Welcome, everyone, to the first annual Meeting Symposia, Combination Treatments for Stimulant Use Disorders, a Promising New Direction. Dr. Nunes is the moderator and chairperson. Dr. Nunes is Professor of Psychiatry at Columbia University Irving Medical Center and New York State Psychiatric Institute. He is multiple PI of the Greater New York node of the NIDA Clinical Trials Network and co-director of the Chosen Center at Columbia University. He's devoted his career to and is internationally recognized for his research on the treatment of substance use disorders and co-occurring disorders, as well as for his work in the CTN, testing the effectiveness of behavioral and pharmacologic treatments in community-based settings. Dr. Nunes will provide a brief introduction of the symposium. Please begin. Okay, Carol, thank you and welcome, everybody. Let me see if we can get the magic to work. Can you see my screen? Yes, sir. You might want to just put it in presentation mode, but we can see it. Okay, we're in presentation mode. Absolutely. If you want to look up at the top, there's a display settings option right at the top, and you might want to flip the mode. Swap presenter view and slide view. Okay, so welcome, everybody. I'm excited to have been able to put together and recruit a great bunch of speakers for this NIDA Clinical Trials Network sponsored symposium on combined combination treatments for stimulant use disorders. Which we think is a promising new direction. The session learning objectives are to understand the evidence bearing on naltrexone, bupropion, and combined naltrexone plus bupropion for treatment of methamphetamine use disorder. Second, understand the evidence based, bearing on combined, the combination of mixed amphetamine salts, extended release, plus topiramate for cocaine use disorder, as well as combinations of medications with contingency management, a medication behavioral combination. Finally, to understand the rationales for combining these treatments for cocaine use disorder and be able to apply them in clinical trials. The basic idea here is that there have been lots of medications tried for stimulant use disorders. And most of these are trials of single medications. And basically, what you get are some studies are positives, most are negative. And the results are inconsistent. And what this probably means is that if there's any effect at all there, it's a small effect, small statistically, and probably not very clinically useful, too small to be very clinically useful. So the idea is a time honored one in medicine, which is to combine two different treatments that work by complementary mechanisms in order to either get an additive or a synergistic effect that would be large enough to arrive at an effect that's actually clinically useful. So just a little graph to describe this. Imagine treatment A in blue and treatment B in orange or whatever color that is. And the vertical axis here is the effect size or Cohen's D, which is the difference between the active treatment and a placebo in standard deviation units. And we see that A and B have an effect size of 0.2 by themselves. 0.2 standard deviation units is a small effect. That's probably consistent with a number needed to treat of somewhere between 10 or 12 or 15. And first of all, you need a study with about a thousand patients in it to be able to detect an effect that small. And secondly, it's not an effect that's likely to be that clinically useful. However, if you combine A plus B with complementary mechanisms and the effects are additive, 0.2 plus 0.2 is 0.4. 0.4 is a medium-sized effect, which is actually clinically useful. And you can probably detect with a study of about a sample size of a couple of hundred patients. If you're really lucky and you have synergy where the whole is more than the sum of the two parts, then you might have an effect size that's even bigger than 0.4, maybe 0.5 or 0.6 or 0.7, something that would probably be clinically useful and that you could actually detect with a clinical trial at around 50 or 60 or a hundred patients in it. So this is one of my favorite examples of this. This is Helen Patinotti study of treatment of alcohol use disorder with combined sertraline plus naltrexone. We're all familiar with naltrexone for alcohol use disorder. It's useful for some patients, but it's pretty limited. And if you look at the clinical trials that the effect is pretty small when it's used by itself. Sertraline similarly has some positive, some negative trials. She did this study with a combination and what you see here, the vertical axis is the proportion of subjects who don't relapse to heavy drinking over the course of a three month trial. These are survival curves. And what you see is that the best outcome along the top here is the combined sertraline plus naltrexone. None of the other three conditions, either sertraline alone, naltrexone alone, or placebo do as well. And the two individual treatments don't separate from placebo. So this is a, to me, a great example of an either additive or actually a synergistic effect. So this is kind of what we're shooting for in the studies of cocaine use disorder that you're going to hear about today. So you're going to get an overview from Udi Gitsa, project officer in the clinical trials network at NIDA, who's going to give an overview of the NIDA medications development program and also the clinical trials network. Madhukar Trivedi from UT Southwestern is going to talk about the major clinical trials network trial that he led looking at combined bupropion plus injection naltrexone for methamphetamine use disorder, which was published in a New England Journal of Medicine earlier in the year. Frances Levin from Columbia is going to talk about two studies from her group of extended release mixed amphetamine salts, otherwise known as Adderall, combined with topiramate for cocaine use disorder. And Joy Schmitz finally is going to talk about her work that's looked at combinations of different monoaminergic medications with an effective behavioral treatment, namely contingency management. And finally, Steve Shoptoff from UCLA is going to provide a discussion and lead a question and answer session. So let me now pass the baton to the presenters. I'm going to stop sharing and we're going to listen to the presentations and then we'll have a live discussion section. Hello, thank you for being here today. My name is Udi Gitsa and I'm a health scientist administrator scientific officer with the NIDA Center for the Clinical Trials Network. It is a privilege to be speaking with you on an overview of the NIDA CTN Stimulant Research Portfolio, of which I serve as a program leader. I have no relevant disclosures. The views and opinions expressed in this presentation are mine only and do not necessarily represent the views, official policy, or position of the U.S. Department of Health and Human Services or any of its affiliated institutions or agencies. The educational objectives are to summarize aims and status of the NIDA Clinical Trials Network CTN Stimulant Research Program and to indicate strategies for evaluating safety and efficacy of already marketed medications and medication combinations as repurposed candidate pharmacotherapies to treat stimulant use disorders. NIDA's CTN is a nationwide consortium of research scientists, general medical and specialty treatment providers, treatment researchers, patients, and NIDA working together to cooperatively develop, validate, refine, and deliver new evidence-based treatment options to patients and providers alike. The network has become a unique resource for expertise on clinically focused research, bridging the gap between research and treatment delivery. In over 22 years, NIDA CTN has completed over 100 studies, resulting in over 700 publications. Our network is comprised of 16 research nodes with 31 principal investigators in academic medical centers, and we also have large healthcare networks, two clinical coordinating centers, and more than 240 community-based treatment programs affiliated with our network. Here is a map of the CTN displayed. With funding support from the NIH HEAL initiative, the CTN has been able to expand our geographic reach, adding five new nodes depicted here in yellow to develop and to test new interventions in new populations. So in response to the stimulant use disorder epidemic and the lack of established medication treatments for stimulant use disorder, the CTN Stimulant Research Task Force generated a list of research priorities, principally in the treatment area, of which I'm presenting a high level overview here of several aims of this research portfolio. The first aim is to identify and to test the efficacy and safety of already marketed medications and their combinations as repurposed candidate pharmacotherapies to treat stimulant use disorders, including methamphetamine, cocaine, prescription stimulant, and other stimulant use disorders. Mechanisms of approved medications for other indications are mostly known, and they can facilitate targeting specific disease symptoms or biological targets implicated by preclinical research. So repurposing allows for faster testing and implementation than does medication discovery, which is often costly and resource intensive, as the early work of medication formulation and safety testing has already been completed. The main advantage of repurposed candidates is that in many cases they have already been deemed safe and may reduce the time frame of drug development and may reveal new targets and pathways from existing candidates. Another aim is to utilize innovative study designs, including adaptive designs, which more closely emulate real world practice settings, and to consider personalized and whole person treatment approaches to tailor the interventions mechanisms of action to prominent relapse triggers and to barriers to recovery that the patients experienced. Patient treatment matching, attending to medication mechanisms appropriate for subpopulation of patients with discrete risk factors and endophenotypes presenting barriers to recovery, as well as other prognostic enrichment design strategies suitable for personalized medicine and addressing polysubstance use and other comorbidities would be a helpful goal. It would also be useful for prospective trials to identify combination medication mechanisms which collectively address symptoms commonly associated with stimulant use disorder vulnerability and to evaluate them in well-designed adaptive design trials with optimal implementation, reducing placebo responses and preferably utilizing available long-acting extended release formulations which may enhance medication adherence. We also aim to leverage artificial intelligence and machine learning to identify promising drug repurposing medication and medication combination candidates using large fit-for-purpose data sets offering high quality real-world evidence. Another aim is to evaluate medications and medication combinations at adequate doses, attending to the mechanisms driving substance use and relapse, patient choice, treatment response, patient comorbidities, and prognostic enrichment approaches. Another aim is to promote big data mining studies using machine learning algorithms or longitudinal studies using patient registries to validate clinically meaningful endpoints from high quality fit-for-purpose big healthcare data sets. Big healthcare data research to uncover these endpoints indicative of significant and sustained clinical benefit may in turn help to improve clinical benefit may in turn help to facilitate the identification of candidate compounds or medical devices efficacious for addressing symptoms associated with addiction processes and relapse risk. For example, stress response dysregulation, sleep disturbances, depressive symptoms, and other co-occurring psychiatric conditions. Another aim is to utilize such clinically meaningful endpoints to systematically test promising pharmacotherapy candidates and medication combinations with synergistic benefit for patients with co-occurring conditions. For instance, to reduce substance use and symptoms associated with these co-occurring conditions. Medication combinations are of interest in that they may amplify the efficacy by targeting multiple complementary neurobiological mechanisms of action. And this is an overview briefly of the current CTN stimulant research portfolio in terms of upcoming trials that we are going to be supporting. And the first one is an efficacy test of transcranial magnetic stimulation for the treatment of methamphetamine or cocaine use disorder. This pilot study aims to evaluate the feasibility and effect size of 