 Thank you very much for the opportunity to be involved and to speak here today. And I welcome everyone myself. I have a bit of time, so I'll just tell a little story. But I remember I think it was about 1998. We had a meeting through ELSI in April in Washington to discuss severe skin reactions. And there were people there from NIH, FDA, pharmaceutical industry, academia. It was not a large group, but a bunch of them. There were eight inches of snow that day. I remember that. And it was beautiful and quiet and scary as anything. So it was a memorable day. But after everybody went around the table explaining how important it was to be able to better understand, prevent, and manage severe skin reactions, I said, and I was a young man. So I said, maybe naively, but I don't think so. I have one question. Who's going to pay for it? And the FDA said, well, we can't pay for it. I said, well, can you leverage industry? Well, I don't think so. Industry said, well, it's not on a critical path, so we're not going to pay for it. The NIH said, well, it could be translational, but I don't think we're going to pay for it. So it basically said, why don't we just go home? Because as much as we want to talk about this, everybody thinks it's someone else's problem. And I don't know, but in kindergarten or grade one, or first grade, as you might call it, there was a story about the chicken who tried to harvest plant the seeds, harvest the wheat, make the bread. Nobody was willing to help until the bread was made, then everybody wanted to eat some. That story always stuck with me. I was just surprised to come to an international meeting at that time and see it still held true. So let's see. Maybe we're here today, and it's been long enough, and the weather will cooperate, but that's always dicey. So first of all, everything that we sort of know about this, and I'm going to assume people know nothing about it, because I think that's fair, is as with all drug reactions, the names are terrible. Even if they sound good, they're probably wrong. So let's start with this. Stevens Johnson, toxic epidermal necrolysis, sometimes called TENS, and plastic surgeons who look after these patients in burn unit call it TENS. And that stands for toxic epidermal necrolysis syndrome, which is not a bad name at all, actually. But that's not what dermatologists usually call it. And the title here is probably longer than my talk, but I'm just going to cover a lot of things fairly superficially, and you can make skin remarks about that, but superficially, because I think other people will get into the details, and you've already received some reading material. So in terms of conflicts, I've worked with various companies over the years who had trouble with drugs. I don't think there's anything here that is going to be screamingly a conflict, and if there is, I'll be happy to point it out, but I think we'll be covering it at too great a distance, and I can't imagine what there would be. And therapeutic comments for Stevens Johnson TEN, which is part of my remit, there are no approved drugs for treating drug reactions, really. There might be antidotes for overdoses, but in terms of treating hypersensitivity reactions, you sort of imagine what company is going to say, we've got this new drug in the pipeline, and it's there to treat drug reactions. I don't know. It doesn't seem like it's going to do really well with the shareholders, but. So one of the terms that we use is scar for severe cutaneous adverse reactions. And there are ones called dress, which is a syndrome that involves internal organs, and the skin, and fever, and people are sick. There's AJEP, where people get blisters with pustules, coming out of the skin, red and scaly, again, very sick, and with fever. Fever tends to be a hallmark of the systemic reactions. A lot of people just get a rash with a drug, but sometimes a rash is just a rash. And it's not a big deal, and it goes away. Doesn't necessarily mean, as I had taught years ago, that it might turn into something scary later. It doesn't. But fever is a scary finding. And Stevens Johnson or TENS, we do see fever with that. These people are sick. And they have internal organ involvement sometimes, and they have a very high death rate. Maybe 30%, 40%, even with the best management. So you've got a 30% death rate from this drug reaction. You've got a 100% horrible suffering rate from this reaction. And the reason you would even allow it to happen is that the drugs we have there we tend to need. If it shows up, and I've seen unbelievably amazing medications disappear, just as they were about to come to market because there were too many of these. And these were drugs that would have changed, literally would have changed the health landscape. And nothing has come to replace them. And yet, if we could have screened those drugs properly, they could be on the market today. And that would have saved a lot of people, a lot of misery and lives. So Stevens Johnson. It's a favorite term we'll often hear people call us from the emergency department at dermatology to say, we have a patient with Stevens-Johnson syndrome. And usually that's their way of getting us to the emergency department. It's usually not true. But anything is Stevens-Johnson syndrome to them. And it's so