Chapters Transcript Advances in Precision Medicine for PTSD: Diagnosis and Treatment Course: 2025 Annual NYU Langone Health Annual Psychopharmacology Review Uh, it's a pleasure to, uh, to have a chance, and Hannah, thank you for a great talk. My lab, which is now an alcohol use disorder and stress lab, is doing clinical trials on two drugs for alcohol use disorder, gabapentin and a carbo and topiramate. So we're trying to move forward to the next generation. And maybe during the discussion, I know, you could talk a little bit about further next generation like GLP drugs, or uh psychedelics, and so on, because there's a lot of new developments that are exciting. But thank you for a wonderful presentation. So, I'm gonna talk quickly today. Uh, about some advances in the precision understanding of PTSD. These are my disclosures, just to remind you that the first case published single case presentation in the world literature on PTSD was published 3800 years ago, uh, uh, in the ancient kingdom of Assyria. It was published by the The physician to the court of Assyria, and he cared for a soldier, a warfighter, who presented with flashbacks, startle reactions, and nightmares following his combat experience. And the patient presented saying, what bothers me most, doctor, is during the nighttime, I see people on the battlefield. Some are living, some are dead, and I'm very frightened, and I try to scream out to warn. My bed partner of the battle that's going on in his mind in this dream, this half dreamlike state, but, but, but I, but like a man, the physician said, but like a man whose mouth is seized with rancid oil, he is unable to cry out. So this is a silent scream. In a 38 year old case description of PTSD, which is generally better than the DSM committees on which I've served. Now, I would like to also just very briefly read you the single best description that I've ever found in the world literature on a case of PTSD. I Didn't find it. My good friend Aric Shaleb, who uh uh was the chair at Hebrew University and worked in our department, found it, and it's from King Shakespeare, of course, King Henry the Fourth, Lady Percy's famous soliloquy, and she is lamenting that her, that, that King Harry, uh um. That his mental illness has destroyed her marriage. And here is the quote, O my good lord, why are you thus alone, withdrawn? For what offense have I this fortnight been a banished woman from your bed? Tell me, sweet Lord, what is it, comorbid depression, that takes away from thee thy stomach, pleasure, and thy golden sleep? The key symptoms of MDD. Why dost thou bend thine eyes down upon the earth, look downcast? Why hast thou lost the fresh blood in thy cheeks and given my treasures and my rights of thee, sexual intimacy, given my treasures and rights of thee to thick-eyed musing and cursed melancholy. In thy faint slumbers by you as I sleep, I have watched and heard you murmur tales of war. Speak of managing thy bounding speech, courage to the field, and thou hast talked of sallies and trenches and tents and palisades, of cannon culum, of prisoners' ransom and of soldiers slain, and all the currents of the heady fight. Thy spirit within thee has been so at war. Thy spirit within thee has been so at war and so bestirred thy sleep that beads of sweat have stood upon thy brows like bubbles in a late disturbed sleep. Panic attacks, sweating. Depression, trauma, nightmares, perfect description, not only a brilliant DSM-5 description of PTSD comorbid with MDD, but incredible poetry. I recommend it all to you. OK. So that's the history. What about the, the diagnostic criteria in the, in our great wisdom in DSM-5, because uh DSM-5 is a democratic process, and because all win and all must all deserve prizes, we've gone from having 6 symptoms, which are the core symptoms of PTSD to 20, but please note the symptoms in red. The symptoms in red are the only ones that have been considered in every description. Of PTSD since uh Mesopotamia. Nightmares, flashbacks, trauma-related thoughts or feelings, trauma-related uh reminders, things of external reminders triggering trauma, hypervigilance and startle, everything else uh are not highly specific and sensitive, the red ones are, and by the way, the ICD 11 relies exclusively on those symptoms. So, the DSM-5 has massive symptom inflation. Uh, what are the limitations of our current base symptoms? You heard this, especially from Andy's presentation today. Uh, clinical guidelines are Let's say. The best available of the bad solutions we have to guiding clinicians how to practice. They are generally thoughtful summaries of the literature, but they actually do not. Serve their purpose, which is to guide clinical practice, because we can't, we actually can't practice using these guidelines for the patients sitting in front of us. So why is that? Because every one of our disorders, like all medical disorders, is heterogeneous, psychologically, symptomatically, biologically, and until we actually understand. That heterogeneity, we can't make good decisions. No one would treat. Ah, typical depression and atypical depression, the same way, because they're biologically heterogeneous, the family history is different, and they respond to different medications, but we