Platform Trial in Psychiatry with Dr. Husseini Manji

Mike is a, uh, 30 year veteran of drug development, uh, in neurology, and even, uh, has been at Pfizer, YFJ and J and headed up, uh, quantitative sciences at j and j, uh, before joining Berry Consultants this year. So, welcome Mike. And, uh, interestingly for the topic today, Mike was really a driving force behind the creation of EPA D and IMI funded platform trial in Alzheimer's, uh, uh, for that. And I'm also joined today by Dr. Husseini Manji.

Uh, and we could spend our whole day, uh, with, with the, uh, accolades for, for, uh, Husseini. He is currently. Uh, at Oxford University, he's co-chairing the UK Government Mental Health Mission. Uh, he is a professor at Oxford University. He is the past global therapeutic head at for neuroscience at j and j where he and Mike, uh, worked together there. Uh, he, so really interesting. So currently in academia. Uh, government work.

He was at the NIH where he headed up the NIH Mood and Anxiety Disorders Program, the largest program of its kind in the world. Also the laboratory for molecular pathophysiology at the NIH. Uh, he's a, he's inducted into the National Academy of Medicine, has multiple prizes, but I thought this one was sort of interesting. He was one of the 14 inaugural health heroes. By Oprah Magazine. So welcome, uh, Husseini.

Involved with j and j and his ketamine, uh, one of the fir the first, uh, antidepressant in, in 30 years. Uh, so you've been down this road. So tell me your interest in precision medicine and psychiatry and what this platform trial could potentially, uh, you know, uh, solve and be.

So I think mental illnesses really represent a significant global challenge with profound health, economic, and societal impact. One of the reasons they're so devastating is that by and large, they strike individuals relatively early in life, usually late adolescents or early adulthood, and they're lifelong. So they're the chronic diseases of the young.

And not only do they cause direct pain and suffering, but they affect every aspect of people's lives, including physical health, social functioning, economic, economic functioning, and um, academic performance. So there's a real need to come up with improved treatments for these conditions because you know the unmet need is staggering. And yet despite this unmet need over the last few years. Many large pharma companies have actually exited this space.

And it's a, you know, so one could say it's actually a reasonable business decision on their part. And why I say that is that they've been concerned that compared to many other, medical illnesses, these are very complex and, it's not been as easy to come up with this precise target for a variety of reasons. One is that the brain really is the most complex organ we have, and it's relatively inaccessible. Compared to peripheral organs.

So we do a lot with imaging, we do a lot with peripheral biomarkers, but you can't directly do a brain biopsy, so you've gotta be, you know, a little bit more indirect. Secondly, as you may well, um, know that the animal models don't translate very well into humans. And so we've got sort of get into humans early on, and that could be one of the real advantages of this platform.

The other thing to mention is that while they're amongst the most heritable of a lot of medical conditions, they arise at the inheritance of hundreds of suscept susceptibility genes that interact with environmental factors. So there's no single driver mutation like there is an oncology I. And then probably the most relevant for this discussion is that the Neology is based largely on a syndrome of subjectively determined criteria.

So you basically ask people to tell you about their mood, their appetite, their energy, and so on. And there's almost no doubt that we put these labels like depression, bipolar, schizophrenia, on a group of heterogeneous disorders. And not surprisingly that if you take, heterogeneous groups of disorders and basically apply a broad label to them and take a one size fits all approach to them, things don't work as well. And so that's why we're so excited about the idea of precision psychiatry.

I. Can we really get to this away from this one size fits all approach and similar to what's happened in areas like oncology in particular? Can we start to move into more of a precision based approach where we can use patient specific information? To basically divide the broader clinical population into subgroups that are more likely to respond to specific treatments.

So rather than focusing on symptoms, behaviors, neurotransmitters, et, cetera, we can start to look at illnesses based on their fundamental biology. Neuronal circuits other biology and use that as a way to really get into some subgroups who may preferentially respond to different treatments. And that's where I think a lot of the advantages of what you and Mike are driving in terms of precision medicine, biomarker based adaptive trials could be game changing in this space.

Comparative effectiveness, so we're not looking at things already approved. Uh, healthcare, how they treat them. You're thinking about phase two type exploration is where multiple pharmaceutical drugs may come into this, that that's kind of a sweet spot.

So if you could have a platform trial where you could interrogate a number of these novel, pathways mechanisms in these phase two type of clinical trials, et cetera. Even, failures as failing early is also very important. So you can redeploy resources to things that might be more successful. And as I said earlier, there's also a lot of reason to believe that while we've got. Made a lot of advances in the science. These really are very heterogeneous groups of disorders.

