Okay. Hello, everybody. My name is John Hyman. I'm a neuroradiologist at the University of Texas Medical Branch. I've been, I've had the pleasure and opportunity to lecture for the AANS APP course since 2015, 2016, I think was the last year I did it, and it's always a pleasure. Let's jump right in. This lecture is about, it's tailored to an audience that has some experience with looking at their own imaging, as I know many of you do, and it's kind of tips and tricks from a radiologist to understand some of the nuances that go around oncological imaging at a slightly higher level, okay? So let's jump right in. We're going to talk about some of the technical aspects of imaging that are important to understand, then get into diagnosing malignancies, kind of first diagnoses, and some of the additional things you want to look at in a study besides, so it's a GBM, et cetera, and then a couple of issues related to treatment and follow-up, and then if I do the timing well, we'll have a little bit of time for questions at the end. So let's start with technical aspects of imaging. The first thing you always want to pay attention to is study orientation. So what I've shown here is an MRI and a CT with a scout line for reference to where that axial plane is on the sagittal, and what you can see is that the scout line changes, and as it changes, the relative position of a tumor or a metastasis or a hematoma even will change depending on how the patient's head is. So between MRI and CT, there's almost always a difference, but even between CTs taken one day to the next or even hour to hour, it may change if the patient's head is down or the patient's head is back. If it's not like a completely straight shot, you really need to pay attention to that and be aware that this is something I tell my trainees all the time. You get what I call the case of the vanishing and reappearing metastasis. Every now and then, I'll hear people say things like, well, the meth that was previously described in the postcentral gyrus is gone, but now there's one in the precentral gyrus, so mixed treatment response or something like that, and that's why you need to be aware of that, and it's also important to understand conserved landmarks on imaging so you know where you are relative to the brain itself and not to the apparent location of something just based on an axial slice. This is also important, of course, for measuring tumors. If you take just an AP diameter on a certain slice and you take the greatest AP diameter on another study, it may look like a tumor or a hematoma or something has enlarged or gotten smaller when it's really just the way you're measuring. If you can imagine a zucchini or something like that, a cucumber, if you measure it horizontally versus at a more vertical, you know, that length is going to change, so I recommend always measuring on either a sagittal or coronal. It depends, but on an area where it's not going to be affected by potential tilt in the axial plane, and don't be afraid to re-measure on a prior to make sure you're comparing apples to apples. Not all that glitters is gold, okay? This is something that, not I would say a pet peeve of mine, but something that, you know, when I ask non-radiologists and even early trainees and radiologists, what do you see here? And they'll, a lot of times, they'll be like, well, there's enhancement here. I say, well, let's hold on for a second, because post-contrast sequences are always T1-weighted sequences. Contrast is T1 bright. That's why we give contrast, and that's basically what enhancement is. But on a post-contrast sequence, just because it's bright doesn't mean that it's necessarily enhancing. So let's look at the example here. We have a post-contrast image. This is a post-treatment study. You can see the pretreatment effect, I mean, the pretreatment scan on the far right there. Somebody might look at this and say, well, you know, there's still, you know, the cystic part is decreased, but there's still persistent enhancement in this lesion, so the lesion is still here. Well, when we look at the pre-contrast sequence, you see that there's really not superimposed enhancement there. What that bright spot that you're seeing on the post-contrast is actually just intrinsic T1 bright signal from blood products, which, of course, subacute blood products are also bright on a T1 pre. So I would say that's, you know, an effectively treated metastasis. It's gotten smaller. There's no superimposed enhancement, et cetera. So that's very important in looking at post-operative scans and determining residual disease, you know, because there's usually, especially if it was a resection, you know, blood products around the periphery of the resection margin. So you always have to look and see if that bright signal around the periphery actually is enhancement or if it's just blood products. Subacute infarcts can enhance. This is also important in general radiology, general neuroradiology, but also particularly in the assessment of post-treatment tumors because you often have a little bit of post-operative ischemia. It's typically venous and barks from, you know, from the clamping if you had a craniotomy. This was not a post-treatment scan, but it's also an example