Science Friday - A.I. And Doctors, Alzheimer’s. March 22, 2019, Part 2

Episode Date: March 22, 2019

When you go to the doctor’s office, it can sometimes seem like wait times are getting longer while face time with your doctor is getting shorter. In his book, Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again, cardiologist Eric Topol argues that artificial intelligence can make medicine more personal and empathetic. He says that algorithms can free up doctors to focus more time on their patients. Topol also talks about how A.I. is being used for drug discovery, reading scans, and how data from wearables can be integrated into human healthcare. Learn more and read an excerpt from Deep Medicine here. Plus: Alzheimer’s disease is known for inflicting devastating declines in memory and cognitive function. Researchers are on the hunt for treatments are taking a number of approaches to slowing or preventing the neurodegenerative disease, including immune therapy, lifestyle changes, and targeting sticky buildups of proteins called amyloid beta. But at MIT, scientists have been trying something else: a combination of flashing strobe lights and a clicking sound played at 40 times per second, for just an hour a day. Mice given this treatment for a week showed significant reductions in Alzheimer’s signature brain changes and had marked improvements in cognition, memory, and learning. But could an improvements in brains of mice translate to human subjects? Dr. Li-Huei Tsai, an author on the research, talks with Ira, and Wake Forest Medical School neuroscientist Dr. Shannon Macauley, who was not involved in the research, discusses how to take promising research of all kinds to the next level. Subscribe to this podcast. Plus, to stay updated on all things science, sign up for Science Friday's newsletters.

This is Science Friday. I am Ira Flato. When you go to the doctor's office, it seems like the face-to-face time is getting shorter and shorter, isn't it? Sometimes you're in and out in under 10 minutes. The experience can feel very impersonal. My next guest says that one of the keys to bringing back the human touch to medicine is AI, artificial intelligence. It sounds counterproductive, a counterintuitive, does it, that an algorithm can increase empathy. But the less time that your doctors need to deal with charts or sorting through conflicting diagnoses, the more time they have to spend with you. Plus now, you can track your own heart rate with your smartphone. Computers are reading medical scans and detecting cancer. Will this new data make medicine more

personalized, or will it be information overload? My next guest says AI can give you more time with your doctor, if it's done correctly. Dr. Eric Topol is here to talk about all of of this, he's a cardiologist and author of the book, Deep Medicine, How Artificial Intelligence Can Make Healthcare Human Again. Welcome back to Science Friday. Thanks, Sarah. Great to be with you. You say that patients exist in the world of what's called shallow medicine, insufficient data, time context, and presence. What do you mean by that? Well, shallow refers to that lack of human bond, the very limited time to see a patient, the limited time to formulate a diagnosis, review the data, have the context, and most of all, the deterioration of the relationship

between patients and doctors that's really suffered over time because of the big business of medicine. So what are the consequences of practicing shallow medicine? Well, besides the fact that doctors and all clinicians become data clerks and are tethered to keyboards, the patients suffer because the misdiagnosis rate is alarmingly high. Over 12 million serious misdiagnoses a year. And as you know, the errors that occur in medicine are one of the leading causes of even death, no less other complications.

So we have lots of mistakes and we have this broken bond between patients and their physicians. I want to get into that, but let me give out our phone number first. 844724-8255 if you'd like to talk with Eric about his book, Deep Medicine, 844724-8255, or you can tweet us at Cy Fry. Let's get into that because you say that is one of the main themes of your book, is the bond, the breaking of the bond between doctor and patient. Right. That is because, for example, the lack of even eye-to-eye contact in those limited minutes, so that is a real hit.

You can't, both on the patient side and on the doctor side, beyond that, if we know from how an expert diagnostician makes a diagnosis, if they don't have it within five minutes, the chance of it being accurate is 28%. So that's about the time that you actually have with a patient. So we can do better than this. We can transcend this problem of burnout and depression among clinicians who feel that they aren't able to do what they went into medicine for in the first place. You know, we think that just the opposite, that if you bring computers in, it's going to make the time less that you have with your doctors because the computers will be doing the work. But you're saying just the opposite, because the computers may do some of the original scanning, for example, that you as a physician may have more.

more time to spend talking to the patients? Right. Well, there's so much data that no human being could have their arms around it for each person. We're talking about terabytes of data between the records and the scans and sensors and genomics, all these things together. So that's really a critical aspect. But the other thing is about speech recognition is so advanced now. Yeah.