have to do that for all of our disorders. I have a major study. From the DOD to do that in PTSD. It looks like the 45 main subtypes of PTSD are simple PTSD without complication and previously high functioning adults who have a single recent event. PTSD with with marked depression, PTSD with marked cognitive impairment, and PTSD with very high multiple comorbidities often related to childhood trauma, and PTSD with marked association, and it's very likely that those have a different. History, a different biology, and will respond to different treatments which we're going to study, and I, I can say just by way of attraction, in a study I just recently completed with Amit Atkin when he was at Stanford, that we found among men and women, both veterans and civilians who had PTSD with prominent cognitive impairment. They did not respond at all. To the so-called golden standard treatment for PTSD, which is cognitive behavior therapy, perhaps because of learning deficits or other reasons. So these subtypes really matter, and they will really shape, and by the way, Lynn, just for the record, it is the number one recommendation in the treatment guidelines for PTSD is to give CBT treatment even over medication, and it doesn't work at all if you have PTSD with cognitive impairment. Uh, there's other heterogeneity, and then we have all kinds of other problems with fuzzy boundaries with comorbid disorders, etc. Uh, heterogeneity and course, biology. Uh, so the next generation DSM-8 should do the following, address biological and clinical heterogeneity, be based in measurable mind, brain, and behavioral relationships, not by shuffling symptoms, provide a framework for linking multiple studies, not just animal and human studies, but postmortem studies. Uh, studies of neuronally derived exosomes, uh, studies of, uh, induced pluripotent stem cell, brain organoids, and many other approaches, which will be required to understand what is actually going on in the brain. And the number one problem, as I mentioned when I was commenting on, on, on your, uh, wonderful. We'll talk today, Andy, the number one problem we have in psychiatry to actually develop precision medicine is we don't biopsy the organ of uh of the disease. And I should say, if any of you are interested, please email me. We have a wonderful single case study published in 2015 of a man treated. By neurology and psychiatry at NYU for 20 years of treatment refractory depression, during which time he received every pharmacotherapy, every psychotherapy, two courses of ECT and other treatments, and had no benefit whatsoever. Marriage ruined, lost his job, miserable life. Ah, neurology thought he might have a rare neurological disorder that explained this problem. They did a needle biopsy under neurosurgical conditions to make sure it was safe, took a small biopsy from a silent area of the prefrontal cortex. Sent it to neuropathology, found massive invasion of inflammatory cells in the biopsy, put the patient on inflameximab, and the patient was not depressed for the first time in 20 years. And I just leave you with one question. What was more dangerous to that patient? Having his life blown up, spending $100,000 on wasteful treatment. Ruining his quality of his life, marriage, and work, and being miserable with depression for 20 years, or doing a single safe microbiopsy of a silent area of the brain with no risk other than a 1 in 10,000 risk of infection, which is not gonna occur under neurosurgical conditions, and have your life changed. It this is why we don't have precision psychiatry. So we're gonna have to develop ways of essentially doing a brain biopsy, using a series of techniques such as postmortem samples, and exosomes, and brain organoids, and other approaches, so we can essentially perform a biopsy without having to routinely go into the brain. OK, so I, I don't have to tell you why biomarkers are important, uh. Imagine, uh, did any of you, are any of you as ancient as I am, did you train? I, I was trained as a family doctor before I went to psychiatry, and I, I treated cancer in my practice when I was a when I was a kid in Toronto, and I, I treated breast cancer and prostate cancer without the benefit of biomarkers. So it was based on trial and error. You tried radiotherapy, you tried chemotherapy, you tried surgery, you tried the combination, but there were no guidelines for which one, for which person. Now, if you practice oncology today, without the use of biomarkers to inform your diagnosis and treatment decision, it is malpractice. And 25 years from now, when we all meet here again, if you practice psychiatry without the use of biomarkers, it's gonna be malpractice. OK, uh, I'm not gonna go into all the phenotypes, you know what they are. There's there's not only clinical phenotypes which I described to you, but there are biological phenotypes, molecular pathways, circuits, cognitive behavior, and so on. There are, uh, there are lots of, if you're interested, we wrote, uh, we published in the New England Journal of Medicine myself, Israel Lieberson, and Aric Shaleb, a set of models to, to outline the deep neuro. Circuitry of PTSD, it's a little bit complicated, but just to say we now very well understand in animal and human imaging