20 sessions of repetitive transcranial magnetic stimulation versus sham in adults with methamphetamine use disorder or cocaine use disorder and to evaluate the safety of repetitive transcranial magnetic stimulation compared to sham in participants with these disorders at the end of treatment and a 12-week follow-up. And the principal investigators are Drs. Kathleen Brady and Madhukar Trivedi. We also plan to launch a multi-site randomized placebo-controlled trial of the combination of injectable naltrexone and monthly injectable buprenorphine for cocaine use disorder. The study aims are to evaluate the safety and efficacy of this combination for treating cocaine use disorder as compared to placebo. Drs. Madhukar Trivedi and Stephen Schapda are the principal investigators. Next, we plan to launch a multi-site randomized double-blind placebo-controlled trial of monthly injectable buprenorphine for methamphetamine use disorder treatment. So, the fourth wave of the opioid crisis involves increasing polydrug stimulant and opiate use, and there are no effective medications for treating methamphetamine use disorder. And this HEAL initiative supported multi-site clinical trial will test the superiority of long-acting buprenorphine and its kappa antagonist activity over placebo as a treatment for methamphetamine use disorder among persons who co-use opioids. And Drs. Stephen Schapda and Madhukar Trivedi are the principal investigators. We also have a data mining study using a combined strategy of artificial intelligence-based prediction and electronic health record-based retrospective clinical cooperation to identify promising drug repurposing medication candidates for treating cocaine use disorder. And identifying novel pharmacotherapies for cocaine use disorder is a priority, and while it is a priority, it has been challenged by a lack of interest from pharmaceutical industries and by a lack of traditional drug development processes that are expeditious. They are rather lengthy in nature. So, we aim to expedite the drug development processes by having a repurposing approach. Research to repurposing existing drugs may accelerate findings of anti-cocaine use disorder medications, and the goal of this study is to identify and to evaluate potential anti-cocaine use disorder medications by developing artificial intelligence-powered drug discovery technologies in big healthcare datasets, as well as big genomic and other large biomedical datasets. The output of this project will be a list of promising anti-cocaine use disorder candidates that may help to expeditiously translate into clinical trials to benefit patients suffering from this condition, and the principal investigators are Drs. Rong Xu and T. John Winhusen. And then we are also planning a proposed pilot study that would explore the feasibility of using ketamine to treat stimulant use disorder and comorbid depressive symptoms. So, this proposed pilot study will be to determine the effect size for a potentially larger multi-site trial of ketamine in treating principally methamphetamine use disorder, and the principal investigators here are Drs. Madhukar Trivedi and his colleagues. And as an example of a past trial that the CTN has supported in this area is the ADAPT2 multi-site clinical trial, and it is a trial led by Dr. Trivedi that was recently published in New England Journal of Medicine, in which Dr. Trivedi will discuss the primary findings later in this symposium. The trial found that a combination of two medications, injectable naltrexone and oral bupropion, was safe and effective in treating adults with moderate or severe methamphetamine use disorder. In closing, we encourage investigative teams to think outside of the box and to develop high-impact innovative research concept proposals in the aforementioned focused areas that I discussed in the first part of my talk, sponsored by their CTN nodes for CTN consideration. And thank you so much for your attention, and I look forward to answering any questions that you may have. Welcome to the meeting. I am Madhuka Trivedi, I'm a professor of psychiatry at UT Southwestern Medical Center in Dallas, and I am the PI of this study that was published in the New England Journal of Medicine that I'm going to discuss, and I'm also one of the PIs on the big Southwest node, funded by the Clinical Trials Network of the National Institutes of Drug Abuse. I'm very excited to be here because we will be talking about a pharmacological option for the treatment of methamphetamine use disorder. As you all know, we have no approved treatments for methamphetamine use disorder by the FDA. So this is one of the first in a large multi-center definitive trial that was funded by the National Institute of Drug Abuse, and I'm very excited to give you a preview of the results of this. The goals that I'm going to try to talk about are assessing the efficacy of this extended release injectable naltrexone plus extended release bupropion as combination pharmacotherapy for methamphetamine use disorder, and I'll talk a little about the rationale of how and what made us decide on the combination, and then give you some demonstration of the efficacy on other substance use outcomes, depression symptoms, quality of life ratings in the clinical trial we conducted. So this study was known as ADAPT2, it's Accelerated Development of Additive Pharmacotherapy Treatment for Meth Use Disorder, and this was funded, as I mentioned, by NIDA. And as we all know, that in a lot of chronic conditions where individual monotherapies do not work, combination treatments have been shown to be efficacious. With methamphetamine use disorder, we know that we don't have any treatments approved by the FDA, not for lack of trying, there have been a whole range of treatment studies that have been done, but they have not addressed the outcomes and therefore have not produced the efficacy we need. So since there are no approved treatments, but there are some promising candidates that have some clinical utility, and both naltrexone and bupropion have been actually postulated to have some efficacy, they've been individually studied, not really shown much efficacy. There has been a very small pilot trial that we did called CTN0054, which was published several years back, which was an open-label trial where we were trying to basically try to assess the magnitude of possible effect and also feasibility of combining bupropion with injectable naltrexone for methamphetamine use disorder. We learned a lot from that, and those lessons learned from that study, which was an open-label trial, and a number of other randomized control trials went into our decisions about many things, including the population we studied, the study design that we used, and the outcomes we looked at. And I'll talk about those three in the next few slides. Suffice it to say that one of the lessons we learned, and I'll repeat them at the end, we learned that traditional randomized control trials have had a significant problem because of the heterogeneity of the population study. Some patients were using it over the weekends, three days in a row, but not more severely. So there was a heterogeneity of the population study. There's also a heterogeneity of outcomes, and the retention rates was a big challenge and a placebo response rate. So all those components went into our thinking about how best to study this. One thing is very clear that methamphetamine use disorder is really increasing, mortality from methamphetamine use disorder is increasing, and it is a dire emergency, in fact, for our field to really have viable treatment options. So first and foremost, as I mentioned, I'll talk a little about the population we studied, but even before that, I want to draw our attention to the fact that traditional placebo control trials really had two major problems, retention rates, because people were not able to stay in with it. And so there was very poor retention and also high placebo response. So we looked into various different adaptive study designs. One of them has been used traditionally in depression called sequential parallel comparison design. We were involved with some of the depressions trials where we've used SPCD or sequential parallel comparison design. That is what we ended up using, so I'll show you some of that in a minute. But its real goals are that it helps reduce placebo response, it improves placebo difference where it exists. It cannot invent it, obviously. And the number of sample population needs to be studied is actually, it is very efficient, so it may end up being a smaller number. The other trial designs people have attempted to use, which we eliminated from our consideration for this, were placebo run-in, so therefore that can be used as, so to speak, a sieve to assist in eliminating people who are likely to do well on placebo. So that is the idea behind it. We felt that that doesn't really still get it. Same thing with treatment run-in, so that you really focus on the placebo run-in eliminates people who are likely to do well on placebo. The treatment run-in eliminates people who are likely to be non-adherent to treatment, so that they're unable to keep track and be adherent to treatment. And then there is David McCann at the NIDA has really talked about a ramp-up design where the placebo responders are not included in the analyses, and it accounts therefore for professional participants. So you predetermine where the ITT sample will begin, and generally it is still a theoretical point, but I think all of these need to be considered as we move forward with trials for methamphetamine use disorder. One of the things we all know about methamphetamine use disorder is that the pattern of use for patients varies so widely that if you don't really be careful about it, then you end up missing the signal where it may exist. As I mentioned, previous studies have been, a myriad of studies have been done, but only seven, which is 31.8 percent, had retention rates of more than 60 percent. So if they don't have a good retention rate, then it may not be something where you can find a signal. Less than one out of four studies had retention rates above 80 percent, so you see the challenge. Nine studies had retention rates less than 50 percent, so that's just impossible to find a signal there. There have been four studies that showed some improvement with active intervention. There is dextroamphetamine, there is mirtazapine, a couple of trials on mirtazapine, and then obviously contingency management has been shown to be efficacious for meth use disorder. Our study, as I mentioned, really our adaptive study design we used was designed to reduce placebo response. Also, as I mentioned, the duration of the trial, the cost, and the sample size creates efficiency if you use SPCB. We did put in several efforts to improve medication adherence and the population we really focused on were severely affected. Mean days of use requirement was 18, but we ended up actually with 27 days of mean days of use in the last 30 days. Clearly a severe population and how that happened I'll show you because our requirement for those who were randomized had to be very clear, so that really helped us. What we studied ended up being basically a double-blind placebo-controlled trial. As you can see on the right side, the brown distinction is placebo versus active medication combination in the first stage. Those who are on the medication combination, their outcomes in the first stage comparison with placebo is calculated and then they continue. They do not then contribute to additional data. The placebo sample, those who are non-responders to placebo, then get re-randomized to placebo or combination medication. The blue boxes and the brown boxes are the two stages that give us the active signal, and then there is a proportionality given to the two stages as used by the developers of SPCD, and this was a 12-week medication phase. One additional thing we did was instead of the traditional four-week injections for naltrexone, we emphasized the need to do it every three weeks. We got FDA dispensation for that, and so we had oral bupropion to the high dose of 450 milligrams per day and injectable naltrexone every three weeks. We, patients had to have moderate to severe methamphetamine use disorder. As I mentioned, their self-reported days of use was 18 days or more, but most importantly, they had to have at least two out of three urine drug screens within a 10-day assessment period and on the day of randomization. So what ended up happening basically is we got patients into the study who were usually using methamphetamine almost every day. That is what ended up happening, and they had to