 Hello and welcome to Noon Conference hosted by MRI Online. Noon Conference connects the global radiology community through free, live educational webinars that are accessible for all and is an opportunity to learn alongside top radiologists from around the world. We encourage you to ask questions and share ideas to help the community learn and grow. You can access the recording of today's conference and previous Noon Conferences by creating a free MRI Online account. You can also sign up for a free trial of our premium membership to get access to hundreds of case-based micro-learning courses across all key radiologies of specialties. Today, we are honored to welcome Dr. Susie Bash for a lecture on critical updates in the dynamic landscape of Alzheimer's disease and dementia imaging. Dr. Bash is a neuro radiologist and medical director at RadNet. Prior to this, she was on the faculty at UCLA as an assistant professor of neuro radiology after completing a two-year neuro radiology fellowship and residency also at UCLA. Dr. Bash's passion and interest lie in artificial intelligence applications in advanced neuroimaging, which add patient-centric value and quality. She's a recurring guest on TV, radio, and podcasts and is actively involved in AI clinical trials, peer-reviewed publications, and AI-related educational talks and webinars. Dr. Bash serves on the Medical Advisory Board of several AI companies and also serves on the editorial board in the AI section for Applied Radiology and is a frequent contributing author to this journal. At the end of the lecture, please join Dr. Bash in a Q&A session where she will address questions you may have on today's topic. Please remember to use the Q&A feature to submit your questions so we can get to as many as we can before our time is up. With that, we're ready to begin today's lecture. Dr. Bash, please take it from here. Hi, my name is Susie Bash. I'm a neuro radiologist at RadNet. Thank you so much for joining us. Today, we're going to talk about critical updates in the dynamic landscape of Alzheimer's disease and dementia imaging. These are some disclosures. So first, we'll focus on sort of an introduction to dementia. Then we'll talk about dementia imaging, then disease modifying therapy and regulatory updates. And then we'll move into ARIA, which stands for amyloid related imaging abnormalities. And we'll actually do some ARIA training with clinical cases so that after this webinar, you'll feel comfortable reading ARIA cases. And then we'll talk a little bit about the future of dementia and then get into some dementia clinical cases. So there are a lot of different kinds of dementia, Alzheimer's disease, vascular dementia, dementia with Lewy bodies, frontotemporal dementia, logopenic progressive aphasia, traumatic brain injuries, cerebral amyloid angiopathy, and CAA related inflammation, as well as normal pressure hydrocephalus. We won't have time to talk about all these today, but we're going to primarily focus on Alzheimer's disease. And the reason we talk so much about Alzheimer's disease is it makes up the majority of the causes of dementia with over 60% by autopsy. So there are 6.7 million Americans that are suffering from Alzheimer's disease. It's a big population issue costing our nation $355 billion in 2021, and that will rise into the trillions very shortly. One in three seniors will die of dementia, so that kills even more than breast cancer and prostate cancer combined. Death from heart disease is down 7% since the year 2000, but death from Alzheimer's disease is up 145%. And this is something that really progresses as you get older, so it doubles in frequency every five years after the age of 60. Now Alzheimer's disease has two major hallmarks, beta amyloid plaque, which is extracellular and can be directly imaged with amyloid PET, and neurofibrillary tangles, which is intracellular tau, most notably in the hippocampi, and that can be imaged with tau pet. Now amyloid PET study can be positive at the preclinical stage of Alzheimer's disease, in fact, up to 20 years before the patient is symptomatic. Amyloid plaque by deposition in disease, you'll tend to see a lot of it with Alzheimer's disease. You can see some with dementia with Lewy bodies, but you shouldn't expect to see any amyloid deposition in front of temporal dementia. This is an Alzheimer's cell here with plaque burden and neurofibrillary tangles compared to a healthy looking cell. This is a positive amyloid PET here and this is a positive tau pet. Beta amyloid 140 and 142 peptides are produced from amyloid precursor protein. The beta amyloid 42 is the main peptide associated with Alzheimer's disease and beta amyloid 40 is the main peptide associated with cerebral amyloid angiopathy. These beta amyloid peptides are transported across the blood brain barrier and then degraded by specific proteins and enzymes such as apolipoprotein E and MMPS. The accumulation of these amyloid fibrils in the walls of small to medium size arterial blood vessels and capillaries in the cortex really are thought to result from a disturbance of any of these time points, either from increased production and paired transport or impaired degradation of the beta amyloid peptides. This aberrant pathway really helps explain cerebral amyloid angiopathy which is known to cause spontaneous interest cerebral hemorrhages, cognitive decline and transient focal neurologic events in addition to Alzheimer's disease. Now we're now starting to realize the importance of the glymphatic system and the role it plays in the clearance of beta amyloid plaques and other toxins. So the glymphatic system denotes perivascular pathways whereby CSF enters the brain parankama via the periarterial space and then mixes with interstitial fluid. This drives clearance of interstitial metabolic solutes and waste such as beta amyloid plaque, which then exit the brain via the paravenous pathway to drain into the lymphatic system. Now these astrocytic aquaporn for water channels in both the arteries and the veins play an important role in reducing the resistance of CSF to movement between the perivascular spaces and the interstitium. Now, unfortunately, as we age this glymphatic convective drainage pathway and the clearance of waste products such as beta amyloid plaque will slow down. So reduced waste clearance has not just been associated with Alzheimer's disease but also with traumatic brain injury, hemorrhage, hypertension, strokes and microvascular ischemic disease. Interestingly, when we sleep these interstitial spaces increase in size, allowing more effective clearance of the beta amyloid plaque. So good sleep patterns actually are thought to play a protective role in Alzheimer's disease. Tau is a protein that helps form and strengthen microtubules which transport nutrients along the axon. Now in the disease brain there's an imbalance of protein kinases and phosphatases, which cause town to become abnormally hyperphospholated resulting in the disassembly of these microtubules. And this impairs cellular signaling. So free towel molecules then aggregate into insoluble paired helical filaments and straight filaments as well, which accumulate in cells as neurofibulari tangles in Alzheimer's disease and other Tau apathies. So, Tau apathies would include Alzheimer's disease behavioral variant frontal temporal dementia chronic traumatic encephalopathy downs progressive super nuclear palsy and cortical basilar degeneration. There are two main types of Alzheimer's disease early