 Good afternoon, everyone. Welcome to the Second Annual Combined MGH Research Institute and Harvard Center for Bioethics Research Ethics Lecture. I'm Melissa Abraham, Director of the Research Ethics Consultation Service in the Division of Clinical Research at MGH. We created this series to provide the opportunity for interdisciplinary discussions of ethical issues pertinent to the broad clinical research community and that have practical applications in the work. My colleague from the Center, Dr. Rebecca Lee, and I are so glad you are here to hear our distinguished speaker, Dr. Steve Jaffee, discuss an important topic in clinical and translational research, therapeutic misconception. Dr. Jaffee is a pediatric oncologist and bioethicist who is the Art and Eileen Penn Professor of Medical Ethics and Health Policy and Professor of Pediatrics at the University of Pennsylvania Perelman School of Medicine. He is also the Director of the Penn Postdoctoral Training Program in the Ethical, Legal, and Social Implications of Genetics and Genomics and Co-Director of the Cancer Control Program within the Abramson Cancer Center. Dr. Jaffee's research addresses the many ethical challenges that arise in the conduct of clinical and translational investigation, examining roles and responsibilities of principal investigators, return of results, and integration of genetic sequencing technologies into the clinical care of cancer patients and more. I'm particularly delighted that he has agreed to join us today because Dr. Jaffee has made many major contributions and is the leader in this field. He has co-authored over 200 articles in the area, has served on many national ethics related committees, and is an elected member of the National Academy of Medicine. He has always been one of my favorite speakers and writers in research ethics. After Dr. Jaffee's presentation, Dr. Lee will moderate a Q&A session. We encourage you to submit questions in the chat to the hosts. Use the chat feature at the bottom of the screen. Dr. Jaffee, we are so fortunate to have you speak with us today about therapeutic misconception and clinical research. What is it and why should we care? Awesome, Melissa. Thank you so much for that introduction. This is of course a bit of a virtual homecoming for me since I spent many formative years of my career in the Harvard environment. I am also grateful to everybody who has been able to join. I saw many familiar names popping up in the participant list and it is great to have you with us. I am going to talk about therapeutic misconception, which is both a perennial and a fresh issue in the context of research ethics, and one that I think all of us, whether we are involved on research ethics oversight and regulation, whether we are investigators, whether we might be patients or family members, ought to be aware of. If you give me a moment, I will share my slides and we will launch. Because this is going to be interactive, everybody who has a smartphone or context, please have your phones out because we are going to be doing some polling through the phones as we move forward. Just one moment as I share. I also am going to apologize in advance because I am hopefully at the tail end of COVID and will probably fall apart into coughing fits at various points during this presentation. I apologize in advance. Hopefully, everybody can see my slides. Actually, hopefully I am sharing the correct screen. Let me stop and make sure I am sharing the right thing. There we go. So, therapeutic misconception in focusing on clinical research, what is it and why should we care? Before we start, a couple of brief disclosures. I did, excuse me, received research funding from Pfizer through PAN up until almost three years ago. I am currently a paid member of a data and safety monitoring board for a pharmaceutical and biotech company called CSL Bering, and I am not going to discuss any off-label use of medications. I have three objectives for the next 45 minutes or so. The first of which is to define the concept of a therapeutic misconception. The second is to show you some data on the prevalence of therapeutic misconception amongst various constituencies and note I say various constituencies because I don't just mean patients or research participants, but I will argue that the problem of therapeutic misconceptions is broader than that. And then think with you about whether and when therapeutic misconceptions undermine the validity of consent to a clinical trial and what, if anything, we as investigators or others who are involved with the research process ought to be doing to try to prevent them or respond to them. Again, this is going to be an interactive conversation and I have put some polls into the presentation. Hopefully this will be a successful little experiment. And so you will see an opportunity. You'll see these instructions as you come along. But to just sort of get yourself registered in the poll so that you can then respond very quickly to each question that I put up. Go ahead and just grab your phone and text, text 237607, text the word Steven Jaffee 423. I don't actually know if it requires capitalization or not. So capitalized just in case. And polls are anonymous, so I won't know what you say. So I'm just going to give people a moment to do that. Actually, the instructions are on the next slide too. So our first poll, just to sort of get us started, how familiar are you with the concept of a therapeutic misconception? Are you not at all familiar with the concept? You've really never heard about it before. You're somewhat familiar, but you're not really not sure what it is. Or you feel like you're very familiar with it. So go ahead and respond to this poll. And I will just give it a minute and then hopefully on the next screen you will see where people are at. I see I hear people saying that they can't see the text code. So the the you're texting to 37607 37607. And you're texting the word Steven Jaffee 423. All one word and all caps Steven Jaffee 423. And then once you are registered, you'll get a thing that says back you're registered. And you just put in a B or C depending upon which of these is correct for you. There will be more of these just wanted just to get us started. So I will skip past it. All right, good. So we see people responding. You are not at all about almost half of you are not all familiar about a third of you are somewhat familiar. And a smaller group of you are very familiar. And now the answers are changing. So there you get a sense of where folks are at. Fair bit of some familiarity with the group, but lots of people who are not familiar. Okay. Again, there will be more polls. And if Melissa or Rebecca, if somebody could put into the chat, the number to see, again, 37607, and you're texting to Steven Jaffee 423. Okay. So the origins of the concept of the therapeutic misconception actually go back about 40 years and really have the roots in this paper in the International Journal of Law and Psychiatry from 1982, where Paul Applebaum, a distinguished psychiatrist currently at Columbia, and his collaborators did an analysis of people who are consenting to psychiatric research. And noted that and quoting from their paper, many subjects entering research projects in psychiatric settings carry strong expectations that the research, like the therapy they have received previously, is designed and will be executed in a manner of direct benefit to them. Further to the extent that these subjects recognize research aims, many of them see no conflict between research and therapeutic goals. And in particular, they documented that many participants in the study, or in the various studies that they were talking with did not recognize things like the fact that arms or treatments would be selected at random, that there was the possibility that they would get a placebo, that there was, there would be masking of treatment assignment, or the fact that the treatment decisions would actually be constrained by the research protocol. And importantly, this was not limited to a setting of people who are seriously psychiatrically ill. So we can't say, well, this is a phenomenon that's limited to psychiatry