be medically and psychiatrically stable, obviously. And one additional factor to think about is we were trying to study the pure efficacy of the medication combination, so they were not in any other formal addiction treatments as they were being run through the randomized control trial. So what we then end up with is a definition. So primary efficacy outcome measure had to be methamphetamine negative urine drug screen, and so responders to the treatment were defined as any participant who provided at least three or more negative drug urine screens out of the possible four methamphetamine urine drug negative urine drug screens. And this was done for stage one at week five and six. So this, as you can see, stage one is six-week trial, so week five and six of stage one, and week five and six of stage two were the two time points when they were giving us two urine drug screens every week. So they had to have at least three out of four negative urine drug screens for both those times to be counted as responders. And we also looked at adverse and serious adverse events. So this is a sample. We studied 109 were randomized to combination and 294 to placebo in the first stage. And as the numbers show, of those 294 placebo randomized patients, participants, 225 placebo non-responders were then re-randomized to medication combination of placebo. So these blue and the brown ended up being our sample of calculation for efficacy. And the most interesting thing is the outcomes, obviously. That is, in stage one and stage two, drug versus placebo consistently produced a significant benefit for the combination versus placebo. In effect, basically, our placebo response was very low, 3.4% in stage one and 1.8% in stage two, as you see on the right side. And their weighted average was 2.5%, and weighted average of combination medication was 13.6%. So what you really end up with is almost five to six times the response rate on the medication combination versus placebo, which translates to a number needed to treat of nine. And as you all know, with number needed to try treatment of less than 10 has been used by a lot of guideline driven treatments to require or suggest clinical change. Like the NICE guidelines in the UK uses number needed to treat of 10 or less to guide clinical change. Additionally, one must remember that a number needed to try treatment, number needed to treat of nine is in the range of what we use for alcohol use disorder for naltrexone, for smoking cessation for bupropion. And so this is really the first ever large 400 plus patient trial that shows clear unequivocal efficacy for the combination versus placebo. These are the raw data for stage one. As you can see, those on the combination up on the red line in stage one and the maroon line in stage two are those who were on medication combination and the green and the maroon lines down here are on placebo. And you can see that the effect actually starting to show at week two for the medication combination in the first stage. Same data, much more in boxes. You can see the greens as you see in the combination arms are beginning to show much more broader effect in the first in the first stage at week two to three. Other, the sample characteristics, so you can understand is most of two thirds were male, about half of them had never been married. Interestingly, about 40% of 38.7% were employed when they were randomized and 85% were non Hispanic or Latino, were not Hispanic or Latino. The primary outcome availability was actually very high 75 to 67.9% at the low end to 83.3% on the high end and tying this to the data we already have from randomized control trials. Remember, I showed you that most of the trials at less than 50% retention rate in most of the previous trials. This really got to the point where we wanted, and so that is what is the other issue we found to be beneficial. Medication, you can see the treatment exposure by stage and treatment arm again really, especially the tablets part was the one that we had to make sure that people were taking the medication because that was dispensed to take it at home. The injections were at the treatment site, and you can see a pretty close to 90% exposure to injection so they got most of them got most of the injections and then about 68 to 74% 75% 75% adherence to the medications taken. So this is a well exposed and retention was very high and the efficacy is strong in that sense. If you look at the primary outcome by subgroups, you can see again consistency across almost all the subgroups, whether they are by gender male and female by younger versus older. The only place where there is less of an effect is for African Americans and that I think is really mainly attributed to the fact that if you're only 48 patients were black or African American in white population, which was much larger at 287 we still see the effect and in fact for the other population which is not white nor black or African American, you see about the same effect so this was really an Hispanic non Hispanic population so the consistency of this finding and I'll return to this issue from other outcomes secondary outcomes also, but by subgroup we did not find any group that actually placebo beat and the combination. Finally, if you look at some of the self reported measures of methamphetamine use and craving they decrease so if you look at the craving vast scale you see the same effect with the same degree of benefit for medication combination versus placebo. If you look at timeline follow back where you measure the days of use again same thing almost 27.2% for medication combination versus 14% for placebo and stage one and the same effect persists in stage two, so the medication combination continues to be superior in both of these in timeline follow back also same thing with cigarette smoke abstinence, as well as reduction in depressive symptoms same magnitude effect in favor of the common medication versus placebo. So I think that if you think about it and put it in the context of previous trials multi center trials or even Phil coffin single site trial in JAMA psychiatry recently. The effect sizes, either not not existent or a very low effect size in some of the trials, even in the single site, and this was a multi site trial that had a consistent effect. These are the street treatment sites that were participating, so there was a node site in New York. In Hennepin, Minnesota and the southern consortium, which is in South Carolina. There were two, three node sites in Texas, two nodes in Texas. One is Houston and Dallas and then UCLA. And then finally CODA and the San Francisco site. These were the nine sites that were participating in the study. These multi-center trials, as you all know, are big, major endeavors. So all of my partners at the big Southwest node, and especially Dr. Steve Shapta and Walter Ling, as well as the data coordinating center, the clinical coordinating center at MS and our NIDA CCTN partners, Dr. Sparenberg, Dr. Gitsa, Dr. Petra Jacobs, and Betty Tai were all instrumental in really helping us do this. So I think that this really provides a very valid, potentially very immediately useful treatment option for patients with methamphetamine use disorder. And we'll spur more new research in this because as you all know, one of the major drawbacks for methamphetamine use disorder in the literature and in current research is that pharmaceutical companies have been slow to come to this. And this is something that may spur more interest in pharma manufacturers also. So I am very, again, very thankful for this opportunity to present to you, and we'll address more questions in the Q&A session. Thank you very much. Hi, everyone. I'm delighted to be here today to talk about research that I've been working on for at least a decade on looking at combined pharmacotherapy and specifically extended release mixed amphetamine salts and to paramate for cocaine use disorder. These are my disclosures, both federal, New York state, and industry support. And I'm putting this slide up just as a starting point to make the disclosures just as a starting point to make the point that while certainly the peak use disorder for cocaine was in the mid-80s, it still has remained a persistent indolent problem with about a million individuals meeting criteria in the United States for cocaine use disorder, and another million individuals who meet criteria for methamphetamine disorder. And so while certainly these two groups are not as large as those who have problems with either alcohol or cannabis, it represents a group that is in dire need of treatment for a proportion of individuals who are suffering from cocaine and methamphetamine use disorder. When I give this talk in the past, particularly when I talk just overview about treatments for cocaine use disorder, there are usually these overarching categories that I focus on, and that's agonist treatment, antagonist treatment, or indirect antagonist treatment, novel mechanisms which can include things like ketamine, and certainly something that I've spent many years evaluating and looking at is psychiatric comorbidities. So individuals who have, let's say, both a cocaine use disorder and depression or ADHD, and developing treatments that may work towards reducing the psychiatric comorbidity such that they may be better able to utilize psychotherapeutic interventions, or in fact in some cases where there may be medications that would work directly towards the treatment of the cocaine use disorder as well as the psychiatric conditions such as mixed amphetamine salts as an example for both ADHD and cocaine use disorder. But today, sort of the theme of the symposium is on combination pharmacotherapy, and I can recall maybe 15, 20 years ago when if you thought about using combined pharmacotherapy, it was considered that you would not be clear enough in your treatment approach and that basically why would you use combination therapy? It might lead to more side effects and it makes treatment more complicated, but as we know in many areas of medicine, whether it's treatment of hypertension or even diabetes or treatment of individuals suffering from AIDS, that often you need to use combinations of medications with different mechanisms of action in order to see a therapeutic effect. Now this is not something that's new under the sun. There was a very nice review, it's now about five, six years old, by Stoops and Rush in which they looked at all the different pharmacotherapies that they could find that have been evaluated for the treatment of stimulant use disorder in the literature, and you know some of this may not be that familiar because the reality is in many cases, with the exception of a few, most of these interventions have been only studied in one clinical trial or the results were mixed so that the decision was not to move towards a larger, more definitive study. Now it was interesting is back in 2014, John Mariani and I and other colleagues had published a paper in biological psychiatry on toparamate and mixed amphetamine salts, so that made it into the literature at the time that they were doing this review, but we hadn't had the second study so it wasn't like there was a confirmation study at the time, and the other large study which you just heard about from Dr. Trivedi is the large study that was done as part of the clinical trials network looking at bupropion and extended release naltrexone, and that was not obviously in this review, it came out just recently, that paper, but I'm very grateful that that study came out because I think it solidified this idea that even if we don't have FDA-approved treatments for either cocaine or for methamphetamine use disorder, there are medications that are available that might, either alone or particularly in combination, might be effective for this treatment population. So I'm very excited that his work has been done and gotten a lot of attention and hopefully will lead to more interest in looking at combined pharmacotherapies. Now what led us to look at mixed amphetamine salts extended release, and again I think it's very important to say that I'm a big supporter of using long-acting formulations, so there's less abuse liability and better, in some cases, tolerability, and you don't need to dose as frequently, which is very important in terms of adherence, but there has been a pretty solid line of evidence suggesting that amphetamine analogs might be useful for the treatment of stimulant use disorders, and particularly cocaine use disorders. On a recent meta-analysis that I had the honor to be part of with Dr. Tardelli from Italy, we found that, in fact, that amphetamine products were, or formulations, were superior