onset which is familiar that's less than 5% much more rare, and late onset which is sporadic and that's 95%. Non modifiable risk factors include age, gender as so it's more common and women probably because women live longer, and also genetics. Apo E4 is the most common genetic risk factor associated with Alzheimer's disease. Now modifiable risk factors include hypertension, diabetes, smoking, excessive alcohol use, obesity, poor diet and for exercise, lack of cognitive engagement in social isolation, depression traumatic brain injury. And then as we mentioned before, sleep deprivation due to the lymphatic system. So, Apo lipoproteins play a role in lipid and cholesterol homeostasis in the blood. There are three Apo E gene alleles E2 E3 and E4 of which E3 is by far the most common greater than 50% of the population has this Apo E4 however is the strongest genetic risk factor for all Alzheimer's disease, because it's involved in impaired clearance of that beta amyloid plat. So 25% of the population carries one Apo E4 allele. And so they're called heterozygous and they carry a three times risk of Alzheimer's disease. But two to 3% of the population carries to Apo E4 allele so they're homozygous and have a 12 times risk of developing Alzheimer's disease. Apo E4 is a risk factor not just for Alzheimer's but also for cerebral amyloid angiopathy, CAA related inflammation, dementia with Lewy body, tau apathy, micro vascular schemic disease and vascular dementia, multiple sclerosis, poor outcome following traumatic brain injury and also ARIA. When it comes to dementia imaging, typically a patient will present with memory loss and undergo cognitive testing often by the neurologist, and then an MRI is typically ordered in part to rule out other pathology that could be causing the memory loss. So for example, posterior cerebral artery strokes can often infarct the hippocampi and result in some memory loss. So can heavy burdens of micro vascular schemic disease. Sometimes quantitative volumetric imaging is done. We actually do this pretty routinely in my clinical practice. We've been doing it for about 17 years now. We find this very helpful in the evaluation of dementia. And then if there remains clinical ambiguity, sometimes a PET study is ordered. FTG PET is by far the most common PET study ordered and this is reimbursable by CMS. Amelie PET is hopefully right on the brink of being reimbursed right now, which we'll talk about in a minute. And then tau PET, which is so far not reimbursable, but can be used in the evaluation of dementia. This is a positive FTG PET. This is the PET MR fusion. And these are both PET CT fusion images. And you see cortical hypometabolism in the bilateral temporal lobes here. And this patient had Alzheimer's disease. This was statistically significant when I ran it through the MIMNR analysis software. Now this is amyloid PET. Just for example here, this is a negative amyloid where you see a tree and branch type pattern and no binding of the amyloid tracer to the cortex, as opposed to a positive PET where we have diffuse binding of the tracer throughout the cortex. This is color fusion right here. Quantitative volumetric imaging identifies and labels anatomic structures in the brain and then quantifies the volume of those brain structures and compares that to an age and gender matched normative database. And this provides volumetric tracking to assess for rate of change over time. There are different companies that perform quantitative volumetric imaging. That was one. Here is another. This is another one here and another here. And then finally this is one that's in development. And there's different reports that these companies produce that but basically they track different pertinent areas of the brain and compare that to a large normative database. Anything in the pink zone here would be more than two standard deviations outside of the mean. So you get an idea of what is important and statistically significant for that patient's age. This is another report from a different company that's also tracking the volumes over time and comparing to a normative database. And this is what another report looks like here. Now, with when it comes to PET imaging, again, we tend to use amyloid and towel can be used and FDG pet. And this is these are what these studies look like in the normal scenario and here's abnormal. This is a positive amyloid pet, a positive towel pet, and this is a positive FDG pet. This patient happens to have Alzheimer's disease. Now, the FDG pet patterns would look different across different diseases. Alzheimer's would be temporal pridal and posterior senior gyrus frontal temporal would, as it sounds, tend to be frontal lobes and temporal lobes. Dimensional Lewy bodies can look a lot like Alzheimer's disease or sort of a diffuse pattern but will often involve the occipital lobes as well. Now, towel pet, unlike amyloid pet, towel pet accurately predicts the location of future atrophy. So here is an amyloid pet where we have diffuse binding, but this is a towel pet where towel accumulates in the areas that have atrophy first. So towel, so this is a towel here and this is atrophy. So towel is actually a better predictor of the timing of imminent cognitive decline. Now, I'd like to talk about these important regulatory updates. This is a very exciting time for dementia. Following the FDA's decision on July 6, 2023, to grant traditional approval of lekembi. The generic name is lecanomab. CMS then immediately affirmed that they will cover this medication broadly. So this is a very pivotal step that really is allowing means for drug access to the millions of Americans that are suffering from Alzheimer's disease. And the only criteria for CMS coverage is you have to have Medicare, you have to have mild cognitive impairment or mild Alzheimer's disease, which must be documented as evidence of amyloid plaque. And then you also have to join a registry which your neurologist will help take care of for you. It's on the CMS website and it's a very simple registry for your neurologist to sign you up in. Individuals with original Medicare will pay the standard 20% co-insurance of Medicare approved amount of lekembi once they meet their part B deductible, but Medicare will cover the other 80%. And actually the largest health system in the United States the VA started covering this drug even before CMS CMS was waiting for the FDA can traditional approval before they decided to make the decision to cover. And that is very exciting. Now there are different disease modifying drugs that you have heard about so far at a helm is really, really a historical at this point because the nanomab and lecanomab have really outperformed at a helm but these are all IGG monoclonal at a helm and received accelerated approval on June 2021 to nanomab has not yet achieved approval yet and lecanomab is received accelerated in January 2023. And lecanomab never received traditional FDA approval. Donanum have hasn't yet but we'll file soon for this and lecanomab received this just on July 6 2023 so this is the first disease modifying drug to have traditional approval. Now all three of these drugs did a great job at their secondary endpoints which is to clear beta amyloid plaque from the brain. And it comes to the primary endpoint which is slowing the rate of cognitive decline at home did not do as well so only in one trial they had 23% improvement but not in the other trial. Donanum have actually had 35% slowing and cognitive decline, their