Narrator - Dr. Abel 00:00 Welcome to HelixTalk, an educational podcast for healthcare students and providers, covering real life clinical pearls, professional pharmacy topics and drug therapy discussions. This podcast Narrator - ? 00:11 is provided by pharmacists and faculty members at Rosalind Franklin University, College of Pharmacy. Narrator - Dr. Abel 00:17 This podcast contains general information for educational purposes only. This is not professional advice and should not be used in lieu of obtaining advice from a qualified health care provider. Narrator - ? 00:27 And now on to the show. Dr. Sean Kane 00:31 Welcome to HelixTalk. Episode 70. I'm your co host, Dr. Kane. I'm Dr. Schuman, and unfortunately, Dr. Patel is unable to be with us, but we were fortunate enough to have Dr. Abby Leiden, join us today. Thank you so much. Yes, it's a pleasure to be here, and today's episode is entitled, does rosuvastatin really cause 42 adverse drug reactions identifying true ADRs to improve your patient counseling. Essentially, the purpose of today's episode is to talk a little bit about how ADRs are reported in mostly tertiary drug references, like UpToDate, Micromedex and things like that. And how to help discriminate when you're trying to counsel a patient, what is a true adverse drug reaction versus what is kind of background noise, such as headache or insomnia or diarrhea, things like that. So if you kind of imagine you need to counsel a patient on a new prescription for, let's say, rosuvastatin, we're just picking a random drug. And as we talked about in episode 69 if you're in Illinois, this is now a requirement that all pharmacists are now counseling all patients on all new prescriptions. And it's been a little while, a couple years since you had Dr. Patel's dyslipidemia lecture. And you pull up, up to date to try to figure out, what am I going to talk about? I remember my algids, but what else do I need to talk about in terms of rosuvastatin? And you find 11 adverse reactions that are listed, and then 31 rare, less than 1% adverse reactions. How can you go through that, that list of 42 adverse drug reactions, which ones are actually caused by the drug which ones are kind of chance findings or background noise. So really, in this episode, we're going to talk about four key things. Yeah. Speaker 1 02:04 So the first thing is, why is it so hard to actually describe adverse drug event rates? Can't we just look in there and find all the data we want? Speaker 2 02:11 Why is it so hard to prove that a drug actually causes an adverse drug reaction? Dr. Sean Kane 02:17 We'll also talk about how tertiary drug references like UpToDate, Micromedex, and Clinical Pharmacology can actually be misleading and cause problems in terms of you understanding whether the ADR actually happens. Speaker 1 02:29 And then finally, for students and clinicians alike, we're going to talk about how you can actually combat that problem, learning advanced ninja skills, if you will, regarding adverse drug event identification, and then also really improving your patient counseling using kind of what information to focus on. Dr. Sean Kane 02:42 So Dr. Leiden, why don't we start with why is it so hard to actually describe an ADR for a drug? And I think apixaban is probably a great example of that. Yeah. Speaker 2 02:52 So I think kind of just looking at adverse drug rates, these differ based on different patient populations. So considering apixaban, as you mentioned, patients may be taking this for AFib or also venous thromboembolism. And the ARISTOTLE trial, which looked at AFib patients, the major bleeding rate with apixaban was 3.6% but how can we extrapolate or compare this to the rates when apixaban is used for VTE, and the AMPLIFY trial, the major bleeding rate with apixaban was 0.6% — so now we're looking at 3.6 and 0.6%? The same pattern seemed to be in the warfarin arm in both studies, where the AFib patients had a higher rate (5.1%) of major bleeding and VTE had a 1.8% rate. These two studies really enrolled dramatically different patients, so AFib patients being older, with more comorbidities, which is why the major bleeding rate is so different. Dr. Sean Kane 03:56 So then, of course, if you were to look up, you know, major bleeding risk of apixaban, you're going to see anywhere from, let's say, point six to 3.6% if we only looked at those two studies, but really like that doesn't give you a full picture of what's going on, because it doesn't put it into context of what was the comparator bleeding rate, or what kinds of people were included in that trial. To be able to know, you know, my patient in front of me does or does not have that elevated bleeding risk, right? Speaker 1 04:20 So some of those things, you know, things like age, other comorbidities. So again, is within studies, you know, we'll talk