 Okay, well, I'll take the next 15 minutes trying to run you through an Israeli program, which actually comes from a slightly different angle. It's going to be the experience of implementing a personalized medicine program within an age-mo-like organization, a non-for-profit organization. Okay, let's go for here. Okay, so just a couple minutes on the Israeli health system. We are, it's again like what we heard before in Belgium, or we know from the U.K., and other Canada, and other, it's a health system where there's full coverage of the whole Israeli population, only that the service here is provided by four providers, say age-mo-like. They're non-for-profit, and the people in the country have the freedom to move between the providers every three or six months, and actually the providers only compete on a level of service, and you know, the length of lines, of waiting lines, and things like that, because the content of the services provided is dictated by the government, and it's fixed for all the age-mo's. Okay, so once every year there's a process that updates the services basket and says what will be included and what not, but once it's included, all four providers provide the same service on the same items, and everything is free of charge, or with a very tiny copay. So we have four services, and I'm going to talk today about the program of Clalit, which is the largest one covering about close to 60% of the population, and with about 1,000 primary care clinics, about 3,200 primary care physicians, seven general hospitals that belong to Clalit, but the insurers of Clalit can go into any other hospital in the country. So these are just hospitals that specifically belong to the organization. In a sense, at least for the Americans, it's a Kaiser Permanente type of organization, size-wise, but it's not for profit. Okay, so and the interesting thing is that we all pay our health insurance to the government. We do not pay to the provider. There's no financial contract between us and the provider, and the government disperses the funding between the different providers, according to various parameters. Okay. Now, when the organization that Clalit, I mean that organization, HMO, whatever we call it, decided to move on towards the subject of genomic medicine, personalized medicine, whatever name we give it, they approached us. We were then active as the National Cancer Control Center of the organization of the HMO, and we had a large team of disciplinary team of a variety of professions. We had quite a large lab. We were involved in a lot of research with tens of thousands of participants. We had a biobank of more than 400,000 samples, of a variety of samples, and we had experience with a lot of genetic testing, research based genetic testing, so it made sense to come to us as an expert group and suggest that we try and develop a program for the HMO on these grounds. We also have the clinical counseling service, and we have very large series of carriers. Some of them are a reflection of the fact that the Israeli, especially the Ashkenazi population, is a founder population, so you get a lot of mutations and you really reach very large numbers. So as a whole, we were a facility that was right in place with a lot of hands-on experience in the genetic aspects of research and large-scale research, so we were approached. I just want to show that within our study, that's one of our studies, our corrective cancer study, and that's the first 5,000 participants in the study, et cetera. We were in a stage when we started designing things. We were at a stage of really separating populations very clearly and very dramatically within Israel. You see the Ashkenazi population here in black. You see the Sephardi population in red. These are the Jewish populations. You see the Arab population in blue. In light blue is an Arab sect that are the Druze. It's interesting. We got an influx of non-Arab Christians who are non-Arab. We have Arab Christians and non-Arab Christians that came in, especially from the Russian republics in recent years, and you see they're here in green in between. So it's very interesting to see the separation on a paper that showed similar data come out from Duke. Who's it? David Ginsberg, right? Goldstein. Sorry, Jeff. Sorry. I'm allowed to do that mistake. That's a paper that he did, and he again could separate very clearly the Jewish population here from the Druze population, Palestinian population, and Bedouin population, and here's like the U.S. population. Population is separable if we were talking about population certification and need to know the genetics of certain populations where you operate your medical system. This is clearly a case where it shows that you need to have a lot of information about that. Our colleague from Kuwait talked before about Arab populations. We actually have tons of data on the genetics of various Arab populations, just because we're doing all most of our studies in northern Israel, and that's where most of the Israeli Arab population resides. So a lot of genetics, a lot of non-genetics data on that, that could be of help to groups. Family history was mentioned before. We put a lot of emphasis in trying to introduce family history into our medical records. Every activity that we do on a national level would try to collect data on that. So for example, here you see the results, the reporting of family history of breast cancer among Jews, Christian Arabs, Muslim Arabs, Druze in women who came in to have a mammogram. So these are the reports of a million women in a tiny country. We only have four million women in the whole country. So a million women reported on family history, the same thing with every screening activity. Everything we do, we make sure to incorporate family history data into our databases. Not only that, we push it into the doctor's computers, and all the doctors are computerized on the same system. So we push it in, and the next step that we took only recently was we identified first-degree family relatives through the population registered and pushed the day information of family history into their medical records without disclosing who the family case is, because then you run into all this, all the privacy issues. But we pushed, we really are trying to push this data wherever we can, and we have registers on many other diseases. We also run large registers of cardiovascular diseases and others, so we can really push in a lot of information this way. In any case, so we set up to build a program, an organized program, really a committed organized program, and there were issues of policy, of education, and provision