studies, the circuits that are disturbed in PTSD, and they are disturbances in emotion regulation and threat. I love this typo salient detection, threat and salience detection. Uh, contextual processing, meaning inability to differentiate a dangerous from a safe context. If I've been sexually assaulted, and I'm now going on a dating experience, and I'm trying to become comfortable again with sexual intimacy, I have to be able to update. Context so that I'm not feeling I'm about to be assaulted again. Let me give you a clinical example from my veteran practice. I've had the great privilege of my whole career caring for veterans beginning with World War II, prisoners of war, all the way up to current veterans. I was treating a returning Iraq War veteran. Uh, who was blown up on, uh, when he was out driving by IED and an improvised explosive device, and a close comrade of his was killed. He survived. He was not badly injured. He had a master's degree in, in computer science and was working in a high-paying job on Wall Street, and he was driving uptown to see me uh uh uh for a consultation. It was a beautiful Saturday morning like today, a sunny day, and he drove up Madison Avenue from downtown to my office at One Park Avenue, and by the time he arrived, he was totally exhausted. Anybody want to guess why? Why was he, why it was a beautiful day. It was a pleasant ride. He arrives exhausted and sweating in my office. The answer was at every intersection on Madison Avenue, he had to check the garbage cans on all four corners for explosive devices. So he had failed to update the context. He still thought he was a convoy operator in the war zone, and so context updating has a specific brain circuitry, and of course the one we understand best is the fear learning. Circuitry, actually, Momala, who is a close colleague of ours, contributed to this literature, which the amygdala. So the three main brain areas are the amygdala, the, the medial prefrontal cortex, ventral medial prefrontal cortex, and the hippocampus, and the cingulate. OK, so I'm not gonna go into all the molecular phenotypes of PTSD. We don't have time today, other than, you know, that it's very complex. Andy, to your point. The ridiculous idea. Promulgated by NIMH that we can only receive funding if we have a single target for a single treatment for a single disorder was an idea promoted by an individual who never treated a patient or had any psychopharmacological experience. OK, so we know that the adrenergic system is dysregulated, so is the HPA axis. So are many other systems. The renin angiotensin system is disturbed in inflammatory and hypertensive responsive and excitatory and inhibitory amino acids are disturbed, endocannabinoids and dopamine, and others. Actually, in my recent study. From NIAAA we have identified 8 different molecular pathways that we think there's very strong evidence are disturbed in PTSD, AUD. There are inflammatory, oxidative stress, reward, reward hor serotonergic signaling, dopaminergic signaling. Gabba glutamatergic signaling, mitochondrial dysfunction, and several others, and we're finding clear evidence that they're all 8 are dysregulated, and all 8 are targets for AUD PTSD treatment. Uh, these are some of the molecular phenotypes. Uh, just to remind you that PTSD is highly, um, uh, uh, prometabolic syndrome. It disrupts, if you compare people with and without PTSD, including those who've been exposed to trauma and don't develop PTSD, we found abnormal glucose levels, insulin signaling. Uh, homo IR, a good measure of insulin resistance, hypercholesterolemia, increased BMI and weight, increased pulse and every other feature of metabolic syndrome. So, disorders like depression and PTSD, leaving aside the treatment with medications, are already pro uh causing major metabolic problems. And uh of interest, uh, these are the pro-inflammatory signals we found in PTSD. Uh, inflammatory cytokines are all up regulated, and one of the most interesting things we have found is we have studied, uh, immune functioning in PTSD cases and trauma exposed controls, and we found that natural killer cells, which play a key role in in managing uh invading pathogens, uh, prematurely age in PTSD. And so that there is senescent with all of us as we get older, one of the reasons why older people are more vulnerable to infection and had to be obviously of great concern during the flu season, COVID, and so on, is that there is a natural aging of the immune system with time. And one of the key features that ages are the effectiveness of natural. Killer cells, and natural killer cells are a second line of defense. They go inside cells that are infected, and, and clean them out and kill them. So if you have weakened natural killer cells, you have very greatly weakened immunity, and stress, severe stress weakens them, so that the natural killer cells of middle-aged veterans look like older Americans. OK. So I'm gonna jump ahead. Uh, I wanna give you a couple of examples of our precision medicine studies. So, We, uh, we, one of the things, one of the things we're trying to do is develop biomarkers that could be used in everyday practice, including in clinical practice and primary care practice. So we, we've, we've now studied this and