about this later, is the disease states of those who are presented within the studies. Also looking at other medications, those with maybe pharmacokinetic or pharmacodynamic interactions, can dramatically impact the incidence rate as well. So really want to focus on that as you apply it to your given patient. Speaker 2 04:42 Another reason why it's so hard to actually describe ADR rates is that many ADRs occur in normal, everyday life. This is kind of like a background noise in that it's hard to prove or disprove association with a specific drug. So for instance, perhaps. Is a patient who's had headaches in the past month, a patient that has a history of diarrhea or nausea, if you were enrolled in a clinical trial, any complaint like this would count as an ADR. Speaker 1 05:13 And one example of that, again, is if you're testing it in a certain disease state. And there are certain effects, you know, or symptoms that may be part of the disease state as well. And so if you have, for example, you know, a depression study, you know, yes, we are obviously concerned about worsening depression from the medication. However, could it just simply be a piece of the disease state itself that's contributing and may have nothing to do with medication that's potential? So a lot of Dr. Sean Kane 05:35 times, you really have to know what is the comparator, ADR rate in order to put it into context of the drug that you're looking at. So for example, does rosuvastatin actually cause headache? The only way to answer that would to be to look at the placebo rate of headache and compare that to what the incidence rate was with the active drug. And oftentimes, as we'll talk about later, we don't have that context to be able to compare the two next to each other, to know is this background noise, or is this a true effect of the drug Speaker 1 06:03 and then, unfortunately, a third point is that sometimes drug companies will play dirty tricks, especially in phase three studies, that maybe mask or kind of smooth over the Adverse Drug rates. Dr. Sean Kane 06:14 One of my favorite examples of this is the run in phase so this is essentially where before patients get randomized in a trial, they have a run in phase where they're either given placebo, active drug or even both sometimes, and basically they get kicked out of the study if they don't tolerate drug therapy, or if they're non compliant with drug therapy. It depends on the protocol. And the problem with that is that when you look at the side effect profile in a phase three trial in the table, that table basically does not include those who were excluded during that run in phase. So if I'm getting Ramipril and I have hyperkalemia during my run in phase, you won't see that ADR reported in the manuscript because that patient dropped out before randomization during that run in phase. Speaker 2 06:58 One other example that I think it's important to look very closely into is looking at the inclusion and exclusion criteria that generally pick healthier patients. One of the more common examples of this is excluding patients with CKD or hepatic dysfunction. But in general, phase three trial patients are healthier than real life patients. Speaker 1 07:21 And then this is where, again, with working with individuals with mental illness, this has always been a big pet peeve, is when we look at different disease state managements within a population with, say, depression or schizophrenia, and with schizophrenia, there's, you know, both. There's a ton of comorbidities. And again, how do we know how these medications interact? We don't fully know, because these individuals are excluded from Dr. Sean Kane 07:41 those studies, you know. And then the other thing to think about is sometimes a very severe adverse drug reaction doesn't occur in phase three trials because we only have a couple 100 or a couple 1000 patients, but really, we start seeing some of these more rare but serious side effects once the drug is on the market. And I think everyone in this room can probably think of a couple drugs that have drugs that have made it to the market and then later been removed from the market because of a side effect that was observed after the drug made it to the market. So we do see phase four trials that are done. These are also called post marketing surveillance studies. They may or may not be required by the FDA. It depends on the context. And an example of this could be PPIs, and whether PPIs can cause pneumonia, C diff, osteoporosis, things like that. If you want to know more about that, if you go back to Episode 57 we talked a little bit about that. Another great example is rosiglitazone (a TZD) and the potential risk of MI, and we also talked about Invokana (canagliflozin) and the risk of lower limb amputation in Episode 67 — so there's a lot of examples where we don't even