There are over 20 companies that are already starting to get in the clinic to use the, the voice to synthesize the note and also whatever it needs to be done after the visit, to basically liberate from keyboards, which are mutually hated by both patients and doctors. The three principles, three components of deep medicine you write are deep phenotyping, deep learning, deep empathy, and connection. Tell us about those three things. Well, the deep phenotyping refers to gathering all that data, that's appropriate for each individual.

And so today we don't do a very good job of that because the data is in so many different places. Each person goes to lots of different doctors and health systems. And we'd like to have that from the time a person is in the womb until the present moment. But that's deep phenotyping, and that would include all the medical literature about a person's condition.

And then basically deep learning is what's so exciting today. This is the most radical jumps we've seen in the history of AI, which has been going on for decades, this deep learning, which is this neural network that can process data with remarkable accuracy for speech, for images, and for text. And so if we use that appropriately, we can outsource to get to this deep empathy state, which is to restore medicine the way it used to be decades ago when there was this precious relationship with. the presence, with the trust, and with a really tight bond. You write, and it was so surprising to read this in the book, that you write about the failure of electronic health records to actually live up or actually be very, very helpful. You say the use of electronic health care records leads to other problems. The information they contain is often remarkably incomplete and inaccurate.

Electronic records are very clunky to use, and most, on average, of age, 80% of each note is simply copied and pasted from a previous note, so mistakes go along with it. Right, Ira, it's amazing. You know, that's been documented in the literature, in recent literature, that these notes are error-laden, and the errors just get propagated from one note to the next. The software is just beyond the clunky description. I mean, I recently had to get retrained in Epic, 25 hours. of training to use the software.

I mean, this is amazing. And it's just so many steps to do such simple things. This would never work in the real world of technology, but it's the way we have it. Companies like Epic, CERner, and many others, that's the way medicine is practiced and burdened in this country. It's been an abject failure without question. I want to get into that a little bit more, but Matt in South Bend, Indiana has a question related to that.

Hi, Matt. Hi. Go ahead. Hey, well, thanks for taking my call. Yeah, I'm just calling about, you know, the question is regarding the data itself. Where does the data stored, you know, and is this something that's automatically happening when I go to my doctor? Right. That's really important, Matt.

I think where we want to be is that each person owns their data. That ought to be a civil right, because no one has all their data, and it's coming from multiple sources. It used to be, you know, just with the doctor's office in the hospital, and it was hard to get to. But now it's with your sensors, increasingly so. It's with your genetic studies or your microbiome and so many other different places, even environmental sensors. So that all that data belongs with you. You have the most vested interest. It's your body.

You've paid for it. And we've got to get there. And by the way, you can't have AI without great inputs. Deep learning requires all your interest. inputs to get that output to help you, whether it's prevent an illness or to manage something that you're working with, you know, a condition that you have. Wasn't that the original computer mantra when I was young? Right.

Garbage in, garbage out. Guy go? You got it. Yeah. And if we don't have a complete data set, you know, that's what I opened the book with, you know, being roughed up with my knee replacement. And the doctor looking for, the surgeon, didn't have all my. data, and that led to some, you know, serious adverse outcome for me.

Let's talk about that in particular, about your question about deep phenotyping. Are we going to be entering a day where you would bring into your doctor's office, if it's not going through their health records, a thumb drive, that they will have your whole genetic score on there and can look to see what's the best personalized medicine to give you? Well, you could do that or you could, of course, transfer it in advance of a visit, but the point being is this algorithmic processing of that data, which always will require human oversight. But that will be distilled. We just don't have time to go through. It's voluminous.