replicate this in a, a number of samples. We started with 343 male Iraq and Afghanistan veterans, and we, um, we sampled their DNA and RNA. And their proteins, and we started with 1 million molecular features in their blood, and we used a technique called wisdom of the crowds in which we had a series of consultants meet with us, Andy, we might try this in the bipolar study. You collect the DNA and RNA on everybody. It's inexpensive. So if we could get everyone to agree that during their patient in the in the bipolar network is during the annual physical exam, we could draw blood for DNA and RNA and get the patient's permission for that, and some symptoms, we could do this. So we did this on all these subjects. And we also got plasma for proteins, and we started, uh then we sequenced all of that, and we had over 1 million features. We had a panel of experts, including our own consortium, which was 76 universities in the Army's molecular biology lab. It's an effort that I led with Frank Doyle at Harvard and, and Rasha Hamamaya and Marty Jedd at the Army, uh uh and Walter Reed, and we Got a consensus from those million features, which 343 would be of most interest as possible blood biomarkers for PTSD. And you can see some of the candidates here. I'll just shoot ahead. And uh we then compared 100 cases and 100 controls, and developed, we did, we use machine learning to develop a discover model to get a more limited set of 25 or so uh molecular markers, and then we discovered them in one sample, and then tested them in a completely new sample that did not participate in the training phase. And here's what we found. Um I'm not gonna go into all the details. It shows you the performance of the features, but the most important thing to show, these are the most important features in the random forest. We found 28 markers. We found 28 markers that actually carry. Um, the majority of the single. Not sure what is is. Uh, it doesn't matter. Um, so I can show you. Can you see this? Yeah, yeah, yeah, OK, good. So, um. Charlie speaking to the mic. We, we We're able to find a small number of a small number of molecular features, uh, about 28 or so, uh, that carried the main signal from the original large number we started with, and using uh a very sophisticated testing to determine the area under the curve of the receiver operated curve, we found an overall area under the curve of 80% meeting. Using these 28 features, we were able to classify PTSD cases and controls from the simple blood test with 81% accuracy, and with a high sensitivity of 85%, and Uh, I don't, I don't know why this is not. Well hold on it, that's it. OK, um, So, uh, this is the first demonstration that we could, uh, and we now have replicated this in new samples. We now actually have a, I've just, I wanted to show you this, but we're filing a patent for intellectual property in the consortium on a set of 12 proteins, which are much easier to ascertain than all these DNA and protein. We have 12. proteins that actually perform this well or better in defating PTSD cases from controls, and the FDA is reviewing that now, and we expect to get approval to be able to test this in a large number of military personnel and then civilians. Um, what else to show you? Uh, 11 other precision medicine study, which I think is very interesting. Uh, we're, we're very interested in finding low cost, ubiquitous, non-invasive biomarkers, and, and, um, Ayana, to your point about advancing practice, we must always be concerned about the wonderful points you raise, which are stigma, patient burden, and patient trust. And we think that we'll have more, we think that most patients in primary care would agree to a routine blood test. That's part of what it's part of standard practice. But even better, we've been looking at, we've been recording the voices of uh participants in our studies with PTSD and the voice quality of participants who've been exposed to the same kinds of trauma. Civilian and military trauma, but who've never developed PTSD, we record their voice with high quality voice recorders. We ask them to talk about something very stressful, that their traumatic event and a neutral subject. We found we only need 3 to 5 minutes of their speech to do this test. We send the speech to. MRI, we send the samples to Stanford Research Institute, and you know how research works. It was pure serendipity. One day I got a call from Dimitri Verry at and Paul in Menlo Park, and she said, Charlie, I'm very interested in your work with veterans, but you probably don't know who I am. I said, Dimitri, nice to meet you. Who are you and what do you do? She said, I lead a team of 25 bioengineers at SRI who specialize in voice quality and the analytics of voice, and I said, well, what have you been doing? Well, we've had a few small projects along the way. One of them is we created Siri for Apple. So I thought, well, this might be a pretty good group. They also in their spare time, created the software for Dragon Naturally Speaking, so you can dictate and transform. So these are the world's leading bioengineers of voice, and they said, if you send me our voice, your voice samples, we will run them through our pipeline. We will, we don't want to know anything about the