to even other stimulant interventions for cocaine use disorder in promoting abstinence. So there's certainly the line of evidence with amphetamines, but there also is the case that Toperamate shows promise. I mean, if you look at meta-analyses, they'll often say the results are mixed, but the studies that have actually looked at abstinence, Toperamate has outperformed placebo. So you have mixed amphetamine salts, extended release, which we've used extensively over the last decade or so, and it has been shown through many different neurobiologic models to improve or increase dopamine transmission, and on the other hand, you have Toperamate, which actually increases gavinergic transmission and reduces glutamate neurotransmission, and might, in fact, through this mechanism, reduce craving when exposed to condition cues. So the thought was, a number of years ago, John and I did a lot of brainstorming together, was that maybe together, when you use these two medications together, maybe they would mimic the partial agonist effects that you see with other successful pharmacotherapies for addiction, such as buprenorphine or varenicline. And so our hypothesis is that perhaps combining these two agents with different and distinct mechanisms of action might be additive or even synergistic, and we would see significant effects. So the first study that we conducted, this was part of an R01 that I received a number of years ago, in which we looked at the combination compared to placebo, with the primary outcome being abstinence, which again is a fairly high bar, and the abstinence over three consecutive weeks. And then the secondary outcomes were measures such as retention, proportion of abstinence weeks, which is a little easier of a bar, because you don't have to be continuously abstinent, although even a week is pretty good, and looking at medication adherence and tolerability. This was a single-site study, randomized double-blind placebo-controlled trial. It consisted of a one-week placebo lead-in phase so that we could assess the severity of the cocaine use disorder and ability of a patient to comply with study procedures. So if they weren't showing up and they weren't complying with the assessments, then they would not get randomized. However, their cocaine use, regardless of what was that first week, they were randomized into the study. So we stratified based on whether or not the three urines that were collected during that week, if one of them was positive for the metabolite of cocaine, benzolecanine, then they would be stratified based on that. And it was a one-to-one allocation of ERS, MAS, which is the mixed amphetamine salts, and topiramate to placebo. They were titrated over two weeks to the maximum dose of 60 milligrams of the mixed amphetamine salts, and topiramate, as you probably know, is done much slower because it has some tolerability issues with it, unless you give it slowly. And it was titrated over six weeks to a maximum of 150 milligrams twice daily. All the patients received weekly medication adherence therapy, which is interesting because sort of the gold standard 10 years ago was to use cognitive behavioral therapy, thinking that that would help people be retained in treatment. Our experience was that patients were quite content or pleased with coming in once a week to meet with the psychiatrist to discuss their progress, get supportive therapy, talk about their adherence to the medication, without having perhaps raising the placebo effect by giving people cognitive behavioral therapy, even though in many of our trials, patients who got CBT, absence was still a fairly high bar. So it didn't seem to have much of a placebo response. Now in eligibility criteria, because of time, I'm not going to go through all eligibility criteria, but one item I want to make sure or criteria I want to make sure you're aware of is that in this study, individuals had to use at least four days during the 28 days prior to study entry because the major outcome is abstinence. And if they're not using it all, you know, or that much, even if for some reason they meet criteria for cocaine use disorder, we wanted to make sure there was some baseline level of use. However, this is a pretty low bar for people coming in for treatment, but this was sort of a standard amount that are days that we were using at the trial at that time. We also excluded individuals who had significant major depression or bipolar illness or schizophrenia or psychosis or any physiologic dependence on a substance such as alcohol that would require a medical intervention. We also had the standard exclusionary criteria and we also excluded people if they were on carbonic anhydrase inhibitors. We were worried about metabolic acidosis with diparamate and kidney stones, so therefore this was an important exclusion criterion as well. The psychosocial intervention, like I said, was manualized enhancement therapy. We did pay people not so much to have negative urines for cocaine, but more for attendance and completion of assessments and also for the return of bottles. And if somebody did everything in the study, they attended every session and returned all their bottles, they could get almost $600 in vouchers. Most people got about half that during the trial. In terms of entrance into the study, we interviewed a lot of people. This is sort of standard for these trials. Almost 100 went into other trial and over 100 went into another treatment trial. The 81 that was randomized was allocated to 42 to the placebo arm and 39 to the combination medication arm. The common demographics that we saw was people were about in their early 40s, primarily male, from different racial and ethnic backgrounds, and perhaps not surprisingly using between 12 and 13 days a week, mean days a week, which is more in line what you'd expect. Even though we had a low bar of four, most patients were using substantially more days in the month coming in for treatment. We were happy to see very good retention. This was a complicated protocol in which we were titrating both medications up and modifying doses based on side effects, and we still managed to have about 70% of patients retained through the placebo period, the lead-in after randomization, the dose titration up, as well as maintenance on the medication during that 12-week period. The primary outcome, like I said, was three weeks of any abstinence, consecutive abstinence during the trial, and we found that the combination arm was twice as percentage likely of getting abstinent compared to placebo, but there was a very strong moderating effect in that the baseline number of cocaine use days, there was a significant interaction between treatment and baseline total number of cocaine use days, suggesting that this was making a difference in treatment efficacy. In fact, when you look at this graph, these tertiles were decided by our statistician actually prior to us looking at the data. This was suggested on the way that the data that was presented for the patients coming into the trial in terms of their severity of use. Interestingly, what you see is that individuals who were using what we would consider low levels of use, only maybe two days a week per month for the month coming into the study, they actually did a little better on the placebo arm than the active treatment arm. On the other hand, as they started to be heavier users prior to coming in more than nine or more days a month, then we saw substantially higher rates of abstinence in the active treatment arms compared to the placebo arms, and this odds ratio of those of using at least nine days was an odds ratio greater than seven. We were struck by these findings, but obviously felt that they needed to be replicated. The adverse events, most of them were things you might expect. This was all adverse events, and what you see is that individuals who had insomnia, increased appetite, were more likely to be in the combination arm, which is perhaps not surprising. Parasthesias is not an uncommon side effect, which is paramate, again, more common in the active treatment arm compared to placebo. We did have two serious adverse events. One was in the active treatment arm in which a patient developed a substance-induced mood syndrome that did require hospitalization, and another participant had to be voluntarily discontinued during the final week of the trial because they decided to enter a residential substance abuse treatment center. So the results of this was that our hypothesis was found in that the combination did produce higher rates of continuous abstinence, twice as much than in those receiving placebo, and suggests the preliminary evidence that there may be something beneficial about this combination. The questions we had, of course, is, you know, should we do a larger trial, which, of course, we always wanted to after we get a preliminary finding with a pilot trial, and what do we do about the low-frequency users? Clearly, they actually might have done better on placebo, and that would be, you know, useful for clinicians because many clinicians might say, well, somebody's using once or twice a week. Do we really need to go to daily use of, you know, two powerful medications? So we went back to NIDA and requested funding for a two-site trial, which was actually funded and led to the next study, and this was a two-site study working with people at UPenn with Kyle Kampmann and his group, and we did the same study again, primarily the same or pretty much the same methodology, but this time we had as our primary outcome three weeks of consecutive abstinence at the end of the trial period. Now, this is a pretty high bar, and often we've had a lot, there's lots of discussion going on about changing outcome measures in clinical trials for cocaine. Maybe that's the reason we've not seen an effect in these studies, but we felt that this would be a powerful bar if we could get over it, because, you know, people aren't going to just automatically put people onto paramedics complicated or even agonist treatments such as mixed amphetamine salts, so we chose this outcome measure, and what it also means is that if somebody gets to the end, doesn't get to the end of the trial and doesn't, isn't there in those last three weeks, then they're considered a non-treatment responder, so that even makes it obviously a higher bar, but we also had the secondary outcome that we had in the prior study, which was three weeks of abstinence during the trial, continuous abstinence, and looking at other measures such as craving, tolerability, and again, adherence, so the study was pretty much the same. One difference was if patients self-reported no cocaine use during that placebo lead-in period were before we stratified, this time we did not include them in the study. We did randomize in a one-to-one allocation, but this time we stratified by the presence of alcohol use disorder, because there's some evidence in some studies with stimulants, particularly modafinil, that patients may have less of a positive response if they have an alcohol use disorder, so that was pretty much kept the same. We did the titration the same with with the mixed amphetamine salts, but with teparamate, this time we only went to 100 milligrams twice a day, because you learn from your earlier study that's the benefit of a pilot study, and what we learned was that people didn't really tolerate or need 300 milligrams of teparamate, and most were getting less than 200, so this time we went to 100 twice a day, and again, titrated in a slow, controlled way. There were three study visits per week. We used medication adherence. We had urine samples tested for quantitative benzolecanine, used timeline follow-back to assess for cocaine use, as well as confirmation from the benzolecanine results. We had calendar procedures and pill counts. It was prior to the use of a mocha or other ways of visually watching people take medication, which I think is more the state-of-the-art now, and I think is what was used for the Trivedi study, but we did collect teparamate levels and qualitative amphetamines for each urine sample that was collected. Now, what was different in this study, like I said, is that in this case, we only took individuals who were using at least nine days going on the lead that we had from the prior study. We also wanted a BMI over 18, given that, frankly, both of these medications can lead to appetite suppression. And we had very strict criteria regarding hypertension, not just about study entry, but also dropout. And in fact, a good percentage of our patients were taken off of medication, a minority though, because of persistent elevated blood pressures. And again, this hurts