And so you want to have this distilled as much as possible. And what would then, how would the doctor then be a co-partner with deep media and with the data? Give us an idea of how they would work together and what kinds of different tasks each would do. Well, there's lots of different scenarios. You know, right now we already are seeing radiologists having the scans read first by a deep learning algorithm, which assures that it won't miss things like a knowledgeable. on a chest x-ray or a fracture in a wrist or things like that. So that's one scenario where it's a pre-read,

but then you have the context of the radiologists and experience to provide that oversight. There's many different ways that this can be folded in. You may have your data being monitored. Let's say you have a condition like diabetes and you have various things that your data are coming in like your dants just your glucose, but your sleep, your physical activity, your stress level, what you're eating, and it's basically coaching you as to have better glucose regulation.

These are what we call multimodal data inputs, but that's the best way, as opposed to today, where we have these dumb algorithms that just tell you whether your glucose is going up or down. So we can do so much better than that in this deep learning algorithmic era. And when the computer makes that first pass, then the doctor has more time to talk to you about what the meaning of that result is. Right. And I think what's so astounding, really, Ira, is the fact that we can train machines now to see things that humans will never see. And that's really quite extraordinary. And it's almost as if things you never have thought would be possible are now attainable.

So whether it's showing a retina picture to top retina specialists, and when they look at it, is this from a man or woman, they have a 50% chance of getting that right. But the machine algorithm is over 97% accurate. And so many things like that. So for the gastroenterologists, polyps are frequently missed, but now they can be machine vision, can find them all. My guest is Eric Topol, author of Deep Medicine,

how artificial intelligence can make healthcare human again. We're going to take a break when we come back and talk more with Eric. Our number 844-724-8255. Stay with us. We'll be right back after this break. This is Science Friday. I'm Ira Flato talking with Dr. Eric Topal, author of Deep Medicine, our number 844-724-8255. Let's go to the phones.

Yeah, let's go to SEShi in San Antonio. Hi, Sashi. Hi. Hi there. Go ahead. So I am a third-year medical student, Dr. Roy Cople, and this seems to pertain to the near future of my education and career. My question is, well, initially, electronic health records were thought to be, you know, the save all. We're going to get all the patients' information.

It's going to help us, and then we get dozens and dozens of companies creating it, and now we'll have a whole other beast to deal with. How do you foresee the development and implementation of AI? Do you feel that this should be a public endeavor where it can integrate all these different electronic health records or inputs of information? Or do you see the private industry again taking over and possibly creating another beast of various AIs that various hospitals use? Right. Well, that is a great one, SESHE. I wish I could go back to third-year med school because medicine in the future is going to be so much better in this regard.

But what we had, the debacle that occurred with the electronic records was just unacceptable software. But now we have the tech titans and so many really innovative startups that are in this space and the kind of functionality, user interface, whether that be for doctors, and now for patients, the problem really gets down to that these EHRs were made for billing. They had no business, I mean, all business, but no patient care, no, there was nothing the patient centered about them or doctor centered for that matter. So this was the fiasco, I think, explained. Now we're changing that, and we've already seen in other countries that they've been able to adjust for the software to be patient-centered. So I do think it's achievable. This is basically,

you know, software algorithms dealing with data. And I think one thing to keep in mind, Sashi, is that we as people have early satiety with data, but when you get working with the right algorithms, it has insatiable hunger. Can't get enough data. And it can do things that we'll never be able to do. There is a trend in the Internet of things. There's so many people. personal sensors now. I know that you're very familiar with these on our phones and watches. In fact, my brother was motivated to go to the hospital when his Apple Watch showed heart fibrillations. Aren't these sensors good ideas? I mean, I know that you've worked with sensors, and actually I remember watching you tweet how you diagnosed one of your own kidney

stones with your own sensor. Right. So what? What? What is the path we should take? Should we depend on our lives on these sensors and feed data in, or are we going to be too dependable on them? And what's the best way to integrate them? Right, Ira. I think the problem with sensors is the appropriate use. So if you have risk for atrial fibrillation or symptoms and you're in a group of people that

would be high suspicion, that's one thing. But it might not be something, an atrial fibrillation detection watch for everybody. body. When I diagnosed on my smartphone that I had a dilated kidney, you know, when I showed up to the emergency room and the emergency room doctor thought I was an alien when I showed him the picture and still sent me for a cat scan. So that kind of shows you that's emblematic of not fully trusting the sensors and the things that we are working with today. But I think over time, we'll figure out really who are the right people, the right circumstances to apply these things. We don't want it done in a willy-nilly way because then you just wind up with more incidental findings, more trouble.