results of their trial were just recently released in July 2023, and lecanomab had 27% slowing of cognitive decline. Now the price tag for lecanomab is big at 26.5 K, but you have to realize, you know, also that this is a lethal disease, and to be able to give time back to our loved ones by slowing their rate of cognitive decline is a big deal. And again CMS is now covering as is the VA, and then private insurers tend to follow. So in the clarity ad phase three trial which was published in the New England Journal of Medicine on January 5 2023 lecanomab met the primary efficacy endpoint by slowing the rate of cognitive decline by 27% over an 18 month period, based on the clinical dementia rating sum of boxes. Now in the trailblazer alls to phase three trial which was published in JAMA on July 17 2023 so just published. Donanum have met the primary efficacy endpoint by slowing the rate of cognitive decline by 35% at 18 months. Additionally, nearly half so 47% of the Donanum have participants had no disease progression at one year with the CDR SP rating scale so that is a really big deal. Patients with less advanced disease so patients that had lower medium tau burden outperform those with more advanced disease that had higher tau burden, and that is just confirming that that earlier the treatment is initiated the higher the clinical benefit, which is why the labels are suggesting this at the early stages of disease. So again it's important to note lecanomab is the only one that's approved by traditional approval to treat this early Alzheimer's disease and approximately 1.5 million Americans will be candidate for on label therapy with lecanomab so again 6.7 million Americans have it but when you just look at the early stage disease, we're probably talking about 1.5 million. Now numerous other trials are also ongoing including disease prevention trials such as the dominantly inherited Alzheimer's network trials, and then there are several trials that include pharmacologic targeting and monitoring of other disease biomarkers for Alzheimer's such as neurofibrillary tangles, which can be imaged again with tau pet. Now the National Institute of aging is currently funding more than 450 active clinical dementia trials so there is a lot of activity in this space. This is looking at the primary endpoint for the clarity trial and here you can see lecanomab here slowing the rate of cognitive decline compared to the placebo. And then if we look at the secondary endpoint which is clearance of the beta amyloid plaque there's just no comparison here's placebo here, no change and slight increase in plaque. Whereas look at how well lecanomab is doing and dropping the amount of amyloid plaque in the brain. So again here's placebo on an amyloid pet looks the same a year later, but the treated limb is really doing a good job of clearing that plaque. Now the efficacy of clearing the beta amyloid plaque is dose dependent the higher you go up in dose the better job it does here's 10 mgs per KIG, and this is the, it's, you can see that much more plaque is cleared at the higher dose. But the problem is, is, like most medications, you know, there can be side effects so with disease modifying therapy and Alzheimer's disease with using monoclonal antibodies, the side effect is aria. Aria E stands for edema or so called fusion that can develop so here's edema that's developed and here's a so called fusion and aria H, H stands for hemorrhage represents micro hemorrhages that can occur in the brain. So you see these tiny little dots on GRE or superficial cirrhosis. Now so called fusions with aria are thought to represent a leakage of protonaceous fluid from the meningeal vessels, and the area edema is thought to represent leakage of protonaceous fluid into that prankimal interstitial compartment. If we look here across the disease modifying therapies, you can see that the incidence of aria is much higher with Adahelm with LeCambie it's the lowest it's 21% overall. You can see that the aria E is about half of that for Donanumab and the symptomatic aria which is the one that we really care about, again much higher for Adahelm with Donanumab at 6% and then less than 3% for Lecanumab so about half that of Donanumab. Symptoms of aria can be varied but it's often could be headache, confusion, dizziness, nausea, vomiting or visual disturbance. And the risk factors include that initial treatment period. So when you first go on treatment at first three or four months you're at higher risk for developing aria. Also the higher doses are associated with an increased incidence of aria. And if you're an ApoE4 carrier, you have an increase incidence and so the label Lecanumab label actually recommends that you get tested for ApoE4 just that you can go over your risk with your neurologist and have a decent idea of what your risk of developing aria is. So right now patients are paying out of pocket for that genetic test and you know maybe in the future that will be covered but that's why not everyone is getting it but it is useful if you do get it. Other risk factors for aria would be cerebral amyloid angiopathy and then if you are on anti-coagulation you are also at a greater risk of bleeding. So the aria E in ApoE4 carriers occurs in 5% of heterogeneous carriers and 50% of homozygous carriers again why it's useful to get the test prior to starting medication. But again the good news is that symptomatic aria which is what we really care about is in less than 3% and the aria E nearly always resolves within one year. So it's exactly typically within four months but usually always within a year so again 97% are asymptomatic. Here's a look at aria E temporal change over time so this was the initial development of aria E and it sort of got worse and then it started getting better and by 300 days had resolved. Here's another case here's the baseline MRI. This is post dosing the patient developed aria E but on the post dosing follow-up which is typically the follow-ups are typically done at one month. It had completely resolved and the patient could resume treatment. Here's another patient here's their baseline they developed aria E post dosing it got a little worse at the one month stage and then by the two months and I got better. But this is the GRE here. You notice here at the post dosing they developed some aria H just two micro hemorrhages just very mild that got a little worse over time and was even worse again at the second follow-up study. So let's talk about how you grade aria. There's mild moderate and severe and there's specific criteria for this. Aria E grading is all about the size. Pay close attention to this slide because we're going to go through some cases and I'd like you to be able to choose what you think the grading is. So mild for aria E is five centimeters and monofocal. Moderate is less than five centimeters and multifocal or between five and 10 centimeters and that can be either mono or multifocal. Severe is greater than 10 centimeters and again can be mono or multifocal. So once you get into multifocal you are by definition moderate or severe in degree. Now aria H unlike aria E is all about the count. So it's not about the size it's about the count. So micro hemorrhage is if you have less than or equal to four you're considered mild aria age. If you have between five and nine you're moderate and if you have more than 10 or equal to 10 you're severe. Now superficial cirrhosis would be mild as one focal area. Moderate is two and severe is greater than