our findings because if somebody drops out of the study or you take them off of medication, they're not getting exposed to the medication and therefore reduces the likelihood of detecting an effect. Nonetheless, we assessed a lot of people again at the two sites. We randomized 127, 63 was in the placebo arm and 64 in the active treatment group. Again, the average age, a little bit older, in their mid 40s, again, predominantly male from different racial and ethnic backgrounds, mostly high school or less in terms of education. Columbia had slightly more enrollments than UPenn, not that much different. The number of days of cocaine use in the trial prior to coming in was 15 or 14, a little bit higher than the earlier study, which is expected since we had this change in the inclusion criteria. Again, we had pretty good retention, a little lower than the other study, but still between 60 and 70%. And when we looked at how many individuals achieved three weeks of absence at the end of the trial, about 14% in the combination group and zero in the placebo group. This was an odds ratio of about 14. The secondary outcome, again, about three times greater proportion of individuals which are achieving three weeks of consecutive absence compared to placebo. So odds ratio of four. This was also heartening. You can see these results. Some people like figures better than numbers. I like figures. And again, either you're a half-full, half-empty person, but consistent with the Trivedi study, the Trivedi CTN trial, we got about a 14% difference in individuals for that last three weeks of the study. The Trivedi study, it was two weeks and two urines. We had three urines and we had three weeks. And also ours was a longer trial. And like I said earlier, the more people drop out, the less likelihood you are to see people doing well or having the opportunity to determine if they achieved absence at the end of the trial. So you can look at this as well. It's only 14% or you can look at it as, well, this is a fairly robust finding. So what about craving? Another outcome measure we found significantly better in the active treatment arm. And self-report, of course, of taking pills was quite high, which is not surprising. That's found from other studies. But when we actually looked at dosing, most people, the mean dose was 60 milligrams on the mixed amphetamine salts, but to Pyramid again was under 200, suggesting that we don't have to maybe push to Pyramid as high or patients don't tolerate the higher dose. So look at mean doses, you see it's even lower. The only side effect that was common, that was more common in at least, which we looked at at least 5% of participants would report a side effect. The only significant difference was dry mouth in this sample. And then in terms of serious adverse events, we had four. I won't go over them because of time limitations, just simply say that none of them were thought to be specifically related to the medication, obviously in the active treatment arm. And then I'm also putting this slide up because this idea of that maybe frequency matters or severity of use matters in that this was a study done by Walter Ling in which it was looking at long acting methylphenidate, Concerta. And the dose was actually fairly modest in my opinion, it was for methamphetamine use disorder. And when they looked at days of use or change in days of use, there was really nothing different between the two treatment arms. But when they took people who used greater than 10 days based on the 30 days of baseline coming into the study, there was a significant difference between the two treatment arms in the heavier users, suggesting that there may be something to this heavier use pattern, which is also probably the group that was picked up by the way the inclusion criteria were set up in the Dr. Trivedi and colleagues study that was described before. So again, our results support the hypothesis that this extended release mixed amphetamine salts with diparamate might be superior to placebo in achieving three weeks of abstinence, either at the end of the study or during the study. Medication management seems to work well and suggest that abstinence is that these two medications can be used to help promote abstinence, but perhaps baseline cocaine use also is an important factor to consider. So where do we go from here? Well, always I'm thinking the next step may be a larger multi-site trial. Between the two trials, we had over 200 patients, but the study done in the CTN had over 400 patients, although the findings remarkably are very similar. One of the questions I've gotten criticized about or people have brought up is well, maybe what's driving this abstinence? Well, there have been trials done with both these medications individually, not actually the mixed amphetamine salts are accepted individuals with ADHD that our group did a number of years ago. But the question is, do we do something that's the most practical, which is what the CTN did and just combine them and then not be so driven by disentangling the effects or is it critical to have forearms and therefore have a substantially larger trial? And I leave that as a point for discussion. I'd like to thank NIDA. All of these grants were funded by NIDA and would not have been possible without their support. The people who work at STARS, both the research coordinators, Amy Mahoney and Daniel Brooks right here, but also the STARS PIs and the statisticians that help us and a special thank to Ned Nunez, who I've worked with for many years, who is a great colleague and a joy to work with. And John Mariani, of course, who none of this would be possible without working with him. So thank you very much. Okay, thank you. It is great to be part of this symposium. I'd like to thank Dr. Nunez and the Academy for this opportunity. We've been hearing about some promising new medication combinations and I'm gonna shift the focus slightly and talk about combined treatments that involve medications and behavioral therapies. In particular, I'm gonna be focusing on the strategy of using dopaminergic medications in combination with contingency management. That has been a longstanding area of interest in my research. First off though, I want to acknowledge and thank NIDA for their funding support over the years and I have no relevant disclosures. For educational objectives, I am first going to describe the rationale for combining medication with behavioral treatment. I'm gonna spend a little bit of time talking about how there's been kind of a paradigm shift in the field in terms of how we conceptualize these type of treatment combinations. Second, I'm gonna review a set of studies that have focused on the strategy of using dopaminergic medications as a way to augment the response to contingency management treatments. Third, I'm gonna talk about some current challenges that we face in terms of translating scientific evidence into clinical practice. And I'm gonna end by talking about some areas for future directions in terms of moving this research forward. So I know the previous talks have clearly described why this is such an important topic. Stimulant use continues to be an enormous problem in the US. If you look at figures like this, there has been a clear increase in the numbers of overdose deaths, often involving stimulant drugs like cocaine and methamphetamine. Not having an approved medication for these disorders has certainly put us at a disadvantage when it comes to having treatment options for our clients. We do have effective behavioral treatments, but the benefit is moderate at best. And we are still without a single highly effective standalone treatment when it comes to helping patients with stimulant use disorders. So when you think about combining medication with behavioral therapy, the field has undergone what I would describe as kind of a subtle but significant paradigm shift. So if you go back to 2004, the paper here that came out, Kathleen Carroll, Costin, and Roundsville, they put forward this paradigm in which the goal was to use behavioral therapy to enhance or increase the efficacy of the pharmacotherapy. So in other words, using behavioral interventions to improve the likelihood of detecting a medication effect. Fast forward and compare that to a more recent paper in 2016 by Dakwar and Nunes, where a slightly different paradigm is presented. Here, we're talking about using medications to enhance or facilitate the effects of the behavioral treatment. So two paradigms, they differ depending on the type of synergy that you're trying to achieve, but both share kind of that same underlying assumption that combining two treatments is likely to produce a stronger effect compared to each kind of monotherapy. So under the original paradigm, again, if you're thinking about combining medication with behavioral therapy, here the goal is to look at ways in which to reduce kind of extraneous factors or sources of variability that could mask or reduce the likelihood of detecting a medication effect. So for example, you'd be using interventions to enhance patient retention or to increase medication adherence. Or to keep patients in treatment if the medication is likely to have adverse or delayed effects. You could use behavioral interventions to manage co-occurring problems that are secondary to the drug use. Or in certain conditions, using the behavioral interventions to reduce substance use if the medication is likely to have a stronger effect under more abstinence conditions. Alternatively, under this other paradigm of combining treatments, where the goal is more about facilitating or optimizing the response to behavioral treatments, one of the probably best studied approaches here is the idea of cognitive enhancement. So here you're looking at using medications that might have cognitive enhancing effects that could improve some of the deficits or impairments that we often see in our patients and which we know can be negatively impacting treatment outcomes. So for example, medications that might improve working memory or attention or impulsivity, these are all important in terms of fully responding to some of our behavioral therapies like cognitive behavioral therapy. But what I'm gonna talk more about is this second example here. And the idea of using pharmacotherapy to enhance reward system functioning in order to improve response to incentive-based contingency management treatments. So why contingency management? Well, let me go into just a quick overview of the number of evidence behavioral treatments that we currently have to treat substance use disorder generally and stimulant use disorder. And the list has been growing over the years. We have 12-step facilitation, CBT, MI. There are now mindfulness and acceptance-based therapies and CM as well as CRA, community reinforcement approach, which is kind of a multi-component approach that includes CM as kind of a major component. But which of these appears to be most effective? And what we are finding pretty consistently when the evidence is reviewed is that contingency management is currently the most reliably effective behavioral therapy that we have. This was the most recent systematic review in 2018 by DiCresenzo and colleagues. They conducted a network meta-analysis. They included 50 RCTs, over 6,900 participants. They included studies comparing 12 different behavioral treatments, targeting cocaine and or methamphetamine addiction. And what they found was that CM and CRA were the two most efficacious and acceptable treatments. Here, I am showing you kind of the network of comparisons for each of their outcomes, abstinence and acceptability, which they defined as dropout due to any cause. The size of the circles and the thickness of the lines indicate the number of trials comparing each intervention. So what you can see is that most of the trials involved comparisons of CM, CBT and treatment as usual. And here you're seeing the results expressed as odds ratios for each of the treatments. I know there's a lot to see here. So the yellow arrows are indicating where the strongest effect sizes were found for the comparisons. And what we found is that contingency management alone or in combination with CBT or CRA showed the strongest effects. So there is general consensus in the field that CM is currently the most effective behavioral treatment, but it doesn't work perfectly. There are patients who don't respond to CM rewards, even when high value, high magnitude rewards are used. One idea is that this individual difference in reward responding has to do with dopaminergic tone, such that individuals who have lower dopamine tone or dopamine function may be less likely to respond