content of speech. We want the biophysics of speech, and we will fractionate. This, the waveform of human speech into 40,000 unique biophysical elements, which is how we solve the Siri problem, and we will tell you how to differentiate your PTSD cases from control. So I thought that's very interesting, and we can talk about why speech, but obviously why did speech even originate for humans, probably to signal safety and danger and attachment, so, But you can see all the advantages, and there's speech everywhere now with Ayana, you, you caution us about cell phones, but one of the great things about cell phones, not to compete with your lecture, but we have speech everywhere now. It's cheap, it's abundant, it's non-invasive, and we can study it, and we're doing that in a DARPA study. Now. Anyway, so we've done this, and we ran, this is the, this is the SRI. A pipeline that developed Siri and other applications. Uh, we took the speech, we fragmented it using Fourier analysis. We got all the elements, the 40,526 features. We used machine learning to find out which ones would predict PTSD or not, did a training sample, a test sample, and we found. Any guess, Andy, how accurately can we classify cases from controls using our best speech features? 90%. 89.1% speech blind they did the SRI group did this blind to any clinical assessment with no knowledge of any of the content of speech or any clinical feature. All they had were 3 to 5 minutes of speech samples from the cases and controls, and they got 89.1% accuracy. On the basis of that, Thea Gallagher and I just received a DARPA grant. To study this in 1000 stressed healthcare workers, and uh we're gonna see if we can replicate and expand this. And it's interesting, I, I would have guessed the the features would have been anxious arousal features, but it turned out the features, at least in chronic PTSD were more flat tone, muffled tone. Difficulty in transition and tonal frequencies, so it's kind of flat monotonal muffled speech, which may, may be related to numbing of emotions, I'm not sure, but those are the features in the first study. OK, psychopharmacology, I think when you asked me to give a talk on psychopharmacology, so I, but of course, yeah, you, you know, it's a very odd irony. Of how someone I I become the chairman of one of the largest departments of psychiatry in the country. I'm responsible for supervising, managing, mentoring, supporting, and loving all of my faculty and trainees. Uh, but if you read My 3rd year report card, which I should have put up as a slide because I saved it, my 3rd year teacher said Charlie is an unusually bright student, but it's important for his parents to know he seems to work best under close supervision. And that's what happens. I became a supervisor. What can I tell you? So, uh, Lynn, I'm somewhat unmentionable and unsupervisable, according to my third year teacher, so I'm doing my best here, OK? All right, so what do we know about the drug treatment of PTSD? So that's my way of, I wanted to start with what's the next generation. We're committed to developing, and in our work with you, Andy as well, to get reliable, non-invasive. Sensitive and specific diag uh biological features to help us with diagnosis, and also more importantly, to guide which treatment for which person at which time and which phase of their illness. So I, I'm hoping in the future, when you give your beautiful slide of the, the, you know, 3 by 3 or 3 by 5 treatments, that it could actually be more nuanced and we could then say, which ones at which time for which person, with which history, with which biology. That's where we're going. OK. But what do we know from the broad literature? First of all, how many FDA approved drugs for PTSD in the world? Any guess? 2. Ah, for a disorder which affects. Uh, 4 to 5% of men, 8% of women in their lifetime, we have two FDA approved drugs, Paxil, paroxetine. This original study showed that paroxetine separated well from placebo, by the way, 20 mg was just as effective as 40 mg. And sertraline, which is probably the most, this is Kathleen Brady's 2000 study, sertraline is probably the most broadly used drug in the world for PTSD. It is very effective and well tolerated. And uh just to remind you of something that you already know from your psycho pharm career, in addition to um Ayana's excellent advice to start low and go slow with many of our psychotropic drugs, so you don't run into too many side effects. I, I, I used to be quite friendly with Don Klein, uh, uh, we collaborated with him at one point. He was on Uh, kind of a scientific senior external advisor to our, to Marty Horwitz and myself when we were at UCSF, and I used to go out for lunch with Don Klein, and he used to say, remember Charlie. There's some very simple, and he was the father of psychopharmacology. He said there's a very few simple principles of psychopharmacology. Most patients don't do well because therapists, psychopharmacologists start at too high, too high of a dose. They dose up too quickly. They get the patient to a treatment effect, then they discontinue them too quickly. That's in psychiatric practice. In primary care practice, it's the opposite. They don't dose up properly. So between between primary care