two. So you need to follow along the soul side and see if it's all one area of involvement. Now let's talk about there. Let's look at some clinical cases where we can go over grading. So what I'd like you to do is as you're looking at this, think in your mind what you think the grade is. So here's baseline and here's post dosing. We see one area here of aria E development. It looks like it is less than five centimeters in size. So this is mild. It's monofocal and it's less than five centimeters. Here's another case that you have three areas of involvement on post dosing. So that would put you into the moderate category because it's multifocal. Here's a patient at baseline. Here's post dosing and you see here there's extensive involvement. In fact, you might be wondering did they have a big stroke, which is why DWI is a required sequence for aria screening. And no, this is not a stroke. This is just shine through artifact. So this is severe grading here. Here's another patient. Here's the baseline and here's post dosing. We see a little area here of a so called fusion and another little area here. So this is moderate because it's multifocal. And so we're in the moderate category. It can't be mild. Here's another patient here and we see multiple so called a fusion here on both sides of the brain. So this is going to be a severe aria E case. And here is again baseline. Here's post dosing. We have this area of so called involvement. And this is actually a mild case because it is only involving one area of the brain here. And it is less than five centimeters in size. Now here's another case here where it's a little bit larger. It looks like it's probably just under 10 centimeters in size for so called a fusion. So this is moderate in degree. Here is a baseline. Here's post dosing on GRE. And we see some micro hemorrhage here. And here's baseline on the flare and here's post dosing on the flare. This is mild for aria age because there's less than five foci. And it's also mild for aria E as well. Now here's a GRE here post dosing. You see multiple little foci. If you count these, these are actually, I don't know if you can see it, but there's actually five little foci here. The number matters. And even though it's small, you have to count the individual little foci. Here's on flare where we see one area. So this would be moderate aria age because we have five foci here and mild for aria E because we just have one small area. Here's another patient baseline and then the post dosing GRE. We clearly have over, you know, 10 foci here. And on the baseline flare and then the post flare, we see an extensive area involvement. So this is severe aria age and severe aria E. Here's another. This was actually post dosing. And this is all on the same patient. Multiple little foci of micro hemorrhages. So that's severe aria age. Here's another case here. This is the baseline. Here's post dosing where we see a superficial citerosis here. And then we have two foci of micro hemorrhages. These are actually just vessels. That's not hemorrhage. And so how would you grade this? This is actually mild aria age, which you might be surprised about, but it's because we have just one focal area of superficial citerosis. If you follow these down, these are all contiguous soul sigh. And we just have two little dots of micro hemorrhage. So we're mild in category for micro hemorrhage and we're mild in category for superficial citerosis. So it doesn't upgrade our call. The call stays as mild. Here is the baseline. Here is post dosing. We have superficial citerosis here and a second area of superficial citerosis. Again, it's important to make sure you know if you're in the sulcus or if you're in the brain, this was judged to be in the sulcus. So it is moderate in degree because we have two areas of superficial citerosis. By the way, if this had been in the brain, then we would have downgraded to a mild because we have if we would have called it one micro hemorrhage and one area of superficial citerosis. But because this was judged to be in the small sulcus here, we are in the moderate category. Now, here's a baseline here post dosing one in two months. Again, you see it getting worse over time, but let's try to work on what we would call the grading. Here's the flare. Here's the GRE. We see a two foci here. Some more foci here and even more foci here. So this is a severe aria E at this day here and then resolved aria E for the aria H where it mild for this one, severe for this one and severe for this one. Now let's talk about aria E mimics and pitfalls. This is very important to note in this patient, this was actually a suboptimal protocol where we're really sort of getting some artifact here in the posterior occipital lobes. When this was, they actually weren't sure about this, had the patient repeat with optimal protocol parameters and this went away. So that was just artifact. Now here this patient was scanned initially a baseline and this vendor, they came back and were scanned on a different vendor and now we see this. And the question is, is that development of aria E? So what they actually did was put the patient back on this vendor scanner and it looked just like this. So this was just artifact and it was not development of aria E. Here is a patient, here's the baseline. And then you see here on the follow-up study, you see all this sort of so-called flare hyperintensity, but it turned out the patient was on supplemental oxygen. They then took the patient off supplemental oxygen, repeated the study and it looked exactly like the prior exam. So that was a fake out, it wasn't really an aria E case. Now this is a patient here, this is the baseline study and then you see this very abnormal looking study, but they felt that the parameters might not have been optimal. There was poor CSF suppression, they actually repeated it with optimal parameters and it all went away. So this actually turned out to be a fake out, but that would be very hard to know that, but that's what it turned out to be in this case. Another thing to be aware of is metal artifacts. So this is the initial exam and this exam, the patient had their hearing aid in, this is the aria E and this is the flare, on the flare you can see all this abnormal signal intensity, but when you look at the aria E, you can clearly see its artifact from the hearing aid. And then this was a patient that this was their post dosing, they did develop a small area of aria E, so mild aria E. And the question is was this aria E as well, which is why it's always mandatory that we do DWI when we're doing therapeutic imaging because this turned out to be an acute infarct. Then the dosing was originally suspended, but the aria E actually just went away and that little focus stayed because that's just temporal evolution of the infarct. Now, when we talk about aria H, these are some pitfalls that you can see. This was the initial MRI, you see the small micro hemorrhage. What happened to it? Did it go away? It turns out this is just differences in slice sampling and a different scanner, but that little focus is still there. But you can see how in reverse order, you may think that a micro hemorrhage developed when it didn't. The other thing to be aware of is vessels. And so you have to track up and down and make sure you're not looking at a vessel. SWI imaging is much higher sensitivity for both vessels and bleeds, so it's a little bit harder like you're finding a wearer's Waldo in a sea of vessels, which is why SWI was not