easy to make that mistake if you don't see it. These are sick people. And I usually just ask, how did the person get to the emergency department? They say, I don't know. And they come back, they say, they took the bus. I say, well then they don't have Stevens-Johnson syndrome. So that's my thing. So what about Stevens and Johnson? In 1992, Stevens and Johnson described what was sort of a variation of what had been called erythema multiforme before in the late 1800s. And it was a bit different. They just had this mucus cytos with mouth sores and spots on their skin. Two kids, black and white pictures in the journal. It's not what we're talking about. And Toxin kept a dermal necrolysis in a wonderful paper by Albert Lyles, a Scottish dermatologist who really described a whole bunch of diseases that had this feature pathologically and clinically. But they weren't all what we call toxic epidermal necrolysis now. In fact, much of the literature on toxic epidermal necrolysis includes staff-scalded skin syndrome and probably other diseases, too. And it's really been a bit of a mess. And yet, in many parts of the world, this is called Lyles syndrome. So Stevens-Johnson syndrome and Lyles syndrome, they're not what we're talking about. So again, we have bad names. We need to probably rename these, which is another good reason for a meeting. But these are bad names. So what do we have? We have a spectrum of a lot of different things going on. In the upper left, we have erythema multiformi. Now, for non-dermatologists and physicians, erythema means red, multiformi means it could look like anything. So basically anything could be erythema multiformi. But that's not true. Erythema multiformi describes a fairly benign disease that's usually post-viral. And people have these three-ring targets. Sadly, Target is leaving Canada, but I like seeing their bags because they had those three-ring targets on their bags, and it always warned my heart to see that. But the problem also is not just in the naming here, but the pathology, the lymphocytic infiltrate that you get, the interface dermatitis, a pathologist will call that erythema multiformi. But it doesn't mean the disease the person has is erythema multiformi. It just means that they have that. We see that in that dress disease. We see that with Stevens-Johnson T.E.M. or TENS when we look at the edge of lesions. So that's a bit confusing. When the SCAR group got together in Europe, over 400 million population were surveyed for the severe reaction that really fits this. And this group is Stevens-Johnson T.E.M., and Stevens-Johnson here was the mild end of a very severe disease. It was not trivial. And what you see is that patient there has this hemorrhagic mucocytus in their mouth with crusting and blood in there. It's in their nose, in their eyes. It may well be on their genitals. They don't have a three-ring target. They have a little too, too huge target. It's a little bit dark red in the middle, lighter red on the edge. We call them atypical targets. The naming of this was actually controversial, but it had to go through many languages and come out making sense. So atypical targets was the bottom line. And then it starts to coalesce and blister, and these blisters get worse and worse. Now, depending when you see that patient, they might meet a different spectrum there. But just like staging with cancer, that's sort of who they become is that's their stage. That's the prediction of what their outcome might be. The further along you are, the more likely you are to have morbidity. Mortality is an outcome. So I'll tell you a quick story for a clinical thing that we saw just late last year. Got a call. A woman has Stevens-Johnson syndrome in the emergency. Indeed, she did. And they said it was caused by a moxel. She had a sore throat. She took a moxel. So a resident goes down, and they know that they have to get a better history, because they know I'm going to ask. And the fact was that she had taken tigrotal carbamazepine three weeks earlier, given to her for headache in a walk-in clinic, which makes no sense, which is often the case for these things. Now, does she have Stevens-Johnson syndrome? Yes, she did. What caused it? The carbamazepine is far more likely. It was a sore throat from that Stevens-Johnson syndrome that she got the moxel for. But it'd be so easy to make the mistake there and say, it was this drug, and never know it was the other one. Could be a huge mistake. So this is her. And she's in the hospital. She didn't go to the intensive care unit. She didn't have that much blistering. But she was in hospital for about four weeks. She was quite sick. She had to be fed with a nasal gastric tube because of the severe sloughing that she had in her nasal cavity and her aural pharynx. All that from that tegrotol pill. And that's bigger than it really is. But she took carbamazepine. Doesn't have to be tegrotol. And this is what she ended up with. And she's Indian. And because she was from South Asia, we asked her, did you have a genetic test done? She should have. She did not. So we did it. And why do I say she should have had it done? Because in 2009, the FDA said that there are genetically at risk populations. We'll hear a lot about that today. They should be screened. She doesn't sound