We have to be really particularized the way we apply things. I want to talk about what I mentioned at the top of the hour. I was talking about radiologists who couldn't see a man in a gorilla suit. on a scan. That really happened. Right. Tell us about it. Now it's quite an experiment where it shows that humans, and in this case was radiologists, their attention, their ability to see things can be impaired. And they missed the man in a gorilla suit, you know, 80 some percent of the time.

Now, why is that important? Well, we get tired. doctors and all nurses, all clinicians. We have bad days. We have moods. We need time off. And, of course, machine algorithms can take on things all the time. They can get sick, too, of course.

But for the most part, they're not distractible. And they can get trained. I think one of the things to note is that they have exceeded already so quickly what we had expected to see in the health care scene. And it's just going to get more impressive over time. A lot of this still needs validation, replication, and surveillance. But I think this is the point that's quite noteworthy is, you know, people can only do so much, and we need the complementarity. It'll augment human performance.

And then, as I mentioned earlier, outsourcing so we can have that human-to-human bond. You did say before we were getting to the break that AI is able to. to see granularity in the data and that we can't, that people can't see, picking out stuff that's very, very hard to see and would be very significant. Yeah, no, the data flood from high-resolution images, from continuous sensor, wearable sensor output, from the electronic records, from all these other sources, it's overwhelming. And in general, you know, the whole world, we're already exceeding Yada bytes, we're moving into, we need hell of a bite.

You know, it's really, so we need help. And this is, this is a rescue for that inability to cope with this overwhelming flood of data for each person. I mean, each person is in the high numbers of terabytes already today, and that's just going to increase. Well, here's a relevant question to that, Mike, in Sugarland, Texas. Welcome to Science Friday. Thank you.

Go ahead. So the question is, I am in AI. I'm in ML field, I'm in IT. I'm the liver of deep learning, so I get that part. The question is we have so many data coming in from Fitbit, Apple Watch, each specialist, each hospital system has their own data system and repository. So how are we going to get all this data into one analytical repository so we could do this deep learning? Yeah, great question, Mike.

It's been done in Estonia, of all places. They have all the data that sits on every citizen there owns their data on a blockchain format, and it's continually updated. If they can do it, I think we can do it. But you're absolutely right. This is the problem we have right now, things are so fragmented. And in order to, as you know, to get to work through the neural net, you've got to have those inputs, and we are not well positioned for that to help each person. It's really a vital step that's necessary. Well, do you think that AI and these systems will be adopted faster in countries that have universal health care where things may be more centralized?

Absolutely. You know, I just finished a year and a half commissioned by the U.K. government to work with the team to review the NHS. And I saw they already are taking off with AI. They're already in emergency departments using voice to synthesize notes. and not using any keyboards. And so they're planning ahead for the AI workforce, which is going to have very substantial impact. So universal health care does help this.

In China, where they have all the data for each person, and of course that brings up the issues about privacy. But they are moving much faster because they have all in one place, and they are way ahead in implementing AI. Justin and San Antonio, hi. Hi, you're next. Welcome to Science Friday. Hi, yes. Actually, right off of that comment, are you concerned that we don't have the civil rights in place

to deal with this type of AI technology coming into mass use? And also are you concerned about the society as a whole relying more on diagnostic medicine because of these extreme improvements in the diagnosis? Thank you. Okay, as opposed to preventive medicine, I guess. you know well i mean i think i am worried and you know wrote quite a bit in deep medicine about the 27 reasons why everyone has to own their data uh we just talked about how we can't really do

i well without that so this is something that you will we have to support it's going to require activism but eventually it's it's because a lot of the data today for each person is homeless you don't have your sensor data sitting in your electronic record and you don't have your don't want your genome sequence or other genomic data in your electronic record. So we don't even have a place for it all, and we need that. So, you know, eventually we'll get there. But in order to get to that dream of prevention, we will know a risk of many common conditions early in one's life, but in order to actually prevent it, you have to have that data continuously brought in so that the neural net can work with it. And so step one is having all your data.