of service. And in policy, first thing was to get the commitment of the management. That's always the toughest part. You're always talking in words of cost-effectiveness, and when you compete only on quality of care, because there's no, you know, there's no other grounds for competition in our system, you want to be innovative, you want to show up as a modern service. You want to, I mean, all these words kind of excite the CEO, okay? So this is what you come and sell. You try to say that you'll provide better health and you'll be cost-effective. The contents were defined as dealing with molecular medicine to, for risk reduction, for disease detection, for efficient treatment, and for prognosis determination. Finances are always a big issue. These are, you know, these technologies are very costly, and also the fact that they are outdated before they actually hit your bench is also a major issue. We had to struggle with issues of homebrew testing versus commercial kids. We decided to go for homebrew testing wherever we could. For example, just because, you know, a founder mutation panel for Ashkenazi Jews in the U.S. would be five to $700 in our lab. It's about $40, and I've done about 50,000 of them. So, I mean, so wherever we can, but when you do homebrew, you have to invest in R&D, because otherwise you cannot rely on the development of a kid of others. A lot of legal and ethical issues. Yes, of course, entry into databases and everything is computerized in our system. It's a big issue. Education. We invested a lot in education of the medical teams. We are aware of the fact that they are scared, terribly scared by this whole field, and many times confused. This is in an illegible language for you, but that's like our KRAS brochure, which tells the doctors, and it's actually aimed at the primary care level to say, you know, what is the disease we're talking about? What drugs will be influenced by the gene? What tests are we going to do? What's the gene test interaction? What's the population variability in the gene? What is the clinical evidence? Who should be tested? And what should you do with the answer of the test? Okay, so this is kind of a, and we have it for numerous, many, many different tests. Provision of service, we established an expert team of geneticists and pharmacogeneticists and pharmacologists and a variety of physicians who are at the service of all the doctors of the organization. They want to ask somebody, they have a phone call, they can ask about whether to order the test, what to do with the test result, what the test is about, and everything they want. So there's a support group and there's a centralized lab as much as we can. We're not limiting it to the centralized lab, but the centralized lab takes most of the toll and does most of the tests. I'll skip these long lists of genes that we are able to do. This is one panel that we gave in and we're starting to operate now the TrueSec on our MySec. And just as an example, doing EGFR testing in lung cancer tissue, we could report close to 2,000 tests within about a year of activity, because we, as a centralized lab, get the samples from all the hospitals. In Israel, we can say how many were positive, we can say what is the distribution of the mutations just by being centralized and being with good databases. We just have all the data right there. We can look similarly at ALC, which follows a negative EGFR, and again, see the proportion in our population that are positive. We can see if we want to look at the whole picture of lung cancer, that 33% will be EGFR mutated, 25-care rats, etc. etc. And again, this is critical data for our decision making with regards to treatment. We can look at multiple mutations in a tumor to see what's mutually exclusive and what's not. We can look at survival patterns here, the people with EGFR mutations and without EGFR mutations. So the EGFR mutate are doing better, but are they doing better because of treatment, or are they doing better just because the mutation gives them an advantage? Well, no, if you have the mutation, but you're not treated in blue, you're like without the mutation. So the advantage is really only if you are mutated and you receive the TKIs to treat the mutation. Similarly, with ALC, we could show the differences. And here, if you have ALC mutation, you're really faring badly. We were even able, because we have all the pharmacy data, everything is computerized, every single element in the system. So we could actually look, if you had an ALC mutation, but you wrongly received TKIs, what would happen? And also, if you did not have the ALC mutation and you received gazatinib, what would happen? And we can see that it's actually of no value. So it's true that gazatinib should be reserved only for out mutated and that TKIs should be reserved only for EGFR mutated. You can show it with rather big numbers of information. So I mean, that's the conclude. I mean, it's a centralized program and we think that the centralized program is a good approach. Whoops. Already? I think I still have a second. Okay. Can you bring it back? Okay. Anyway, we think that centralized program has its advantages. And by having big volumes, you can really make sure that you have very good quality control for your tests. You can make sure that your turnaround of the response of results will be fast and quick. And you have tons of data that actually come to a central database where you can control and see the stuff. I wanted to acknowledge the team. You are not going to see the picture of all the team members. But in any case, this is just an example of how to try in a very organized manner, take something from A to B and under the responsibility of the organization. So this is really covering a whole population. Thank you very much. Mary Reling St. Jude. So you showed the incredible ancestral diversity that you have in your population related to race and ethnic groups. Do you have any examples of a variant that's actionable in one group and not actionable in another, either somatically acquired or inherited? No. So I have all these specific... I'm still here. No, I'm okay. Okay. Okay. Okay. Dr. Rhodes. No, I mean, because you're asking whether it's actionable or not. And you're referring to the result of the action to show that something worked in this group and did not, although the two groups had the same mutation. Right. Right. Or non-existent. So for example... That's a different situation. Yeah. No, but that's very, very surprising to us that we have, you know, within the Jewish population, you know, we could find the 185 DELEG, whatever, BRCA, in Ashkenazi Jews, also