or people with serious psychiatric illness. Probably the more famous paper is this one from the Hastings Center report from 1987, where they returned to the issue of therapeutic misconception. And I think to find it for the first time, although I think there are better definitions out there over time, and I'll show you one. And they argued that to meet where they defined a therapeutic misconception as to maintain one, it's to deny the possibility that there may be major disadvantages to participating in clinical research that stem from the nature of the research process itself. And I want you to sort of just keep in mind this idea that there may be major disadvantages to participating in research that come from the nature of research, because it's going to be a theme that I'm going to return to in something we're going to wrestle with. Now, as I said, I don't think that this definition is helpful, but it doesn't quite nail it down for me. My preferred definition comes from a report from the National Bioethics Advisory Commission. So this is the commission that President Clinton put together during his administration. And in this NBAC report, they said the belief that they defined the therapeutic misconception as the belief that the purpose of a clinical trial is to benefit the individual patient, rather than to gather data for the purpose of contributing to scientific knowledge. And they went on to say that it is not a misconception to believe that participants probably will receive good clinical care during research. But it is a misconception to believe that the purpose of clinical trials is to administer treatment rather than to conduct research. And so the focus is really a misconception about the purpose of research, not about the likely effects of research, not about the likely benefits or risks of research, but the underlying purpose of research, which is to advance generalizable knowledge, to learn something for the benefit of future patients, rather than to necessarily benefit the individual, although again, individuals may well benefit from being in research. So coming up on another poll, so hopefully people have been able to register again, you're texting 37, you're texting 237607, and the word you're using to register is Steven Jaffe 423. So here, imagine that you are an investigator, and you're talking with an adult patient, and they have metastatic cancer, and you're inviting them to join a randomized clinical trial. And this is a trial that compares a standard chemotherapy regimen. If they're randomized to regimen A, then they're going to get something that's very standard for their cancer. Or if they're randomized to regimen B, they're randomized to the same regimen, but there's also the addition of a novel agent, which the hope is will improve the efficacy of that treatment regimen. And there, of course, are phase two trials of these combinations, and the addition of the novel agent appears safe. And it shows some promise of efficacy and phase two trials. But as we know, phase two trials don't usually give us the definitive answer about safety and efficacy. And this patient who has this decision making capacity, wishes to consent to the trial. But despite your strenuous efforts, the patient appears to maintain an unshakable belief that the purpose of the trial is to benefit her. And again, I'm emphasizing this is a patient who has decision making capacity, and you have tried really hard to shake this belief, and yet you have not been able to shake it. So as the investigator, and I'm sorry that this is a format that doesn't show you the, the sort of texting instructions at the top. What should you do? Should you A, okay, I've tried my best, this is a patient who wants to join the trial, they have decision making capacity, I'm going to accept their consent, if that's your answer, put an A. If or do you say, well, they are operating under a therapeutic misconception. And therefore I'm going to decline to enroll in the trial, I'm going to say to them, I'm sorry, I can't enroll you, or C, you're not sure. So go ahead and answer that question, I'll just give you a moment. And then we'll start to see what people are seeing on the next screen. Interesting, a lot of not shores. Significant minority would say accept the consent, and the next biggest group would say decline to enroll the patient in the trial. So you can see that there is both uncertainty and disagreement amongst those who are in the audience about what you would do in response to this challenge. Now let me give you a different scenario. There's the final. Imagine instead that instead of this being a patient who's being offered a randomized controlled trial, you're inviting a patient with multiply relapse cancer to take part in a first in human phase one trial of a novel agent. This is a single agent. This is a first in human trial. Typically these are dose escalation studies looking for the appropriate dose and looking at safety of the drug and not yet looking at efficacy. So this is a phase one trial and this is a patient with multiply relapse cancer. And again, this is a patient who has decision making capacity. They wish to consent or he wishes to consent. You've made strenuous efforts to disabuse him of his belief that the purpose of the trials to benefit him, but you still get the sense that that's really what he believes the purpose of the underlying purpose of the trial is about. So here's your poll. Hopefully everybody is registered and A, B or C. You'll accept the consent. You'll decline to enroll him in the trial or you're not sure. A lot of not shows. I don't know what the underlying numbers here are, but the single group here, unlike the last one, interestingly, is to decline to enroll him in the trial. But there are some who would ultimately accept his consent and many who were not sure. So again, you can see that there's a lot of uncertainty even amongst people who think about these issues a lot, as I'm sure many of you do, about what to do in a difficult situation like this. Okay, so to find for you the therapeutic misconception, but I want to distinguish it from some related concepts with it, with which it's often confused. And I think this is a very nice and very important paper from Sam Horning, who at the time was a host baccalaureate fellow actually at the NIH, along with Christine Grady, who's now the director of the biothic center at the NIH or biothics department. And they distinguish three concepts that I will define for you in a moment. The first of them being therapeutic misconception that we've talked about. Second being what they called therapeutic misestimation. And the third being therapeutic optimism. So according to Horning and Grady, and I'll sort of hone in on each of these. These are three distinct concepts. The first one, you should already be familiar with therapeutic misconception, where the research subject is conflating research with clinical care. They argue that it's rarely tolerable, because in order to make an autonomous decision to participate in research, you have to understand what the nature of the research is. And if you have a therapeutic misconception, then you don't, and therefore you shouldn't potentially be able to join the study. Perhaps you should be prevented from being in the study. At least I take it that that's what they mean by rarely tolerable. I take it that they mean something stronger than we should simply try to avoid it or try to work against it or try to disabuse it. They distinguish that from therapeutic misestimation. For those of you who joined late, I'm apologizing for my cough. This is the lingering effects of COVID. The therapeutic misestimation is where the research subject underestimates their risk, overestimates their potential for benefit from a study or both. And according to Horning Grady, this is sometimes acceptable, because understanding the exact probability of the harm and benefit might not be necessary for an autonomous decision to participate in research. And you can think about whether you agree with that statement or not. In other words, if I massively overestimate my potential for benefit or if I massively underestimate my risk, am