utilized for the trial and also why it is not recommended for therapeutic screening. That's why we use GRE and also to remain a consistent protocol. But this just turned out to be a vessel here. Now, when basal ganglic mineralization is symmetric and chunky like this, it's very easy to see. But when it's little punk-tape foci, it might be harder to tell. I would just assume that's mineralization, but again, it's something just to be aware of. And then also around the sinuses, you can get dropout susceptibility artifact from the air, and so you don't want to overcall those as bleeds. And particularly with sometimes around the sinuses, you know, mastoid air cells here, you can get little punk-tape foci, but those aren't bleeds there. And then the other thing to be aware of is this particular patient. You see this, what appears as susceptibility artifact bilaterally in the occipital lobes, but this was actually phase encoding artifact. So you just look for the march across the line in the phase encoding direction to differentiate true hemorrhage from just artifact. Now when to suspend dose? So the criteria is different for ARIA-E and ARIA-H, but starting here with ARIA-E, the main thing to remember for ARIA-E is moderate and severe. Because whenever the degree of ARIA is moderate or severe, the dose must be suspended. And also when the clinical symptoms are moderate and severe, the dose must be suspended. So anything in the moderate and severe category, whether ARIA-E or symptoms, you're going to suspend the dose. If they just have mild symptoms, you don't need to suspend the dose. So this is very important for a neuro radiologist and any radiologist that are reading this study to remember because it's up to you to convey back to the referring neurologist what the degree of severity is and whether the dose must be suspended. So please do not forget to call the neurologist if you see an ARIA category that would require a suspension of dose. Now this is ARIA-H here. If you're in the moderate or severe ARIA-H category, you're going to suspend the dose. But unlike with ARIA-E, any patients with ARIA-H that are symptomatic, you're going to suspend the dose. So all across here, if there's symptoms and you see any ARIA-H, again, this is over baseline. So let's say if they had on their baseline MRI five micro hemorrhages after their next imaging study, once they're getting a dose, if you see five, you're good. But if you see anything a therapeutic, you know, micro hemorrhages that has developed, that's when you start your ARIA grading. Now, what is the impact for imaging facilities? The primary impact is going to be increase MRI scan volume. So an MRI is required by label at baseline and it should be recent. The old label said within one year, but now the new label has been changed to say recent. I don't consider recent a one year, obviously. So probably within a couple of months, and some institutions are going to sort of require it, but others just if you've got it within a couple of months, then you're good to go. Then an MRI is required prior to the fifth dose, prior to the seventh dose, and prior to the 14th dose, these are all at the high dose categories, which is 10 megs per KIG, administered by IV infusion every two weeks. So MRIs must be acquired prior to these. These are scheduled MRIs. The AUR guidelines also recommend an MRI prior to the 26th dose, that's at 52 weeks. So that's not on the Lacanimab label, but it is a recommended schedule. So this is five MRIs here, and then plus additional MRIs, non-scheduled MRIs whenever the patient develops a new neurologic symptoms. So if you assume maybe the patient might have five headaches a year, the neurologist is probably going to order an MRI each time and probably on an urgent basis. So if you assume about 10 scans per year, roughly, and let's say there's 1,500 treatment candidates in the United States, we're looking at 15,000 new MRIs a year impacting imaging facilities in the US alone throughout our country. So this is a very big number. Imaging facilities really need to be prepared for the increased number of MRIs that they're going to see. Now, another impact for imaging facilities would be consistency, and that is consistency in protocol, field strength, and ideally with vendor. The protocol is very important. So the required sequences as recommended by the ASNR, Alzheimer's ARIA study group is a GRE, a FLARE, and a DWI. I can tell you at our institution we're just going to do a routine brain. Now, some people may be doing an SWI as part of their routine brain and some neurologists may always require an SWI again because of the increased sensitivity. If that's the case, please do a GRE in addition to that because when you're doing therapeutic screening, you need to do the count off the GRE. So in your dictation, you would report the number of micro-hemorrhages you see on both the GRE and the SWI, but just put a note in there that you're using the GRE for the ARIA grading. And then field strength, ideally really should be done on a 3T or 1.5T and preferably maintain the same field strength between each visit. And then ideally if you can stick with the same vendor, that would be wonderful, but I know that that's not possible in a lot of cases at our imaging enterprise. We're the largest outpatient freestanding imaging enterprise in the United States. We have 355 freestanding imaging centers that we read from. So we have a big diverse fleet of scanners, so we can't always maintain the same vendor. Now the other impact for imaging facilities would be an educational initiative for neuroradiologists for training for ARIA. We're going to do ARIA training here in this webinar, but you know, all's imaging is a good website for additional information. And educational initiatives are also ongoing through the ASNR and through RSNA and through other educational webinars and things like that. Now, another impact is quantitative volumetric imaging. We're already, at least in my area, my clinical practice, my referrers order this for almost every dementia patient, but in other areas it's not ordered as frequently, but this is a very useful AI tool. It can allow you to track the hippocampal volumes and it can help with automated ARIA screening report and is also was used in some of the trials. The other implication for imaging facilities would be the beta amyloid conformation, which is required prior to treatment. This can be done through LP, which is covered right now, but again, amyloid PET would be a preferable way for patients to be screened for a beta amyloid conformation, and we're hoping to see that that will be covered in the near future. And then PrecivitiAD and also Quest has a new blood test as well. These are both blood tests. At this point, there's still some variability in the reliability there, and also no coverage as of yet for blood testing. Now, let's talk about ARIA reporting. So ARIA screening and follow-up can be reported manually. The neuroradiologists are again very important because it impacts the therapeutic decisions. It's in terms of the interest level for ARIA and Alzheimer's, it's very interesting because the ASNR American Society of Neuroradiology has an all ARIA study group. There's about 1,000 neuroradiologists that come a year. There's actually 2,700 that are members of the ASNR, but about 1,000 make it to the meeting every year. And they thought that there might be a small number who would sign up for the study group. Well, it turned out 800 neuroradiologists