know in the clinical trial itself, the phase three trial. But then as we start studying it, more, as it's been on the market, more people use it, we start seeing signals of adverse effects that we didn't even know about before the drug made it to the market. Speaker 1 08:57 Oh, another one that sometimes you know, another darker one that has come up recently in kind of critiques of previously published studies, is the idea about potentially coding side effects differently in order to avoid sending a strong safety signal. So if, if, as you're reporting a side effect, somebody reports it in a study, you determine how do you want to classify it? And so, for example, 30 patients out of 200 report sedation. You know, 15% rate that that can be alarming. However, if 15 of them are coded as somnolence and 15 as sedation, or even something else more general, like a CNS depressant effect, now with each of those, you see them as like a 7.5% and if somebody is only focusing on those greater than 10% both of those have now been swept under the rug, whereas if they were combined into one endpoint, you would have noticed that. Speaker 2 09:39 I think another example of this is how major or minor bleeding is reported, same as kind of overall cardiovascular events really important to look into, especially if there's an end point that has multiple outcomes. So cardiovascular events, are we talking about? Stroke, recurrent, MI? By death, looking at composite endpoints and what kind of they entail is very important when you're looking at the extent or how common it is to have an adverse drug reaction. Dr. Sean Kane 10:12 So Dr. Schuman, one other component here is that if you just think about it for a second, it's actually really difficult to prove, definitively prove that a specific adverse drug reaction does or does not occur as a result of a specific drug so what are some of the reasons that come to your mind with respect to the inability to definitively prove that? Speaker 1 10:31 Well, one of them is, again, is power is with studies, the goal is to enroll enough individuals to reach power to be able to determine an end point. In these cases, it's efficacy. However, safety is something that if a certain smaller number of individuals have that side effect, it's a much smaller number. And so because of that, they're not adequately powered to detect those kinds of differences. So they may report something, but it's not quite to that same, you know, robust, you know, nature that we're having a significant P value, and Dr. Sean Kane 10:57 even if you don't have a significant P value, so either you do have one and you're at risk for a type one error, because maybe you've looked at 20 different adverse effects and you're saying, Yes, this one does occur. Type one error is a high risk there, or maybe you don't show a significant P value because it's under powered, and you're at risk for a type two error. So no matter how you slice it, these are never, almost never, primary endpoints of your study, therefore, they're always at a higher risk of a type one or a type two error. Speaker 2 11:25 And then, as Dr. Schuman had just addressed, kind of the overall background noise in terms of a patient population or a specific patient's history, so do they have a history of headache or diarrhea or depression? And looking how that could affect the reports of adverse drug reactions in Speaker 1 11:44 the clinical studies. One thing we can sometimes do is use a number needed to harm, or an NNH will sometimes be reported, again in a summary of multiple studies, and they can be used to compare multiple different medications to one another. But the one issue with that, again, is you're only picking one specific side effect. You say, I want to know the number needed to harm for this really bad side effect of the medication. You're not looking at the other side effects as well as again, back to the idea that we're not comparing apples to apples. Those two studies may have tried to enroll similar patients, but they probably did not enroll the exact population. So to say side effect rate is, you know, 8% and therefore you know number needed to harm is six here and eight here, therefore this drug is safer than the other. Can't really say that. Dr. Sean Kane 12:24 We even see that as we talked about at the beginning of the episode with apixaban, if you were to do a number needed to harm from the AFib apixaban study, it would look like harm is much more likely versus in the VTE study. And again, it deals with baseline characteristics of the patients their risk of having the adverse effect on their own. There's so many different things to think about that makes it really hard to compare apples to apples between two different studies. Speaker 2 12:48 So the next kind of thing that we wanted to talk about is how tertiary drug