Hello everyone, it's a pleasure to be here today and to talk to you about a topic that I think is incredibly important and that's imaging of neurologic decline following traumatic brain injury. We have an incredible panel of speakers today. The first is the internationally known and credible speaker important topic of brain injury, the chair of rehab at Harvard Medical School, and the president of Spalding Rehab Hospital. Our next speaker is Dr. Ross Zavant, and he'll be talking to us at first about some of the clinical aspects associated with, with imaging neurologic decline following brain injury. So with that, Dr. Zavant, take it away. Thank you for having me. So the first topic for us for the next few minutes is to talk about the clinical presentation of neuro decline. We're hearing a lot about the neuroimaging and what other features we are looking for in this degenerative pathology. But now let's talk about the linkage to the clinical presentation. So this is a slide based on Kathy Satmon's work, which shows that every brain injury is unique and it's very difficult for us in the old way to categorize things as mild, moderate, and severe. All of these neuroimages are that of a different severely injured person, and you can see the heterogeneous nature of this. Now in this slide that I've shown before is some work by Raquel Gardner, who's at UCSF, and Raquel worked with us to do latent class modeling in a review of the COBRIT study that we were a part of. And what we found is that in a scenario of people where there should be one path, there really, really was eight separate paths, suggesting there are many factors that go into recovery and then potential decline or different trajectories of recovery. Now when we think about outcome, we really should look and say there are genetic variations, but there are also pre-injury variations. Those things have to do with things like allostatic load, and as McEwen said, many of those events early in life can lead to a six-fold change in mortality or number of years lived. There are host responses, which are inflammatory or responsiveness of mitochondrial DNA. There's neuroplasticity. There are neurodegenerative mechanisms. There's neurocognition and neural reserve, right? How much education did you have? How much social interaction? And there are host factors related to simple issues of the type of brain injury, which are obviously, as we showed, not all the same. So in this slide by Ginger Polish, what you can see is that TES and neurodegenerative disease exist on an X and Y axis with greater exposure and more symptoms, being more likely to have traumatic encephalopathy syndrome and neurodegenerative-like phenomena. But we also have two other factors to consider. Much of what we know about this entity is really based on a little bit of bias sample from donated specimens, and 73.4% of those cases are from football. We also have to realize that the more we tell people that if you've done something, you're more likely to have an event, the more they will have that experiential of that symptom or event. And that's a nocebo effect. They're very powerful in pain, and they're very powerful in other behavioral symptoms. So we have to titrate pre-existing risk, exposure, symptomatology, and what are we telling people and how are we diagnosing this? And this is why this has been so complex for years. If we look in general at brain injury outcome, and here I'm going to refer to those with mild, moderate, and or severe injury, what we do know is that people get better and then they decline. And they decline over time. And there's prior data from Scotland that suggests that even if you go to the ED, this is Tom McMillan's data, with a mild head injury, your lifespan is altered. We know from Cindy Harrison-Felix's work that those model systems patients have a nine-year lower lifespan. So there is something going on. That something may be complex. It may be related to neurodegenerative disease, yes, but it also may be related to other comorbid or other phenomena that the brain injury suggests vulnerability for, or biopsychosocial risks for brain injury suggest vulnerability for. So what is the link to the neurodegenerative disease? And Dan and others have discussed the specifics of the pathology, but I'll just review here that this is not brand new. Harrison-Martland first described so-called punch-drunk syndrome in JAMA in 1928. Milspod talked about dementia pugilistica in 1937. Original characterizations were usually in boxers, many of whom also had a large cavern septum callus in them, thinning of the corvus callosum, and ventricular enlargement. And we know that there is a sport that traditionally and historically had been associated with this, and this is boxing, right? Boxing's primary job in part is to cause a head injury. It is to defeat the opponent, and therefore it is among the things that may have very specific risks. We know from some work from Barry Jordan that career lengths, the number of bouts, and poor showing, you're not doing very well in the ring, you fight a lot and you're not so good, as well as APOE genotype, having that at-risk genotype of 3-4 or 4-4 suggests risk for neurodecline. But boxing has long, as we'll talk about in a second, been associated with a motor phenotype, a bradykinetic phenotype, in those who've participated and show some element of decline. So how common is the pathology? Well, this is an older study from Lehman et al. in neurology in 2012, which they did is they looked through the National Death Index of a cohort of NFL former players, and they chose the year 1960 forward to 2007. The reason they chose 1960 was the inculcation of the hard shell helmet in absolution. When you look on the table two, what you see is all degenerative disease and cardiovascular disease causes. And here they just compared American-style football players or NFL players to controls, the general public. And they saw a standardized mortality ratio in cardiovascular disease of 0.71, but neurodegenerative disease, let's look at this, 3.62, dementia 3.86, Parkinson's 1.69, and ALS 4.31. Substantial. And Jesse Mez, who works with Dan closely, used a convenience sample of 202 American-style football players. 87% had CTE using pathologic criteria. Now, these are donated brains. Three of 14 high school players and 110 of 111 NFL players, essentially 99%. Stunning. But it depends in life when things happen. This is some work from Gu Nguyen, myself, and others from the Football Players Health Study, in which we decided to say football versus baseball instead of the community. And the community is a problem because there's a healthy worker bias. There are people who could never get on the field. There are people with innate illnesses or impairments that suggest they're very different than athletes. So now let's look at another group of athletes that don't typically hit their head as much and aren't as heavy. Don't do the same lifestyle things. I turn you to panel B in this slide. Panel B shows early on that there is a significant cardiovascular mortality difference. Remember, this is a National Death Index study as well. But later on in life, there is a significant neurodegenerative disease separation or survival curve, suggesting that neurodegenerative diseases, one might expect, occurs generally later in life and that that separation gets larger as life goes on in time. Early life, many of these things are cardiovascular, and one might poise it that many of these pathologies interact at some point with neurodegenerative pathologies. Now, let's take it outside of American-style football and look quickly at professional soccer. This is work by Danny McKay and Willie Stewart's group in the north of Scotland. And what you see here in figure one panel is a hazards ratio. The x-axis is years of life by decade. The y-axis