try to anticipate and address issues in translation that we think we can be facing. So we did a lot of work with primary care physicians doing intensive focus group with physicians in D.C., Boston, Chicago. We did one-on-one interviews. We did cognitive testing of all new survey elements, because at this time, this is about seven years ago, folks were really not conversant in genomics, particularly primary care physicians, and we wanted to make sure that this would work. We also did a pilot under actual survey conditions. And what we found, which was similar to some other research in smaller populations of physicians that were out at the time, is that only 5 percent of PCPs felt confident in their ability to interpret genetic test results, and only 4 percent fell prepared to consult patients considering a genetic test. In terms of ever-ordered, in their practice history, ever-ordered any genetic test, we weren't surprised that it was 60 percent because of the OBGYN-related tests. And those tests need to be made available for Medicaid providers as well. But we had about 30 percent of physicians who had ever ordered a test for VRCA 1 and 2, even once, at about 16 percent for colorectal cancer and Huntington's disease as examples of other tests. Obviously, they don't have the infrastructure or the knowledge and should not be ordering tests primarily. They should be referring to specialists. And so here we saw that 62 percent of primary care physicians had ever referred a patient to genetic testing for a testing center or counseling. And 62 percent the same had ever referred a patient for a specialist for the patient's condition. Interestingly, only 16.7 percent of PCPs nationally had ever referred a patient to a clinical trial. In a more recent survey by Stanek and colleagues that was in nature last year, we can see here in terms of pharmacogenomics in particular that a very, right now we, you know, the understanding of pharmacogenomics is not the problem. 13.6 believed genetics affects drug response. And 30 percent have received PGX training in medical school or postgraduate training. What's interesting here is that you see the last three bars here. You see an interesting pattern where 26 percent anticipate ordering a test in the next six months. However, only 10 percent feel informed and only 13 percent have ordered a test in the last six months. So there's this kind of optimism about their capacity to be able to deliver pharmacogenomic treatments. So the barriers now are not so much understanding the basics of PGX medicine, but rather what tests are available, how do they procure them, when do they use them, how do they interpret the results, and how to apply results to an individual patient sitting in front of them in real time. And here EHRs, Electronic Health Records, could be a major facilitator to clinical translation of these novel treatments. Our team at Harvard was the research team for the Office of the National Coordinator for Electronic Health Records and had hit adoption in the U.S. And we had done a lot of work with physicians, the first physician survey and hospital survey about hit adoption. And now we're at the point where about 70 percent of primary care physicians are in a practice that has an electronic medical record. And this is really tremendous progress due to the focused federal support for this. And a very interesting wrinkle that we'll get to when we talk about community health centers, which is a safety net provider serving exclusively almost, minority, uninsured, medicaid, insured, and other underserved patient populations, because of the recent funding at the federal level, nearly all health centers will have an EHR within the next year or so. So that's really a great thing to take advantage of. So in terms of looking at the variation in early adopters of pharmacogenomic testing, I want to point out here that if you look at these oncologists, not surprisingly, have had much more, have had much more experience. But you can see here that family and general practitioners, primary care providers still have had very little, this is from last year, have had very little contact with pharmacogenomics medicine in their actual practice. So in terms of the work when Karen and I first started out with the work with a T-Turk, what we were doing was trying to understand how would physicians respond to a new genetically informed test to tailor smoking treatment. So we had a randomized scenario in which the new test, half the group received a survey in which the new test tailored treatment was described as a protein, something based on serum protein, and the other half the test was described as a new genetic test. And just that one difference, just the word genetic reduced PCP's willingness to offer the new test to their patients by 11 percent. That might not sound too dramatic, but when we convert these percentages into daily smokers, it's actually four million daily smokers who would not be offered the test, just because of that reticence around the word genetic. We also were interested in thinking about the genetic variants that were under study at the time. We created a clinical scenario that was very realistic, it was based on Karen's research and other research at the time about the likely test characteristics, and we gave them three additional facts about the test after getting their first response. Once we said in addition this particular, and again these were still randomized, this particular genotype or protein has been shown to predict an increased risk of becoming addicted to nicotine in the first place, other things to consider, that the prevalence of this genotype or protein is approximately 43 percent among white smokers and 53 percent among African American smokers. And the same genotype or protein is also associated with an increased risk of alcohol and cocaine as well as some other conditions. So they're getting at the pleotropy of the variants that were under study at the time. And you can see here that you're controlling for a range of physicians' characteristics, demographics, their own smoking status, what year they graduated from medical school, their practice setting, their current smoking treatment practices, and their general attitude about genetics, that each of these additional facts reduced physicians' willingness to offer the test to their patients. And again I should say this is stated willingness because we're trying