doctors who prescribe 70% of all psychotropic medications, not getting patients to a therapy. Therapeutic dose and psychopharmacologists being kind of cowgirls and cowboys and flying up the dosing schedule, uh, we don't do a great job, so you have to start low, go slow, but remember you have to stay long. So, uh, if you look here. You can see this is simply what happens to people at the completion of a 12-week trial, double-blind trial of sertraline, the patients went into an open trial for another 24 weeks, and they continue to improve quite significantly from week 12 to week 36. So our recommendation is, if you treat a patient who's had chronic PTSD with an SSRI or another drug, there are others. Venlafaxine, and bupropion, there's a number of choices. Uh, you should stay on a minimum of 6 to 12 months at a therapeutic dose, and you should taper over another 6 months. OK, uh, venlafaxine, uh, this is a study by Jonathan Davidson. Venlafaxine is just as good. Uh, this is an extended release form as sertraline and paroxetine, and, uh. Ayana, the reason that venlafaxine is not FDA approved, is the same reason you mentioned money, cause it was off patent by that point. But it's a terrific choice, and venlafaxine and bupropion. May be somewhat better for men. And, uh, and women may do on average a little better, especially during reproductive years on SSRI's and, and women as they go at menopause and in the postmenopausal period. Often then don't do quite as well on SSRIs again, unless they're on estrogen replacement therapy. So there's some preliminary data from Alan Schatzberg and others, and this is not strong evidence yet, because we haven't done what you're describing, which is these very large studies, so we can actually answer this question. But one of the things we should at his sex effects. And one of the things that seems to be one thing that needs to be looked at very closely is whether there that men and women actually differ in the benefits they get for stress, anxiety, and depression from SSRI's versus more mixed receptor drugs. That's just something for the future. Uh, MAO inhibitors, how many of you use MAO inhibitors in your practice? Wonderful, yeah, the, the, it's, it's a measure of how many years you've been in the trenches as a psychopharmacologist. Uh, those of you who trained with me, uh, which was, you know, maybe a couple 100 years before the big meteorite hit that took out the dinosaurs. So those of you, those of you who are long of tooth to, here's a little fact that we don't wanna talk about. MAO inhibitors are mostly safe, particularly with reliable patients where you can be careful about monamines. There are reversible MAO inhibitors, which are even safer, and pop quiz for Daniel Soescu, head to head, if I compare MEO inhibitors, tricyclic antidepressants, SSRIs, SNRIs, and other drugs, which are the most effective? I mean, there's uh one study where an SSRI plus mirtazapine was a little bit better than MAOI. In general, MAO inhibitors are the most effective treatments for stress disorders, anxiety disorders, mood disorders, and every other disorder. So, another one of the great mysteries of life, why we don't prescribe them. Uh, trazodone, not very good as a full antidepressant, but trazodone between 50 and 150 mg, is very useful for insomnia and PTSD. And remember, our, our first two cases of PTSD I, I gave you, my Syrian warfighter and, and, uh and my Shakespearean quote. Sleep is the most profound disruptive symptom in PTSD, and we have to be able to deal with nightmares, and insomnia, and trazodone 50 to 150 mg is quite good. There are a few studies for nightmares. Um, there's some suggestion in some, but not all studies that prazosin is specifically helpful for night for trauma-related nightmares, and the recommended dose is 5 to 15 mg. Uh, I have a lot of experience. This is Murray Raskin's work from Wash from UW. I, I have a lot of experience in my own practice with prazosin, just a cautionary note, prazosin. Um, causes a lot of hypotension when a person, if a person is naive to it, so I generally start with 1 mg. And I asked them to sit in a chair or get into bed, take 1 mg, be careful of falls, and it takes a week or two to acclimate. But once you do that, you can dose them up to 5 1015 mg. But you must don't push it too quickly and be careful on the initial doses. And also, you have to be a real doctor. You actually have to take people's uh standing and sitting blood pressure, or it's not safe to prescribe prazosin. Murray Raskin found, by the way, in the interest of precision medicine, then you'll give me a 10-minute warning, right? Am I there? You're not there. OK, you'll give me a 5 minute warning. I, I, I can go fast. Um, Murray Raskin found that patients who, um, have the biggest blood pressure changes comparing lying to standing blood pressure, who are most adrenergically reactive to that, do best on, on prazosin, so that's something to keep in mind. Uh, uh, Andy, our old friends lamotrigine, gabapentin, uh, tiagamine, uh, some studies, we don't know a huge amount about anticonvulsants in PTSD. I use a fair amount of gabapentin, particularly where, uh, for PTSD comorbid with alcohol use or substance use disorders. Um, atypical antipsychotics are