is hazards ratio. And here, this is risk for neurodegenerative disease versus age-match controls in the north of Scotland. And look, at age 70, you start to get a change, and by 80 and, of course, 90, there's a higher neurodegenerative disease risk among soccer players. That said, if you make it to 90 in your late 80s, that's not necessarily a terrible outcome. When they looked at match controls, and remember, they can do this because they have the NHS, they have a common record from which to follow people. I want people to look at these hazard ratios. Alzheimer's disease, 5. Dementia, 3.87. Motor neuron disease, which we have to assume is mostly ALS, 4.33. Parkinson's disease, 2.15. These are all of some interest. But we have to remember one thing. Age makes pathology dirty. It suggests PART or RTAG or other comorbid pathologies, as well as microvascular disease, that plays into differentiating both pathologic diagnosis and, I would say, phenotypic diagnosis. When you get someone who is in their 90s or in their late 80s, it's very difficult to discriminate all of the comorbid conditions that have contributed to the phenotype that we are seeing. So, let's look at that phenotype and traumatic encephalopathy syndrome. So, as I said before, much of the early data historically had come into boxing. This was typically motor-impaired people with slowed speech, bradykinesia, and pseudo-Parkinsonian-like features. And if we look at a series of core clinical features, as had been proposed by Montenegro and others, we see cognitive features, behavioral features, mood features, and motor features. Those can result in a conundrum of symptoms, some of which are related to motor impairment, verbal explosion, mood disorders, anxiety, apathy, insomnia, and motor features. But there are parts of these criteria that were problematic. And there were many criteria, Montenegro, Reims, Viktorov, and Jordan. And what LAFI did was they compared inter-rater reliability of these, and honestly, it wasn't so great. So, then we have to take on another feature. And the other feature is that this work by Grant Iverson in our group is rather important. If you take older men, age 40 to 60, and you ask them, how many of you are jealous? How many of you seem irritable? How many of you might threaten people? These are people without head injury or repetitive head injury exposure. The amount of folks endorsing some of these is pretty stunning, over 20%. So, how we impute some of these effectual symptoms is very problematic, and there's a strong correlation with depression and anxiety, which seems to be rising in middle-aged men. So, if we think about this from a conceptual perspective, right, we have all of these risks on the left-hand side of your slide. Could be age of exposure, which has a possibility of risk, although controversial. Cognitive reserve, these various loads in one format or another, genetic or otherwise. There are comorbid factors, how you lived your life. There are caveats related to vascular health, obesity, probably anesthesia exposure, diet, sleep, that result in this syndrome. And this syndrome is characterized by progressive decline, motor features, cognitive features, mood features, and behavioral features, and I would say the latter two still require further clarity. Now, what was done recently has been really important. Bob Stern and Doug Katz led an NIMDS consensus group to look at diagnostic criteria for traumatic encephalopathy syndrome. And we'll go over some of these core paradigm features, but in this nice paper that was published in Neurology this year, what you can see is there are things that you have to meet to go to each stage that are required. That can be substantial repetitive impacts. That could be clinical core features such as cognitive impairment or behavioral risk regulation. That could be not fully accounted for by other disease processes. In other words, if you're going to fully account for it and get rid of it with something else, it's not TES. And these criteria were originally developed to do one thing, allow us to push forward in understanding the entity by developing research-based criteria, not necessarily clinical criteria. And if this is TES, we'll go over this quickly, what is our certainty level and what do they have? So if I was going to leave you with one thing, it's got to be a progressive course. This can't be boom. It's got to be a progressive decline. It can't be fully accounted for by other diseases. Neurobehavior cannot be accounted for in any way by other diseases entirely. Could be a co-contributor. And corporate diagnoses does not exclude TES. So you could have pain, sleep apnea, other issues. It doesn't exclude TES. What you need is a core feature, substantial exposure, cognitive impairment, and neurobehavioral impairment. So let's look at these for some required elements. So there's got to be substantial repetitive exposure to head injury. There is some argumentative in the literature in different schools that suggests that perhaps form frosts of this can come from a single blow. But for the TES, NINDS criteria, it's repetitive impacts. That could come from military service. That could come from high exposure to contact, collision sports. The general belief, because this has been imputed from pathologic data, is that it's probably five years of organized play or longer. I don't care about you tackling in the backyard when you're eight. I care about organized, practice-oriented collision sport. And other sources are also possible. That could be domestic violence, where there is a significant risk. Head banging or other vocational activities, such as those people who do door blast breaching. And now let's look at the clinical features. There's got to be cognitive impairment. And that cognitive impairment can be reported by self or informant. It's got to represent a change. It's got to be substantiated by prepared informants on standardized mental status or neuropsychological tests that score at least one and a half standard deviations below appropriate norms. There's got to be reporting and subsubstantiation of neurobehavioral dysregulation. That is impulsivity, violence, fused mooding, emotional liability, massive mood swings. But they cannot meet criteria for bipolar disorder. And they can't have features that are behaviorally a bit more linked to frontotemporal dementia, i.e. real disinhibition, inappropriate language or behavior. Now, in C, I emphasize that point, but this is the key issue. These features have to have been progressive and worsening over at least a year's period of time. And in any of the suspected cases, really often longer than that. There are supportive features. Those include delayed onset, begin at least two years after the repetitive exposure. So, if you have a repetitive exposure and these features begin right after that, we've got to think about that. Because we want to space it from persistent post-concussive syndrome or post-concussive spectrum of symptoms. Supportive feature B is that of motor signs. Those can be Parkinsonian, radicanesia, rigidity, motor neuron-related disorders. As we said, there appears to be some linkage to sarthria, action tremor, and ataxia. And in the motor neuron disease, there is a predilection among potential TES cases and others with repetitive exposure to be bulbar and onset. Psychiatric features can also be supportive. That could be significant depression, anxiety, and a profound apathy of individuals. There is also a sense of paranoia, delusional beliefs, or suspicion, persecution, or unwanted jealousy that did not occur early in life. And it can't be accounted for by other disorders. The pattern of cognitive deficit is not most likely accounted for by preexisting or established neurodegenerative disorders. The behavior and mood disorders, if present, are not accounted