signed up for the study group. And then we sent out a poll to all 2,700 neuroradiologists that are members of ASNR, and of the responding polled neuroradiologists, it turned out that 63% of them were interested in using an automated proprietary computer audit detection, screening software, AI software for ARIA detection, longitudinal follow-up, and automated reporting. So we will talk about that and look at that. Now, automated ARIA reports are in active development right now from some of the major vendors and pending FDA approval. So this is kind of what it looks like, this automated segmentation for quantitative MRI. This is the white matter overlay, which would you use for ARIA e-screening on the flare sequence. And the other input sequence would be GRE, where you'll see automated detection of the foci of micro hemorrhage. So this is the GRE beforehand. You see these little foci of micro hemorrhage, and they can be very small. And I'm not sure if you can appreciate this, but there's actually two right here. And then there's a few more along here, one here and one here. And this is what it looks like after the automated detection is applied. Here's an ARIA e-case, automated detection through quantitative MRI. Again, ARIA e and automated detection. Here's an ARIA e with so-called fusion case that is color-coded after it goes through the QMI software. This was originally. So again, it's very useful for detection of ARIA e and ARIA h. There are different screening reports that can be generated. This happens to be one from one company where it will, for ARIA e, it will give you the lesion burden and the count, what's new, what's enlarging, what's shrinking over time for ARIA e and for ARIA h. Again, the count, for ARIA h, the count is mandatory. So it's very important here. We'll do the same thing for superficial cirrhosis and compare dynamically since the prior study. So here's sort of a summary report here for ARIA e. Lesion one, it's giving the diameter and size because this is what's required for the grading system. So you get the diameter for all four of these lesions and you get the change from the prior study. It tells you where it is, right frontal lobe. And then it gives you a severity score for ARIA e. And this is based on this criteria that we talked about before for mild, moderate and severe. Here's in our sample ARIA h report that is in development. And again, counting the current micro hemorrhages as well as the baseline and you can get the change in the number. So baseline had one and then there was nine new ones at the follow up imaging. We'll tell you where they are in the brain and again, we'll give you a severity grading for both superficial cirrhosis and for micro hemorrhages. So having to be moderate for micro hemorrhages and mild for superficial cirrhosis according to this radiologic severity chart. This is another ARIA screening report from a different company. So let's talk about sort of the future of dementia imaging. ASI is actively working on a subcutaneous auto injector for the Lacanumab type molecule to bypass the need for bi-monthly infusions. So trials were in progress with that and they aim to file for FDA approval in April 2024. And actually another great benefit of that is not only the ease of use to be able to do it from home but it will also cut down on the infusion reactions. Over 10% of patients will develop an infusion reaction which sort of presents like flu-like symptoms, fever, flu-like symptoms following infusion. That tends to happen more at the beginning of treatment. So that will also help mitigate against that. In the future, we also like reimbursement for blood tests like the Quest AD detect test. That's $400 out of pocket and the Preciviti AD which is $1250 out of pocket. And then we really need CMS coverage for that amyloid PET for beta amyloid conformation to bypass the need for a lumbar puncture. Patients do not like having to get a lumbar puncture. So currently right now one amyloid PET is covered per lifetime but in a trial setting only. But again, as we spoke about earlier on July 18th, 2023, CMS proposed to rescind the national coverage determination that is currently restricting amyloid PET coverage. So that then permits the individual Medicare administrative contractors, these MACs, to make coverage determinations. So right now it's unclear when CMS will start reliably reimbursing for amyloid PET nationally and at what rate. But we're hoping that this is coming down the pipeline because CMS did make movement on this with this decision. Other things in the future will be new drugs on the horizon. We're expecting to nanomab to apply for conventional approval. Hopefully this fall, but if not at least by this winter. And then we still need a cure. There's no cure, but prevention trials, as I mentioned before, are in progress. Okay. So let's look at some cases here, just dementia cases. This is, there's several cases that we can review, but I'm going to limit it here just primarily to Alzheimer's and a couple of the other major dementia disorders. This is 79 year old with memory loss. They came in in 2013 here and you see some mild to moderate right temporal lobe atrophy. They had quantitative analysis done in 2013. When they returned three years later, their atrophy had significantly progressed and repeat quant was done. So here's a look at this font study in 2013. There was statistically significant hippocampal atrophy in the, I'm sorry, hippocampal atrophy and enlargement of the inferior lateral vendors. That all progressed at the follow-up study. Now we have all these areas of statistical significance and the HOC, the hippocampal occupancy score, which is a measure of meso-temporal atrophy, was statistically significant. On the normative reference charts, you can see here how the hippocampal volumes are wrapping rapidly off the normative curve, as well as the entorhinal cortical volumes and other measures here in the temporal, parietal, frontal and occipital cortex. This is a triage brain atrophy report, which is again broken down into substructures and lobes of the brain. Anything in blue here is getting smaller. So the cortex is getting smaller. Anything in red is getting bigger. So the ventricles in the sulci are getting bigger over time. This is the type of thing that you would see sent to your packs when you order quantitative MRI imaging. It's a very useful visual indicator here. This is another company's report here, where again, it's measuring the volumes of the pertinent substructures over time and comparing over time. So you can see things that have progressed in 2016. This is also from their report, where it's pointing to areas of statistical significance. You can see in 2013 here the hippocampi on segmentation and significantly smaller in size three years later. This patient also had on GRE multiple little micro hemorrhages. So they had cerebral amyloid angiopathy. This was not a treatment patient. And then they went on to get an amyloid PET, which was positive diffuse binding throughout the cortex. So this patient had both Alzheimer's disease and cerebral amyloid angiopathy. This is an 86 year old with memory loss and mild right temporal lobotrophy. There was statistically significant mesotemporal hypometabolism on, this is the PET CT images, and this is the PET MRI, which I fused to the flare sequence here. And I run all these through the memneral analysis software. So again, it was statistically significant. Look at the dramatic progression and atrophy by 