I making an informed and autonomous decision to join the study? And finally, they distinguish that from therapeutic optimism, which is they define as hope for the best personal outcome. But I think actually probably better would be to say I'm like overly optimistic about the chances that I might benefit and their definition of it towards the end. Thomas hopes he will be one of the five percent who benefit from the phase one cancer trial. If he said, will Thomas really believe that he will be one of the five percent? Of course his statistical odds are precisely five percent, but if he really believes that he's going to be one of the five percent and there's a 90% chance he's going to benefit, then he's being overly optimistic. And the question that we are left with is whether that is compatible with him giving informed consent to that study. And we'll come back to that. Now data, to turn to the data on the therapeutic misconception, I think the prevalence of them is sort of contested and not clear in the literature. And I think that is partly true because it turns out to be quite hard to define and it turns out to be hard to turn that definition into a clear measure that everybody can agree on. That has been a challenge in therapeutic misconception research for the past 20 years. So I told you that Applebaum and Litz were some of the early sort of architects of the concept or sort of founding parents of the concept. They published this paper back in 2004 in the journal IRB as an effort to kind of come up with a measure and then to measure its prevalence. They constructed a two-part definition or two different sort of dimensions of therapeutic misconception. One that they call TM1, which they defined as an incorrect belief that they're meaning a patient or a research participant's individualized need, will determine assignment to treatment conditions or lead to modification of the treatment regimen. This is an incorrect belief about the sort of personalization or personalized nature of clinical research. And the second which may actually be more like therapeutic misestimation as defined by Horning and Grady is an unreasonable appraisal of the nature or likelihood of medical benefit from participation in the study due to a misperception of the nature of the research enterprise. And then in order to assess the prevalence of therapeutic misconception, both sort of the TM1 and TM2 flavors, they interviewed 225 participants with a whole bunch of different diverse diseases. So these were not anymore all psychiatric patients from two different academic institutions. And according to their judgments based on these qualitative interviews, they believe that 31% showed evidence of TM1, 51% showed evidence of TM2. And if I should have put this on the slide, but if I remember correctly, it was about 60% that showed evidence of one or the other or both. So they argued that most, certainly not all, but most research participants show some evidence of therapeutic misconception. Now in my view, TM1, their definition of TM1 is closer to what I consider a therapeutic misconception. So maybe the prevalence is 30% if you're more stringent in that way. And they concluded the majority of the participants in this study manifested some degree of misconception about the extent to which entry into a clinical research project would be likely to benefit them directly or would allow individualized decisions to be made about their care. As a result, they may have failed clearly to appreciate the risk-benefit ratio of the research project to which they were being asked to consent. And the implication may be, and I just want to emphasize this last sentence, is did that mean that although they consented to the research, their consent was not valid. And if to take that to its logical conclusion, maybe even shouldn't have been accepted. And I'm not going to argue that their consent should not have been accepted, but I think this is the question that they are raising. Now they went further in a follow-up study and published in 2012, some of the same authors, plus some others, where they attempted to develop and validate not a sort of qualitative interview-based measure, but now a quantitative sort of short survey measure. And so what they did was they administered a 28-item survey to 220 trial participants, again a diverse range of underlying conditions now at four academic medical centers. And their goal was to access the presence of beliefs that were associated with therapeutic misconception. And from this initial 28-item survey, they then did some statistical psychometric analyses that allowed them to exclude 18 items that weren't contributing to the overall measure, leaving them with a 10-item survey, which they then compared the results of against this gold standard interview that they also did as part of the study. And that allowed them to say what is the sensitivity, specificity, performance characteristics of this measure and how do the quantitative and the qualitative compare? So again, remember they had also done an interview and based on this interview, 51% of participants manifested a therapeutic misconception, either the TM1 flavor, the TM2 flavor, or both. Now remember they were also, they came down to this 10-item survey, and this survey successfully distinguished between those who manifested the TM on the interview, those scored a mean of 33 on the survey. And if I remember correctly, the survey could have gone up to 50. So it was a mean of 33 out of 50. And those who did not manifest a therapeutic misconception on the survey, whose mean was 23. So you can see that the mean scores of those with and without a TM on the qualitative interview were very different. They did some other statistical analyses and said based on our optimal cutoff in our survey, 55% of our sample showed evidence of a therapeutic misconception based on the survey. Now you could choose a cutoff at a higher or lower level and that would change the estimate, but this was their best estimate is that roughly half or a little more of their sample showed some evidence of a therapeutic misconception. Now this, their work has not been uncontroversial and others have taken issue with whether their measures and in fact in general measures of therapeutic misconception are really capturing the real thing. So an example of that is this work from Scott Kim, who is now at the Department of Bioethics at the NIH Clinical Center. And they've done work in a range of different populations. This is a particular paper from patients with amyotrophic lateral sclerosis where what they did was a web survey about a hypothetical first in human clinical trial, but with real patients who all were suffering from ALS. And in giving the sort of hypothetical consent process to this hypothetical trial, 10 of the 72 patients said that this study that is mostly intending to help me and that's I think a reasonable statement of therapeutic misconception. Now obviously that's a much smaller estimate than what Applebaum and colleagues were coming up with. But what Kim and colleagues then did was ask a number of follow questions. And most of the people who gave this response, the study is mostly intending to help me, were actually asking a question or answering a question that they thought was about their own motivations for joining the study. And if I recognize that the study is about answering a testing hypothesis, answering a research question, but my motivation for joining it is to help myself. I'm not under a therapeutic misconception. Now there was a bit of a back and forth about this paper and other papers from the Kim group. And Paul Applebaum and his colleagues were critical of it. And just a quote from their criticism, like any empirically derived construct, therapeutic misconception should be and has been subjected to testing and the possibility of disproof. For us to have any confidence in the results, however, these effects should involve appropriate subjects in realistic research settings with valid measures interpreted fairly. The study by Kim et al disappoints on all three counts. And they criticize it. And there's certainly some validity to this. Obviously, testing