So even if you had, quote unquote the right target, if you apply it to all comers, you're probably gonna lose a signal. Whereas if you could apply it to a targeted population, then you'll really know whether it works or not. And again, based on the work you and others have done with biomarkers, would allow us to, move in this direction. I'll just mention one more thing, related to your comment.

So I think in my opinion, it's absolutely correct that the sweet spot may well be the, Phase two proof of concept trial types of trials. But there is reason to believe, so there's a lot of, scientific data to suggest just like in neurodegenerative disorders. In neuropsychiatric disorders, the immune system, I. Actually seems to play a major role.

Once again, it's not going to be in all comers, but if you can identify people based on some sort of immune diathesis, they may respond to an immune targeted drug. And the reason I'm bringing this up here is that, there are a number of these drugs that have been developed for, example, autoimmune disease like rheumatoid arthritis, inflammatory bowel disease, psoriasis, et cetera. That may hit the right targets in terms of say, IL six, TNF, alpha IL 1223, et cetera.

So what we're thinking is that there may also be the possibility of using this for, approved drugs, but in a sort of repurposing, approach where we can see, with the immune, neuro immune being one of the best examples, I think. Okay.

So they're present in bipolar disorder, schizophrenia, et cetera. And what we've been trying to do as a field is exactly what you said is try and map some of these symptom domains. Onto biology or pathology so that the idea might be an IL six monoclonal antibody, or IL 23, et cetera, could target the circuits, neuronal circuits responsible for these symptoms. Whether we call you depression, bipolar. Schizophrenia and so on.

And so there's something that's been, um, the term that has been used is trans diagnostic. And the very good news is that regulators, um, both the FDA and in the uk, we've been interacting with the M-H-R-A-I. Um, sort of get it. They understand that, you know, sort of our old classification system probably isn't as conducive to coming up with novel treatments, so they're very willing to look at this trans diagnostic approach. Clearly, regulators need to see the evidence.

You know, they can't just take a word for it. But, so for example, there's a lot of interest in the possibility that if you could show that something works, for example, for this domain called anhedonia, which is very impairing, you could potentially, potentially get the approval to use it to treat anhedonia, whether your official diagnosis is major depressive disorder, bipolar, schizophrenia, and so on. So what you said is spot on. That's the way we're thinking about this.

We have, uh, GBM Agile, uh, is a trial looking at glioblastoma, where in that case are some, uh, strata that are broken up, but there a drug can come in and say, I'm interested in mutation X.

What do you think the, the, the, the drugs out there are going to be and what, what should the trial try to answer?

And, um, a lot of, um, byop biological information was collected, various kinds of omi et cetera, to say, can we sort of see, you know, what the. Determines response to this modality of treatment, et cetera. But I think the former aspect you mentioned is actually going to be more exciting. You know, prospectively saying, look. We have a drug that affects this neuro immune cascade or this hippocampal plasticity cascade, or you know, this sort of, um, excitatory I inhibitory balance, et cetera.

We think it should work on this sort of subtype of people. Can we recruit those types of subtypes of patients and in fact, demonstrate that this holds up as you know, you know, sort of companies. Um, so one of the things about pharmaceutical companies is that I would say, you know, probably 15 years ago or so, people wanted to treat all comers because you had a larger, you know, population.

I think now there's a recognition that if you can treat a subgroup really well, and maybe your drug becomes the treatment of choice for that subgroup. That's actually pretty good. So pharmaceutical companies have gotten away from this notion of, let's treat every depressed patient and hope for the best. So they're much more interested in subgroups and so I think they would be more interested, you know, in this sort of notion that, okay, if we think these sort of immune mediators as an example.

Might work. Can we identify those people? And you know, I think we should come back in a moment to, you know, what biomarkers are, um, ready for prime time, if you will, in psychiatry with those sort of biomarkers and see if it holds up so that we can sort of say, yes, our um, drug. It's preferentially going to res, you know, treat these patients.

As you know, one of the challenges in medication development is also, um, getting payers to pay for medication because there's so many generic medications. So if you could show with data, I. That using this biomarker, it really increased the probability of success that this population will respond. There's the possibility that the payers would be willing to say, okay, we'll give you a higher price because you're not going after all comers, you're going after this targeted subgroup.