to non-drug rewards. That begs the question as to whether dopaminergic medications that can normalize dopamine signaling might thereby increase the salience of rewards that are offered as part of CM. There's some really nice neurochemical evidence supporting this hypothesis. So this was work done by Diana Martinez and colleagues at Columbia some years ago. She looked at cocaine users, cocaine using patients who were receiving CM treatment. She classified them into either CM responders or non-responders based on the amount of rewards that they earned during treatment. And then she then compared the two groups using PET imaging and found significant differences in that the non-responders had lower dopamine signaling compared to the responders, which goes along with and kind of supports this hypothesis that dopamine signaling may serve as a medication target to improve CM response. So in line with this, over the years, there have been a handful of clinical studies that have tested this notion, this hypothesis about pharmacological augmentation of CM by targeting dopamine, either directly or indirectly. What I'm going to do now in the next set of slides is quickly review these studies and summarize the evidence regarding the potential efficacy of this treatment combination strategy. One of the first studies was by polling and colleagues, and they used bupropion as a dopamine augmentation strategy. So what they did in their trial was kind of a full factorial design where participants received either active medication, bupropion or placebo, and that was crossed with either a contingency management condition or a non-CM voucher control condition. And what they found quite clearly is that the group of individuals receiving the combination of bupropion with CM show the better response in terms of reduced drug use. And you can see that here, the darker line on the bottom, representing that combination of contingency management with bupropion showing superiority compared to the other conditions. A few years later, we conducted a study based on the same rationale, but in our case, we use levodopa carbidopa as the pharmacotherapy strategy. Again, the same idea and the same factorial design in which patients received either active levodopa or placebo, and that was crossed with three different therapy conditions, one condition being a kind of treatment as usual control condition, another being cognitive behavioral therapy, and the third being CBT plus contingency management. And what we found consistent with the study I just described is a significant interaction such that contingency management with levodopa showed significant effects in terms of both reducing cocaine use and increasing consecutive cocaine negative urines over time compared to all of these other treatment combinations. We then did a follow-up study because we were curious about this kind of CM, levodopa treatment combination. So in this study, we tested the medication effect across three different contingency management conditions. So one CM condition targeted attendance. Patients were rewarded for attending their clinic visits. In another condition, the contingency management targeted medication compliance. And the third condition was the more traditional abstinence-based CM condition that targeted negative urine specimens. What we found was that the medication effect favoring levodopa only was detected under the CM condition that targeted abstinence, suggesting to us that there's kind of a more nuanced synergy on here between contingency management and this dopamine enhancement approach. There's also this study by Koston and colleagues. In this case, the pharmacotherapy intervention was disipramine, same type of factorial design. Patients received either active or placebo medication, and that was crossed with either a CM or a non-CM condition, and the results showed that the combination of CM with disipramine was superior to the other combinations in terms of reducing cocaine use. And then finally, this study, we conducted this a few years ago, this time looking at the SSRI citalopram in combination with contingency management. The design was different. All participants received contingency management. What we found quite clearly is that those receiving citalopram versus placebo had a significant reduction in their cocaine use, again, supporting the notion that this combination of CM, in this case with citalopram, was superior over the other condition. So altogether, I've presented, very quickly, five studies that have tested this hypothesis, this kind of general idea that dopamine-enhancing medications in combination with CM have efficacy, and the results are showing kind of a pattern of promise in terms of a signal being detected with this combination. It's always nice when there's a systematic review that reaches the same conclusion, and so this paper by Tardelli came out a few years ago, which essentially did the same thing. They reviewed the body of literature, including studies comparing contingency management with pharmacological treatments for stimulant use disorders. They identified a total of 10 studies, which included the ones that I've just reviewed, and they also concluded that the combination of CM and medications was superior to each intervention alone, and that these two interventions seem to act synergistically to enhance each other's effects. So what next? Again, we seem to have now kind of a pretty clear signal of the efficacy when using this combination, but what I want to do in conclusion here is talk about areas in need of additional research, kind of to help promote or move forward with this treatment strategy, and I think one very important next step is to consider combining CM with more potent dopaminergic medications, in particular prescription psychostimulants. We've heard Fran Levin talk about some of her work showing very positive results with extended release mixed amphetamine salts with or without topiramate, both in terms of the efficacy and the safety of this approach. As far as I know, there are no studies being done looking at the psychostimulant medications in combination with CM, which again seems like a logical next step to kind of boost the efficacy to see if we can't get a stronger effect with more potent medications. I think another area in need of additional research is to look more closely at identifying the patients who are most likely to benefit from this pharmacotherapy augmentation approach. I talked in the beginning how the hypothesis is that those individuals showing lower dopamine signaling might be more likely to benefit, but this has not really been tested prospectively, and to do this best might be through using some of these adaptive trial designs, and so there are a number underway. If you look at clinicaltrials.gov, these kind of studies are being conducted. The general idea is to do it in phases, so during the first phase, all participants receive contingency management. At some midpoint of treatment, you classify the individual as being a CM responder or a non-responder, and then the non-responders are those who then might go on to receive the pharmacotherapy augmentation during the second phase of treatment to see if you can't get a greater response to contingency management. Another very active and much-needed area for additional research is to use technologies to overcome some of the barriers to CM delivery. We've reached a point now where having patients come in repeatedly over weeks to provide kind of an in-person urine drug test is really not feasible and really presenting kind of a barrier to so many individuals who don't have the access or the time or the means to make in-person visits, so we've got now a number of mobile phone-based apps that can deliver CM remotely so that you can have the individual self-video, medication-taking, drug screening, and you can deliver the CM rewards kind of automatically and immediately using the kind of debit card payment, so I think this is likely to be a game changer when it comes to moving forward with CM interventions in the future. Even if, though, technology can overcome some of the practical barriers, there's still a lot of issues in standing in the way of disseminating and implementing CM. Studies report that only 10 to 25 percent of clinicians are using CM in clinical practice. There are a number of reasons. Some of the criticisms are more philosophical, having to do with the notion of bribery or that it's unethical to pay patients to make these behavior changes. The effects of CM can be short-lived in some cases, and it may target more on external sources of motivation rather than internal motivation that promotes more long-term behavior change. Probably the most common concern with CM has to do with cost and staff time, and there's where I think a number of initiatives have been launched by NIDA, by SAMHSA to make CM easier to implement in real-world community treatment settings. This is an example, the Motivational Incentive Suite, which provides programs with free, no-cost, downloadable tools and step-by-step instructions, including software programs that can be used to track behavior and implement the CM program as according to practice. I think there are signs of progress. I think the best example is the U.S. VA hospital system where CM is now recognized as the treatment of choice for substance use programs. This paper in 2014 describes the whole process of implementing CM across many VA sites, how providers were trained, and how the results are showing efficacy in both retention and treatment outcomes. There are also more and more community programs that are using CM, and there are some bills that, if passed, would allow state Medicaid programs to pay for CM services for substance use disorders. I think in summary, while we continue to search for an approvable medication to treat stimulant use disorders, we do now have a good amount of evidence to support this strategy of augmenting contingency management with medications that target dopamine function. The results so far look promising. There is definitely a need for more research to replicate these earlier studies, as well as to test more potent dopaminergic medications. I think as we continue to overcome some of these barriers to implementation, CM will become a more viable treatment option to use in combination with pharmacotherapy. So, with that, I will stop, and thank you, and take any questions. Well, thank you all for those terrific presentations, and I now want to introduce our colleague, Steve Shoptoff from UCLA, who's going to lead us in a discussion and question and answer session. Steve, take it away. Thanks, Ned, and if the conference presenters can bring themselves up on camera, that would be great. There's a number of questions in the chat. I just want to say, first of all, it is so wonderful to be part of a conference where we're talking about bringing medications into the treatment system and the timing of that. That's basically what I heard today, is we got something to talk about. For the first time in 25 years, we got something to talk about. So, I'm understandably super excited about this, but I think there's some real issues that we need to think about, not only the vulnerabilities of the appearances and the optics of using contingency management to get people to get control over their substance, their stimulant use, but also about the treatment system and whether we have the system in place to be able to implement efficacious medications for treating stimulant use disorder. The truth is, there's a lot of stigma that's involved in being a person who uses stimulants at the level of addiction that maybe other people don't face because the treatment system is sort of set up to deal with them, but I think the behavior sets that people bring in with stimulant use disorder present real challenges to, say, family medicine settings, to primary care settings, and often to the specialty settings with psychiatry settings and counseling settings. So, we got a lot of work to do to massage and prepare the field, and I know that's not the work of scientists, but it is part of what we have to do because policymakers don't really pay much attention to our data. They really pay attention to the arguments that we put forward, which is why I want to be excited today because if there's a policymaker in the audience, I want you to reach out, and I want to help you make decisions to bring these treatments to people who are dying because of stimulant use and exposure to fentanyl, whether purposeful or not, as a result of this. So, I'll can my enthusiasm and start asking some questions that came in through the Q&A. So, let's start with