very powerful. She doesn't must be screened, should strongly be considered for screening. But still, it was there. And I'll tell you the impact of this. If you Google Stevens-Johnson Syndrome, and many of you may have done that before this meeting, you would have probably had to go through the first eight pages of lawyers to get to anything actually medically close to Stevens-Johnson Syndrome, because it's big business for lawyers. And they all have expertise in this, and they'll get you a big settlement. Shouldn't be there. We shouldn't be there. This person never should have got the drug, perhaps. And we should be able to protect people, much in the way we've seen that can be done in a country where you theoretically can do that. So she was B-star 1502 positive. And so what does that matter? Because we already know she had the reaction. Well, it does, because now the amoxyl looks even less interesting to us. Nobody can convince me the amoxyl did it. And believe me, people are always trying. Three weeks, how could it have caused it? Well, if you believe the tipping point idea, that's sort of it too. If you keep piling sand in a pile, there's a point where you eventually get an avalanche of sand. And there's something that happens here to happen as well. So the HLA test was very supportive of the diagnosis and the etiology, and would be very good for genetic counseling in the family. Now, most people in that family, if she was your sister, you probably would not want to take that drug. What looks hypothetical and rare is clearly not so hypothetical and rare. So her family history, which I thought I had next, is actually quite interesting, because her brother had been taking Tegretol for years without any problems, three years. And we finally got him to come in and have his testing done. And he, too, was 1502 positive. So it doesn't mean that everybody who has that gene is going to get the reaction. It's an important concept in several areas. One, we're talking about genomics and screening populations, but we're also talking about genetics and individual risk. And while it's not 100%, it's the best marker we have of risk. And it might not be the only marker, but it could be the best right now. And we know that if we screen, and Wenhung Chung and his group have done that, if you screen you can prevent this disease. But there are people there who could have taken Tegretol and not had a problem. But if you really, really needed carbamazepine or allopurinol, another drug that could do this with a different gene, if you really, really needed it, then you wouldn't just have what is currently in the product monograph that says there is a risk, you might have a risk. You'd say you really do have a risk. And we know the best way to treat this drug is to stop it early. So if we see this reaction and we want to stop the drug early, we have an impact. There's a lot of things that come from knowing that. Now this is something that I've been building up over the years. There's many components that go into this, and I'll get into it. But you'll be hearing a lot during this meeting about the mechanisms from people far more knowledgeable than I. And you'll be seeing pictures about how T-cells and keratinocytes, when I started in dermatology a long time ago, keratinocytes, there's a whiff, perhaps, that keratinocytes did something and just be there. And they do a lot. And they're not the only cells in the skin who are doing a lot. And then you've got dendritic cells, and you've got lymphocytes, and cytokines, and it's a much more exciting story now about what's going on. But the bottom line is that the genetic testing still has great, great relevance, and I hope I showed that. Now James Reason and Kali came up with what was ultimately called the Swiss cheese model. And I remember speaking in Switzerland, I asked them, what do you call Swiss cheese? And they said, cheese. OK, it's good to know. So in this cheese model, if you're from Switzerland, the idea is that whether it's airplane safety or anything, that you really need a lot of things to line up to make it a problem. We have built-in safety systems. We are here because we beat the odds in so many ways. And drugs make it to the market by a survival of the fittest. They also make it to market after being screened extensively and then being used in clinical practice extensively. And the bad ones generally get gone because they do cause too many reactions. So for the people who have this, a lot of things have to line up to make it happen. So if we pretend that we're going to look at what's going on in a patient, here you have this wonderful story, which is what attracted me to this area in the first place. You sort of know what caused it in terms of the trigger. And you know what the outcome is. And now we have better definitions. And you've got the drug. And then you know the structure of that drug. And maybe the metabolites of that drug. And this is where cytochrome P450s come in. And so you're getting not only that parent drug, but metabolites. Or you might be getting more of the drug because the metabolism is blocked and it's not as active as it should be. And then it interacts with the immune system through the HLA, but maybe through other components. So you've got the structure. You've got the