2016, at which time quantitative post-processing was done and multiple areas of statistical significance on this quant study. And same with here on the other, this is another company's quant study here. And so this patient went on to have an amyloid PET, diffuse binding throughout the cortex, a positive amyloid PET. So this patient has Alzheimer's disease. Case three is a 73 year old with memory loss. And we see here moderate right temporal lobotrophy. We see some reduction in the volumes on quant imaging in 2016. So here's the reference charts here on one company. Here's another company triage brain atrophy. This patient went on to have amyloid PET in 2015. This is the amyloid PET CT fusion and pet MR fusion, which I fused the T2 weighted images. And this was positive. So this patient has Alzheimer's disease. This patient is a 70 year old here that had some atrophy. But not a lot. And if you look here on the flare images, we see this cortical infarct here, another small cortical infarct here. This one's on the primary motor cortex, a few little lacuna infarcts in the cerebral white matter. This was their DWI in 2020. They came back in December and now we have an acute infarct in the right cerebellar hemisphere on the GRE. Again, this is a non-treatment patient. We see superficial cirrhosis and other little micro hemorrhages scattered around. On the FDG PET CT in 2020, we see cortical hypometabolism diffusely in the parietal lobes and the temporal lobes, and particularly in areas where we have infarcts here in the brain. This is the FDG PET CT cortical hypometabolism that was statistically significant in the temporal lobes and in the biparietal lobes. Here's the FDG PET CT again with color fusion. We have all the temporal lobes or hypometabolic as are the parietal lobes and the posterior cingulate gyrus. This is the FDG PET MR fusion, which I fused to the flare and again statistical significance temporal and parietal. That was a patient obviously that had Alzheimer's disease. Moving on to this next patient, this is a 73-year-old male with profound visual hallucinations, delusions, gait difficulties, resting tremor, frequent falls, and only minimal memory loss. On the MRI here, we see atrophy, moderate atrophy in the bilateral parietal lobes, as well as atrophy in the bilateral occipital lobes. We will look here on quantitative imaging. We have some atrophy here in the occipital cortex, as well as within the hippocampi. And amyloid PET was done. Here's the PET CT, amyloid PET MR. And this was positive, diffuse finding of the tracer throughout the cortex. So the patient went on to have a tau pet in 2019, and we do see some tau deposition here in the occipital lobe on the right, a little bit in the posterior cingulate gyrus and a little in the temporal lobe. The patient then came back for an MRI in 2021, and we have some progression of atrophy in the parietal and occipital lobes. Here's an FDG brain PET CT that was done in 2021, and we see statistically significant cortical hypometabolism in the bilateral parietal lobes, as well as within the bilateral occipital lobes and in the temporal lobes. And this is the FDG brain PET MR. Again, where you see statistical significance in the occipital parietal posterior cingulate gyrus. And here's the fusion with the MRI as well. This is the FDG PET surface map in 2021, where again we see parietal and occipital statistically significant cortical hypometabolism. So this actually was a patient with dementia with Lewy bodies. We had abnormal deposition of alpha-synuclein, where these Lewy bodies here in the cells. Now, you notice that this patient had a positive amyloid PET, which typically you would think of Alzheimer's disease. But as you probably remember, when we went through one of the earlier slides, you can get beta amyloid deposition with DLB. And in this case, it was enough deposition to turn the amyloid PET positive. It's much less common than Alzheimer's disease, 1 million in the US versus 6.7 million with Alzheimer's. Survival rate is typically five to eight years after diagnosis. The important thing is 80% of them will present with visual hallucinations. So if you get a history of this, be thinking DLB. And so that's the presenting symptom in most cases. And then again, you can have that amyloid deposition and neurofamilary tangles with DLB, but not as much as with Alzheimer's. This last case is a 74-year-old woman with personality change and memory loss. This is the patient's FDG brain PET CT in 2018. And we see here statistically significant cortical hypometabolism in the bilateral temporal and bilateral frontal lobes. This is what it looks like when I run it through the MIM neural analysis software. Anything in red here is above the midline and the blue is below, I'm sorry, below the mean. This is statistically significant if it's more than two standard deviations below the mean in the left and right hemisphere by Z-score analysis. So we're statistically significant here in the temporal lobes and in the frontal lobes. And then this was a negative amyloid PET on this patient in 2017. This is what the CT looks like. You see the dramatic progression and atrophy in just one year. I mean, look at these temporal sulci compared to what they were a year ago. Look at the size of the temporal horn. That's really has progressed over time. And then in 2017 and 2018, look at the frontal lobe sulci. Here's the superior frontal sulcus compared to what it was just one year earlier. So this was a frontal temporal dementia case. There are three main subtypes, the behavioral variant and non-fluent and semantic. This is the most common, this behavioral variant here, which manifests as personality and behavioral changes, which is what this patient had. 30% are hereditary. Most notably the behavioral variant, which is an autosomal dominant type. And there are different genetic mutations that can occur with this. This tends to present in younger people between 45 and 60 years of age. It's much less common than Alzheimer's disease, 55K with frontal temporal dementia in the US as opposed to the 6.7 million with Alzheimer's. But interestingly, it is the most common dementia in young people less than the age of 60. The survival rate is typically 68 years after diagnosis. They should have a negative amyloid PET. You should not get amyloid deposition with FTD. And other diseases that are part of the FTD spectrum can involve motor changes such as ALS and critical basilar degeneration and progressive super nuclear palsy. So in summary, the dynamic landscape of Alzheimer's disease will have a profound effect on both patients and on neurologists, as well as neuroradiologists and imaging enterprise. So we all really need to be geared up for this. Neuroradiologists are going to play a key role in clinical decision making about the eligibility and continuance of disease modifying therapy. So it's important that neuroradiologists are trained and how to read these ARIA cases. I can tell you that at our imaging enterprise we are doing internal, I'll do an internal webinar which all of our neuroradiologists will participate in the training and they'll actually get a certificate if they complete both ASNR and RSNA are also actively thinking about issuing certificates after training programs as well. Now imaging centers will need to have agility to adapt to the significantly increased scan volumes. And finally just remembering that this is a really exciting and pivotal time in the history of neuroradiology. I feel like