in Iraqi Jews, which are completely a different area, but in none of the North African Jews. So... Right. Right. Not before we tested. So I have a very specific question about a germline variant, and that's the D36Y variant in V-Core C1. So one question was in V-Core C1, so the warfarin gene. So one of the questions I actually had was whether you have a focus only on the tumor genome or whether there's a germline as well. No, no. We're definitely doing germline testing too. I've done thousands of V-Core C1 as part of you know, I mean, research-based, because again, we are constantly involved in R&D. We have to, because we have to, if this comes up to be an important gene, we have to be ready to supply the service nationwide. There are data, there is really data that suggests that there's a 5%, there's a 5% or a rare non-synonymous variant in V-Core C1 that confers relative or absolute warfarin resistance. So these patients take, instead of taking 5 milligrams a day, take 15 milligrams a day. And I'm just wondering whether you have encountered that one and whether that is an implementable variant. We probably would not encounter it. I mean, specifically in our lab, because otherwise I would have known about it. But I do know that we tested for the whole, you know, cardiovascular panel, including, you know... So the cardiovascular panels have to include this one. I would have thought they had to include this one, because this is an Ashkenazi variant. Wait, okay, sorry. No, no, no, I mean, I just, you know, you're the one who delivers the service. It's not me. So I was hoping that that was going to be an answer to Mary's question, a variant that is ancestry-specific and actionable, but maybe I don't know enough about it. That's not her question. Her question was if I can find it in one ethnic group, in two ethnic groups, but it's actionable in one and not in the other. That is a tricky one, because I mean, finding it in... You're right, that was my question. Dan was wrong. I wanted to ask you about the data that you showed on Crisatinib not being effective in EGRF negatives, which is what one would expect. You would think that those would be unimportant data, but actually in talking with our Centers for Medicare and Medicaid Services, one of the things that they find is that physicians will do these tests and will give the drug anyway, even if it's negative, you may have the same thing happen in your place, because it's the best choice. You see, that's why it happened. I mean, how would it... How would you otherwise have it? It should be sequential. Right. You have the mutation, you're eligible for the drug. Correct. Many patients elect to use the drug, try to do whatever, and that's how we can show. I mean, otherwise, it would not have had anybody who would knock and deal with it. So what I wanted to say was, A, we need to get physicians to actually follow up and act based on the tests that they do, but that's one issue. Another is that it's very helpful to have that kind of evidence, because actually some of the ways that our Centers for Medicare and Medicaid Services CMS makes decisions on payment is not necessarily on having evidence of benefit, but just having some influence on medical care. So it doesn't necessarily have to have an evidence. It's being written up. Great. Great. Okay. Yeah. So just letting others know that those kinds of data are very helpful, and they said, if you have any evidence that things don't work, we want to know that. So that would be great. Maybe I'll ask, or make the last comment. What I heard you describe was an incredibly powerful system for evidence generation. You describe it as a centralized means of providing quality service provision, but you've got the genomic data, the pharmacy data, the outcome data, and because you're an HMO, you have obviously the cost data. So to me, this would seem like the perfect system to do what is thematic in this meeting to really begin to generate evidence at scale. Do you want to comment on that, or do you disagree? Well, I agree for a statement on a perfect system, but... No, no, no. No, it is very powerful, I just make it very powerful. And actually, a lot of different agencies are making use of our data. Like now we're happy to have the FDA access question about certain drugs, not the genomics of drugs, but like, you know, side effects and stuff like that. Whenever the issues come up, we can immediately, you know, query our databases. And this is a massive database. I mean, there's billions of points, and it's actually quite hard to handle. But yes, I mean, it's special. The only thing that worries me a little bit when we do our work is to what extent is it generalizable? That's the only issue. I mean, if we are really genetically different, to what extent is that the case and because it's really, I mean, naturally I don't have the time to show a million examples. But you know what I mean? Like, we have half the lung cancer rate per same smoking than non-juice. Now it's true in Israel and it's true in New York. If you compare Jews and non-juice in New York and studies with 10 to 15,000 participants, big stuff, and you know, with all the effort to find the genetics of that, including whole genomics, you know, not all, I mean, yeah, whole genomics and everything, we can't put a finger on what's happening. Okay, so there's a lot of issues. And the question is really to what extent is it generalizable, you know, because there are many issues, you know, I mean, like the BRCA gene, very common in our system. Does anybody know why, if the penicillin is 50%, why do two women with the same mutation, one gets attention, one does not? I mean, 100 women, 50 get it, they do not. Why? I mean, the simple consortium is trying to answer this question for 10 years. No answer. So your mention of the BRCA gene makes me channel Heidi again and just raise the issue of it's been very difficult in the U.S. to get data on sequence variation in BRCA1 because all the data are locked up in a proprietary database. But there are efforts to try to either, you know, liberalize or liberate some of those data or report them individually. But are you guys working on BRCA1 sequence variants, depositing them somewhere, making them available with phenotype information, because it would be critically important, I think. Frankly, the answer is no, and it's no because we are so spoiled. We have the founder mutations. We don't need to look at variants. We just go directly for the founders. So we don't sequence. We're going to start sequencing now because we have enough reason to do it. But actually, although we have found the variants.