metabolism. And we know about the cytochrome P450 or CIP genes. We know about the human leucosanandrogen or HLA genes. And at the end, we get the full expression of the immune toxicity. You might have problems in several of those areas. And at the very end, if your immune system doesn't want to listen to it, it doesn't listen to it. So we don't know all that, but we do know that there are many steps that go on, and we can actually measure those. So this is about her brother. So her brother, guess what? So he indeed was HLA BSTAR-1502 positive. So it's not everything. So that's where this multi-step process comes in. And that's why we're interested in the CIP genes as well as the HLA genes, among others. Now the epidemiology. If you signed up for the meeting, you would have received a paper in epidemiology that we did with Ronnie Dodio-Gad, who's from Israel, Phil Laws, who leads now in the UK and myself. And what the bottom line is, to make it very simple, that it's a rare disease, if you will, on purpose. It's a rare disease because we get rid of drugs that cause it a lot. However, there are areas and populations. Nyveripine in Malawi, drugs are being used there to treat AIDS. You get a lot of toxic epidermal necrolysis. And you might tolerate in other situations. I don't want to make too much political statement there, but it's an issue. But it's rare. But we do know that there are risks in certain populations with certain genes and certain drugs. And so we know we need to pay attention to those. I mentioned carbamazepine because it's probably one of the best worked out models. I don't think it's the most dangerous drug there is. That's not why we're talking about it. But it's the best worked out model. And allopurinol is also a similarly dangerous drug for these kinds of reactions. But the warnings are not quite as clear. And there are so many warnings for patients, for physicians, product monographs, labels, that are sort of not right there until you actually get the disease and look it up and you go, oh, that's what they were talking about. And we have genes for that. And I see we just had a person in a burn-in who had that, a woman who was admitted with dress who has it as well. Now here's the genomics. You've got this paper. There's a lot of drugs and a lot of genes that we know, all with different relevance and different strength. Diagnosis of this disease. Well, these are sick people. I showed you a picture before. I won't dwell on it. But this is painful. And people are in a coma usually when we're treating them. We keep them in a coma for a long time. But I have to tell you that when they come in as patients later, they've suffered horribly. And they don't want to see these pictures. You have to hide them. I don't show them to the residents and say, oh, this is what this person looked like in hospital. It's terrifying for them. So how do you make a diagnosis? You need lots of things. And I was at a meeting with somebody called the Sheer's Diagnostic Triangle, which I liked. But I'd like all of these. I'd like to know, is it a blistering disease? Does it have those spots? Is it systemic with fever, maybe respiratory, and other involvement? What about the histology? Do we see the epidermal necrosis or apoptosis or whatever we're seeing in the epidermis? Do we see that? And then when you want to stage it and define where it is, you can go back to this thing that we made up many years ago or the early scar years in Europe to say that these are the features that define what stage you're at. Causality? Causality assessment is something that clinicians do very, very poorly. And this is from XKCD, NASA engineers, Russell Monroe, who does this. I used to think correlation applied causation. Then I took a statistics class. Now I don't. Sounds like the class helped. Well, maybe. Correlation does not equal causality. And causality assessment is very much enhanced by testing. But it's a shame to have to do the testing after the fact. So the treatment is hugely variable. We did a survey, our group, with Ronnie Dodie again. We surveyed all the Canadian US burn units. Had about a 35% response rate. And results were all over the map in many different areas. We looked at how people treated it. If you want to summarize what the standards today may well be, they might be high-dose corticosteroids, cyclosporine, and even anti-TNAF therapy. Intravenous immune globulin, which is expensive and was touted early on, may not be that effective. Does have risks of hemolysis and clotting, et cetera. So it's sort of fallen away. But for legal reasons, people will still use it because you don't want to be sued that you didn't treat the person fully. For the eyes, maybe cyclosporine eye drops will turn out to be useful. But for now, I think the gold standard are amniotic membranes. And they need to be paid attention to. To survive this and be blind for the rest of your life is not exactly a happy ending. The genitals, we see women who come in with genital scarring that could be permanent. Children, boys, and girls don't have their genitals examined necessarily carefully, end up with permanent scarring. Again, something that needs to have attention. And post-traumatic syndromes have been well-reported. And we're studying it now as well. But I'll tell