references can often be misleading regarding adverse drug reactions. So going back to that, rosuvastatin — UpToDate cites headache as an adverse drug reaction occurring 6–9%. Does Crestor really cause headache? I'm willing to bet that four out of five pharmacy students will counsel on headache. Dr. Sean Kane 13:12 I can tell you, this is one of my pet peeves, to be counseled on headache, diarrhea, any kind of generic adverse event when the drug really doesn't cause that problem. And as much as I love up to date and use it frequently, this is one of my pet peeves in terms of misleading the pharmacy student, in terms of understanding that adverse drug reaction rate. Speaker 2 13:34 I think another kind of pet peeve of mine, when looking at up to date and other tertiary drug references is they may report an adverse drug reaction, but it's hard for us to extrapolate which ones are the important ones. So there may be a drug reaction that's a little bit more rare, although it's very clinically significant. And so just looking at these tertiary references doesn't always help us figure out what's the most important thing to counsel a patient. So if you Dr. Sean Kane 14:02 actually look at the phase three trial for rosuvastatin, depending on which one you pick, the headache rate did occur about 6% of the time in one of the studies as an example, but the placebo rate was 5% — so if you put 5% next to the context of 6% it doesn't seem like it really causes any headache, if at all. It would be a very, very small signal of headache. So hopefully, if you knew the context of the comparator, that would really change your impression on does rosuvastatin truly cause headache or not? Probably not. Speaker 1 14:31 And the same trend can be seen with other adverse reactions for rosuvastatin. So nausea, 3.4% with the active medication versus 3.1% placebo. Constipation, 3.3% with rosuvastatin versus 3% placebo and dizziness, 4% versus 3% — so again, these are not huge differences. And you know, one way to do it would be to provide the placebo or, you know, with somewhat tongue in cheek, or have placebo listed as a separate reference within there, with its own side effect rates. Dr. Sean Kane 14:57 So Dr. Schuman, are you trying to tell me that rosuvastatin may not cause nausea, constipation, dizziness or headache? I'm afraid I am mind blown. I think this is such a huge point for especially pharmacy students to understand is how these tertiary drug references can be misleading if you don't have that context. Speaker 2 15:18 So the next thing we want to talk about is, how can you combat this problem? Learn advanced ninja skills regarding adverse drug reaction identification and improve your patient counseling. So Dr. Kane, why don't you talk to us a little bit about daily Med, daily Med, daily med. Dr. Sean Kane 15:34 So daily med might be one of my favorite drug reference tool kits that I have out there. Daily Med, if you don't know, is a website. It's produced by the US National Library of Medicine, and it's basically the official FDA approved package insert, which you can get from the FDA website. And it is the same content, except it's in a very usable format. It's easy to access. It's easy to kind of skim through that content versus like a PDF file, for example. The cool thing is that in the package insert, which I think is a very under appreciated drug reference for pharmacy students and pharmacists, you're going to see a lot of cool stuff, like the boxed warnings are always at the very top in a box, which is how it got its name, you'll see warnings and precaution sections. You'll see an ADR section, and in fact, if a drug was approved after 2006 The FDA has much more stringent guidelines in terms of how they report some of these sections. So you're going to see very robust sections, especially in newer drugs versus some of the older drugs approved, especially before 1960s they're not going to be as good, but they can still be helpful. But for sure, newer drugs are going to have really robust package inserts, which is actually where these tertiary drug references get a lot of their information from, is the Speaker 2 16:45 PI one clinical Pearl, I think, when you're counseling patients or trying to identify what are the most important adverse drug reactions, is start with looking at the boxed warnings and warnings in precautions section. These sections are present in tertiary drug references, and often in highlight rare but serious and common ADRs that are probably or definitely associated with the drug. Speaker 1 17:10 And so sometimes these sections can be a little bit wordy in terms of all the information they give. But the nice thing is, because they give it in a paragraph form, it gives you a little bit of context of that warning. It does, for example, it doesn't just say, you know, oh, these, Anna, you