Dr. Gietze. Udi, could you talk a little bit about the role of government as the sort of the developer of last resort for medications and treatments for stimulant use disorder? Why don't pharmaceutical companies step up to the plate like they have for some other addictions and take on the role of developing medications and bringing them to market? Is that the role for government here? Well, thank you for that question, Dr. Schoch. I think a good role for the government is to facilitate public-private partnerships, PPPs, or product development partnerships, and to incentivize industry to more actively develop compounds to treat these disorders. So, there are PPPs or product development partnerships for other conditions, and so I think there are models out there to accelerate these partnerships. And, for example, it has worked wonders with the partnership to develop a COVID vaccine, and we can use a similar model here if we have the proper incentives in place. Great. Thanks. Dr. Trivedi, there's a question in the chat about folks who were non-responding, 59 non-responders to the placebo condition and not being re-randomized. Can you talk a little bit about that, or is that top of mind? You're on mute. Yeah. Say that again? Yeah. So, there were 59 people who were placebo non-responders in the ADAPT2 trial. They were not re-randomized at the end of stage one. So, can you comment on why they wouldn't have been re-randomized? So, I think that the consort diagram describes it. The 32 of those 59 discontinued in the trial, stage one of the trial, at some point during that period. So, therefore, they were declared as non-responders because they didn't meet criteria, but they were not available to be randomized. And then the other 27 who could have been randomized, but they did not make the appointment at the time of randomization. So, therefore, they missed that window when we could have re-randomized them. So, as a result, these 59 who could have been, but they didn't get re-randomized. And related, not related to this at all, but sort of a question that we all have to deal with is, you know, as we're not, there is no pharmaceutical company that's moving this forward, these medications forward for approval. It means that docs have to write for exemptions from Medicaid or other insurance companies to be able to use, say, Vivitrol off-label or these other medications. Can the physicians on the panel, can you talk a little bit about what's the strategy of getting, do you have any words that can help facilitate providers who may want to use Vivitrol and bupropion or topiramate and mixed amphetamine salts off-label and to be able to treat their patients with it? What sorts of things best get medications approved through these insurance companies? Well, it is approved for alcohol use disorder. So, I mean, at least in my studies, half the patients meet criteria for that as well. So, you can go that route cutely. And in terms of bupropion, I mean, many patients have dysthymia or depression. And to Dr. Trivedi and you and Ned are all on that study to your brilliance, you included people with depression. I mean, that's the other problem with clinical trials is they're often excluding. When I read that it was like at the clinician's discretion of safety of whether people could come in, I thought that was brilliant because you put in a grant that you get sometimes axed for including what goes on in the real world. And so, bupropion is used a lot clinically. So, I think at least those subsamples are probably in the sample anyways. It doesn't take care of everybody, but it takes care of a good portion of people without having to even fight with the insurance company. So, from a practical perspective. I think Dr. Levine is absolutely right. I think the part of the reason for this very question ultimately is stigma, right? Because there's such a huge amount of prescription in the United States for non-indication that is off-label use in the U.S. for a lot of medications. Then the condition is seen to be serious, then there is not such a debate. I think so. So, a little bit of this is stigma, but I think that Fran's suggestion of approaching it at least for these populations that have depression, dysthymia, or for that matter alcohol use, I think is the first starting point and that is something that we should be able to use. One of the questioners asked about oral naltrexone also, which would probably be easier to get the insurance company to approve. Any thoughts about that? So, the short answer is, you know, first, we did not use oral. So, therefore, we are extending it from injectable into oral. There are pharmacokinetic issues that are different. And the second part is, unfortunately, if you look at some of the earlier oral naltrexone data with net use, the efficacy has not been that strong. So, you have to really consider that as you select the treatment, but as a last recourse, I guess you should try. I mean, you really emphasize the importance of adherence and there were a lot of efforts in that trial to secure adherence and, you know, a month, an injection that lasts a month is a great way to get adherence. And, you know, one of the other sort of challenges in previous methamphetamine and cocaine trials has been, and Fran and I both alluded to it, is the studies have been, have struggled with retention and adherence to treatment. So, therefore, this population has a challenge. And so, if we give oral, it might work, but you have to be very careful in making sure they take it regularly. And that's a challenge. Great. You know, kind of related to that issue as well is like the people are asked, some people are asking questions about, has it, you know, do you have a comment on the every three week dosing practice and whether there's pushback from insurance, anything on that topic, Madhukar? I think that, again, my sense is that I don't think insurance companies, at least in the people that I have experienced to this, people adopting the advice that Fran gave about using injectable Naltrexone, insurance companies are not that rigid at following so that you can't give it every three weeks. So that is what I would try to push the envelope on trying to do it, because if you look at the pharmacokinetic data for blood levels with injectable Vivitrol, you see that the drop-off starts at week three. And so you want to avoid it if you can, but again, you have to work with the pairs, obviously. So I would like to turn the attention to the behavioral side of things. One of the things I like about this conference presentation is we're not just talking about medications. We're talking about behavior therapies as well as medications. We know that people are more than just the medications they take, and there's no way to medicate our way out of stimulant abuse or stimulant addiction. It's just not going to happen. We got to have some behavioral approaches, and it's getting even worse with COVID-19. The points of care and prevention for all of our participants and our patients are eroded or shut down entirely. So there's just so much of the social determinants that are aggravated by things like you're talking about, Madugra's stigma, that make it very difficult to help people to make real moves in terms of outcomes here. And Joy, one of the things you've been doing that I've really watched you develop over your career is this issue of dopaminergic agents and incentives and seeing that link between these medications being able to enhance the effect of a very potent behavioral therapy. So I'm wondering, can you talk a little bit about, say obviously with Dr. Levin's approach, it's pretty clear, you know, you could bring in contingency management, as you said, with a more potent dopamine agonist, which I think is an outstanding idea. But what about like bringing contingency management with Madhukar's idea about the issue of the combination of naltrexone, which is an opioid antagonist. Would you expect that there would be a bit of, one of the questions in the chat was about, what about naltrexone and incentive salience, you know? And, or would it work with the bupropion side of that? I'll shut up, you should answer. Well, I've already thought about that. We were assigned in the Madhukar study and working with those participants. I mean, the focus was on the medication, rightly so. And, you know, it had great results, but there's so much going on in their lives. And I thought certainly adding on some type of supportive therapy, whether it's CM or, you know, some kind of psychosocial supportive therapy can only further benefit these patients who have so much disruption in their lives. So I would be extremely in favor of that being the next step. Hopefully it will be at some stage. Great. Dr. Levin, there's a couple of questions in the chat about ADHD and there's another question in here about prescription stimulant misuse, or can you talk a little bit about your approach and how it might differentially help people with ADHD? And can you talk about like treatment and prevention for this other fairly limited stimulant use problem, the stimulant prescription stimulant addiction? Yeah, you wanna give me an hour? I mean, you just gave me the frame of an hour there, Steve. Thank you. I'll try to, okay. First of all, I mean, ADHD, as you know, is near and dear to my heart because I think it's terribly underdiagnosed. And despite having a training program for 25 years, I know that there are fellows out there who still run programs and don't even ask whether their patients have ADHD. So that's my advertisement for please ask about ADHD because repeatedly ADHD can undermine treatment, more impulsive group, they don't stay in treatment, they drop out and most, virtually all treatment trials have ignored it. Even the CT and STRIDE trial did not diagnose for ADHD, which I thought was unfortunate. So I think that it's something that has to be looked at. My experience in both my private practice and in my research is that you need to treat both and that stimulants offer one of these unique situations where high enough dosing treats both the ADHD and the stimulant use disorder, at least cocaine use disorder. There's been work by Maya Constenius looking at methylphenidate at high doses and doses that usually would not be used in the United States, 180 milligrams of Concerta, which is twice the FDA approved dose and worked for both the ADHD and the amphetamine use disorder. They use amphetamines more than methamphetamine in Europe. So yes, it's a big perhaps confounder, but it also can be the group that may be most responding to treatment. With our original TAC trial, the first one we did that I mentioned in the talk, we actually excluded people because we had another clinical trial going on that was taking the patients with the ADHD and cocaine. So in that study, virtually none of them had ADHD and they still responded to the combination. The second trial, we weren't really assessing for it because it does take time to assess for ADHD correctly. Just doing a screener is not gonna be a good idea and it can take almost 45 minutes to an hour to do it on top of a full psychiatric eval. So we don't really know what percentage in the second trial had ADHD, but I do think that that is a group that we should not be avoiding treatment of their ADHD and avoiding stimulants if it's done in a careful, controlled fashion. And as you guys know, you're the CTN experts, that the two CTN trials that use methylphenidate, Concerta, they did not find... Therese Winhausen did a review of the data, both from the adolescent study and the adult study, which was nicotine users, which might be different, but did not find that there was any evidence of misuse or abuse of the medication. The adolescents lost their pills more than the adults, but among the adolescents, there was no difference in losing pills, whether they were on placebo or active medication. My own experience with patients and also in clinical trials by and large, if anything, patients wanted their dose reduced. We've never really had patients say, I want my dose higher. If anything, they wanted, because of the side effects, they wanted the reduction. So that's my own personal experience, but also the experience of other clinical trialists. That's very different than misuse or abuse of prescription stimulants in high school and college students. That's a significant problem in the United States. And that's something that has to be looked at and dealt with very carefully. Most of the kids who are misusing or abusing stimulants are kids who are not in treatment actively for their substance use disorder. Kids who get prescriptions for their ADHD are often, a percentage of them