Thank you, Vic, for inviting me. How do I make the slides move? I work for the federal government, and I have no disclosures or conflicts of interest, I hope. I do write liver talks, so that's my contact. Liver Talks is a website, freely available, that deals with drug-induced liver disease. It's a joint effort of NIDDK, my institute at the NIH, and the Library of Medicine, without whom I couldn't have done it. The aims are to advance knowledge about drug-induced liver injury, but also to stimulate research and interest and fill this room up, maybe, with more people interested in this topic. Well, Liver Talks, it's important to point out that Liver Talks includes the drugs that cause liver injury, but it also includes the drugs that don't. And this is the difficulty that we have as knowing the denominator of what we're dealing with. It's largely dealing with U.S. drugs. There are very few drugs in Liver Talks that were not approved in the United States. It's largely prescription drugs, non-prescription drugs. It has selected herbal products. It has some drugs of abuse, some metals, and so forth. What doesn't it have? It doesn't have drugs that are not absorbed systemically, topical drugs. It doesn't have drugs that are things that are given once, like vaccines, plasma products, and so forth. So it's not everything that the FDA does. What are its features? One thing is that Liver Talks is formulaic. Every chapter is structured the same. The uses are very much the same. It does give you, also, brief things about mechanism and actions and common side effects. This is what Liver Talks used to look like. You probably have seen if you use it. It has three components that we point out. One is this green box on the side. These are introductory overviews about drug-induced liver disease. Also descriptions of causality systems, descriptions of severity systems that have been discussed today. And also an ability to submit a case report to Liver Talks. Then in the middle are where you put in the drug that you want to search for, or you can use the alphabetized list. Well, something happened here. Last month, something happened. Liver Talks migrated, they say. It has a different appearance, but it has the same content. A different search engine, but it's vastly improved. You can search on anything, any word, not just the drug name. It also can use brand names, and if you misspell the drug, it will try to correct it for you. So it's a great advance there. Liver Talks is now cited on PubMed, if you do a PubMed search. Very helpful. But we no longer accept submitted cases, not allowed by the Library of Medicine. This is what it looks like. It looks a little different. It's important to search... I don't have it there as the pointer. It's important to search where it says, search this book. Don't go up here and search in Bookshelf. This website is now on what the Library of Medicine calls Bookshelf, and it's where it puts books. Well, this isn't a book, but if you put your search up there, you'll get lost. This is what it looks like. A page. I show isoniazid, a drug we talked about a lot here. We have an introduction of two sentences about what it is and whether it causes liver injury. Then a background material which talks about in detail what it is when it was approved, its current indications, and so forth. Following that, the hepatotoxicity, in a formulaic way. Each paragraph deals with a different presentation of injury. Isoniazid, as you know, causes transaminitis. Very common. That's the first paragraph. The second paragraph is the acute hepatitis that it causes, which can be severe, can cause acute liver failure. At the bottom there, they show... I wish the pointer worked a little better. It shows the likelihood score, which I'll show you later on. This is a likelihood A. No question. We have a short paragraph about a mechanism of injury and then a shorter paragraph on how to manage it. This is what needs to be improved in the future. How to take care of liver injury once it's occurred. We have case reports on liver tox for the drugs that cause liver injury. And try to have a case report for every type of injury that occurs. This is a severe acute hepatitis. It occurred in a Korean lady who was treated for positive PPD and almost died. We choose cases where there's nice documentation and serial lab results. We also have product information. You can go here and get the actual current product label, the package insert, for the precise information about indications and so forth. And also the chemical structure. That links to other parts of... other websites on chemical toxicity. And finally, we have the references, which are now called the annotated bibliography. And they're not references because they're not cited in the text. There's no citations of the references in the text. But they're given at the end. And the first one is generally High Zimmerman's book. It's now 20 years old. And following that, Neal Kaplowitz's and Lori DeLev's multi-authored textbook on drug-induced liver disease. And following that, pharmacology books. And then the references, which are in chronological order. So it starts with the oldest. And you can see here the first report of isoniazid hepatotoxicity. And the reason why we don't cite the references in the text is that there are too many of them. Isoniazid has 430 references. So