And at least in this country, very few people, if any, have that. I have a tweet from Kelly who says, will AI really lead to a doctor spending more time with patients or will they just schedule more patients per hour? Yeah. Well, you know, Kelly and Ira, that's my fundamental concern. A lot of people, as I am, are worried about privacy and worsening inequities and security data and, you know, bias.

But the biggest thing for me is if we don't stand, stand up for patients. This is the time because there's going to be this big revving up of productivity, efficiency, workflow. And if we don't say that's got to be the gift of time to spend with patients, which is so vital and has been lost, if we don't do that, we're going to lose perhaps the biggest opportunity that we're going to see for a long, long time. Yeah, you say that is one of the main messages of your book, the takeaway you hope people have. Let's talk about nutrition because you mentioned nutrition as one area where the guidelines keep changing. How can AI be used in nutrition?

That's fascinating. Ira, because the chapter on Deep Diet, I get into the point that we didn't know how to individualize a diet until we had machine learning. And the group in Israel at the Wiseman Institute led by Aaron Siegel, what they did is they now have studied thousands of people. they got all their data that we've been talking about, plus their gut microbiome, glucose sensors, everything they ate, and what they were able to show is you could predict from all that data what would be good for you to avoid glucose spikes after we eat. And that's something that although we don't know that getting rid of these spikes when you eat will prevent diabetes, but it certainly is suggestive. And of course, now we're learning about other things that are very heterogeneous. genius. So if you and I ate the exact same thing, the exact same amount and time, we would have

very different glucose in response and triglycerides and other labs. And so the question is, can we individualize a diet and we're chipping away at it? But if it wasn't for AI, we wouldn't been able to bring all this data together to fashion, to have a bespoke diet if one wants to follow it. Talking with Eric Topol, author of Deep Medicine on Science Friday from WNYC Studios. You know,

if you can have a book about AI be a page turner, Eric, you certainly have done it. Let me go to the phones. To John in Denver, interesting question, John. Welcome to Science Friday. Hi, thank you, Ira. Yeah, I read an opinion

piece in the New York Times back in January by Dr. Kool- LAR, assistant professor of health policy. Anyway, the gist of his opinion piece was that if we incorporate into AI, some of the biases we already have, racial biases or sexual biases, we know for a long time the medical profession has struggled getting accurate research during, say, women for heart attacks or other minorities, not include in research studies. Is there a possibility that AI could kind of incorporate those biases without us even realizing it? Well, you're absolutely right, John.

But it isn't the AI that's doing it. It's us humans. So all that bias is part of the inputs. And so that's the problem. If we put – and it's already – you know, I go through many examples in the book of where that bias has shown as inputs. And of course, you can expect that the bias is coming out as well through the neural net. So I think this is something where it's interesting. All these problems of AI, now they're starting to use AI to deconstruct to prevent the bias from being inputted.

And so that's going to be interesting to see if we can get our arms around it. But this is a serious problem. Let me, in a couple of minutes we have left, we have a lot of people who've tweeted this. They want to know from you, Eric, what is the roadmap? How do we get all of this done, like Estonia did, for example? Right. Well, there's lots of different things that we need to do. But the biggest thing for sure is that as we embrace this potential rescue for so many clinicians

and for patients together, we have patients taking on more responsibility in charge, with their data that they're generating and doctors that in outsourcing some of the things that they don't do well or don't want to do, like being a data clerk. We can see this flywheel effect. So the biggest thing is we've got to stand up and use this properly to get back the care and health care. And we can do this. Do doctors have to stand up themselves?

Like the Parkland High School students stood up. As you say in your book, the doctors have to take the lead in this. Yeah, we just lie down when the EHRs came around and all these other things. But as you've seen recently, when doctors stood up for stay in my lane, for the guns and the NRA, we can do this. And I think it's going to be vital if we're going to get this moving in the right direction. Because it could make things worse. We've got to acknowledge that when you have all this benefit that we don't actualize for patients. How do you make this a campaign issue in the upcoming?