for by a primary disease. Let's say, for example, FTD, as I just gave you an example. Concurrent neurodegenerative diseases do not exclude TES. So, for example, you could have CTE plus another microvascular pathology resulting in TES, or Lewy body dementia, which may be associated with Parkinsonian-like symptomatology. And concomitant diagnoses of substance use, PTSD, or mood or anxiety disorder does not exclude TES unless they account for all the features. Now, this is going to be a hard one, and a place where researchers who do both the behavioral work and the behavioral linkage to not only the phenotype, but the pathology and biomarkers, be they imaging, or be they blood-based biomarker, or even physiologic, are going to have to draw certain linkages over time. Some of those might be best characterized by the extremophiles. And now we look at the level of function. We may have people who are independent. They're, they have some of these features, but they're still holding a job. They're able to engage in community activities. And they would be A, those who with subtle, indefinite, or functional dependence, slight reduced performance, some difficulties in instrumental ADLs, a difficulty with telephone management. They're going to be in this subtle, indefinite functional dependence. C is those with mild dementia, definite impairment of instrumental ADLs. And they may be functioning in part at home. They have more difficulty in doing some basic activities of daily living. Moderate dementia is those that are not completely dependent. They can be taken to some functions outside the home, but really restricted to simple activities. And severe dementia, they really can't participate in outside activities. They can't function well in their own home setting. So what is very interesting about these criteria is the movement to levels of certainty, because in a difficult to define disease process, I think this level of certainty is very contributory. These are things that are suggested. They meet TES criteria, but don't meet the additional criteria for possible, probable, or definite. Possible CTE is exposure for greater than five years, cognitive impairment, as we just defined, and a minimum of three of the following five. Delayed onset, motor signs, more significant psychiatric features, a level of functional dependence, and the onset of cognitive impairment with or without neurobehavioral variance regulation before age 60. Remember, we're coming back to that age makes it more challenging. And then C, probable CTE, exposure to American football for greater than 10 years, cognitive impairment as defined previously. Delayed onset, motor signs, possibly one of the psychiatric features, severity of functional dependence is mild or worse, and if onset of cognitive defined before age 60, a negative amyloid PET scan. Now, be careful. There have been some discussions about using T807 in a New England Journal paper, which is a tau ligand. We're not talking about that. We're talking about you've excluded via Pittsburgh B or other amyloid analog on PET that this is a Alzheimer's disease pattern, suggesting perhaps if you have all of these things, you're in the probable CTE category. And definite CTE is a neuropathologic diagnosis now, and it always has been, and we can't make that definitively in the clinical setting. What's important is that we now have research criteria to link markers to phenotype. So, if we look at this path for certainty, we see in this cartoon, they meet TES criteria. It meets cognitive impairment criteria. No, you come down here. Yes, we determine the level of exposure, the number of years as we just reviewed. It meets two or more of the following, suggestive of CTE, possible CTE, probable CTE, and then here, it meets the TES criteria. Definitely, boom, TES with definite CTE. This has to be pathologic. But we have to be careful. Not everyone is going to get behaviorally sick. So, this is Emma Russell, again, from Willie Stewart's group in the north of Scotland, in which they matched former professional soccer players and population. All these individuals take a great deal of header-based risk, and we didn't see higher levels of anxiety and stress, higher levels of drug use or alcohol disorder in the professional soccer players. In fact, there may have been a lower risk. So, not all exposure in all populations will lead to a behavioral risk over time. The elegance of what they have is data from the NHS, which, as I stated before, is longitudinal. And we did this commentary in neurology, neurosurgery, and psychiatry about does contact sport always lead to despair? Because perhaps in lower doses, perhaps for refined phenotypes, perhaps for individuals who led a different path of life, their amount of brain reserve or cognitive reserve doesn't produce a meaningful phenotype. You can have pathology in your heart without necessarily having cardiovascular-related syndromes at all. In fact, many of those people can run long periods of time. The issue is there are potential benefits to also consider of that group behavior where being a part of a team may outweigh the social risk over time. And this has to be weighed out very, very, very carefully and understood. When is there a dosimetry-related issue that crosses with other comorbids to produce a TES-like phenotype? So, as I said, other comorbids may be important. And as we've thought about these things, we've thought about exposure, pathology, comorbids, age, human cofactors, a functional decline, and loss of early middle age. So, TES has many important pathologic paths. Further defining unique aspects is needed. There may be a lower boundary of pathology that doesn't progress, and we don't know what its clinical meaning may be. Repeated head injury is not good, and brain injury is linked to degenerative disease. The phenotype and subphenotypes and certainty thereof will be informed by these new criteria. How it relates to biomarkers in full and how we can titrate that into levels of certainty and how we might use that to develop targets for therapy is unclear, and we may be able to see some of these features in other things, such as ALS, temporal lobe epilepsy, or schizophrenia. I thank you for your attention and kindness to this presentation. Well, thank you very much, Dr. Zavant. That was an incredible talk. Dr. Zavant will be available with me for questions at the end of the talk, but with that, let's move on to the second part of our three-part talk, and that is going to be my section on the actual pathology associated with these issues. So, I have no conflicts of interest, and as Dr. Zavant and others have talked about, there are one of the foci that we have when we're learning about traumatic brain injury, especially during residency and throughout training, is on the immediate injury. And so, this is a schematic, then, of the hospitalizations and deaths that occur after traumatic brain injury. But what we often don't really discuss as much is, as Dr. Zavant very clearly pointed out, the long-term sequelae associated with these head impacts. And so, this is a meta-analysis looking at individuals with any brain injury of any severity that involved a loss of consciousness. And these 15 case studies were aggregated by a meta-analysis, and it was found that there's a 1.5 odds ratio associated with any kind of a traumatic brain injury and issues afterwards. So, it basically is a 50% increased risk of having dementia after a TBI with loss of consciousness. So, what are these TBI due to pathologically? Well, there have been a number of different pathologic processes associated with that trauma, the first of which we'll talk about briefly is Alzheimer's disease. Now, a recent review found that only 15 of 55 studies evaluating this link between Alzheimer's disease and traumatic brain injury actually had neuropath