it's all the literature on isoniazid. Beginning with Searching High's book and then Neal's book. So along with this transfer, we've had some structural changes in liver toxin. We now have been asked to have an oversight board that reviews the contents and gives us advice. It's a three-member committee appointed by our institute. We just started a month or two ago. They are charged with reviewing all new content. They, as well as external reviewers. And if you in the audience would like to serve as an external reviewer for new content on hepatotoxicity, please let us know. Here are the current members. And here's our link that you can email us and tell us what's wrong with what it is. We can change it. That's the other advantage now. I can change it in a day or two. Before we had to wait several months before changes. So after seven years of availability, we have 1,200 drugs on liver tox. They're mostly prescription drugs. As far as completeness, we believe that we have about 95% of drugs that are approved since 1980. And it's a very active webline. So you can analyze the data in liver tox. It's pretty hard for you to do it, but it'll become more easy with this transfer to Bookshelf. So here's three ways I've analyzed them. By the class of agents, by the year they were approved, and by their likelihood of causing injury. And I'll show you those things. And I'm also going to show you something about drugs for liver disease. Here are the classes of drugs in liver tox. It's just the number of them. The most common type of drugs approved that are in liver tox are antibiotics. So it's not a surprise they're the most common cause of liver injury. Followed by central nervous system agents. And so forth. And you can see liver GI is down there at number 6. So how about likelihood? How do we judge likelihood? This is very difficult. There's no good way to do it. And this is the only way that we could come up with. It was published a couple of years ago. We put drugs into five categories, A to E. B means there's at least a dozen, but less than 50. C is up to a dozen, more than three. And D is very few. One to three cases in the literature. Now in these categories, C and D, where they're very few, Dr. Bjornsson actually got out the paper and reviewed it and judged it based on root chem whether it was at all reasonable. Then we have E, which means no cases. But no sooner had we made the system as we realized we needed another category, which is E asterisk, which means there are no published cases, but we suspect that it causes injury. And here's the breakdown in liver tox. 10% of the agents in liver tox are category A. There's more than 50 cases in the literature. 9% B, and 12%, 15%. If you add them all up, 46% of the drugs in liver tox have been implicated in at least one case of drug-induced liver injury. All right, now, so then we look at these in regard to what they are. And here's the slide of the causes of liver injury of the 515 drugs that cause liver injury. What are they? And you're going to say, well, this looks like the slide with the same antimicrobials. Now, antimicrobials also, this is the other difficulty. That's antibiotics plus antifungals, plus drugs for TB, plus antivirals. Very important. So it's not just antibiotics when we say antibiotics are the most common cause. Now, I'm getting a very mixed-up slide here, but this gives you the class of drugs that causes liver injury. As I said, 46% of all drugs have been implicated. And here it is by class. And you can see most of them is like 50%. It's about the same. Well, are there any that rarely cause? And here are the ones that have a lower-than-50% rate. Gastrointestinal, they're generally safe. Respiratory, why is that? Well, the reason is these other drugs are used to treat diseases, whereas drugs for GI are not for nausea, for diarrhea, for acid reflux. They're not for serious disease. They're for symptoms. The same for respiratory and urologic. And the ones that do cause injury, you see a little bit of red down there at the bottom, those are diseases. Those are ulcerative colitis. Here's the distribution of drugs in liver tox by the year of approval by decade. And it shows you something very interesting, and that is there are more and more drugs being approved in America today. And it keeps going up. In fact, for the last decade, the decade isn't over yet, and I'm sure it's going to go above 300. So anybody that thinks that our pharmaceutical industry is going to die, the answer is no, they're not at all. There are more and more drugs being applied. And that makes your problem bigger all the time. Do you know the 50 drugs that were approved last year? Oh, my goodness. Now here's that same slide, but this is by class of drugs. To show you what class of drugs are more commonly approved now than before. And this shows you what all of you know is the big one is cancer. Cancer now represents almost a third of drugs that are approved in the United States. And before, in the previous century, it was like 10%. That's what's getting you. What's falling by the wayside is very interesting. Cardiovascular, CNS drugs, and analgesics. There's none been approved the last 15 years. We have enough analgesics, I guess. Now this is drug induced liver injury by decade. And this is what was talked about by the previous things. There's less and less of the drugs