to anticipate and we don't actually have an intervention to offer them. Just to further complicate things, we also found significant differences in patient and physician attitudes about willingness to incorporate genetic testing to tailor smoking treatment into practice by the self-identified race of the physician, him or herself. And here you can see that Hispanic physicians were much less likely to state willingness to try this new test. And then also there was some variation across the different facts that we provided in terms of test characteristics. Since my work is in disparities, I was also interested in looking at how adoption might be different or how the experience and trying to anticipate issues for minority serving physicians. And we define minority serving physicians. We looked at the percent of their patient populations that were from minority communities. And we ranked them nationally. And those in the top 20% were identified as minority serving in terms of the distribution. So we saw here that in terms of ever-ordering a genetic test, that those that were minority serving in the top quintile were significantly less likely to have ever ordered a BRCA 1 and 2 test, a test to assess colorectal cancer risk, and a test to assess Huntington's disease risk. Now as I said, we would expect them to be referring. But here we found some interesting results. We still are in the direction that we think, although the first two were not significant. But ever referred to a clinical trial was really interesting, particularly given the trouble recruiting minority patients to the study populations. So ever referred to a clinical trial, this is ever in their practice history. Maybe half as likely to have ever referred a patient to a clinical trial. And then also Medicaid. So Medicaid providers, Medicaid generally reimburses at rates lower than private insurers. And people who are high concentration Medicaid providers definitely are more under-resourced settings. They see a lot more complex patients, poorer patients with multiple comorbidities. So we also saw that this is another factor that we need to keep an eye on. And I want to talk a little bit about community health centers as an example of what the IOM report identifies as safety net providers. So these are providers who either by mission or who by mandate, as in the case of CHCs, they run out of the HRSA program here. And basically community health centers by law have to take all comers. They have no option to turn a patient away. So when a new novel therapy or some new capacity comes online for CHCs, they operate at practically zero margin. So they really need special targeted financing. The only way that CHCs are ahead of physicians in general is because of targeted financing for EHR capacity in CHCs. So anything special needs a special pot of money for CHCs or it will not happen. And as you can see here, fully, we're looking at close to 80% of their patient population, 85% that are either Medicaid or uninsured. Okay, this is a very depressed group of patients in terms of finances. And that in terms of reaching the minority populations, minority smokers, you can see here that basically about 60% of their patients are a minority. So if we develop some effective strategies that were targeted at community health centers, we basically have a ready infrastructure. There are about 1,000 community health centers right now. We have a ready infrastructure to disseminate to community health centers. We're capturing 10% roughly of all minority, uninsured, non-English speaking, and migrant patients in America, just through this one provider group. So we did some pilot work funded by NHGRI where we were asking them, they're mandated to provide OBN, GYN related genetic assessment. So we're interested in what other kinds of genetic tests they might be providing. And this is either directly or through referral, because again, they have very little capacity on site. And we found that there was 5.3%, that was it, that had had a capacity to get their patients, BRCA 1 and 2 testing, if it was indicated by clinical guidelines, about the same percentage for colorectal cancer, Huntington's disease even less, sickle cell more, largely because of the population that they're serving. So minority patients are likely to face greater challenges in accessing the quality genetic services. They're overrepresented among low literacy patients, so there's implications for informed consent. They're more likely to be served in settings with less skilled personnel and lower quality care. And so there's poor access to health services in general, due to health insurance issues. They're disproportionately represented among underinsured and uninsured, okay? Underinsured where you have some bare bones plan that will cover you if you get hit by a truck. But short of that, you're probably out of luck. So now let's move to the patient perspective. And the question is, if we build it, will they come? If Karen and Rachel are successful in coming up with this biomarker that's really ready for prime time, will patients be willing to undergo the assessment in order to NMR testing? And then will they be able to follow the treatment results that are indicated by the test? So with work, this is funded through the LC program. We have completed recently a national digital survey to investigate lay understandings and beliefs about the role of genetics and other factors in addiction. And we didn't want to look just at nicotine. We also wanted to see if there were cultural differences. Is this between black and white adults? Because we didn't have the funds to have adequate sample size for other racial and minority groups. So basically we wanted to look at whether alcohol and cocaine, these patterns, differed by those substances. And among smokers, we wanted to assess their willingness to undergo testing to receive tailored treatment. So as I said, 2,400 adults in the US, this is a random digital survey. We achieved a 40% response rate among those who we actually were able to get to as a live person. I mean among those in general, which includes people that we just got an answering machine. But among those who we actually made any kind of personal contact with, we had 81% participation. And among those, we had 393 smokers that were roughly half black and half white reflecting the sample. So we did a lot of work. We're trying to understand what are the consumers, what