therapeutic misconception in a hypothetical setting is not nearly as good as testing it in a real setting. And Applebaum and colleagues also criticized them for their, what they said was an oversimplified measurement instrument. And in particular, their failure to include measures of overestimation of benefit. But I hope I've convinced you that whether overestimation of benefit is part of therapeutic misconception, it certainly is as the measures that Applebaum colleagues have developed include it or whether it's a separate concept as sort of articulated by Horning and Grady I think is something that is still under discussion in the field. Finally, I want to show you a study that asks, can we reduce therapeutic misconceptions? This is a randomized trial that Applebaum was involved in. Paul Christopher is a psychiatrist now at Brown led on reducing therapeutic misconception. Of course, this one was in a hypothetical clinical trial, so potentially subject to some of the same criticisms that Applebaum and colleagues leveled at the Kim work. Here what the Christopher and colleagues did was they recruited 154 patients with diverse diseases to consider a hypothetical clinical trial that was specific to their disease. So the cancer patients got a cancer trial, etc. Patients were randomized to either a standard consent disclosure with or without what the investigators called a scientific reframing intervention. And what this reframing intervention included was a description of how the purpose of the research is to test whether the experimental treatment is better than the control carefully described randomization made some explicit descriptions of limitations that investigators have on their flexibility to modify treatment talked about the fact that the trial involved masking or blinding in the rationale for doing that and emphasize that these procedures were there not to improve the care of the participants in the trial, but rather to enhance the validity of the research answer. And here is the I think the main figure that shows the sort of effects of their intervention. This shows it by different levels of baseline education of the participants. But you can see that the control group had across educational levels, TM scores on their measure that were higher by a difference of an average of about six points than those in the scientific reframing group here in the solid line. And so it does suggest that the intervention was effective by a six point statistically significant amount in reducing therapeutic misconception amongst participants pretty much across all educational levels. Although it seems like there might be a greater effect at higher educational levels that did not sort of pan out statistically. The other thing that is really notable here is the fact that therapeutic misconception seems to be much more common, whether both in the intervention and the control group amongst those with lower baseline levels of education and much less common amongst those with higher levels of education. And I think that that has been a consistent finding across multiple studies. One question that I've always had about that is how much of that is because people with higher levels of education are better at interpreting the question that they're asked on the survey versus are they really operating under less of a therapeutic misconception. And that's a question that I don't think the literature gives us the answer to just yet. All right. I have focused on therapeutic misconception. We do know that therapeutic misestimation is very common, certainly particularly in the sort of phase one trial setting where it's been best studied. And so quickly this is a paper from now almost 20 or 20 years ago from Kevin Weinfert at Duke and colleagues where they surveyed 328 adults who had been offered participation in a phase one trial at five different major U.S. cancer centers. Of those 328, 79 percent chose to enroll, but the respondents to their survey included both enrollees and non-enrollees. And so the question that was asked is what's the probability that the treatment in this clinical trial will control your cancer? Again, this is not hypothetical. This is a real trial. And here are the results. And the thing I want to point out to you is first of all, the reality is down here, zero percent to five percent, maybe 10 percent is a high, but not completely unrealistic estimate. But you can see a very large number of people, 25 percent said 50 percent chance that it will control my cancer. And here's a bunch of people saying 70, 80, 90, or even 100 percent. And these are if people really believe that the chance that the cancer or the trial will control their cancer is 80 or 90 percent, then they are operating under a profound therapeutic misestimation, whether it's a misconception or not, I think is less clear. So get your phones out and hopefully everybody's registered for the poll. Imagine that you're the investigator for a phase one cancer trial of a novel single agent. And during the consent discussion with a patient, and this is a patient who you have no doubts and baseline about their decision making capacity, you learn that they believe that the chance the phase one trial will control their cancer is over 50 percent. And as I mentioned on a couple of slides ago, meta analyses suggest that response rates in single agent phase one cancer trials average about five percent and are usually pretty short lived on the order of months. It's rare for people to have responses that extend longer than that. So your which of the following would best describe your preferred approach? One is your attempt, of course, to correct the patient's misunderstanding or misestimation of the likelihood of benefit it. But unless you're successful, you don't enroll them in the trial. In other words, you have to be able to correct it else. You're not going to enroll them. You're going to say to them, sorry, I can't enroll you. Second option is, of course, your attempt to correct their misunderstanding. But ultimately, you agree to enroll them even if you're not successful at correcting it. And the last possibility is you're just unsure. This is too hard. And I recognize that it actually is very hard. So let's see what people are. The largest, the majority of you believe that people should not be able to, if they're really under this much of a misestimation of what the likelihood of benefit is, and you know that as the investigator, you should not be enrolling them. But there are some who would be willing to enroll them, even despite this having made a good faith effort to correct it. I am, by the way, not going to suggest that there are clearly right or wrong answers to these difficult questions. Okay, we've talked a lot about patients and research participants. What about investigators? Do we as investigators contribute to a participant's therapeutic misconceptions? Here, Sam Horning again from his time at the NIH with Christine Grady as the senior author and other colleagues looked at consent forms for phase one oncology trials. They abstracted consent forms from 272 clinical, phase one trial sponsored by the NCI as well as major pharmaceutical companies. And in their conclusion, but I'm going to show you the data, they concluded consent forms for phase one oncology studies almost never promised direct benefit to subjects, rarely mentioned cure and usually communicate the seriousness and unpredictability of risk. Although there's room for improvement, the substance of these forms is unlikely to be the primary source of misunderstanding by subjects in phase one oncology trials. So they were pretty reassured by their own interpretation of the data. And I can see why if you look at the first line here, almost all of them said explicitly the study is research and in most cases that was a pretty prominent statement and almost all of them said that this is a dose escalation safety or toxicity study as opposed to presenting it as an efficacy study. However, only fewer than half clearly differentiated research procedures from clinical care procedures which in my view allows therapeutic misconceptions to take hold and almost all of them refer to the