follow-up. And so, it's possible that the dementing process that was occurring that was labeled Alzheimer's disease in life was in fact another process. But even without that caveat, there's increasing evidence that there's amyloid deposition that occurs after traumatic brain injury. And in fact, we've observed that in our own brain bank, where we find that a substantial portion of our individuals, so this is 114 individuals and military cohorts, so military veterans exposed to traumatic brain injury or repetitive blast injury, they had the amyloid associated with Alzheimer's disease, and then a subset of them had CTE, as Dr. Zafant was talking about, in addition to the Alzheimer's pathology. So, one thing that was observed though is that in addition to age being significantly related to the presence of Alzheimer's pathology after a brain injury, the brain injury itself was also associated with this pathology, as was duration of repetitive head impact exposure. So, all these factors were associated with both CTE and Alzheimer's in this study. In addition, Parkinson's disease has been linked to head trauma and Parkinsonism, and this is something that's been in the public consciousness really for some time. I mean, in Rocky IV, Rocky was developing a Parkinson phenotype, and in fact, the first description of CTE was in Parkinsonism Features by Harrison Martland, as Dr. Zafant reviewed in 1928. But a lot of animal models also suggest that traumatic brain injury is associated with increased alpha-synuclein deposition, and that that is also in favor of the resolution of Alzheimer's disease. And this is a very interesting study that was done with 694 autopsy participants looking for Lewy body disease in individuals with either exposure to repetitive head impacts, exposure to TBI, or controls. And what we find here is that this depicts the different types of features associated with the different pathologies, pathologic processes. But what you can see is in a substantial cohort of individuals, there's a good deal of Lewy body pathology. And what we found in this study, again, is that duration of contact support exposure was associated with the presence of Lewy bodies. Additionally, there's been a link between head trauma and ALS, and Dr. Zafant reviewed some of that literature for us just now, but we'll briefly touch base on some of the work. So we talked about the NIOSH study and the Harvard Football Player Health Study that Dr. Zafant just now, but there's also a 14-study meta-analysis that found that there's a 40% increased odds of having ALS if you had ATBI, and the severity of that brain injury increased the odds of ALS by 70%. So even though these other studies showed 4x increased risk of ALS in elite contact sports participants, there's evidence as well that with severe brain injury, you can have a very substantial increase in your risk as well, although not as high as these elite athletes. In addition, we had two studies that were published in J&EN that we ended up finding that there was a relationship, again, between contact sport exposure, and actually in this case, sorry, it was blast exposure, and the presence of Tdb43 pathology. So again, there's a link between ALS pathology and both TBI and repetitive head impact exposure. And finally, then there's chronic traumatic encephalopathy, or CTE, as Dr. Zafant went into a good deal of wonderful detail about. But we've talked a bit about the epidemiology of CTE. That's namely that in biased samples, it appears to be very prevalent in this biased cohort of NFL players, but that still represents 10% of all NFL players who passed away and whose brains were studied. The other 90% of individuals whose brains were studied, presumably some subset of them would have CTE as well. In addition, it's not something that's present in individuals without... Is there contact sports exposure? So this is 66 former contact and collision sport athletes and 200 controls within a neurodegenerative disease brain bank. And there's no CTE in individuals without contact or collision sports exposure, whereas about a third of individuals with contact collision sports had CTE. In this VITA study, which looks at individuals from the left bank of the river Danube that were born in 1925 and followed to autopsy, there is no CTE in these 310 individuals without history of dependent exposure. And as Dr. Zafant said, there's this 4X increase risk of dementia in NFL populations. So CTE is certainly linked to TBI and that tau process. So how is this actually occurring? Well, there are a number of different possible mechanisms. The first two involve neuroinflammatory processes. So basically you can have reactive astrocytes being transformed in response to blood-brain barrier dysfunction, or you can have surveying active microglia entering into the tissue, again, because of blood-brain barrier breakdown that occurs when you have the actual collision in contact causing a shearing between the different tissue types of the endothelial cells and the neural tissue. You can also have actual shearing of the axons through a stretching of the axons themselves. And that can cause problems with the myelin sheath that could lead to neurodegenerative disease. You can also have the microtubule breakdown. So the tau is the building blocks of the microtubule. And when that breaks down and it can cause the aggregation of that tau phosphorylating it, and that could result in a pathologic process. And finally, you could have an actual plaque formation due to APP acutely increasing immediately after a brain injury and that causing A-beta peptides and pre-stinillin and other processes, again, causing the amyloid deposition that you see in some cases after a traumatic brain injury. So all of these processes independently or in concert could actually result in pathology that's observed after brain injury. Computational modeling actually predicts the location that we see the bulk of this pathology in following a brain injury. So basically this was a study by Kajari and Brain a few years ago, where they ended up finding that the maximum trauma-induced strain following a injury is actually at the depths of the sulci in the brain. And that's in fact, where we see a lot of the pathology, particularly CT, but with Alzheimer's pathology after a traumatic brain injury as well. This is another computational study that again, found that the maximum amount of strain was located at the depths of the sulci, which is where we find the CT pathology, as well as around blood vessels, again, where we find CT pathology. And so these study in PNAS really supports the idea that this strain is what's actually causing the pathology that we later observed. The neuroinflammatory portion of that model I showed earlier is further supported by evidence in our brain bank, where we find that the co-localization between inflammatory biomarkers, inflammatory cells, and neurofibrillary tangles associated with phosphorylated tau, that they actually are basically touching each other. And so that indicates that there might be some process, whether one causes the other, it's hard to say from these data. And the tau that form as a result of that are in fact different, is different than the tau in other neurodegenerative processes. So the tau following a brain injury is unique. But the overall prevailing factor then is that the total amount of exposure, the number of brain injuries or repetitive head impacts in this case are associated with increasing pathologic murder. So in this case, green is no pathology and darker red is pathology. And this is every case in our brain bank. So 600 some odd cases and the number of years of play on the X-axis. And you can see a clear progression and this is just football. So, you know, the number of these individuals also on military exposure or other exposures that explains