used a lot in complex PTSD or treatment refractory PTSD. I've not been deeply impressed from the oral literature. There's lots of side effects, and in general, the efficacy is not great. So, I, I, just as a general principle, I Discourage the use of atypical antipsychotics and the use of benzos for PTSD. They're both not very good. Benzos interfere with fear extinction learning, and anti atypicals have a lot of side effects, and they're not, they're not, they're generally not well tolerated. Uh, there are some special cases, PTSD patients who get, who get have sudden anger outbursts. One of the great pleasures in my practice, I've been treating PTSD, uh, since before it was a diagnosis. And I cannot tell you how many times in my office, people have blown their top, but mostly caused by my unempathic misunderstanding of their life. But, and, but my, my non PTSD patients will say to me, uh, Doctor Armor, that, I don't, that doesn't really resonate with me. Are you sure? And my PTSD patients go ballistic. So there's a sudden go from 0 to 100 with irritability and anger, and there's some combination of medicines you can use for that, including mood stabilizers. Uh, PTSD and alcohol use disorder, I think Ayana, you covered this brilliantly, disulfiram, the Camprosate, naltrexone, gabapentin, topiramate, and CBD is interesting. Maybe we were, we're looking at it. Esther Blessing in our group is looking at it for PTSD and of course, uh, Uh, we're going to look at, at, uh, some other psychedelics as well. Uh, PTSD and substance use disorder. Avoid benzos period, not just in the context of substance abuse disorder. Avoid using them during it for acute stress disorders. If a patient presents to you 6 days after a sudden traumatic event. Supportive therapy, and also the use of adrenergic blocking drugs are much better than benzos, uh, for, uh, and these are the combinations of drugs we use for substance use. Uh, children, interestingly, children and adolescents may respond better to adrenergic modifying drugs like clonidine, guanfacine, and propranolol than to the SSRIs. That's very interesting. Um. You know, to go start low, go slow, stay long. New directions, there's all kinds of, it's totally disgraceful that a disorder that affects 6 or 7% of the entire world's population. Question, I, I've only, I never used to give pop quizzes during talks till I met Ayana. And so you can, you can thank her or not, for, for, for expressing the pop quiz gene in me, which was latent, but is now fully expressed. So I'm gonna ask you a pop quiz. If you were able to survey primary care doctors' practices in sub-Sa Saharan Africa, what would be the most prevalent psychiatric disorder you would find? Any guesses? Everyone says major depression. It's PTSD. PTSD is about 1/3 more prevalent than depression in sub-Saharan Africa, because of massive lifetime trauma exposure from every natural disasters, HIV, but number 1, interpersonal violence. Another cop pop quiz. Andy, this one's for you. See, one of the things my, my, my residents all, all know this above me now, so they try to sit in the farthest seat away from me during seminars. And of course when they do that, I always direct the first question to the farthest seat away, any of my, yes, Thomas, he's a very easy target, Thomas, he sits right across from me, he gets blasted with pop quizzes. So, uh, Uh, what was the question I was gonna ask you? Something I've lost my train of thought on this. Uh, it'll come back. Yeah, yeah, yeah, yeah, this was another really fun pop quiz. Uh, OK, I'll come back to it, um. There there's a lot, a lot of new drugs that we're looking at. And uh we're looking at, uh, you would think with the profound dysregulation of the HB axis, we could find some agents, uh, CRF antagonists, and, uh, or, or, or related drugs that will be helpful. So far, not too exciting. We're looking at endocannabinoids, oxytocin, neurokinin, uh, ketamine, NMDA, CBD. Uh, and, uh, and psilocybin. Uh, IV ketamine has been studied by Adriana Federer at at Sinai. Adriana, a wonderful investigator. She was, I had the privilege of being one of her mentors when she was my resident at UCSF, uh, NMDA assisted psychotherapy. You probably all heard recently that the FDA declined the NMDA application, uh. Uh, the, uh, I, I read that very closely and followed the story, because at NYU we actually did, we participated in the phase 2B, I think, trials, uh, Steve, Steve Ross. Unfortunately, there were a number of ethical breaches in the, in the original trials, including A therapist in British Columbia, who treated one of the patients in the, in the original trials with NMDA uh entered into a sexual relationship with that patient during the course of the treatment, and then retained that patient in the outcome trial, and then on investigation, had been found to have lost their license to practice psychology in the state, in the province of British Columbia. So very major ethical breaches, and the FDA is also worried about abuse potential and uh placebo responding, a lot of problems. Um, this is the data from one of the MDMA assisted treaty which looks promising, but there, there's ongoing ethical concerns. Uh, I'll skip over that. Uh, neuromodulation. So, in addition to when we