that are approved cause liver injury. So you can see that in the previous century it was about 60%. Now it's down to, in this decade, 18%. So why is that? And look, also, importantly, the type A drugs almost disappeared. I list some of those high risk drugs. When I call them millennial drugs, those... So this is a CME course and you're going to have to take an exam afterwards and you only get one question and you're going to have to get it right or you don't get any CME. So the question is, why is there this decline in hepatotoxic agents being approved? So answer number one is industry is better at identifying agents that cause hepatotoxicity before seeking approval. So all those people out there from industry will tick number one. Number two, the FDA is better at identifying agents that cause hepatotoxicity, as Dr. Rubin has pointed out. So those from the FDA in the audience, I see a few of you, you'll tick number two. Number three, it takes years before there's cases of idiosyncratic liver injury are recognized and published and therefore it's just an artifact. So number three, all those cynical hepatologists out there are going to tick number three. I think the answer is all of the above. There's a little bit of both. Certainly it's because there hasn't been enough time, but also there are fewer drugs that are being approved that are toxic. There are some that are toxic. I think maybe the FDA hasn't withdrawn them. They should have. I think they want to keep their record clean. That's a bad precedent. So you say, well, it's falling off and I shouldn't be interested in this topic anymore. But no, actually it's a bigger challenge than before because there's an increasing number of drugs being introduced, especially for cancer. They have very difficult names and very difficult mechanisms of action. They often have very specific indications, very restricted. It used to be anti-cancer drugs were like Vincristin. Good for solid tumors, right? No more. Now the drugs are approved for breast cancer or for colon cancer or for soft part sarcoma. Whatever that is. I don't know what it is. But there's a drug for it. So two different, two special kinds of agents that are very frequent now are shown here, which are the tyrosine kinase inhibitors. The first one approved was in 2001, and Mantanib, a major breakthrough in cancer research. But now we have 66. And so the other CME question is, name all 66 tyrosine kinases that are available in America. You all fail, I will too. Monoclonal antibodies, the other ones, they were introduced in the last century, but it was only in the last 15 years that the monoclonal antibody technology is such that they can make safe ones, and they are quite safe, actually. But some cause liver disease, strangely. There are 76 monoclonal antibodies approved for use in the United States. And these together account for 30% of the drugs approved in the millennium. So, you're a hepatologist, what about drugs for liver disease? The problem is there's no WHO classification for liver disease. So I had to go back and look through all 1,100 drugs and tick off which ones were due to liver disease, and this data may not be accurate for that reason. I didn't include diuretics and beta blockers and drugs for transplantation. But interestingly, most of these, almost half of these drugs were approved for other indications, not for liver disease initially, and only later found to be useful in liver disease. And so I list some of them there, and pink are the ones approved for another indication initially, and only later used in liver disease. And notice there's a transition from the pink to the white. And that's shown here. And black and white are rather in green and red. And this is the decade of approval of these drugs for liver disease, and whether they were approved initially for another indication in green, or for liver disease in red. And you can see in the previous century about one drug for liver disease approved per decade. And that's changed, and it's changed dramatically, particularly in the last decade. Now you're gonna say, oh yeah, but that's just hepatitis C, and indeed, most of it is hepatitis C. Of the 20 drugs approved specifically for liver disease, 16 were direct-acting antiviral agents. But there are some others. Drugs for cancer. In fact, the slide is out of date already. I believe there are five, five or six, five drugs for cancer approved, liver cancer. And do they cause liver injury? Yes, and here's the data. 27% of these 52 agents have been implicated in at least one case. And look at the A's, the ones with 50 or more cases. They're all drugs. Previous century. So there's an increasing number of drugs for liver disease. Some can cause or worsen liver injury. What is this? Well, as you heard, there are two types of injury, direct and idiosyncratic. All right, whatever. Here's direct. This is hepatic decompensation in a patient receiving obetacholic acid for primary biliary cholangitis. Probably direct injury. Very interesting. You see, she was on erso, but her ALKFOS was still quite high, about five times elevated, and her ALT was high. On obetacholic acid, these decrease, and become, they decrease by about 40%. They don't become normal, but they certainly decrease. And then after two to three months, she comes in with jaundice, fatigue, itching, vasitis, decompensation. But