why there are some subset of individuals with very little football exposure but have relatively severe CTE. So duration of football appears to be a strong mediating factor. And this in fact is the data behind that figure that I just showed you, which was published in Annals last year. So all of this together supports this idea that there's a relationship between tau pathology and age, which has been well-established for decades at this point and neuroinflammation and tau pathology as well. But the year's exposure can also increase neuroinflammation and can also result in tau pathology. So all of these processes together can result in neurodegenerative decline that we deserve that Dr. Savant referred to. And so it's important that when we take all of these this into consideration that we don't incorporate some magical thinking. Like, so for example, what Tom Brady here said in response to the question about the hits he's accumulated over his career. So he says, well, first of all, I've been doing it a long time. So your body gets used to the hits. The brain understands the position that you're putting your body into and my brain is wired for contact. I'd say in some ways it's becoming callous to some of the hits. So I'd say that's, you know, based on the mechanisms that we just described, that's, this is, like I said, a magical thinking. And so with that, we have a whole team of individuals that have put a bulk of this work together and we would both then like to continue on to Dr. Korte's presentation. So we'll be taking questions at the very end though, but I do appreciate the questions. I see a couple in the chat now, so very excited for our chat afterward. And with that, I'd like to hand it over to Dr. Korte. So Dr. Korte is a brilliant MD-PhD who's one of the world's leaders in studying the neuroimaging processes associated with the pathology I just described and the clinical phenotype that Dr. Savant previously described. She is, has a dual appointment as a professor at the Ludwig Maximilians University in Munich, but is also here at Harvard Medical School. And with that, I'm very excited to hear her talk. Good afternoon. My name is Inga Korte. I'm going to speak about neuroimaging repetitive head impacts, and I'm going to focus particularly on the long-term effects of exposure to repetitive head impacts. My disclosures are displayed here. My spouse is employee at Siemens. I receive research funding from EBIT as well as from funding organizations in Europe and in the US. A special thank you to Ross Savant and Martha Shenton who have introduced me to the field of traumatic brain injury research many, many years ago, and who have seen me grow from a postdoc into a full professor. Before we dive into the aspect of neuroimaging, I want to spend a moment on the terminology. Traumatic brain injury has been characterized and categorized into mild, moderate, and severe TBI based on the initial symptom presentation. The term concussion is often used as a synonym to mild TBI and often used in the context of sport. And then there's the terms of concussive head impact or sometimes also called repetitive head impact. And they, by definition, do not yield acute symptoms. However, there is evidence that if they're sustained over time, they may have cumulative effects on the brain. Subconcussive head impacts are often sustained in context sports such as American football or soccer. And the reason why we're studying the effects of repetitive head impacts are because there's evidence that there is an association with later life neurodegeneration. One of the most prominent neurodegenerative diseases in the context of exposure to repetitive head impacts has been chronic traumatic encephalopathy or CTE, which has been diagnosed in football players and boxers, but also in soccer players. Importantly, CTE is a neurodegenerative disease. It can only be diagnosed post-mortem. It is characterized by a deposition of hyperphosphorylated tau in neurons and astrocytes. It usually starts in the depths of the sulci. You can see this here where the brown stain, and oftentimes starting around the vessels. And then later, in later stages, it spreads across the brain. Potential mechanisms include repeated shear injury. And this is a computer model that shows that shear injury is most intense in the depths of the sulci. This is here indicated by the red spots. And this, interestingly, is the same area where we see CTE starting in the depths of the sulci. There are other sports where repetitive head impacts are common. One of them is soccer. And there's also evidence that in former professional soccer players, there is an increased risk for neurodegenerative diseases later in life. This is a study by Willie Stewart's group in Scotland, and they showed that there was a lower likelihood of passing away in soccer players in their 40s, 50s, and 60s. So there seems to be something protective about being a professional athlete. However, in the later stage of the life, so starting at 70 and 80 and 90, soccer players were more likely to pass away. And when we look at the primary cause of death, interestingly, ischemic heart disease was less likely compared to a large group of matched controls. However, neurodegenerative diseases were more likely in former professional soccer players compared to controls. So this is alarming and matches also the evidence that we have from American football players. This is a very simplified model that my colleague and friend Alex Lin and I came up with in 2015, where we said there are acute effects from traumatic brain injury, and you have an acute decrease in quality of life, or here indicated by the dotted line with repetitive head impacts, where you may not even experience any symptoms, you may not have any decrease in quality of life. In most of the cases, people recover completely from this decrease in quality of life or the symptoms. However, there are some, and the literature varies between 10 to 30% of people after traumatic brain injury, they have chronic or static symptoms, such as fatigue or headaches or sleep disturbances. And then there's a smaller group that will develop after a period without symptoms or steady symptoms, a progressive neurodegenerative disorder. Currently, we do not entirely understand why someone go on to be completely fine and others go on to have chronic symptoms or then even go on to develop neurodegeneration. And then this is part of all of our studies that we're currently performing to identify the risk factors for why people go on to have a neurodegenerative disorder. But I think it's important to have this very simplified scheme in mind when you read studies and hear about study results, so you know where we are approximately in the realm after traumatic brain injury or exposure to repetitive head impacts. So what is going on while we sustain either an impact to the head? It doesn't have to be a hit to the ground, but it can also be an acceleration-deceleration injury. We think that the mechanical force to the head leads to a deformation of the brain tissue. It's basically a stretching, and the stretching can then lead in particular places to a tearing or even a complete shearing of axons. We know that this often happens at the gray matter-white matter intersection. So this is the cortex, and then this is the white matter, and this is an area of high vulnerability because of the different consistencies of gray matter and white matter. We also know that some places in the head are more likely, or some places in the brain are more likely to get injured, such as the center where the corpus callosum lays, for example. Neuroimaging helps us to identify not just in the acute setting, those who need neurosurgical intervention, but for this talk particularly, helps us to identify long-term outcome and long-term brain alterations