Well, you know, we don't have in this country universal health care, which we need to have. And we also need to get honored the fact that each person should own their data and they shouldn't have to struggle so hard to just get little pieces of it, which is just absurd. So it's not just about the universal health care that our country is an outlier with the worst outcomes of the 37 richest countries and the only one that has such gross inequities. but it's all these potential benefits that can be derived from providing care for each person, but not just health care, also the care. The book is Deep Medicine, How Artificial Intelligence Can Make Healthcare Human Again, Dr. Eric Topol, a cardiologist, and you can read an excerpt from his book on our website at Science Friday.com slash deep medicine. Eric, fantastic book.

I mean, there's so much information in there. It's a great read for everybody. for the health care profession. We're going to take a break, and then a new innovative potential treatment for Alzheimer's that's being tested out in lab animals, fast, pulse, light, and sound. It worked in mice. Can it work in people? Non-invasive, got some results.

We'll talk about it after the break. Stay with us. This is Science Friday. I'm Ira Flato. Chances are we all know someone with dementia, like Alzheimer's. disease, and we are all worried, at least in the back of our minds, about our chances of falling victim to it, because we know there is still no cure for it, and we eagerly listen for any hopeful news about advances in research, because so far Alzheimer's treatments have not really

worked. Well, we now have some of that hopeful news from a team from MIT that has been thinking differently. Here you go. Imagine a therapy where you sat down with just a flashing light and a click sound, all pulsating at 40 times a second for just an hour a day. If you're a mouse engineered to model Alzheimer's, it turns out this very alternative therapy has some striking results. Cognitive function improves. Memory and learning improves. These pesky, sticky amyloid plaques clear up, and your brain produces a very specific pattern of neuron firing called gamma waves. As Alzheimer's research know too well, translating research, Research results from mice to human brains can be a tall order.

Tall order. So what's next for this light and sound idea? And where does it fit with other things? Researchers are trying in the quest to reduce the devastating effects of Alzheimer's disease. Here to explain her research is Liwe Sai, a cognitive neuroscientist and director of MIT's PICOWR Institute for Learning and Memory. She's one of the authors of this study published in the journal's cell. Welcome, Dr. Sai.

Thank you. And also with us is Dr. Shannon McCauley, assistant professor of gerontology and geriatric medicine and a member of the Alzheimer's Disease Research Center at Wake Forest Medical School in Winston, Salem, North Carolina. Welcome, Dr. McCauley. Thank you. Let's talk, Dr. Sy, first, about this study you did. What made you think sound and light could have any effect on Alzheimer's symptoms? Yeah, this is a very good question.

So when we look at Alzheimer's disease, we kind of look at it from a different angle. So we all know that there are plaques and tangles and cell loss, but a few people look into the brain waves. And we found that this brain waves, especially at the 40 hertz frequency, we call it gamma waves. They are impaired in the very early stage of the disease. So we ask, what will happen if we increase the 40 hertz gamma waves in the brain of Alzheimer's disease mouse models? So you sat the mice down with the 40 hertz waves of sound and light at the same time, just flashing lights in front of their eyes and sounds through their ears? Exactly. So really much to our surprise, we found that if we just flash 40 hertz light and presented 40 hertz sound to the mice, then their brains started to produce this gamma waves very robustly.

So we realized that we can easily use this very non-invasive approach. to induce gamma waves in the brain. Just to let our listeners hear what those clicks sounded like, just the warning your radio was not broken when you hear this. Wow, those clicks plus the flashing light, they were able to clear out the brain of amyloid and tau and all these physical indicators of Alzheimer's in the brain? You know, it sounds crazy, I know.

But, you know, we just, you know, just obtain this, striking results every time we do it. And how long does it last for? Yeah, this is a very good question. So far, we believe that we do have to do it continuously. We presented the light and sound to mice one hour a day, every day, and we started to see this effects after about a week. And if we do it longer, we see even a better effect.