you, the outcome is not that people are afraid to take drugs. They know doctors give them drugs, so they're afraid to see doctors. And they will just not go to see the doctor. And their stories are highly impactful. They're powerful stories. So we can do better. We know that patients should be screened, but they're not being screened. We see that over and over and over again. And it's disconcerting. Better communication has to come from all the advances in science that we're having to get perhaps better regulatory and other influential effects to talk about this, to get patients to understand it, to get doctors to understand how to communicate all this, so it doesn't become too skewed one way or the other. And patient support groups are pretty good at staying involved and being in touch. But what we want to do is get doctors to be in the mode to be able to do this. But maybe it does need to be systematic and systemic across the health care system. Care guidelines are needed, and we have a meeting coming up where we're going to discuss that. I just put this slide together as a bit of a model. And I hope Terry's comfortable with it. It's about partnering for success. You can have models that are purely funded on a public end. Or it could be, say, with NIH, or purely on a private end with industry. And FDA would have influence, perhaps, at the industry say, at least to make things happen. And there could be research applications that go to the university that are studying the science of safety and the efficacy of these drugs, and coming up with actual clear structures and models to make it happen. So I'm going to conclude with this slide. And to Terry's shock and surprise, I'm actually on time. On Tuesday, June 8, before the World Congress of Dermatology, we have a meeting of the international severe cutaneous adverse reaction group. Wen Hongchun and I are co-chairing that. And it's going to be dermatology heavy because it's a dermatology meeting. But we have speakers. Many of you here may be probably be speaking there on severe cutaneous reactions with much of the emphasis on Stevens-Johnson TA. And we'll be polling, as well, for people's attitudes to therapy and treatment. So I'll just leave that up there in case anybody wants information about it. This is the link that will give you the program and information about how to register for that meeting. Thank you very much. Thank you, Dr. Scheer. We have time for questions. Please, Paul. Yes. So, Neil, I ensure that review. Some would disagree with not including IVIG in your treatment list. But I actually agree with your assessment. But what I want you to do is address the histopathology. Because in Stevens-Johnson, some people call this all a spectrum. But there's a noted differential in terms of the histology with inflammatory cells at one end with virtually no inflammatory cells at the other end. How does one evolve into the other? Or are they really two different entities? I know now they're always lumped. Everybody lumps them. But has anybody looked to really see what that is? So it's a very good question about the spectrum and the immunologic response. I think, Wenhong, you're going to be talking about some of that? Yeah. So we'll hear more about the granuliscin story and others about the histology. So the histology, so when we have on that slide with the spectrum of disease, if you just look at the skin and you see this infiltrate of lymphocytes, then you want to know are the CD4 positive, CD8 positive. Might look for other markers, fully to understand those. So now we can differentiate those. So that's sort of helpful. And you're looking more at the CD8 ones that are going to lead to blistering. And then you're looking at the cytokines that come from there. But I think if one wanted to look at a nice model of it, when you get into a far right where we had people with this diffuse blistering without all the spots and stuff, some of it called pure plaque, there've been other names for it. But it's just kind of thing. There's a large differential there. But one of those generalized bolus fix drug eruption. And in fixed drug eruption, we've always wondered why it's specifically in that area and not this area next to it and how the patch testing will be positive in one area, but not right next to it. And Tetsuo, Shihara, and Tokyo has done a very nice job of looking at the stages there. And one sees the stages you go through. There are memory cells there that will start a whole cytokine storm. And then they're not really that necessary after that. But they call in the cytokines, call in the CD8 positive cells. And then you get more going on. And then those cells sort of disappear. Does that answer the question about that whole spectrum? No. But it does show that there are all these stages that are going on. And it's a story that's hard to recreate. We don't really have good animal models. So I think there's a lot of work to be done in that area. So you have mentioned the metabolizing of the drug. And my understanding is that the metabolites contribute to antigenicity. And that may trigger CTLs, et cetera. But what else about the metabolites? I mean, there's a whole theory outside the immunological aspect of this disease, right? Yeah, so our early work was all on the pharmacology