following mild traumatic brain injury or repetitive head impacts. We are using not just the conventional MRI techniques that we have available in the clinical setting, but we're also using in TBI research a number of complementary techniques. We're, for example, looking at volumetric changes in gray matter-white matter, global gray matter and white matter, but also regional gray and white matter. We'll look at structural and microstructural alterations using diffusion imaging. We're also looking at the connectivity, how different areas of the brain communicate and networks work. We can also look at function using fMRI. We'll look at the biochemistry of the brain using MR spectroscopy. And in some cases, we're able to use PET imaging for metabolism and particularly uptake of tau. In most of our neuroimaging studies, we're using not just one technique, but actually as displayed here, a puzzle of different techniques that explain different aspects of brain health. And I'm going to show you a few of the findings that we have in those exposed to repetitive head impacts over a longer period of time or those with history of mild traumatic brain injury. When we use conventional MRI, and this is, you know, those are the sequences that you would usually get in a clinical setting. We have T2-weighted imaging and FLIR imaging and T1-weighted images here. And you see that there are changes that you might have seen in different other neurodegenerative diseases. And you can see that they're very unspecific. One would be here, there is a reduction in brain volume. You see much more CSF than you're supposed to be seeing. And you can see in the depths of the cell site here. You also see enlarged lateral ventricles. And this is not specific to the exposure to repetitive head impact, nor is it specific to CTE. It is a sign of neurodegeneration in general. And then there is, for example, white matter hyperintensities. You see those in the FLIR sequences. They are usually located around the lateral ventricle, sometimes here directly, and then sometimes they confluent together. Sometimes they're like spots, spread it throughout the hemispheres. And also those are not specific. They are often seen in neurodegeneration. They're also often seen in association with cardiovascular risk factors, for example. You see an enlargement of lateral ventricles as shown here. You also sometimes see specific regional atrophy, such as here in the hippocampi on both sides. But as I said, conventional imaging will only show us very late in the progress changes, and it will not help us tremendously or significantly in differentiating different neurodegenerative diseases, which is why we're then looking at more advanced neuroimaging techniques. One of the signs that we often see in patients with traumatic brain injury, and which has also been associated with CTE post-mortem, is echivum septum pellucidum. That is shown here in a post-mortem brain, and can also be identified on MRI. We measured the CSP in a group of former NFL players, and were able to show that former NFL players who have been exposed to repetitive head impacts throughout their career have higher rates of CSP. So it's more common in NFL players than compared to controls. And when there is a CSP, it is also greater in length, and it is, so it's greater in size. And also if we correct this for the size of the corpus callosum. And importantly, there was also an association with cognitive functioning in those players. So do we think that the CSP is causing cognitive decline? No, we don't. But we think that it is a sign of brain injury, probably due to mechanical injury, and that the decline in cognition is also a sign of declining brain health. Another sign that we see and we can measure, we can quantify using advanced neuroimaging is brain atrophy. It has been seen in cases of CTE postmortem, and we can also see it in those exposed to repetitive head impacts, for example, in American football players with reduced brain volume. And particularly important are regional brain atrophy. For example, here in a study of American football players from the DETECT study, we saw that amygdala volume, cingulate gyrus volume, hippocampal volume were significantly decreased in the American football players compared to the controls. And importantly, their reduction in brain regional volume was then associated with a decline in cognition, for example, here in attention and psychomotor speed, as well as with visual memory. So those are all signs that there is a neurodegeneration going on. We can measure it with neuroimaging. We can also relate it to behavior and behavioral outcome. Another important aspect was a reduction in thalamus volume that we saw in those exposed to repetitive head effects while playing American football. And importantly here, for every year athletes played longer, they had a significant decrease in thalamus volume. So this really indicates a dose-response curve. The longer you play, the lower your thalamus volume. And interestingly, when we looked at the time when they started to play football, when they were first exposed to repetitive head effects while playing football, they had an even more impressive decline with almost 65 per year. The year they started playing earlier. And the reduction of thalamus was again associated with decline in cognition here, particularly worse visual memory. The other study that points in the same direction that the age, a younger age at first exposure may be a risk factor for later life brain alterations is the study here that we performed using diffusion imaging where we looked at the corpus callosum. And again, the corpus callosum is an area of high vulnerability within the brain because of its location. And it's very, very long fiber tracts. And here in this study, we compared those who started playing football before the age of 12 and those who started playing football after the age of 12. And we found that those who started earlier, so that's the AFE group here in green, had lower values of fractional anisotropy. And fractional anisotropy is a measure of integrity of microstructure of a fiber tract. So those who started playing earlier were exposed earlier, had a higher risk of showing lower numbers here and lower integrity of the fiber tracts of the corpus callosum. Now, going from those who are already later life athletes who used to be exposed to repetitive fat impacts to those who are currently exposed and they're young and they're healthy. This was a study where we investigated 15 to 17 year old soccer athletes that were competitively playing soccer. It was one of the earliest studies from 2012 where we compared them to swimmers and we made sure in the study that they were not exposed and they had not experienced a mild traumatic brain injury previously. And what we found, and that was impressive at the time and was also concerning, is that we could already see changes in their brains compared to the swimmers at this early age without the history of a concussion or mild traumatic brain injury without any symptoms or cognitive decline. And we could already see very similar signs in diffusion imaging, and those are, for example, radio diffusivity, axial diffusivity and also fractional anastrotropy that are very similar to what we know from and knew already at that time from studies on mild traumatic brain injury. So this study indicated for the first time that those brain alterations may actually precede cognitive decline. We may be able to see them very, very early on in an athlete's career. We then also looked at the important tracts that we already know are highly vulnerable to shear deformation, the corpus callosum, the corticospinal tract, and the cingulum bundle. And we were able to verify that even for those specific tracts, we had very, very similar results. So this was a result that then