And it affected their learning ability and their memory and help them in those ways? Indeed, indeed, yeah. We, you know, we not only saw reduction of plugs and tangles, and we also see that their memory becomes better and they learn better. It sounds almost too good to be true. I understand. I understand. Dr. McCauley, what does the rest of the Alzheimer's research community think about this work? Is it too out there?

You know, I think this is a really highly innovative approach and it's creative. And the findings, you know, I agree are fairly provocative, right? You know, to just think that these series of flashes and auditory clicks could clear plaques and tangles and restore cognitive function is exciting. but a little bit out there. But if you start to think about the pathogenesis of Alzheimer's disease and the fact that the Alzheimer's disease brain has a problem with hyper-excitability. We know that neurons aren't functioning properly. We know networks are impaired.

We know there's a great deal of synaptic dysfunction. And so if this approach can almost provide the brain with a reset button to normalize brain function, you know, it could be a possible treatment for Alzheimer's disease. So, you know, the findings are absolutely provocative. I agree, but they're very convincing. And the fact that there was this stimulation of microglial clearance. So these are the, you know, your garbage men of your brain that can clear away debris. And that Dr. Seid's work shows that there's this, you know, this increase and this ability to clear bad proteins from the brain really does suggest it has traction, and I'm very excited to see it translated up to humans.

Now, let's talk about that, Dr. Sy, when do we have human trials of this? Because you're not really doing anything invasive, you know, when do we start a clinical trial? Yeah, it's a very good question. I'm very excited about this approach, precisely because, because this is so non-invasive. And in mice, we know that even after we presented the light and sound, every day for months and months, these mice look completely healthy. They get better and better with their memory and learning.

And, you know, they seem very healthy. There's no deleterious effects, as far as we can tell. So I think that with this kind of... kind of very safe and very non-invasive approach, I think, you know, it will be relatively straightforward to translate to humans. But really, you know, the $100 million question is whether a human would respond similarly to how the minds responded to this treatment. Yeah. And, you know, but you had two very simple things. You have some flashing lights, and you have a 40-hertz sound.

What's to prevent any physician from just setting this up in her own, you know, office to try it out? So that's why here at MIT, we actually started some very early, you know, feasibility and compliance kind of study in human subjects. primarily we just really want to figure out a good condition and a good device to induce gamma waves in human subjects so that, you know, we can, you know, we can make sure that when people do it, you know, we have the best condition for them. Dr. McCauley, there was some other big Alzheimer's news this week, and that was biogen drug. Andyucanamab has officially been pulled out of phase three clinical trials. There were, as you know, lots of hopes writing on that drug, right? Yeah, absolutely.

So I think we just found that out Thursday that, you know, that adiucanamab, they decided to stop their phase three clinical trial. And, you know, that was a pretty big blow to the field, you know. And again, as we try to work very hard to get something to patients and caregivers, you know, it's a major setback for us. Reports, you know, even within the last year have suggested that, you know, this, this immunotherapy that targets amyloid, one of the, you know, earliest changes in the Alzheimer's disease cascade or thought to be, it shows that it can't bind the toxic forms of amyloid.

It can remove amyloid from the brain and even showed some promise on cognitive improvement, which is, you know, kind of our holy grail for trying to get. you know, a therapy to market. You know, so the fact that they decided to stop this trial and and the good news was it's not because of a safety issue. But unfortunately, the bad news is it was because they didn't think they could reach their primary endpoints for, you know, slowing cognitive decline. So yeah, that's a pretty big blow to us at this point. Is there any thought that we might be aiming at the wrong targets here? Absolutely. You know, I think So I think one thing that we have to keep in mind with all of these amyloid-based therapies is, you know,

we've learned so much in the recent decades about how Alzheimer's disease progresses in humans. And the fact that amyloid does seem to be one of the first things that changes, and you do need to see amyloid entangles at autopsy to give a diagnosis of Alzheimer's disease. So I think originally the idea that if we could stop the amyloid from accumulating or remove it, after it's accumulated from the brain as a way to stop Alzheimer's was sound. But I think what we're finding now is after all of these trials where we've made some great, I think, improvements in clinical trial designs and the antibodies and what type of amyloid beta we're targeting, you know, we still aren't seeing that holy grail effect on cognition.