side, the reactive metabolites story, whatever. And I still think that's very relevant. But it isn't the only story. And so you might have people who are creating more of what would be a reactive metabolite that might be more toxic. However, you might also have people who are of decreased metabolism of the parent drug. And the parent drug is the problem. So for allopurinol, for example, almost all those patients have renal insufficiency, higher levels of allopurinol, hydroxyallopurinol. And when you patch test, for example, we did a ton of those in a group in Portugal did a lot too. And we did it with the drug, with the metabolite. No one's ever patch test positive. What's going on with allopurinol? I don't think we really understand. But the idea is that the entire molecule might react with the HLA system or other ways of triggering the immune system. So there's a lot to do. But I think you're right, the idea that there is pharmacologic components, immunological components. And depending on the puzzle and other genetic and epigenetic aspects might lead to what the actual disease is that you see. So even though we know the cause, here's a drug and here's the outcome. The outcomes might look similar, but the story in between might be very different for individuals. Yes. What is the cost of conducting the test? What test do you use? And then how quickly do you get the results back? I can only speak for our site. It could be different all over the place. But we did look into that when they said, why are you ordering so many of these genetic tests? You're ordering too many. And I said, well, everybody else should be ordering them. We're ordering them because nobody else did. And they said, OK, fine. For us, it would probably be in US dollars, about $200 US dollars to have the test done. It would be nice if that came down. But I mean, honestly, if you showed somebody a picture of toxic epidermal necrolysis and you said it'll cost you $200 to find out if you're going to be at risk for that, I think most people would pay it. But then they wouldn't take the drug anyways because you just showed them a picture of toxic epidermal necrolysis and they said, never mind. I mean, if it was food, you wouldn't eat it. I mean, you just don't go near there. And so I think it's on balance, though. One would have to do a real economic analysis to see to the health care system, especially if you're using IVIG. And as Dr. Katz said, doesn't necessarily support it either. But on the other hand, it has used a lot. But it's a very expensive therapy even for a short time. And then if you need repeat dosing, it can be quite expensive. And then it's a disease with high mortality and huge morbidity. So I don't know what it's worth. I don't know what it's worth. And how quickly do you get the result back? People will get the results in a day or so. So we had a patient we recalled about in one of the teaching hospitals, major teaching hospital. I was emailed by one of the internists who said I have a patient who had a reaction to phenitone. We want to give him carbamazepine. And the residents all told me that they should have genetic testing done. Is that true? And I said, I'm surprised the residents actually knew that. But that's right. And she said, how do I order it? I said, well, how do you order a complete blood count? You write down CBC. And you write down HLA BSTAR 1502. And you get it done. And they had it within a day, and he was positive. So they're very happy not to have given him the drug. So it can be done very quickly if you need. So why this disease is attacked by your tissues? Why this skin and mucosa? Is there a simple explanation of this? It's much more complex. Well, there are a lot of cells that will have markers that show that they do hone into the skin. But it's still a good question for a lot of these reactions. Many of them, for instance, are in areas that have been sun exposed. And we know about reactions to many drugs, including chemotherapy, where it will attack previously sun exposed or burned areas. Others are pressure exposed. We don't know why completely. But I think there's a lot going on there in terms of what the skin cells want to do. However, it depends on a crossover with dress. Some of these people will get liver involvement. Some people will have kidney or other blood involvement, but not as much as the other systemic syndromes. And I don't know if that's going to be discussed today about the skin homing or a bit. Yeah, probably a bit. We'll hear a bit more about that. The point is that it is skin and other stratified squamous epithelia. It's not all epithelia. So we have many precedents for that. The best precedent probably is in the area of Pemphagus. For example, we know what the antigens are, and it may be that that secondary need for the disease, in addition to the HLA, is something else that has specificity for stratified squamous epithelia. Because you go down the esophagus, you get to a point where there's no longer stratified squamous epithelia, and you don't see the lesion. You do see the lesion in the general mucosa, which up to a point where there's stratified squamous epithelia. So that must be a clue. I don't know the answer, but it must be a clue. I like the lady there. She had esophageal involvement. Many