are they thinking, and how might that thinking affect their willingness to undergo NMR testing. So we did a lot of focus group work in rural Alabama and in the city of Baltimore and Boston to try and come up with language that reflected the way that the average patient that was less than a college education would be thinking. And we came out with this question that took a lot of work and a lot of field testing. And this had to be something, remember, that could be asked easily on the telephone. Please pick the choice that you think has the greatest influence on a person's ability to quit smoking. Would you say willpower, having family or friends who support the quit attempt, a person's genetic makeup, using medications or smoking cessation counseling and having God's help? And these were generated again by this extensive qualitative work. The results are very striking. As you can see here, African American respondents, 50% rated having God's help as the most important factor in a person's ability to quit. And 58% of white respondents rated willpower as the most important factor. And in multivariate analysis, controlling for age, sex, education, smoking status, self-described religiosity. So we're already controlling for that. And health status. Blacks were still more likely to rate God's help as the greatest influence with an odd ratio of 4.7 and less likely than white smokers to rate, sorry, this isn't the smokers, this is everybody, and less likely to rate medications as the greatest influence. 0.41, a very striking finding. And we found also that these views, that this cultural difference really persisted across substances. So you can see here, in terms of having God's help as the greatest influence on a person's ability to quit, it persisted across nicotine, to quit drinking, to quit using cocaine, very consistent results. And in terms of rating medications as the greatest influence on a person's ability to quit, the results for black were very less consistent, but all in the same direction and all significant. So in terms, let's turn now to those results were for the whole entire RDD survey of 1,200 blacks and 1,200 whites in the US, a nationally representative sample. Now let's turn to our 392 smokers. So you can see here, not surprisingly, that the portion of smokers that had ever used medication for smoking cessation, 32% black and 57% white. The proportion would ever use counseling, again, a racial difference. But then look at the next one. The proportion rating medications and counseling is very important. So here, you have black smokers rating medication as much more important, even though they're much less likely to ever use them. And then, of course, we have this difference in terms of having God's help. And in our regression model, with willingness to undergo genetic testing to be matched optimal smoking cessation treatment, we had an odds ratio of 8.94 in terms of those who believe that medication or counseling on the ability to quit, if they believe that that influence was very important, they were nine times as likely to say that they'd be willing to undergo genetic testing. And then look at the odds ratio in terms of the influence of having God's help, a completely opposite effect. Now, we need to address these cultural differences in consumers and smokers' views. We need to identify new strategies to engage black smokers, because what's at stake here is, particularly among black men, you see that they are much more likely to die of smoking-related illnesses. And as I said, if we don't solve these problems now, these racial disparities are likely to increase. Almost the more successful PGX smoking treatment is, the more dramatic the black-white racial disparities gap will be in smoking-related illness and death. So smokers' attitudes and beliefs, I believe, are the linchpin to realizing the promise of genetically tailored treatment for smoking. And it's these two things. I want to come back to this point. It's the potential of pharmacogenomic treatment strategies. It depends on smokers' willingness to undergo genetic testing to be matched to optimal treatment, but this other piece about using medications indicated by the test results. That's the other piece that we need to keep our eye on. And concerns about the translation have really focused more on the genetic testing and racial differences and willingness to undergo genetic testing in various contexts. But the real barrier, I believe, is in the bread and butter issue of smokers' willingness to use medications in general in order to help them quit. And so here we have pharmacogenomics research actually pointing us back to a very fundamental issue in tobacco control. And with black smokers, there's a particular conundrum. So black smokers are far more likely than white smokers to express a willingness to undergo genetic testing. And this is a nationally-represented survey with a very good response rate for a random digital. But African-Americans are less likely to rate medications as having the greatest influence. And this is substantiated by actual past use. Black smokers are less likely to have ever used medications than a previous quintetempt. So we have this disconnect. So they're willing to undergo genetic testing, even though the purpose of the test is to match them to optimal medication. So what is this having God's help thing about? What does this mean to black smokers? And how does this affect pharmacotherapy use? Might this be a new link for trying to understand low pharmacotherapy use among black smokers? 79% of African-Americans, according to a recent Pew survey, say that religion is very important in their lives compared to 56% of US adults nationally. So to try and probe this, with help from NHGRI, we've conducted some other formative qualitative research in order to direct our future research in this area. We conducted 12 African-American extra focus groups, sorry, six extra focus groups in rural Alabama and inner city Baltimore, going back to the same sites where we went originally in designing our survey. We had about a dozen African-American smokers per group. And we really zeroed in on those who have less than a college education had never used pharmacotherapy in a previous quintetempt. So if we can move this group of black smokers into a willingness to use medication to help in a quit attempt. And on a screening survey, they rated having God's help as the most important.