of the same thing. So this was a 42-year-old male with a known brain mass. This is the history I got with questionable stroke versus lymphoma. He had been, you know, neurosurgery had accepted him as a transfer and the prior outside hospital MRI, you know, had described an enhancing mass in the temporal lobe. This was the study we got. And, you know, if you look at this, there's enlargement of the atrium, volume loss. In other words, you know, this is exvacuodilatation. This is encephalomyelation surrounding gliosis. And again, there is a little bit of, this is a post. You can see the vessels are enhancing. This is the pre. And like I was saying, this looks like it's enhancing, but it's actually just bright on T1. So this is all probably cortical laminar necrosis. This is very clearly an infarct. You know, this is a resolving now subacute to chronic, I would say chronic infarct. So you don't need to worry about this. And probably on the prior study, you know, they were seeing it in the subacute phase and they were seeing it as enhancement that can look like a mass. So that gyroform enhancement in particular, always be aware that subacute infarcts enhance. The other thing, and this very much depends on where you work, what kind of quality you get. In tumor imaging, consistency is really important. And we don't always get it. Like at my institution, we have multiple different scanners across campuses, across hospitals. So we don't always get the consistency we would like, but we do recognize the difference in technique. So we don't get tripped up on how we interpret studies. So again, we're looking at post-contrast sequences here. On the left, you have spin echo sequences. On the right, you have gradient-based sequences. Spin echo sequences are black blood. So like look at the posterior aspect of the superior sagittal sinus here. This is a post-contrast sequence. You can see a little bit of enhancement in the choroid plexus, but the flowing blood is black. And on the gradient-based sequences, it's bright. Okay? That will, that can dramatically affect different areas. Like for me, the one that really frustrates me is post-operative imaging of pituitary adenomas, because the cavernous sinuses obviously have a lot of enhancement and trying to determine residual tumor versus vascular enhancement sometimes can be complicated. Traditionally, I think typically, most places will do spin echo for tumor evaluation for metastatic disease anyways, because theoretically they say, you know, these little enhancing vessels can be confused for metastases. I don't really think that's the case. But just be aware of that. Sometimes you have to get a gradient-based sequence. If you're doing something like volumetric software, like NeuroQuant or what have you, they need the better gray-white matter differentiation to do the segmentation. So you may be getting on some of your protocols, a GRE-based sequence as opposed to a spin echo. So my point is just to be aware of that, because the appearance of a scan, even though there may be no change in the tumor, like if you look here, can look dramatically different based on the technique with which it was acquired. So just be aware of that. This is also kind of a technical aspect. Fat saturation is helpful in post-contrast imaging. So this was an example of a clival meningioma involving Dorello's canal. The patient came in from an outside hospital. He had an MRI that had been read as normal, and he had a very overt cranial nerve 6 palsy. My resident actually picked this up without any problem. This is a fat-saturated T1 post, and you can see this retroclival. It's in plaque, so it's kind of thin. You can see how it might be difficult to miss right here. Easy to see. This is the outside hospital study, and you can see what their problem was. They didn't do a fat saturation, so the normal fat in the clivus, the clival marrow, is the same intensity as the enhancement in the tumor itself. So, you know, I don't blame them other than, you know, if he's got a 6 palsy, you should evaluate 6. This is certainly detectable on this study, but, you know, to a cursory glance, you can see how that got missed, and whereas here, you know, the enhancing tumors are quite visible. So, fat saturation is useful for post-contrast imaging of the brain. This one is, it's important for clinicians who are interpreting their own studies. This is something that I really harp on for my trainees because now there's often not a filter between the imaging report and the patient. In other words, patients now can directly access their report. It's not like the neurosurgeon necessarily or the, you know, advanced practice provider is going to first look at the report and then tell the patient what it is. You know, they can go into MyChart and see it directly. So, you have to be very careful about what you say. This was a case where we were evaluating for altered mental status in a patient with known malignancy, and I think the report said no intracranial metastatic disease, and I know it seems like a minor thing, but I would never say that. I would say, you know, there's no intracranial metastatic disease within the limitations of the exam, but CT is not the most sensitive for the evaluation of metastatic