And we need to change our approach to look at different things that go wrong in AD. We know when individuals start to go through cognitive impairment and cognitive changes, that it's tau that's rising, you know, the protein that's found in neurofibrillary tangles. We see this huge neuroinflammatory response, changes in metabolism. And so there's a whole host of other targets that I think we need to, you know, go full force on in order to treat this disease. Dr. Cy, the fact that your study with the light and the sound was effective in removing both the amyloid and the tau? Do you think that was what might have made a difference? You know, I think our approach still yet to be tested in humans. But I, you know, I agree with Dr.

McCauley. I think at this point, we still simply don't know enough about the disease. And the disease clearly is very complex. And probably people get Alzheimer's disease from, you know, different ideology. And so, you know, I think that in addition to this individual molecules and genes, we really need to also look at the brain as a whole system. The brain functions like a computer. So, you know, we need to look at the circuits. We need to look at the network and figure out what's wrong with the network and how can we change the whole network to change the state of the brain. Interesting. This is Science Friday from WNYC Studios. We're talking about Alzheimer's disease here. Dr. McCauley, we talk about in this business about changes happening so many

years before we actually see symptoms. Do you hold out an idea that there might be early detection getting even easier? And if we do detect it earlier, does it make it? doesn't make treatment any easier or more effective? Absolutely. I think you hit the nail on the head. You know, in the last, again, over the last decade, we've seen a tremendous jump in the technologies we can use to actually stage this disease.

We have new neuroimaging biomarkers that can see plaques in the brain and can see tau accumulating in the brain. We have CSF biomarkers that, again, can look not only at tau and a beta levels, but can also get at changes in neuronal integrity. And now, even within the last couple years, we're developing serum-based, so blood-based biomarkers of disease. And so what this is allowing us to do is really get a composite of what's happening.

And as Dr. Sides said, move beyond just amyloid and tau, right? So we can look at neuronal integrity. You know, when do these networks start to, you know, their network at terror. activity become aberrant or neuronal loss starts to happen. And that might guide when we intervene and what we intervene with. So I think the big thing right now is really understanding how the human clinical course of disease happens and understand the heterogeneity within that data set as well. We had Dr. Eric Topol on just now talking about what his ideas about deep medicine and using

artificial intelligence. Do you think there is a use for artificial intelligence? going through all the data of patients? Absolutely. I mean, I think that, you know, we're creating from genetics to cellular function to, you know, I mean, to the point we're getting, you know, single cell RNA data, you know, it's such massive data sets out there that there needs to become a way that we can look at a signature for Alzheimer's disease and one that incorporates function.

You know, again, what Dr. Sai says is that we really need to make sure that we're bringing this back to preserving cognition, network connectivity, and also an interaction between the brain, the vascular, metabolism, and all of this. So, yeah, I do think that we need to start to think about this in terms of big data. Dr. Sae, Dr. Sae, where do you go? I'm sorry, where do you go from here with your work with the mice now? Yeah, so wisdomize, you know, we so far demonstrated that both light and sound can induce gamma waves, but, you know, we have, you know, more senses capacity. So we would like to see whether by other sensory stimulation, such as smell or, you know, tactile stimulation, that can also induce gamma waves. And also, we would like to further, you know, understand how this works.

So as Dr. McCauley said, we show that the microglia, you know, the brain's immune cells respond robustly to this gamma waves, and they become much more active. You know, they restore their normal function to clear the amyloid. and presumably other toxic waste. And we also show in our cell paper that the blood vessels in the brain also respond. I'm going to have to leave it there and have you back and talk about the rest of it because it's very exciting, Dr. Sai. I want to thank Dr. Llewe Sine, Dr. Shahnem Nicali for talking about their work on Alzheimer's Disease. That's all the time we have again.

run out of time. BJ Liedemann composed our theme music, and if, you know, we're on social media all week long, you can also do our podcasts, and go on to our website for great stuff. As always, I'm Ira Plato in New York.