notice that her ALKFOS and her ALT still haven't gone up. This doesn't look like drug-induced liver disease, does it, to you? But this is, I think, drug-induced liver disease. The other characteristic's very slow to recover from this injury. Quite severe, but her enzymes remain low for a long time. So it has a different phenotype, doesn't it? A different feel. Now you say, well, that's just special case. Well, no, it's not completely special. Here's idiosyncratic drug-induced liver disease in a patient with liver disease. Had autoimmune hepatitis on low-dose steroids. The ALT is still high. Put on methadone, on 6-mercaptopurine. Had a nice response, the ALT came down nicely. But then a couple months into therapy, comes in jaundice, quite jaundice. Notice the ALT is still low. The ALKFOS is normal. And he doesn't recover, has a transplant. Okay. So this introduces the third form of drug-induced liver disease. You've heard about it today, and that is, this is not due to direct toxicity, and it's not due to idiosyncratic reaction against the liver. It's due to the action of the drug on something else. It induces an underlying liver injury, or underlying predisposition, we're not sure. The examples of it I show here. Most important one is the triggering of an immune reaction, which you see with the drug, with the checkpoint inhibitors, but also with other cytokines, with beta-interferon, alpha-interferon, with anti-TNFs. Now the importance of this third form is that it's not rare, it's not common, it's what I call not uncommon. And it occurs in susceptible subjects, although we don't know who's susceptible in all cases. And important for this audience is, this form of liver injury often can be treated, often can be managed, can be predicted in some cases, and can be prevented in many, like reactivation of hepatitis B. So this is to point out that many of the causes of liver injury and drugs for liver disease are indirect to hepatotoxicity. Others are direct, and it relates to the pharmacokinetics of the drugs that these people are getting too high a dose. So in conclusion, about 45% of currently used medications have been implicated in causing liver injury. Most commonly, drugs have been in use for decades, but all classes of drugs have been implicated, and the diagnosis remains a challenge. And for the audience, I would say management, treatment, and prevention of drug-induced disease is a major challenge for the ASLD community for the future. Thank you. Jay, thank you very much. I appreciate it. I appreciate it very much. We have a few minutes, and I'd like to draw some questions from the audience. Jay, your talk, and what you've accomplished, by the way, there are flyers on your seats. Please take them, take them home, give them to your trainees. I didn't mean it when I said, I don't know now that the past four years we've had one trainee on the wards, a medical student, a resident who hasn't used liver tox as an important resource to care for patients. But you know, you bring up an important point. Why are we seeing fewer drugs causing liver injury being approved? And I go back to not only the things you pointed out, but also I think there's much greater awareness. I think to my own career, back from the early 90s, and I encountered this drug. Adrian, one of my mentors, we dealt with, even on site, this study with troglitazone. And we were seeing these patients develop these really severe injuries. And those kinds of experiences where patients are taking a drug, it's their well-meaning, they become ill, some die, that can be very sort of, you know, earth-shaking to a clinician. So, questions from the audience? Ari? So, Ari Regev, I enjoyed all the talks. They were very, very good. I specifically would like to address Jay's comments regarding the decreasing number of apoptotoxic drugs out there. And I think it's a little bit a matter of how we define apoptotoxic drugs these days. Because out of the large group that you mentioned of oncology drugs that are now, I think you said, a third of the new drugs out there, there are quite a few that would be, can be viewed as significantly apoptotoxic. And they're, we're just not, they're not being discontinued because they're in a different paradigm of benefit-risk. And another interesting thing is that they are, they have developed some kind of a risk management approach which is also quite new. And I was wondering if the panel has an opinion about this. We now take these drugs, not to mention treating steroids, treating with steroids, the immune-mediated drugs. But we take some of these drugs at significant levels of toxicity. And we stop them and we wait a little bit until ALT goes to a grade one or whatever we call it. And then we restart them. And in many cases we are able to continue many more than we ever imagined. And these approaches are actually written in the labels of these drugs. And they seem to be working. And I was wondering if there's any thoughts about that new paradigm. Adrian, you brought up the concept of adaptation. Any thoughts? Jerry? Yeah, it's, adaptation is diagnosed after the event. It's adaptation if the numbers go down. And the problem occurs when you're monitoring is what level do you lose your nerve? And stop it. And that's really very difficult. Now if it was a statin, where we know that the risk of serious injury is extremely