patients will have bladder involvement, like in Pemphagus, too. People get bladder involvement. But I think in the beginning with a big disease like this, I feel like we're the investigators for a fire department. When you're seeing this whole house on fire, you don't get down to the nuances of where the fire started and what other damage has been done and what's missing. And so when this disease comes in, they're often very bad, especially at a referral center where we get them after they've been played around with at other centers. So yeah, it could come up later because we haven't looked for it. But it's not as well staged as dressed where you sort of see a lot of atypical lymphocytes, and then you see a lot of eosinophils, and it's got a different staging. Actually, we'll talk more about that at the ice guard meeting. But it may well be happening early on. But I don't have the same sense of exactly when. You don't necessarily look at them in such great detail. So excellent talk, Neil, as usual. From a liver point of view, I always understand why drugs harm the liver. That's where they're metabolized. But the skin, bit of a mystery, isn't it? Are these drugs concentrated in the skin? Are they metabolized in keratin sites? I mean, we wondered about that from the very beginning. We tried to metabolize some drugs with skin cells in culture. And I started off with acetaminophen, which is a good example of thinking of the liver, that we could do with hepatocytes in culture. In those days, it was still early days about culturing them. It wasn't that long post-green. And we could see. We did HPLC, and we could actually see some metabolites. But I think the metabolic contribution is probably pretty small. It's not like the liver, and it would have probably passed through the liver before it ever got to the skin anyways. So I don't know that the pharmacology in the skin is quite as critical as the immunology in the skin. That's just my quick summary. We may hear different today. The other thing is we always think of a gene as the bad thing. This is the bad thing. But it may be the wild type. And there's something that protects the other people. So when you find funny reactions to a drug, sometimes it's the wild type rather than less than HLA. An example for that to me is with sulfonamides, where acetylation might be important. And slow acetylators may be at higher risk of some of these reactions. But I often look at it the other way. I think fast acetylators are protected. You might have a funny 2C19, but you might be protected because you're a fast acetylator. So sometimes the genes protecting people and sometimes genes are hurting people. And then it's the balance. That's why I like the Swiss cheese model. So Neil, I was wondering, you commented a couple of times that the nomenclature is difficult and the diagnosis even can be difficult. And I think we've seen that in other fields. And sometimes the answer has been what's been referred to as molecular taxonomy. So if you can define a disease molecularly either on a genetic basis or otherwise, sometimes things fall into place. And breast cancer is a good example of that. Probably charcomery tooth that has all these different types. Do you think there's now an avenue for molecular definition, not just genetics, but the genetics and the immunology and that? And is that a direction we should try to pursue? I think it's a direction you should try and pursue. I don't know exactly what that menu is going to look like with how many different genes and how many different. It was interesting years ago, I remember, with epidermolysis bullosa I was speaking to genetics group of sick children. And they said, well, you don't think that could ultimately be a single gene causing those. I said, well, why not? Well, it's so complicated. They said, well, zinc deficiency is just from zinc. I mean, it looks complicated too. But somehow with this, I think it's just sort of, with drug reactions it's different because they're weeded out. You know, we obviously have a lot more information if we kept on drugs that were killing lots of people. But we don't have that with Charcot-Mary tooth and with other ones. And the family history is then perturbed because this woman's children will never take the drug. And her siblings won't take the drug. So you end up with all kinds of challenges, I think, to get the kind of specificity that you might like for that. But it's certainly worth knowing more. And if we can tailor drugs to be based on that, it would be great. And I think the cards that we would have ultimately in the future would be a lot more complicated than just your HLA. Sorry, may I have another question? Yes, you mentioned that in drug development, there were companies that just in very late stage drug development stopped because of SJSTN. Did those trials collect DNA samples? And because the odds ratios on these reactions are so high, they may even be able to find what is the associative allele. Any thoughts on those lines? Oh, yeah. I think about that all the time. We actually were collecting cells in patients we were studying as part of that reaction because I thought we might be able to do genetics one day. But this was before people would get consent to do genetic studies. And the samples, oh, they could