we're up for the challenge and we're willing to play a key role in the struggle against this absolutely devastating disease. Thank you so much for joining us. Thanks so much for your lecture, Dr. Bash. We can take some questions now if folks want to put those in the Q&A box. We've got a few already in here. Dr. Bash, how do we bill for QMRI? Okay, so for quantitative MRI, traditionally people have billed under the old 3D code but it really was not meant specifically for quantitative post-processing in the brain. And so a really exciting event happened on July 1st, 2023, just recently here where the AMA granted specific CPT-3 category codes for quantitative imaging of the brain. So that is wonderful. Those will go into effect on January 1st, 2024. So we strongly encourage everyone to transition from billing under the old 3D code to the dedicated QMRI codes. Now these CPT-3 codes are tracking codes, so there's going to be inconsistency in payment initially. But if everyone starts billing under these new codes, that will really drive conversion to a category one code, which then will become reimbursable. So it's very important for people to be aware that the dedicated codes are out there and to bill under those starting January 1st. What are the biggest practical challenges awaiting us in therapeutic dementia imaging and how do we overcome them? Yeah, so I think one of the biggest challenges is going to be communication between the neuroradiologists and their referring neurologists. Because if the neuroradiologist doesn't know that the patient is on treatment, then they know to screen for ARIA, right? So they see blood products. They may think progression of cerebral amyloid angiopathy. They see edema. They don't know what is causing the edema. So one strategy that you could use that I suggest would be to have the referring neurologists have that conversation and maybe have them order their studies as AD therapeutic baseline. If it's for the patients on treatment, that will alert the neuroradiologists and scheduling that the patient is on treatment and they're doing their baseline MRI. And then AD therapeutic monitoring for when they come back to do their scheduled MRIs or if it's an unscheduled after they are already on treatment. And then if it's a patient that has dementia, but they're either not in treatment category, they're not eligible for treatment. Maybe they're more advanced Alzheimer's. You could just have them order it as a routine non-treatment dementia study. But if you have consistent language, every time when you order the study, that's going to trigger the correct dementia protocol. So that communication is very important. And then also the backwards communication from the neuroradiologists to the neurologists if dosing must be suspended, if they category of ARIA warrants such. And then the other thing is really planning. Consistent protocols should be in place already. So at our imaging enterprise, we'd like to do a dementia protocol on all of these patients. So again, the mandatory sequence, we're just going to do a routine brain, but we'd like to add the 3D T1 instead of the traditional T1 so that the patient is eligible if we want to do quantitative MRI. You need the 3D T1 to do quantitative MRI. So every protocol should include a GRE, a FLARE, DWI in a 3D T1 at a minimum. And then to accommodate for the be aware of and plan for the increase in MRI volumes. So one thing that you can do to mitigate the high volumes is things like deep learning for image reconstruction. This AI tool that will allow you to scan your patients 50 to 75% faster, but still actually have in fact better image quality than if you hadn't applied this AI solution. That's something we use throughout our imaging enterprise. And it's an extremely useful tool which really is moving into standard of care at this point in time. The other thing is to plan for staffing to cover increased volumes. And then the other component is training neuroradiologists. If neuroradiologists don't know what RA is, they don't know how to read RA cases, they cannot play that key role that is expected of them. So training is very important. So again, in our imaging enterprise, we're going to do an internal webinar that all of our neuroradiologists on the East and West Coast will attend and they'll get a certificate. But ASNR is going to also offer training and RSNA. And so training is very key and a big educational initiative is needed there. Great. Okay. What about the controversy surrounding or identifying amyloid as causing Alzheimer's disease? Well, I think everyone's pretty much in agreement that amyloid and tau are both biomarkers for Alzheimer's disease. There's always a bit of a question of the balance between causality and association. But we know that causality is at least at some part a component because ApoE4 homozygote carriers have a 12 times increased risk of developing Alzheimer's disease. And they also have an increased risk of developing hemorrhages. And we know that ApoE4 is tied to impaired clearance of the beta amyloid plaque from the brain. So that is a sort of a causality component. And we also know that from the disease modifying trials that when you effectively clear the beta amyloid plaque, which all of the monoclonal antibodies do, that there is a slowing in cognitive decline. So that is another argument for that. So I would say, you know, I wouldn't worry about it so much. Just worry about what the end result is, you know, is the patient getting better, you know, over time. I'm not getting better, but at least slowing the rate of progression. All right. One more question. And this one's a little bit more specific. We utilize SWI in our routine MRI brain protocol. Just want to confirm that counts for ARIA H must be done on a GRE. Yes. So the ARIA grading scale was based on GRE, not on SWI. The trials purposely only use GRE because pretty much any scanner can do a GRE. Not all scanners have the capability to do SWI. Certainly the newer ones do. And then, you know, also sometimes they, you know, it's a longer sequence. And so they wanted the trials want to incorporate something that everyone could do consistently, which is GRE. Now, if the grading was done on the SWI, it would be probably a different grading scale because SWI is more sensitive for bleeds. So again, it's not a problem if you're doing SWI. It just, if you can also then add the GRE for the grading of ARIA so that everyone is consistent across the board, just because there is a higher sensitivity for blood products with SWI. And then the other component also is, you know, sometimes SWI is harder to interpret. Not all neuroradiologists love it just because sometimes it's hard to tell if something is an end on vessel versus a micro hemorrhage. Whereas on GRE, it's much, you know, you feel much more confident that it's always a micro hemorrhage and not a, you know, a vessel. So those are kind of the reasons behind that. Awesome. Dr. Bash, thank you so much for your lecture today and for everybody for participating in the question and answer session. You can access a recording of today's conference and all our previous new conferences by creating a free MRI online account. Be sure to join us next week on Thursday, August 17th for a lecture with Dr. Petra Lewis entitled Getting Promoted as Clinician Educator. You can register for this free lecture at MRIline.com, follow us on social media for updates on future new conferences. Thanks again, Dr. Bash, and have a great day. Thanks.