low, one can go on perhaps for another week or so measuring. If it was INH, where adaptation is very common, but the risk is quite high, you might lose your cool earlier. And then the answer is well, but there are other anti-tuberculous drugs. So the guidelines don't exist. And it has to be experience and risk-taking and management. So adaptation is lovely when it happens, but until it happens, you don't know that it's gonna happen. You can say it's gonna happen, but you may not be sure. So I think one just has to sort of tough it out. And that's where the problem of how do you get the monitoring done? That's why if we can have point of care monitoring, and the patients having to go to a lab once a week or twice a week then is no longer a burden, then I think our lives become a lot better, even though we may be doing more tests than we really need. Anybody else want to address Dr. Reagan's comment? The important thing, the oncologist. In liver talks, what we were talking about is clinically apparent liver injury, not transaminitis. You know, there are many of the new cancer drugs where 70 to 100% of people will develop transaminitis. And yet they're used, but you didn't see jaundice and significant liver injury. And you just back away. With most of the cancer drugs, they're being monitored. You know, they're seen monthly before their infusions and so forth. We're gonna take another question, but if there's anybody in the audience from our Asian partners, I'd like you to think about the concept of whether you use some of the liver function testing. I know you do. Certainly more liberally than we do in the United States. And maybe step up to the microphone and tell us how and why, how you apply those. Dr. Watkins. Yeah, hi. Great symposium. Just a quick comment about why there haven't been many drugs approved with unrecognized liver safety liability. And a major reason is the requirement for much larger and longer trials now. So troglitazone was approved with 1,100 patients treated for six months. Now to get a diabetic drug approved, you'd be in the thousands for years. And then of course the phase four post-marketing. After zymeligatran, the first oral anticoagulant, rivaroxaban, was approved after 60,000 patients had actually been exposed to the drug. So I think we're patting ourselves too much on the back to say science is coming to the rescue here. And of course, in addition to extending the time when new drugs get to patients, it's also time off patent, which is estimated to be about two and a half million dollars a day for every day drugs aren't approved. So. Thanks, Paul. Please. Orangey Bullet from Tel Aviv. Very nice talks. I was wondering about a little conundrum, and I'm not talking about the autoimmune or the checkpoint inhibitors, but the other drugs. As we're getting a genetic signal from the HLA genotype, I was wondering why don't we get response to steroids or not use steroids, and especially, I mean the easel guidelines have certainly recommended against using them. Andrew, tackle that? Well, that's a very good question, and I'm not a card-carrying immunologist, so I can't give you a good explanation. But you're correct in that there's no guarantee that the patient will respond to that treatment. So I think that's more of an immunologic issue in terms of what the specific targets are and what might be done to maybe reduce or force adaptation to the foreign antigen. We're gonna wind down if there are no other questions, but Craig, you brought up the concept of adducts. And I've never viewed adducts as a predictive tool, but maybe a more diagnostic tool. In fact, we're in the Dillon thinking about using adducts to link a presentation with an exposure, for example, catechin adducts. But do you really think that there's a role for adducts in predicting liver injury? Yes, I think it's a great question. I think in line with this idea of reactive metabolites as well and what that looks on a molecular level of driving some either innate or adaptive reaction, but also downstream binding and modulating other kind of processes that would tip the ship, if you will, more so. So I think that's probably reasonable, but in terms of all the adducts that are formed, I won't say I'm a pharmacologist or a genomicist either. And understanding the thousands to, if not, another order of magnitude of adducts that could be formed from just the metabolism of one single agent. So to wrap up, Dr. Rubin, thank you very much for highlighting Dr. Senior's work in E-DISH. I'm sure he would prefer us to say 90-ish rather than the actual. But thank you all very much for coming. Before we all disband, I want to make sure that we're all aware that from this point, we will retire or move to the Early Career Investigators Lounge. I want our mentors to raise their hand. Paul, I know that you're here, Paul Watkins. Migrate toward these mentors. We have Paul Watkins, myself, Dr. Bonkowski, Dr. Suzuki, Ayako Suzuki, Dr. Ding, Alex Gervais, and Adrian. So for our young people interested in these kinds of topics, if you've heard something that you want to explore, this is the time to come talk to us. Thank you all very much and enjoy your AASLD experience. Thank you.

Liver Talks

drug-induced liver injury

prescription drugs

non-prescription drugs

mechanisms of liver injury

Bookshelf platform

genetic factors in liver injury