investigational agent as treatment or therapy without sort of qualifying it as experimental investigation or a research intervention. So there's some therapeutic misconception even embedded in the language of the study. In the interest of time, I was going to do another poll but I'm going to quickly skip it just in the interest of time to suggest to you that consent forms may sort of facilitate therapeutic misconceptions. I also think that investigators or at least oncologists and physicians harbor therapeutic misconceptions and this harkens back to some work that I did when I was at Dana Farber with my mentor Jane Weeks 20 years ago. What we did was we mailed surveys to a stratified random sample of over 1,100 oncologists who were divided into three groups. There were the medical adult oncologists, the pediatric oncologists and the other group included radiation oncologists and surgical oncologists and GYN oncologists and others randomly selected from the major cancer society director directory. And we had about a 50% response rate which for those of you who've tried to survey physicians is actually pretty good. And we asked two questions. One, there are many reasons why physicians may enroll individual cancer patients in clinical trials. Thinking about the patients to whom you offered clinical trial participation over the past year or were the main reasons why you offered enrollment. And then as a separate question from a societal point of view what are the main purposes of cancer clinical trials? In considering your reply please consider general purposes that are true across phase one, phase two and phase three trials. So focusing first on the main reasons why you as the physician or investigator enrolled individual patients you can see that the largest group said ranked as their first choice reason to improve the treatment of future cancer patients which is obviously not a response that's consistent with therapeutic misconception ranging from 87% of other oncology specialists to 53% of medical oncologists gave that as their answer. At the same time I shouldn't say that at the same time down here the largest actually I read your percentages wrong but the largest group of pediatric oncologists said I did it to ensure that my patients get the most data of the art treatment a little fewer than half of medical oncologists gave that response and only a quarter of other oncology specialists gave that response. So you can already see some differences across oncology specialties in the views about why people are enrolling individual patients and trials but I think the most important part of our study was the response to this question what are the main societal purposes of clinical trials? And here what I want to point out is I think hopefully you would agree after listening to me that the right answer or the best answer to the question what's the main set of purposes of clinical trials is something like to improve the treatment of future cancer patients and you can see that 73% of medical oncologists agreed with that about 60% of pediatric oncologists and over 80% of the other oncologists in the survey agreed with that but a surprisingly large number especially of the pediatric oncologists responded that the main societal purpose of clinical trials is to ensure that cancer patients or trial participants themselves get the most state of the art treatment and we were surprised at the sort of prominence of that answer given what I think is a certainly in the research ethics world a consensus that the reason we do trials is to answer questions to advance science and to advance treatment so even though these are minorities of the totals they're still pretty large groups or large minorities especially amongst the pediatric oncologists and if we are operating under these kinds of therapeutic misconceptions or at least some of us are we communicating these or fostering these for our patients many of you are probably familiar with the concept of clinical equipoise this is one that is frequently invoked to justify the ethics of a randomized trial it traces back to a biothesis named Benji Friedman who is from Montreal who basically argued that randomized trials are ethical if a state of clinical equipoise exists and he defined clinical equipoise as there's no consensus within the expert clinical community about the comparative merits of the alternatives in a trial to be tested so thinking about clinical equipoise the definition that I just gave you again no consensus within the expert community about the comparative merits of the alternatives to be tested and therefore it is okay to go ahead and randomize or offer randomization to participants between these two alternatives which of the following best describes your view a randomized trial must satisfy clinical equipoise to be ethical b a randomized trial can be ethical even if it does not satisfy clinical equipoise or see you're not sure again a very interesting split and this is I think quite common in research ethics some of our most sort of challenging or important concepts are ones that different people can very much disagree about so there's your final and you can see half of you thought it's possible for randomized trial to be ethical even if it doesn't satisfy clinical equipoise but three percent of you thought otherwise Frank Miller and Howard Brody to bioethicists wrote in the Hastings Center report now about 20 years ago that argued that therapeutic misconception may be deeply embedded in the underlying fabric of clinical research and use clinical equipoise as an example of why they thought that even those of us who are sort of think about research ethics all the time may be operating and maybe using concepts that are inherently sort of filled with therapeutic misconception and here's their argument clinical equipoise is a widely although certainly not universally endorsed standard for ethical research and the underlying basis of clinical equipoise is the assumption that clinicians must satisfy their therapeutic obligations to patients when they're offering or conducting research that in order to ethically offer or conduct research you must first satisfy your therapeutic obligations but they point out that there's sort of two problems in practice with this one is that sponsors routinely propose and IRBs routinely approve placebo controlled trials even when a known effective therapy exists now if that known effective therapy is sort of life or limb saving is not going to be approved but if it's a more symptomatic condition or something that is more mild it may well be approved and secondly setting aside the treatments that are being studied in the trial trials routinely include procedures like biopsies like lumbar punctures like drug withdrawal or washout phases that impose risk and burden on the research participants that are done in the interest of science or in the interest of validity and clear answers but don't have a compensating therapeutic justification and so on the one hand we say we require clinical equipoise which implies a commitment to therapeutic goals on the other hand we include things in trials or we allow that trials to move forward that are inconsistent with pure therapeutic goals but that include things that are motivated by research ends and so Miller and Brody say this is incoherent our practices like these on the slide don't line up with our principles at least if we believe that clinical equipoise is ethically important and I know that for at least half of you you believed it maybe at least you believe it's not necessary maybe it's important or a good thing in general but not necessary so Miller and Brody argued that a better approach would be to accept and particularly if we're going to avoid therapeutic misconceptions is to finally acknowledge and accept that the ethics of clinical medicine and the ethics of clinical research are different so for example in the Belmont report I'm quoting from it here the term practice refers to interventions that are designed solely to enhance the well-being of an individual patient or client and that have a reasonable expectation of success clearly distinguishing in the Belmont report from