sparked many other research studies looking at the effects of repetitive head impacts in younger athletes. And with the aim of hopefully identifying biomarkers, neuroimaging biomarkers early, to be able to tell who may be going on to neurodegenerative disorder or cognitive decline later. Importantly, most of the studies that we currently see are based on group differences. However, we know from a clinical perspective that TBI is very heterogeneous, that every injury is unique and every brain is unique too. So you have two sets of uniqueness that come together. And here in this study, based on the interest consortium cohort, we actually looked at the different participant and compared them to a normative atlas that we created based on the controls of the study. And we were able to show that the areas of changes in diffusion measures that indicate microstructure damage are in very different locations here across the different participants. And those are just a few examples from this large scale study. There are, however, areas that often pop up in participants such as here, the corpus callosum that you see in different subjects. However, there are also areas that only pop up in some, and the more you go into the periphery, the less likely that you find overlap. So I think from a clinical perspective, we really need to also move the field of neuroimaging towards individual injury profile, not just relying on cohort group comparisons. Another aspect of current research are risk factors that are based on the individual itself. In this case, we looked at sex-specific differences following exposure to repetitive head impacts and how differences between females and males may look like in neuroimaging. Here, we did a study on ice hockey players exposed to repetitive head impacts over the course of a season, and we compared how much they changed and where their brain structure changed over the season and compared males and females. And what we saw was that females, first and foremost, had more pronounced changes over time and more different clusters in the white matter and also in larger clusters. And here on the left-hand side, you see those red clusters where females had more changes than males between pre-season and post-season. You can also see this displayed here on the right side where the changes between males always left to post-season are not as pronounced as female pre-season to then post-season across the different diffusion measures. So this was one of the very first studies that we published in 2017, indicating that there was a greater vulnerability to the effects of repetitive head impacts in female athletes. So what could be a cause for this greater vulnerability? We know that volume, the brain of females and males is different to begin with. For example, volume is different. It's known that males have a greater brain volume than females. It becomes more complex when you look at gray matter and white matter because there are areas where females have more gray matter and other areas where females have less gray matter. And a similar pattern evolves when you look at white matter. It's also complex. And then there is a new study that came out in 2018 that looked a level deeper and looked at the axons using electron microscopy. And they found that females had a smaller diameter of axons. You can see this here in the upper row compared to male axons. And importantly, it was not just a lesser or smaller diameter, but also a lesser number of microtubules that you can see here. So microtubules are the part of the axon that gives it stability. So female axons are less stable, so to speak. And the density of those microtubules was also less. So it was not just that they were smaller, but also they were in proportion less microtubules. And importantly, when they pulled on those axons, they also saw them rupturing sooner or more easily than male axons. So this could be at least one reason for the greater vulnerability of the female brain to the effects of maltraumatic brain injury as well as repetitive fat impacts. There are certainly other aspects that play a role here, and I do not want to go into too much detail, but I think this is important also from the clinical perspective. It's a study that looked at different networks that females and males are predominantly using. And we know that males use networks that are mostly related to motor sensory and executive function more often, and they rely on those more often than females. And I know this is a generalization, but it's a pattern that has been documented. And females are using networks that are more for social motivation, attention, memory. And I think the important aspect here is that the networks that are being used by males more often are intra-hemispheric connections. So you see them here on the bottom side. You see the networks are within one hemisphere, either the right or the left. However, the networks that females are more relying on are connections that are inter-hemispheric, meaning that they connect between the left and the right hemisphere. And you see them here in the upper row. So if you think of the mechanical forces that result from a hit to the head and that they oftentimes injure the center of the brain where the corpus callosum is, you can imagine that the same hit to the head with the same force resulting maybe in the same type of injury may then lead to a different set of symptoms in those who rely more on this inter-hemispheric network than those who don't. So even this difference in network could also explain some of the differences that we see between males and females. Going back to why we keep studying the white matter in the context of mild traumatic brain injury and repetitive head impacts, in this study from just this year, we looked at American football players and associated the white matter diffusion findings with their level of total tau in their plasma. And we could see that those who had higher values of exposure here measured with the head impact index, they also had higher values of actual diffusivity in the anterior part of the corpus callosum. So we're right here. We also saw the same type of results for the second part of the corpus callosum. So this would be the red area. And importantly, the diffusion changes that we saw in the anterior corpus callosum were also associated with a higher level of plasma total tau in those retired professional football players. So that means that the signs of neurodegeneration that we see in diffusion imaging also relate to higher total tau that we can measure in the blood plasma. And this is a recent study that then underscores the importance of studying white matter in those exposed to repetitive head impacts. Because here, this is a study by Anne McKee's group. They saw from a biopsy that there is a decreased number of oligodendrocytes in the white matter in those with CTE. And also not just a decreased number, but an altered proportion of the different subgroups of the oligodendrocytes in the white matter. And they also saw that this was associated with neuroinflammation and neuroinflammatory transcripts. So this indicates that the changes that we see in the white matter may not only be due to sheer deformation and mechanical injury, but also may be due to neuroinflammatory neurodegenerative processes in the white matter. And interestingly, even six years ago in 2015, Alex Lynn and I published a study on former soccer players where we had already neuroimaging signs of neuroinflammation. We looked at those former professional soccer players, compared them to an athletic control group. And we saw increased cooling and increased myelinositol in the former soccer athletes. And those metabolites correlated with more exposures to those who had been exposed more to repetitive fat impacts showed higher measures of neuroinflammation. And