know, antipsychotics. Don't use them in patients that are geriatric. And just leave it at that. It gives you a reason why, based upon, you know, this population, this study, and references it out, and gives you, again, a little bit more to it than just saying, headache, 5% there's a little bit more background. Dr. Sean Kane 17:37 And really, I think that that is one of the distinguishing factors between a patient and a healthcare provider is to be able to analyze that paragraph of data, synthesize it and then give it back to a patient in a usable format, especially if they have follow up questions on, is it okay for my elderly mother to take this anti psychotic for example, now you are at least equipped with the very basic understanding of Why is that warning there or that potential ADR there? Speaker 1 18:03 So another thing to do, then, is really out of this, is review that adverse reaction section with a skeptical eye. So the first thing to do comparing the drug ADR rate versus a comparator so things such as nausea, headache, diarrhea, etc, with statins. And as I mentioned, though we still even here, have to be somewhat careful, because the populations may be different in those individual studies, and we don't always have head-to-head studies saying we gave half the patients rosuvastatin, half of them atorvastatin, and we compared the adverse event rates. Speaker 2 18:32 I think the other important thing is to check for dose dependent adverse drug reactions. So for Topamax, for example, paresthesias were 6% in the placebo rate, 35% in patients that take 50 milligrams per day, and then up to 51% for patients that take a higher dose, 100 milligrams a day. So here, there's clearly a strong relationship between the Adverse Drug rate and the dose, which really builds a case for the drug truly does cause the ADR. Dr. Sean Kane 19:04 I think the other thing you have to think about is, based on your knowledge of the mechanism, is there a biologic plausibility that that drug can actually cause that side effect? So for example, we know that Lisinopril is an anti hypertensive. We have a good understanding of the renin angiotensin aldosterone system, so we understand why it could cause hypotension and dizziness, and we know its effects on bradykinin, so we understand why it can cause dry cough and angioedema. And apixaban — why could it cause bleeding? Should be pretty obvious, right? So I think that it's easier to accept causation with the drug if there's a clear and known mechanism of that drug, why it potentially could cause that side effect, right? Speaker 1 19:43 And the thing about that is, even if there isn't an obvious mechanism, and again, I am 100% with you, Dr. Kane, is I, you know, both with indications for use and side effects, is if it doesn't relate to the mechanism, I'm a little skeptical, but there are some medications where there isn't a good mechanism, but we do know the side effect — antiarrhythmics being proarrhythmic is one example, and atypical antipsychotics causing metabolic side effects is another. That was one that blindsided a lot of individuals. We've now learned a little bit more about how it may be possible, but that's a big one. And then one that we've already alluded to with box warnings, and we'll talk about later, is SSRIs and suicidal ideation. Speaker 2 20:17 I think the another addition to kind of think about, and one thing that's very frustrating in clinical practice, especially being in the emergency department when we're trying to evaluate whether or not a patient comes in with an adverse drug reaction, is that not all of the adverse drug reactions have yet been reported, and so it's very hard, I think, for a clinician, when you're seeing an adverse drug reaction for the first or second time, you can't always rely just on the tertiary references to identify how exactly that may occur. Dr. Sean Kane 20:47 And I think that all of us in the room have probably multiple times had the question of, does drug X cause side effect y? And to your point, Dr. Leiden, maybe we don't know. Sometimes it's a super rare thing. Sometimes it's this background noise, and that's one of the most frustrating questions to answer, because unless definitively Yes, it absolutely causes that side effect, you can never really, truly rule out a drug causing a side effect. Speaker 2 21:12 So kind of putting it all together with patient counseling, what ADRs are common and associated with the drug therapy. So don't counsel on ADRs that are more background noise, like headache, unless the drug truly does cause that particular adverse drug reaction. And I Dr. Sean Kane 21:29 think like nitrates, for example, would be a great example of this. We have really good evidence that nitrates cause a ton of headache, and it does get better with tolerance and things like that, but we know it