So that's, I think, where more of the action is going to be because, um, a lot of what we are interested in, um, my group is in the earlier phase study. I think pharmaceutical companies would be very receptive to sort of putting a bunch of drugs from different companies into the same platform. You know, when it's a late stage phase three asset, et cetera, they may be a little bit less, um, willing to do that.

But in the early phase where this would hopefully be, um, generating proof of concept and if it works, you can go and do your own phase three study. You know, maybe it might be under the auspices of this initiative, but I think they'd be willing to do that because of the, um. Uh, markedly enhanced rigor, efficiency, speed that the platform trial would, you know, bring to the table.

types other drugs come in and they're specific, and, and the, the potential patients in the trial are having a, a, a set of biomarkers done, which would inclu their, their ability to, uh, be included with any of the drugs. They're then randomized it, uh, to these drugs or a, or a placebo, uh, within that, for those, they're eligible for. We could largely have similar visit schedules and things collected on these patients within this master protocol.

What, what kind of things seem to work or don't work with that?

Um, you know, sort of have this network so you can, you know, both have large numbers of patients, everyone, you know, following the same procedure for the patient evaluation, for the biomarker collection, et cetera. But I think what's also very important is that we also have this in the uk. Um, some of the sites we have are at what you might call the heavyweight sites, the Oxford, the Cambridge, the Kings, the UCL, et cetera.

Many of these sites are in areas where you have larger populations of individuals from different socioeconomic background, from different, um, racial, ethnic backgrounds, et cetera. So you are basically also trying to get the information, you know, if this thing works, does it actually work in general, or what population does it work in, et cetera. As part of that, in initiative, we have a major OMI initiative where we're collecting saliva and plasma in particular.

Um, one of the challenges we have, as I mentioned earlier, is that while there's strong reasons to believe heritability plays a big role, these are genes of small effects. So you don't have the driver mutation, so you need to think more also about transcriptomics and metabolomics, et cetera. And as you know, whenever you're talking about measures that are more sensitive to change compared to sequence variants, um, they're more, um, subject to noise.

So you really have to standardize things, et cetera. One of the things in what we are trying to do is part of this thing called the Mental Health Goals Program is really emphasize that the biomarkers need to work in the trial, but they need to work in the real world as well. And that's where we have one of the additional challenges in psychiatry is that we've done, I think, some phenomenal work with very sophisticated brain imaging, for example.

F, um, fm, FMRI with specific task or magnetoencephalography, et cetera. And you could do it in academic sites for the trials, but in the real world, someone in rural Alabama or in Nottingham, et cetera, it's not gonna go for those things. So they're, these biomarkers have to be related to things that could be routine, inexpensive, accessible. So we're putting a lot of effort in terms of, even when we're trying to look at the brain imaging signature.

Try and relate it to a peripheral signature, which could be plasma, for example, saliva or a digital signature, so that these things can be utilized, be utilized widely. The other thing we're doing that other companies are doing as well, I. Is for example, say you think someone might have a problem with hippocampal plasticity.

There are some cognitive tests and for example, Mike would be very familiar with them that you could do on a computerized battery and in your home, and that would tell you that you have this hippocampal in. Plasticity impairment or with EEG, you can send these caps to people's homes who have these EEG leads connected to them and you, you're using machine learning and artificial intelligence to gather the information from a single lead electrode.

You know, it's not gonna be as precise as a 64 lead, you know, um, EEG laboratory test, but it's good enough. So I think coming back to what you said, um, everything you said is spot on. I'll come back to placebo as well. So that's one of the other sort of challenges in psychiatry is that patients with things like depression or anxiety disorders.

We strongly believe that when you enroll someone in a study and you give them hope and you give them, you know, first time in their life this thing has been explained to them and they're being supported regularly, et cetera, I. There tends to be a somewhat high placebo response, but I think, I think placebo is driven by biology and in the field of pain that's been sort of born out.

So, um, people respond to placebo in pain trials as well, and some studies have demonstrated that if you pretreat someone with an opioid antagonist, you block the. Placebo response to pain suggesting that you're basically actually releasing your endorphins, which are giving you the placebo response. So I think it's not farfetched to think about the possibility that through this platform trial there will be placebo responders. Can we even identify the signature that predicts a placebo responder?

Because that would help companies a lot as well. Okay.

As more and more drugs come in, you have a huge resource to learn about the diseases themselves

Is this a government funded thing? Should this be a nonprofit, should be a profit funded thing, pharma? Where do we go?

Of a pharma, uh, r and d company, also not a academic group, but a very skilled group of, uh, integrators. I think that could work very well. But can I just add, uh, sort of the historical journey, uh, to this?