this is in line with the previous study that I showed that came out years later. And interestingly in those soccer athletes in the same group, we saw that there was a link between the neuroinflammation going on that were measured with MR spectroscopy, also with a decrease in cortical thickness. And now we know that cortical thickness decreases over time in normal aging. And we can see this here in the control group, you see the blue line that slowly declines over age. However, in the soccer group, we saw a steep decline in cortical thickness compared to the control group that we then interpreted as a sign of accelerated aging. And they showed a cortical thinning in those blue areas here on the left and on the right hemisphere. And again, we saw a link between the thinning of the cortex and exposure to repetitive fat impacts throughout their career. So in summary, we have seen evidence of a link between exposure to repetitive fat impacts in athletes of different sports and brain alterations measured with different types of advanced and conventional neuroimaging. So there is an association between exposure and brain alterations. We don't know, we don't have the full picture of the mechanisms that lead to those brain alterations, but the neuroimaging findings that I showed you indicate that there are neuroinflammatory and neurodegenerative processes that may lead to those findings later in life. We know that exposure itself cannot explain all the findings because we often see many other athletes that don't show those signs. So there are other risk factors that need to be identified. What we know from neuroimaging so far is that the duration and the extent of exposure to repetitive fat impacts seems to be associated, correlates with severity of imaging findings that we see. We also see that there may be an association between the age at first exposure, at least this was true for American football players, that an earlier age leads to more severe changes later in life. And another indication from neuroimaging is that female sex may be a risk factor for more severe brain alterations due to the effects of repetitive fat impacts and mild traumatic brain injury. What is desperately needed are more perspective and longitudinal studies using comprehensive approaches. And what I mean by comprehensive approaches is not just one imaging technique or one cognitive test, for example, but really a battery of tests and imaging techniques that help to explain brain health. And of course I'm biased here, but I think neuroimaging has a great potential of identifying risk factors, but also helping in identifying spatial location of brain alterations associated with symptoms and cognitive outcome. I thank you very much for your attention. Thank you very much, Dr. Corte. Dr. Zafant and I are here and excited to answer questions. It looks like we have a lot in the chat. I apologize that we've gone a little bit over. So we unfortunately won't be able to get to all of the amazing questions that we have so far, but we'd be happy to take a few of them. So I see the first one here is, I get quite a few evaluations for cognitive impairment in psychiatric patients, bipolar and schizophrenia. I'm told these patients start lower and might be more liable to a dementing process. Any insights as to cause and or different pathology from the usual dementias? So I can take this one. That's a wonderful question. So whenever we're talking about neurodegenerative disease, everybody has their own baseline functional status. And there's this idea in neurodegenerative disease of cognitive reserve. So what can happen then is if you have a lower baseline cognitive reserve, then you might be less able, less resilient to a certain amount of pathology. And so what we see, we see this in cases of individuals with other neurologic processes. We also see this in folks with other types of injuries to the brain. And so one of the hypotheses for what might be happening with brain injury in general, but also with bipolar and schizophrenia is that when individual's cognitive reserve might be lower and for the same amount of pathology, they might unmask symptoms a little earlier, which is why we have to add to that. Yeah, Dan, I would just add that some of our patients who do have those behavioral disorders also have associated comorbids or other linkages that further make them further susceptible to brain dysfunction on top of their lack of reserve. That and the combination of all these things certainly makes a population potentially more vulnerable. Perfect, so the next question is, I got in a bit late, I don't recall hearing about ApoE4. Is this important or has this been seen as idiopathogenic? Dr. Zavot, would you like to take it away? Sure, I'll take a crack at this one and then let you throw up a lob for you. So ApoE genotype is probably the one that is held up the most in a brain injury risk for a neuro decline. The strongest series of the evidence seems to be in those with more severe injury who are ApoE4-4 or ApoE3-4, which is the at-risk alleles. The question comes in that is somewhat provocative because ApoE is involved in lipid metabolism in the brain is, is there the opposite extremic group? In other words, we know those people who are ApoE2-2 are somewhat protected against dementia. What hasn't been done and what that our group is starting to begin to look at is, is there a protection effect in brain injury from those rare people who are ApoE2-2? And you have to have enough numbers to really begin to look at that. But the linkage tends to fall off in the strength of its association a little bit in milder injury, at least single milder injury. Nothing to add there. We'll take one last question because we're pushing 10 minutes over now. And I know we have to jump off to the assembly meeting after this at 15 past. So how does temporal lobe epilepsy produce a TRS like this picture? Is that a, like a CTE like picture which I think that's what that's referring to. So I do have an answer for that. So there have been a number of studies that report CTE like pathology in individuals with temporal lobe epilepsy. And I think one of the major issues there is that in the current CTE pathologic diagnostic criteria were first proposed in 2013. And then subsequently the NIMDS criteria were proposed in 2016. And those allowed for neuronal and astrocytic tau, phosphorylated tau to be diagnostic of CTE. However, it was not intended. So a strict reading of the criteria allowed for either for just astroglial pathology, tau pathology to be sufficient to qualify as CTE. Although that was how the criteria were written the criteria were revised in 2021 to indicate that it wasn't intended for astroglial pathology alone to be diagnostic of CTE. And in fact, if there is astroglial tau pathology alone that's probably not CTE, it's probably RTEK age-related tau astrogliopathy. And so the cases of temporal lobe epilepsy that were diagnosed with CTE like pathology, probably RTEK. So- The other thing in linkage to what Dan just said which is he's absolutely right. Those cases are more likely than not RTEK in reality is a lot of people with TLE tend to fall and repetitively hit themselves. Right? And so there's a clinical phenomenal logic. How many times did they fall and hit their head on top of the fact that more likely than not the pathology itself was not correctly laid. Yeah. With that, I feel sad to end a little and 10 minutes late but with so many great questions on the board but it was truly a pleasure talking to you all and thank you for this. Thank you all. Thank you.

traumatic brain injury

neuroimaging findings

neurodegenerative mechanisms

comorbidities

chronic traumatic encephalopathy

amyloid beta

traumatic encephalopathy syndrome

repetitive head impacts

brain alterations

individual risk factors

longitudinal studies