are using other drugs of abuse. And that's a group that is at higher risk of misusing or diverting your medication. So you have to know the sample that you're dealing with. People who are coming in for treatment for their substance use disorder, who you identify as having ADHD is likely to do well. Unfortunately, the group that often misuses or abuses their meds are kids who don't have ADHD and getting it for other reasons on college campuses or in high school, or are kids who do have ADHD and their drug use or their substance use is sort of not being recognized and they're getting the pill. So it's not that either one isn't true, but just sort of have to know what population you're talking about. Great, thanks. And I wanna sign up for the course, cause I love that, that was great. And I would love to hear you break it down. So, Ned, we've been at these tables for many years like this, you know, luckily it's been a good career. It's been wonderful. One of the things that makes me excited about this area right now is that these guys have done the hardest thing to do in science, which is the same thing twice, or they've done something with a very conservative outcome variable and being able to use a large adequately powered trial to actually target the signal size of the medication and reducing stimulant use. So I kind of want to hear you riff on where we are. We have, you know, we've had these discussions about should we have the last two or three weeks of a trial be the primary outcome variable? Should we be looking for some sort of reductions in methamphetamine or cocaine use as a significant outcome? And given that our medications are not at the same effect size of say the opioid agonist, at what point should a clinician begin thinking about using this kind of data to drive their clinical practice? You're on mute. So I'm, in terms of the design of the clinical trials, I'm a firm believer in the high bar, the, you know, the outcome measures used by the Vivitrol bupropion trial for methamphetamine by Dr. Levin's extended release methamphetamine plus topiramate trials and some of the data that Joy Schmich, Dr. Schmich showed us as well. And my rationale goes like this. I don't really know. I don't think addictions for the most part are treatment of addictions is served well by reducing the substance use a little bit. I think if you're really gonna get somebody better, you've got to get them abstinent. And so I like abstinence. You're in confirmed abstinence as a hard outcome. And the other thing that you get with that, you get a low placebo response rate. I mean, our placebo response rates in the studies at Columbia over the years are very low regardless of what kind of behavior therapy you combine with it, other than, you know, contingency management, which we haven't done a lot of, which interestingly would raise the placebo response rate. But, you know, you look at both Dr. Trivedi's trials and trial and Dr. Levin's trial, the placebo response rates are very low. The proportion of patients on placebo that achieve abstinence is really low and a low placebo response rate is very good for the design of a clinical trial. If you have a treatment that has a real effect, you're more likely to detect it in a trial with low placebo response. So I like the secondary outcomes or the other softer, if you will, outcomes as secondary outcomes, craving, quantity of use, measures of psychosocial functioning. And I think if you look at those, you also, if the effect is there on abstinence, you also see the effect there on other things. Madhukar, wasn't there, wasn't Walter Ling's treatment effectiveness assessment also done in the bupropion-vivitrol trial, which is essentially a measure of psychosocial functioning. It's a measure of not just abstinence, but overall how they do it. I think that was also a significant advantage for the medication, yeah? Yes, in similar effect sizes, yeah. Yeah, yeah, yeah. So I'm a fan of that hard outcome. The other thing I always cite is the nicotine literature where they've been doing medication trials in nicotine use disorder for many years and they have pretty good consensus among the scientists in that field that the best outcomes are abstinence-based outcomes, whether it's point prevalence abstinence at the end of a trial or something they call prolonged abstinence, which is kind of similar to the outcomes in the trials we've just heard about. Predominantly urine or biologically confirmed abstinence over weeks at the end of the trial. How's that? My understanding also is that if thresholds other than abstinence are to be used and accepted, it would be helpful if one would show that the drug use pattern predicts sustained and substantial clinical benefit. Right, right. And that's, right. So that gets into the whole question of you've got the short-term outcome, a six-week trial, a 12-week trial, which outcomes are the best predictors of how the patient's doing six months later or a year later? And of course, in our efficacy trials, we don't tend to collect that kind of long-term data. In the clinical trials network where it's more effectiveness, community-based effectiveness in some of our trials, there's the opportunity to measure that longer-term outcome. Great. A couple more questions came in. So Dr. Schmitz, one of the questions has to do with, can you speak a little bit about how to find access for incentives, what an average incentive program would cost? How much, can you give us a little bit of the details of not only what it is, what's the sweet spot you want to hit in terms of like bang for your buck in terms of a contingency management program to produce abstinence? And that second part, a community provider, how do you go find access to incentives to be able to implement this? Yeah, and just adding onto that, I was really curious about your slide, the meta-analysis showing different technology-based contingency management platforms. I wonder if you can comment on how those platforms manage to pay for the contingencies, right? Because that's always one of the big barriers in contingency management is how do you fund the contingencies? Yeah, and that continues to be the barrier. You know, of course, I showed the slide with the VA setting being the biggest sign of progress in the field. But we've been working a little bit with some of these smartphone apps, and they've been pretty creative in working with insurance companies, private insurance companies, maybe workplace settings where they can get the buy-in to pay for the CM incentives. And, you know, it's slow going, but I think the momentum is building. In terms of like that sweet spot, that's hard to define a little bit. I think, you know, it depends on the organization and the setting. There are, of course, the lower-cost CM approaches with the prize bowl, which, you know, in terms of how much it would cost, that varies, but much less than more of the escalating kind of traditional voucher-based incentives. But there's where I think some of these tools that are now available through SAMHSA, through NIDA, that you can really, it walks you through step-by-step how to implement this, you know, what target behavior, you know, how to really adhere to the behavioral principles of CM so that it will be effective, and how to estimate the cost per individual, depending on the CM system you develop. Great, thanks. One of the questions that keeps coming back to me when I go out and give these talks is, you know, how do you work with patients with, say, Lisdex or with mixed amphetamine salts? You know, how do you work, how do you prepare your patients to work with you on these medications that have potential for abuse or diversion? And, you know, as a, you know, a doc in a box, maybe in your own practice, you know, what sorts of universal precautions are needed, if any, to be able to manage using stimulants to treat stimulant use disorder? You're asking me, right? Sure. Yeah, okay. Or you can ask Ned, too. Yeah, you can join. Yeah, we're partners here in this. I mean, first of all, you don't have to give a lot of pills to begin with. That's number one. You can give one week at a time. You can check the PMPs to make sure that there's, you know, in terms of running out early or calling you with more prescriptions more frequently. And longer-acting formulations are really crucial. I think that they have a lot less value. I mean, they have some. I mean, they're misused, too, and kids will take it or young adults will take it to stay up for exams just as much as stimulants. But when you start getting in changed roots of administration, like intranasal roots or injectable roots, much less likely with the longer-acting formulation. So by and large, somebody has to give me a very good story, which occasionally people do, of why they need short-acting versus long-acting. So I think that is the factor. There is always going to be someone or some people who are going to scam you. I mean, there's just, you know, with any controlled substance, whether it's buprenorphine, I try to make this case that it's not just stimulants, it's buprenorphine or even benzodiazepines. And you can be a doctor that never prescribes a controlled substance, but you're going to wind up untreating or inadequately treating a subgroup. And, you know, particularly where ADHD, you know, it has an effect size of one, the stimulants. It's one of the most effective treatments that we have as psychiatrists or internists to not treat patients who have, let's say, particularly cannabis use disorder or other substance use disorders is problematic. Now, on the other hand, if somebody has a drug use problem and they just want treatment for their ADHD, I'm not going to do that. Or the stimulants, or if they have just for stimulant use disorder, a comprehensive treatment must be in place. And they have to be willing and at least engage with other forms of treatment other than just simply the medication. And I think that's an important if, and many patients don't want to go over that bar and that's fine. And then, you know, I said, I tell them right up front, this is a controversial thing we're doing. It's, you know, it's off label and we have to do it very carefully. We have to monitor this closely. Some doctors like to do contracts and, you know, you can do that if you feel more comfortable doing that, you know, if you lay out all the specifics of what the expectations are. And I also said that I'm getting urines frequently. That's another thing that often psychiatrists don't think of doing is, you know, addiction psychiatrists do, but general psychiatrists don't think about collecting urine tests. And I'll do that as well, intermittently and randomly. So somebody, you know, I've got them on a stimulant and they come up, you know, positive for cocaine. I know what's going on. I know it's not going to be the same, you know? So I think there's all these measures you can put in place, but ultimately it's the rapport and relationship you have with your patient that's going to depend on whether or not you feel comfortable. And also somebody has had a dependency or abuse of stimulant medication. Well, I would never say, never say never, but that group, you have to have a much higher bar of whether you're going to start on medication. And the final thing I'll just say is that, and this will appeal to Dr. Trivedi, is you have to be very careful to assess for bipolar disorder or bipolar diatheses, because we've had one or two cases over 10 years where it unmasked an affective disorder, particularly bipolar illness. And after it being seen by two or three psychiatrists in our clinic, and it got unmasked with stimulant medication. So that's another thing you have to be very careful about, and particularly with a group of patients in which you're trying to tease it out from an ongoing stimulant disorder as well. So you don't just jump in and do it is what I'm saying. There are all these things you have to have in place. Thank you again for an amazing symposium. There were many questions that couldn't be addressed, but unfortunately we have to end it for now. Please come back at 2.30 for the workshops, and then again at four o'clock for the award ceremony. And then again tonight at seven o'clock for our film workshop on compulsive gambling and cyber gambling. So I'll see you all in a little bit less than 15 minutes. Great symposium.

symposium

combination treatments

stimulant use disorders

Dr. Nunes

effectiveness

methamphetamine use disorder

cocaine use disorder

extended-release naltrexone

bupropion

mixed amphetamine salts

topiramate

personalized treatment approaches

larger clinical trials

behavioral therapies

contingency management