the term research which designates an activity designed to test hypothesis permit conclusions to be drawn and thereby to develop or contribute to generalizable knowledge right so research and practice are different they have different aims they require different oversight and presumably they have different ethical permissions associated with them and if you think about the common common rule the common rule does not require that every risk be justified by some sort of compensating potential for therapeutic benefit to the participant quoting from the common rule and IRB in order to approve a study must find that risks to subjects are reasonable in relation to anticipated benefits if any to subjects and the importance of the knowledge that may reasonably be expected to result so at least for adult patients or adult research participants you can impose risks or offer include risks to the participants that are justified not by benefits to them that would be the kind of therapeutic model but rather by benefits to science importance of the knowledge also important to note that the rules for pediatrics are different and more restrictive than this and in fact in pediatrics almost all significant risks or all significant risks have to be justified by potential for benefit to the participant to the kid themselves and finally just to give you a couple of examples of therapeutic misconception being in the sort of deep in the fabric of research here is a news article from the American Society of Clinical Oncology and I just want you to think about whether there is a sort of implicit therapeutic misconception in this title of the of this article when joining a cancer clinical trial is your last treatment option should we think about a cancer clinical trial as a last treatment option now actually if you read the article itself it's actually quite good and I don't think the article is infected with a therapeutic misconception in the way that the headline of the title is but this is an example of something you see quite commonly and finally finally before I talk about what we as investigators or research staff might do about therapeutic misconceptions this is the national cancer national comprehensive cancer network which is one of the places where oncologists look for authoritative guidelines about treating different types of cancer and I want to point you to this statement which appears in every one of their many many treatment guidelines and CCN believes that the best management of any patient with cancer is in a clinical trial and to me when I read that first of all it's a there's a question of whether it is empirically true the implication is you're likely to have better outcomes if you join a cancer clinical trial than if you don't but also whether there is a kind of implicit therapeutic misconception in that statement statements like this tend to be a lot more common 20 years ago I think with attention to to research ethics they have become less common but the NCCN still includes them in all its guidances all right just a few more minutes does the does the presence of a therapeutic misconception preclude valid informed consent and what should investigators research staff research nurses research coordinators what should we all be doing to try to minimize therapeutic misconceptions and avoid them so I hear I harken to a bit of an article that Frank Miller and I wrote in the Kennedy Institute about the external about 15 years ago with a response from Paul Applebaum and chocolates to the view that we took and one of the things that to sort of form the background to our article is the view that the stakes of research consent vary amongst trials sometimes the stakes of joining a study as opposed to doing something different or very high other times they're much lower so some different scenarios which I think have different stakes implicit in them one might be it realistically a placebo controlled trial of a new anti-depressant for moderate depression second would be a phase one trial of a new anti-cancer agents like some of the things that we've talked about third is a trial of a new anti-apoleptic that includes a lumbar puncture because of the sense that we need to measure drug levels in the CSF to look at penetration and a fourth might be a comparative effectiveness trial of two approved anti-hypertensives either one of which you could reasonably get for your high blood pressure outside of the trial so in the first of these there's the possibility that you might be giving up something that might help your depression resolve more quickly second involves a drug with at least statistically a low likelihood of benefit but with risks and burdens based on meta-analyses of prior phase one trials the third obviously involves a procedure that is painful that offers benefit to science but not necessarily benefit or not benefit to you and the fourth means you would get one or two drugs either one of which you might have gotten anyway so why might therapeutic misconceptions to any one of these studies threaten the validity of informed consent well in their response to frank and me Paul, Applebaum and Chuck let's talk about a dignitary disadvantage the idea that if you don't understand the purpose of the research you're not able to make a meaningful choice about research participation there's something intrinsic about understanding that purpose they also talk about the fact that if you one of the things and joining a study usually involves these constraints on personal care because the protocol specifies so much about how treatment will go and if you are under a therapeutic misconception you're likely to fail to recognize those constraints on the personal care that you might have assumed would be there otherwise a third might be that by having a therapeutic misconception you might be unable to recognize what you're either giving up or risk that you're accepting or benefits that you're foregoing by joining a trial and a fourth might be a substantial possibility that you would have made a different or might have made a different decision in the absence of a therapeutic misconception of course the truth could be any combination of those my own view is that number four here clearly threatens the validity of your consent number two and number three threaten the validity of your consent when the stakes of the consequences are high enough of your decision which probably means that they kind of devolve into number four I'm especially concerned and I'll talk about this in my last couple of slides about circumstances where investigators fail to take any steps to counter therapeutic misconceptions and even more so and the things that investigators and research teams do contribute or encourage contribute to or encourage therapeutic misconceptions and in my approach to this I think on the one hand there's a balancing act to be done because on the one hand we don't want to police people particularly people with decision making capacity I think we have to be careful about policing their informed consent to the level that becomes disrespectful to them making decisions that they want to make for themselves but at the same time professionals working with them we have an obligation to help them make decisions that are consistent with their values and priorities and that's the tension that we find ourselves in in helping people to avoid therapeutic misconceptions so just to wrap up what are our when I say investigators I should really be talking about all research team members responsibilities regarding therapeutic misconceptions number one always be clear on our own minds about the scientific purpose of trials and sort of disabuse ourselves of our own therapeutic misconceptions number two clearly communicate what that scientific purpose is that research question purposes to potential participants number three never contribute and this is perhaps my most important point never contribute to or encourage therapeutic misconceptions on the basis or on the part of research participants number four help people to see those elements of the trial that are specific to the research that are unique and incremental and part of the experimental design