still be in a minus 80 freezer somewhere. What isn't? But no. And then people are long gone, so I don't know what happened. It was a long time ago. But the drug I'm speaking of is Sorbinil, which was an all-dose reductase inhibitor, and had huge impact of preventing diabetic-related sorbitol cause neuropathy, especially neuropathy, and retinopathy and nephropathy. And there were many drugs coming out that were also going to be all-dose reductase inhibitors, and it looked like to be a future that might be better with other drugs. But I don't think it's happened yet. And patients who were on this drug, they're the only people I've seen who had Stevens-Johnson syndrome who said, is there any way you can get me back on that drug? My pain disappeared within days. Could you get me back on that drug? So it's a big loss. And now you'd think you could screen for that if you had the samples. But then you'd have to have people to have the reaction to prove it. Though structurally, it's very similar to phenitone. And Will Cross react in vitro with people's cells that was published in the Annals of Internal Medicine with phenitone. Neil, has there ever been a drug that has been identified before it's gone on the market and has been eliminated because of TEN? I mean, it's such a rare event. Yeah, yeah, drugs. I mean, often if you do a phase one study, if there's 10, 20 people and four people get a rash, that's often the end of a drug. You don't know if they're going to have TEN. I'm talking about TEN, though. So this drug was TEN. Sorbinil was about to be approved. And phase three data, Vera Brill's group, published the neuropathy data with serial nerve biopsies and everything in the New England Journal. And then there were a couple of cases from some of the studies. And that was the end of it, so never made it to market. When you show the case report, you report within the family, the lady has the same gene with her brother. The lady had the same Johnson Central brother does not. Do you have any data show that people carry the same gene without the drug gene interaction? Wenhung, we'll talk to that. But I think the idea is yes. I mean, we know that for every one of these tests, there's a number needed to treat and to test. And you're going to have a lot of people who have a gene who aren't going to have the reaction. On the other hand, if they do have the gene, you can have a very thoughtful approach to the therapy or conversation. And I think that just for us highlighted that it was not was the answer for that woman, but obviously there were differences. There were differences between the two of them in terms of why one of them had a reaction or not. And the HLA was a great screening tool that could have been used, but it wasn't the only answer. OK, thank you very much. So there's one more question. Oh, there's one. The trouble is, if you answer a question, then you might stimulate another question. I just want to answer. If somebody has had a previous reaction, like a simple rash to a drug, does that predict something more serious? Well, it's a very fundamental clinical question. I'll just repeat it so everybody's heard it. If somebody took a drug and had a simple rash, and then they took the drug again later, are they going to be at higher risk of having Stevens-Johnson TN? I will say no, but with a caveat. And the caveat is, I can pull up cases that sound like that, and maybe they do happen. There was a woman who we saw, for instance, who had a history of allergy to aminopenicillin, tampicillin. And she saw a dentist, and a dentist gave we use cloxicillin, not dicloxer, through cloxicandidae. It was given cloxicillin. And she had toxic epidermal necrolysis from cloxicillin. And we patch tested her later to both, and she was positive to both. Would I have said that if you had a reaction to an aminopenicillin and a rash, which 8% of people do, that you're at risk of getting a reaction to cloxicillin? No. But the dentist was sued. I don't know what happened. It's just sort of nice to see, not just doctors getting sued. But I don't know. They asked, they got an answer, and you probably went to give her that drug, which makes it hard to know. People will say, oh, you had a rash from such and such. I'm not going to give you that, ideally. I teach our medical students about a sulfonamide reaction where the doctor said, you're allergic to any drugs. She said, I'm allergic to sulfonamides. She said, oh, what happened? She says, I don't know. It was like 30 years ago. And he said, oh, so it'll be fine. And she ended up in the burn unit with toxic epidermal necolysis. Well, I think the teaching there is without genetics. If the person tells you they had a rash, you sort of look for alternatives unless you really need to use it. So it's going to be hard to collect that data. OK, so now we have to move on. We have 30 minutes for discussion at the end of the two keynotes. So next speaker is Dr. Wenhang Chang, Director of Drug Impersensivity Clinical and Research Center, Department of Dermatology, Chang'an Memorial Hospital, and Associate Professor of Chang'an University of Taiwan, and the title is, as you can see, basic science of pathogenesis, functional genomics, and mechanism. So 20, 25 minute talks, and we can have five to 10 minutes discussion.