come back, and we all have a meeting together in 25 years, which I'm looking forward to seeing all of you, uh, uh, we, in addition to the fact that you'll be telling us that you're using blood biomarkers, brain imaging to decide who, who needs what treatment and what their diagnosis is, your practice is going to be markedly shifted. Over time, from pharmacotherapy to neuromodulation therapy, and the reason for that is that pharmacotherapy is like killing a mosquito with a bazooka. Pop quiz, who's my next? Thomas, how many of the 11 billion neurons in the human brain, Thomas is a very distinguished neuroscientist, a psychiatrist, uh, of the 11 billion neurons in the human brain, what percentage of them have serotonin receptors? Say at least 20, probably more. At least 20, it's 100%. There is not a single neuron in the human brain that as far as I know from my last analysis, that does not have a serotonin receptor. So when you, Happily, and in your very relaxed prescribing style every day in your practice, give someone an SSRI and you have the fantasy that that medication is going into their bloodstream, and it's going through the blood-brain barrier, and it's going to their basal ganglilia or to their VTA and it's tuning up their VTA and making everybody smile and be happy or wherever it's going. Uh, you should also remember it's going to every other neuron in the human brain and changing their signaling properties. So, what we want to do is move away from Medications, which bathe the entire brain with with molecular pathway changes to circuit-based approaches, uh, which actually target specific disruptive circuits to the extent that we understand it. So I'll just make a prediction, 2020, 25 years from now, you'll be doing much more TMS, uh, neurofeedback. And other approaches to medication treatment. And all these are being tested in PTSD. Glenn Adler and I have a, uh, have just completed a five-site open trial of uh FMRI informed EEG neurofeedback for PTSD which is super cool. The patient comes in once or twice a week for 8 to 16 weeks. Their task is to put on an EEG head cap, to watch a video screen, in which they see avatars on the screen, and they have to learn using mental control how, thanks, 5 minutes, they need to learn how to take people who are in. The initial scenario, they're very agitated in a waiting room because their flight has been delayed again for the 3rd time and they're up complaining at the counter that they're very upset that their flight's delayed, and your job is to get them to calm down and go back to the seat. And the way you do that is by engaging, down regulating your amygdala. So this is an EEG driven direct amygdala control procedure. You use prefrontal cortical efforts to down regulate your amygdala, and that's very hard to do with EEG, but our, uh, because EEG doesn't assess amygdala activity, so. What our collaborator who developed this technique did is she put people in an fMRI machine and simultaneously acquired FMRI and EEG signals, says Talma Hedler, Tel Aviv University, and then she had the EEG signal and the amygdala signal from the FMRI. In real time together and use machine learning to find out what features in the EEG were associated with changes in amygdala activity, and then you can use those in the EEG in your office, so patients don't have to go, they have their brain scanned. So this is an incredible technique, and we have about to start a randomized controlled trial of a uh neurofeedback and sham neurofeedback, but it looks really good so far. Adolescent and we have an adolescent trial for PTSD and then you're developing trials for ADHD. Len has an open trial showing that EEG neurofeedback is very helpful for, um, I think for mind wandering and, and some executive, he'll, he'll show that it. OK, great, uh, conclusion. We need to move from practice guidelines to precision medicine. Uh, we need to know what subtypes of patients. With which clinical features and biological features, and psychosocial and cultural features, different and need different treatments at different times, so that when you are referred a new PTSD patient next week, you're not thinking, oh, I better start with sertraline, cause that's FDA approved, and the answer is no, you shouldn't. This patient needs to start with prazosin or something else. OK. And that's what we call heterogeneity, as you know, and this comes into the whole issue that I know that you were talking about the number needed to treat. So if we treat a heterogeneous condition with non-specific drugs, it'll take 20 to 15 and 50 and 18 to treat. But if we could treat a very specific problem, With very specific treatments, it would be 2 to 4 necessary to treat. Buprenorphine is 2 to 4 now. Uh, well, that is totally buprenorphine is totally amazing. Uh, I think I'm gonna wind down. Uh, there's some, there's some information on evidence-based psychotherapy. That's for another day, other than to remind you, if your patient has significant cognitive impairment, if they have major problems in attention, concentration, and working memory, they're likely to do poorly with behavior therapy for PTSD. And I mentioned praises and for you, and that's it. Thank you. Published March 1, 2025 Created by
CME Knowledge Hub