and number five highlight the things that participants are either giving up from the care that they would receive otherwise or accepting risks burdens etc by joining the trial these are some of the most important things that people need to be aware of when they join a study when we think about trials that impose either significant incremental risks or burdens on participants I think we are obligations sort of ramp up beyond that I think we have an obligation to query understanding of key elements of the trial including things related to therapeutic misconception like the purpose of the trial like the incremental research risks and burdens of the trial when we identify understandings I think we have an obligation to attempt to correct them and in some cases and I think the hard question is what are those cases if we're unsuccessful after really vigorous efforts to correct those misunderstandings we have to decline to enroll the patient in the trial sometimes because this raises questions does this okay I thought this patient had capacity but maybe they don't if I'm not able to help them understand this thing that is so important to the consent and so maybe a capacity evaluation becomes appropriate so what underlies my view I think we as research team members are absolutely responsible for our own actions we can't always control how other people make decisions and sometimes attempting to do so might even be disrespectful of them or inappropriately paternalistic but when stakes get high our professional responsibilities mean that we need to ensure people understand the consequences of their decisions before we accept their consent so to summarize therapeutic misconception is a misunderstanding about the underlying purpose of a trial it's distinct from though it's often conflated with misunderstanding of the harms and benefits of the trial it has proved difficult to both define and measure but it's likely pretty common and I hope I convince you that it affects not just research participants but many of us as well and finally it's our job to never encourage therapeutic misconceptions and further to strive to identify and correct them and rarely to go so far as to decline to enroll people who have excuse me who have persistent therapeutic misconceptions with that I will thank you and unshare my slide and take a sip of water and look forward to the discussion well thank you so much Steve that was a really interesting talk we do have a number of questions and I'll start off with the first one a few of these you already answered through your through your lecture so the first one is let me ask you how does the misconception change when as so often is the case the PI is the subject's primary caregiver or perhaps the participants oncologist for example great question not a lot of good data on whether the like who the investigator is or who the PI is in relation to the patient sort of influences therapeutic misconceptions one might sort of imagine okay somebody like new that I've never met before is coming into the room to invite me to join a trial it might help me to signal that this is like a different thing than my usual care as opposed to I've been seeing this oncologist for the past five years there's been a change in the status of my cancer and now they offer me you know just that sort of what what might look to me just like the next treatment in line so I think that when the oncologist or the physician is the patient's physician is also the PI there's sort of an added responsibility on them to like help the participant or help the patient to see the okay now we're talking about something different than the usual kinds of treatments that I've been offering you all along and but I think it's the responsibility of all physicians or all investigators everybody who's doing the offering to sort of make that clear anyway so it's a it seems like it would be a risk factor but it would be but the data aren't really and I would also say that you know two things one is there's some evidence and actually we'll have a paper coming out about this hopefully before too long that patients and here this is from research on the cancer setting actually would prefer that it be their own oncologists that offer them a trial which you know maybe operates from a sort of point of therapeutic misconception and also some other evidence although you know somewhat preliminary that when you're when the person is offering you the trial as your oncologist is actually the PI you may come away more informed because they actually may be in a better position to like educate you about the trial than than others who may not have the same relationship with you may not have the same ability to like talk to you about what this means for you because they know you so I think even if there is a sort of therapeutic misconception related upside to having somebody else do the consent there may be other equally or more important downsides equalizing effects of course that makes sense and then you've offered some excellent points to ponder questions for the consent process one question came up with is should certain questions related to TM be added perhaps to the routine consent process or should that routine consent process even be changed in some way to mitigate therapeutic misconception what are your thoughts I think the there's two parts of that right one is particularly when the okay so let me start at the like lowest state's end of the spectrum right that comparative effectiveness trial to comparing two treatments either one of which would be very reasonable for you and not a lot of strong reason to prefer one over the other like spending a lot of extra sort of time and energy on issues of therapeutic misconception probably is not a good use of the most importantly the patient's time but anybody else's time either so I'm not so stressed about it then and I'm not sure I would do a whole lot different you know clearly stay with the purpose of the research is we're person between these two drugs we're trying to find out if either one of them has advantages if you join the study when you flip a coin language or something else we'll you know we'll decide at random how which of them you get all of that yes I'm not sure a lot more than that is needed when the stakes are higher when it's a you know a novel drug when it's a placebo control trial when it's an early phase trial when yeah let's say it's a surgical versus medical trial something like that when it really does matter then I think a very clear description of you know we want you to understand this is what we are trying to get it in the trial this is what we're not sure about maybe start there like here's what we don't know and here's what we're trying to find out and then some queries about you know sort of a teach back approach to trying to like okay you know you know tell me tell me what you understood about why we're doing this trial tell me what you understood about how being this trial is different than not being in this trial or I think valuable things to do and there is some decent evidence that teach back type approaches are helpful in general I would only add and this actually touches back on your question about the investigator or the the primary physician being the one who seeks the consent that I think that there is pretty good and consistent evidence that I kind of like third party in the consent process who is sort of independent of the not independent but who's distinct from the investigator is a really helpful way of getting to better consent outcomes and better consent outcomes I mean in terms of understanding most of that literature sort of looks at research nurses but I think research coordinators can fulfill the same role and there's a lot to be said for a somebody who's not the treating clinician but whose job it is to help educate the potential participant the patient about the trial I think there's a lot to be said for that model again especially in higher stakes trials of course and there's quite a few papers that advocate for that approach as well yeah so an interesting question around marketing and whether marketing efforts of hospitals or potentially cancer centers