causes headache. That's a perfect counseling point. But not for rosuvastatin, Speaker 1 21:44 yeah, no, if a drug has a box warning for an adverse event, that's definitely and that's going to be something you've got to counsel on those things. So SSRI, suicidal ideation — one I've been doing more and more of lately is the antipsychotic clozapine and neutropenia; these are the biggies. They're, you know, real problematic ones, and you gotta and really to understand the risk versus benefit of these medications, you gotta go through those with the patient. Dr. Sean Kane 22:06 They're box warning for a reason, right? They're so important, the FDA gave them a special section at the very top of that package insert. Speaker 2 22:15 Also, we wanna look at if a drug causes a rare but serious ADR or causality isn't firmly established. It's really up to you as a healthcare provider to assess the importance of including and in counseling. Unlike a drug commercial, you can't list 30 ADRs that might occur triage your time. What are the most important ADRs to discuss with a patient? Right? Speaker 1 22:38 Just again, to throw back to another episode, you know, as we talked in Episode 63 not every serotonergic drug has the same risk of serotonin syndrome. For some medications that is going to be one that you do want to discuss. But not every medication, you know, with cyproheptadine or cyclobenzaprine is going to have the same rate as a full antidepressant SSRI. Dr. Sean Kane 22:57 So hopefully this episode has at least brought to light some of the challenges associated with counseling patients on adverse drug reactions and even knowing whether an adverse drug reaction actually happens with a drug or not. A couple key concepts that I think are big takeaways from today are one, quantifying and showing causality between a drug and a specific ADR is actually really, really hard to do thinking about differences in patient baseline characteristics the study population itself, and then just baseline events of a specific ADR among a placebo group because of their comorbidities and things like that. And also kind of tricks involving trial design that can minimize some of the ADRs and how they're reported, or kind of diffusing them over a number of different ADRs can be really challenging to really quantify a specific ADR Speaker 2 23:46 another important thing to keep in mind is most clinical trials are not powered to compare the ADR rate between a drug and its comparator. They're powered to compare efficacy. So frequently when they're looking at ADRs, this could result in type two errors, where there would be a non significant P value between a drug and its comparator. But be very careful when interpreting it, because the trial was not powered to look at the difference in ADR rates. Speaker 1 24:16 So third, again, we all love some of these tertiary references, whether it's UpToDate, Micromedex, anything like that, but they can be misleading regarding the incidence rate of an adverse drug reaction, especially without that extra background. So you really to truly evaluate, and need to look at the incidence rates of the comparator, look at the incidence rates of a placebo, in order to really know does that drug cause it, and then also keeping in mind you know the mechanism of action that may be related Dr. Sean Kane 24:40 to it. Then finally, daily Med, or just the package insert, however you want to get to it, this provides an exceptional resource that I think is really undervalued by many healthcare providers and even healthcare students. This allows you quick access to the box warnings, Warnings and Precautions and the Adverse Reaction sections that give you the context of that comparator or that placebo rate. To know, does rosuvastatin actually cause headache, or was it a similar number to the placebo arm? And I think that that in itself, that DailyMed or the package insert, is one of the best tools for students to really identify what do I actually want to talk about when I talk to a patient? So with that, I think that concludes today's episode. If you'd like to see our show notes and get a link to daily med you can access us at HelixTalk.com we're also on Twitter at HelixTalk, and I wanted to give a special thanks again to Dr. Leiden for her time today. So with that, I'm Dr. Kane, Speaker 1 25:31 I'm Dr. Schuman, and I'm Dr. Leiden, and as always, study hard. Narrator - Dr. Abel 25:36 If you enjoyed the show, please help us climb the iTunes rankings for medical podcasts by giving us a five star review in the iTunes Store, search for HelixTalk and place your review there Narrator - ? 25:47 to suggest an episode or contact us. We're online at HelixTalk.com thank you for listening to this episode of HelixTalk. This is an educational production copyright Rosalind Franklin University of Medicine and Science.