disease, and occult lesions, I mean, in small lesions could remain occult because, again, you don't want to get into a situation where you have a patient that, you know, just doesn't like interacting with medicine, and he or she was kind of forced to get this study by their spouse because, you know, they were like, you need to get evaluated. Now, they're like, nah, they said it was negative. I wash my hands of it like I'm done. This is that same patient only three days later, and you can see that he's actually riddled with metastatic disease. He just has numerous mets. They're everywhere, and they really are occult on the CT because they haven't gotten big enough to have vasogenic edema, so that's important. This is kind of an aside. It doesn't have to do with tumor imaging, but this is the same, I would say, be very careful with spine imaging. This was a young woman who came into the ED with symptoms of, didn't start out as cauda equina. It started out as low back pain, and then it progressed to cauda equina pretty rapidly, but you can see that CT lumbar spine, it's negative. I mean, it's essentially negative, but I never say the spinal canal is pain on a CT. I mean, you just can't really say that, and especially true in the cervical spine. You can see there's a huge disc extrusion here, and it's acute, and it's edematous, and because it's edematous, the density on the CT is pretty similar to surrounding fecal sac, right, and so you don't really see it, but this little thing here, that's all that's left of her fecal sac at that level, and of course, she ended up being okay. She was decompressed and, you know, didn't have any bad outcomes, but be careful of that, and as providers, be mindful that, you know, especially if radiologists aren't specialists or they're just sloppy, you know, they may go sometimes beyond what they can really say about a study, you know. Be careful of false negatives because sometimes people will go beyond, you know, and say something more definitive than they should really be able to say within that study. Okay, so just to recap that first section, pay attention to study orientation. Not everything bright in a post-contrast is actually enhancing. Make sure you cross-reference with the T1 pre-contrast to see if it's just intrinsically T1 bright. Remember, subacute infarcts can enhance very important around the margins of resected tumors in the acute phase, and it's also why it's really important to get immediate post-operative imaging, you know, within a day or two so you can see that infarct because if you wait like a week and then you have enhancement around the margin, you know, the DWI might have gone away and you may be just stuck with enhancement and you don't know if it's post-operative ischemia or if it's actually residual disease. So it's important to get that post-operative baseline quickly. Pay attention to techniques, spin echo versus gradient echo for your posts, fat saturation is helpful, and don't overstate impressions beyond study limitations when reporting, don't assume knowledge. Okay, so let's look at some first diagnoses, incidental malignancies, and pertinent positives. My first point, and this is more for radiologists but also for surgeons, if it quacks like a duck, you know, it's a duck, and what I mean by that is young people get cancers, right, and sometimes they're pretty obvious. This is a 35-year-old female. This is like a, you know, encyclopedia butterfly glioma, right, ugly, enhancing, going across the corpus callosum, lots of edema. Nobody's going to miss that, but sometimes they look like this. Here's a 29-year-old with a four-day history of migraines and malaise. He's got a predominantly cystic mass, very thin peripheral enhancement. This is a DWI to show it's not an abscess, which you might, you know, think it is. A little bit of irregularity here. This is also a GBM. This was also a biopsy-proven GBM. So, you know, have a high threshold of suspicion, and just remember that young people also get cancers. This was a case from my fellowship when I did not have the experience that I have now. It was a 38-year-old female, previously healthy, 30 weeks pregnant, one-month history of increasing confusion. They found her in the street, you know, taking off her clothes or something like that, and they brought her in, you know, in a vacuum. If you brought this patient in and was like, this is a six-year-old guy, he's been, you know, smoking cigarettes all his life, or, you know, yada, yada, yada. I mean, clearly, I'm going to put meds in the differential here, but at the time, I was, I just went down a different pathway. I was like, well, I don't know, maybe it's, you know, some kind of weird infection, like, I don't know, but it was metastatic disease. So, you'll always remember, like, it can be a tumor. This was a metastatic pulmonary adenocarcinoma. Separate thing to remember is that, unfortunately for women, you know, pregnancy is also an immunosuppressant, and your immune system does have a cancer screening function, so you're slightly more susceptible to getting cancers or to having cancers progress when you are pregnant, which is, I think, played a part for this young woman. Here's another one. Understand the difference between