promoting that they offer perhaps early access to new therapies in clinical trials could potentially contribute to therapeutic misconception this is this a good marketing ploy does it work and does it contribute to this overall therapeutic misconception so I strongly suspected works because otherwise they wouldn't do it or at least they believe it works and I think you know if I can certainly understand if you have something that has either no I mean you've got some you've got a condition that is serious and you know life threatening or function threatening or you know you're sort of really suffering from it and either there are no sort of established effective treatments or you've tried them and they're not working for you you know it is perfectly reasonable to say well is there something out there that may not yet be sort of have met the standard of quote proven but maybe there's enough evidence or there's enough reasons to think that it might even be effective for me or might help me and I'm going to go looking forward and if that means that I'm going to go looking at a sort of academic center or something like that I'm going to have to travel for like I can't argue with anybody who would do that I probably would do it myself if I were in that situation and so I think there's a sort of careful dance between like not being misleading in your framing of what those things offer and at the same time saying you know we we offer things that might be of benefit to you so it's a subtle dance of how to frame things in a way that is accurate because you also don't want to be like so restrictive on how you tell people that they can frame things that they can't you know even sort of tell it like it is so a challenge of course and you know IRBs are overseeing these as well so hopefully they'll be catching some of those yeah I don't know I mean I guess they're overseeing the sort of trial specific marketing but they're probably overseeing the kind of like okay we're a one of the world's leading cancer centers that offers all kinds of cutting-edge treatments in clinical trials right they're not overseeing that general statement that's that's right yeah of course so we have a few that have come in looking at asking about pediatric research so one one is asking about the difference putting the difference in required therapeutic effects in pediatric research compared to adult research account for some of the differences in how pediatric oncologists perceive research goals any thoughts about that if the question is does the fact that our regulations sort of constrain us to be like to ensure that their potential therapeutic benefits justify potential risks does that explain what like pediatric oncologists are saying I think the answer to that has to be no I think first of all I'm not 100% sure that all our most pediatric oncologists actually know that and secondly I think what's going on there is that you know going back to the 1960s there has been this culture of you know it's a unique symbiosis and integration of clinical research clinical trials into pediatric oncology practice and I mean it's on wonders for advancing the field of pediatric oncology and the treatment of kids with cancer right I mean the sort of pace at which most pediatric cancers have gotten to be more effectively treated is remarkable and that's because of the clinical trials enterprise and the cooperation that's gone into it but it has also meant that the usual sort of like because it's so integrated into the fabric we haven't had to have this sort of mental exercise about what are the boundaries but you know okay now I'm doing research now I'm doing treatment it's just you know it's sort of like standard of practice to offer clinical trials is kind of like first line for every patient who walks in the door and I think that is not the case in virtually any other area of clinical medicine or research medicine okay thank you and I think that certainly makes sense and I think you've answered this in some respects is that as a clinician have you found that informed consent conversations you've had about TM differ in some cases such as end of life versus early onset of a more curable disease do allow room for perhaps more TM in certain circumstances than others you've talked about comparative effectiveness research research for example if if NIH were to run a Moderna versus Pfizer vaccine study perhaps there might be more room for TM perhaps not but you've kind of alluded to that versus end of life situations or more critical cases in in cancer for example my you know there's an irony here because in end of life and I speak mostly from my own cancer experience but in in sort of decisions of last resort if you want to call it that or end of life cases where people are saying well I have no other treatment options sometimes people people look at clinical trials as treatment options of last resort you kind of saw that in that ASCO news article that I showed you in the headline and so in some sense it actually like that's setting actually lends itself to just sort of letting the therapeutic misconception play out which I think is very problematic right this is really a time when we need to be as clear as we can be that we're like testing a question we're trying to figure out like what's the right dose of this should we be doing it at 100 milligrams or 200 or 300 milligrams and what are those sort of basic common side effects of this drug and the chances that you will benefit or at least that you will experience sustained benefit are actually pretty low in fact they're very low and yet the perception is in those kinds of cases of people having very little choice whereas actually they should be understanding that they really do have a choice and that they're making an active choice interestingly in the sort of earlier phase or earlier stage of cancer when there actually are choices I could get standard treatment or I could join this trial and then there might be this randomization between standard treatment and something different they're actually everybody perceives that there is a choice and so that allows you us to kind of like avoid therapeutic misconceptions because we can clarify there's a choice between being in the study and not so the circumstances lend themselves to like accepting worse or going with the therapeutic misconception in the end of life setting where to me that is probably the worst place to kind of just let it run yeah absolutely and the data on the phase one oncology trials is just remarkable certainly interesting so perhaps this should be the last question given the time but would it be preferable to assume therapeutic misconception is present unless shown otherwise I mean better to decline a subject who might have understood rather than enrolling on who does have therapeutic misconception I want I mean everybody who's listening I would encourage you to think about okay you're talking with a patient who sees a trial as something that might be good for them or might be good for their kid and would like to join that trial and sees things in that trial that they would like and you say to them I'm sorry I can't allow you to join this trial because you seem not to understand the following like just imagine yourself seeing that there might be some times when it is appropriate to say that but I find it very hard to say that and in my own experience I can only think of two times when I said to a parent you know I'm sorry I can't offer you this trial anymore based on our conversation and we have to like go back to just offering standard therapy neither one of which was precisely because of therapeutic misconception it was because of other aspects of misunderstanding or sort of challenges in wrestling with making the decision on behalf of their kid but that's not an easy thing to say to somebody I'm sorry I can't let you be the decision maker either for your kid or for yourself feels very disrespectful and that's why I've argued that it probably ought to be used pretty rarely thank you I mean very insightful and wonderful talks Steven and thank you for being here today all right this is fun thank you thank you and be well bye bye