And so the question is how can we also, uh, encourage and convince, uh, sponsors, uh, within the pharma r and d world? And so, uh, you know, uh, how much. Innovation is encouraged, and I want to tell about my own experience. Um, Sini, when you first hired me, you said, you know what, uh, bite your tongue, be prepared and get ready. And what happened was that you supported the. The writing of a proposal that eventually became an IMI, uh, proposal, and IMI EPA came out of that.

IMI EU Pearl came out of that. Other things came out of it. None of that would've been possible without a champion who happened to be my boss within a farm r and d organization. And I'm absolutely convinced there are peril situations elsewhere. That's number one. Number two, our uh, boss, uh, Dr. Paul Stoffels used a phrase saying, you take the risk.

But maybe that near term achievement, starting a particular trial or having lots of patients included in a trial in a short period of time are actually is not really meaningful in helping us to get to the end game what it takes there. Are individual protagonists for innovation who bridge the gap between these pressures that exist? Time pressures, budgetary pressures, the difficulty of synchronizing different utilities within an organization and the.

Invitation from senior management to be innovators. one last point in my experience, and again, supported by Hui and others, all the interactions that I look back at with health authorities have been incredibly valuable learning experiences. So it is absolutely not the case that health authorities and regulatory. Experts are conservative by nature. Absolutely not. have such an insight broadly that can help, the right path.

So I would say, the patient-centric perspective will lead us to bridge the gap around some challenges exist. But the invitation is there to be innovators and the type of GBM Agile model that we have, I would think is a good setup.

You know, we want to embrace patients, learn from them, and, you know, sometimes they tell you that, you know, if you, you. You, you as a, you know, medication developer might think this is the biggest unmet need, but actually Prozac is okay for my depressed mood, it's my lack of motivation, my inability to go to work, et cetera. So to have the patient sort of, um, educate, for lack of a better word, industry, this is what I really need.

Um, I. The second thing is that help industries sort of, you know, talk to patients about, you know, there might be some things that we know is a big problem for you today. Um, and say for example, COVID, you know, induced a lot of what we call brain fog. But unless we, you know, right now we don't even know what brain fog is, how to measure it, what the target would be. So it could be something we tackle down the road, but unfortunately today's side doesn't allow it us to do it here.

Coming back to what Mike said and the patient. Um, advocacy, you know, in conditions like hiv, aids and many other conditions really drove things and we think that the same thing could be happening here. They can indicate that I'd be very happy to volunteer. What Mike said about regulators is completely my experience as well. So with Rob Kiff, with Janet Woodcock, with June Rain, who was the head of MHRA, it's been exactly the same.

They want to be, you know, sort of more agile, get things to help patients, et cetera. They need to obviously have a bar for the evidence, but they're very willing to work with you. So the last thing I'd say is, once again, I would agree with Mike in terms of how this could be done. So I think some sort of, you know, public-private partnership, so, you know, um.

You may know that a acne, the Alzheimer's Disease Neuroimaging Initiative would've been one of these great success, um, stories where, you know, we were able to collectively identify people who are on their way to dementia. And based on the rigor with which things were done, there's actually now a plasma marker. Um, plasma two 17 phospho, that is looking amazingly good, but you wouldn't have been able to come up with it unless you were able to have this large longitudinal study, et cetera.

So we've got this initiative in the UK and we're actually talking very closely to the French government, to the German government, and in fact, for many years while. I was at j and j, I was the co-chair of the foundation for the NIH Biomarkers Consortium, the Neuroscience Group.

So I think there's a distinct possibility that, you know, sort of, if we frame this correctly, I. We can get both public and private, and most importantly, patient stakeholders to come together, you know, from, you know, the biopharma companies. Not everyone will join, but if XX number of join, again, you know, Mike referred to our boss, Paul Stoffels.

One of the things he's really also championed is what's happened in the semiconductor industry where things have moved forward so well that if you're not. Part of this consortium, you actually get left behind because all the advances happen here. So I think in industry there may be some early movers. You know, my hope is that companies like j and j and Lily and BMS and Abby, et cetera. We'll join early, but then based on the success, others will want to join as well.

And I think we can, we can pull this off if we put the right people together and the right infrastructure. And the last thing I'll mention, and you don't wanna over promise, so I think that's sometimes, you know what sort of is bad people, you know, say, claim we're gonna cure world hunger with this. No, we're not. You know, we're in an early stage. We need to think systematically, but this is how we'll go about it.