vasogenic and cytotoxic edema. So, cytotoxic edema is the type of edema we see in acute infarcts. The cell, It's cellular swelling and cell death because of the failure of the ATPase. And remember that the white matter are the axons. The cell bodies are in the gray matter, which is, you know, the cortex in the brain, centrally in the spinal canal. And so, when you have cytotoxic edema, when you're describing strokes, you know, your radiologist may say something like, there's blurring of the gray-white matter differentiation, etc., etc. Cytotoxic edema is just leaky, you know, leaky interstitium in the brain. And so, you get what you see here, which is white matter hypotenuation. And the problem is that with neural symptoms, a lot of times they're nonspecific, right? So, this is somebody, a 69-year-old male, neuro-deficit, stroke-suspected. I'm sure he has, you know, a neuro-deficit. And you have to be really careful to not look at this and be like, oh, yeah, there's hypotenuation. You know, this is probably an infarct. What you would look at here is say, well, I mean, the cortex is really preserved here. The hypotenuation is subcortical. This looks like basogenic edema, not cytotoxic edema. We really need an MRI to evaluate for potential underlying lesion. And here it is. You know, this is a first diagnosis of a metastasis here. Here's another kind of similar example to that. Does what I'm seeing make sense? Don't ignore the voice in your head. Again, this is, you know, kind of tailored to people who are dictating reports, but also important to people, you know, who are interpreting the reports like neurosurgeons. This was a, I will tell you, it's over 10 years now, but this was a kind of a terrible miss by neurosurgery at our institution. This was a 20-year-old female. She presented for a trauma workup and she had this here. And again, cortex is preserved. There's some hypotenuation here. The radiologist who read this said, you know, there's some subcortical hypotenuation left frontal lobe. This could be a contusion, but you know, it's a little bit unusual. There's no overlying soft tissue swelling. It needs to, she needs an MRI to evaluate that because there could be a lesion here. What neurosurgery did was they just repeated the imaging the next day. And she was discharged with a diagnosis of unchanged left frontal contusion. Okay. That's problematic for a number of reasons. And the radiologist read it the same way. It was like, hey, you need an MRI here. I never really seen white matter contusions. That's unusual. Like contusions are typically hemorrhagic, you know, and you're not going to get an isolated white matter edema in the absence of any kind of like soft tissue swelling or edema. Like that would, you'd be talking about DAI at that point from like significant shear injury. So this is a weird thing. There's no injury anywhere else, blah, blah, blah. Anyways, she leaves. I learned about this study later because I read this study nine years later. Here is that hypotenuation that she had, which has now become an anaplastic oligodendric leioma. So an unfortunate outcome for her and a reminder that, you know, you need to be, again, remember young people have tumors and kind of you want to make sure everything fits together. And this is a, now she is going to have transcortical. This isn't cytotoxic edema. This is transcortical because it's infiltrated. So another thing about leiomas. This one kind of goes, just makes sense. Be careful when you, when you're looking at somebody with a known cancer, you know, look, look at this, what we call the study corners, right? So this was like a CT head, but he actually had a metastasis to the choroid and this was testicular cancer. So you want to make sure you look at the bones and look at other things. Here's another example, staging exam for lung cancer. The parenchym was normal, but there's this enhancement in the left posterior parietal calvarium with intrinsic loss or with loss or infiltration of the normal T1 bright mirror signal. That's metastatic disease. And again, that's still probably a stage four disease, right? It's in the bone. So that's significant. DWI is useful for more than just stroke. We typically think of DWI as something that we diagnose strokes with, but for neuro-oncological imaging, DWI also is bright for hypercellular tumors, medulloblastomas, meningiomas, even sometimes GBMs, although that's not common, but you know, lymphoma, small blue cell tumors, DWI can help you with that. It's also useful for finding osseous metastases and nodal nets. One of the most commonly misdiagnosed by neuroradiologists are incidental head and neck cancers on the corner of studies. This was an example, again, of a metastatic node of Rouvier and the patient ended up, although you can't see it on the DWI, you could see it on the conventional cross-sectional like T1 post-sequences. He had a nasopharyngeal carcinoma that was an incidental diagnosis. So let's talk a minute about pertinent positives too. Even benign tumors can be quite problematic. And what I mean by that is, in addition to diagnosing the tumor itself, there's often affiliated things that you need to pay attention to. So this was a craniopharyngioma, you know, super predominantly, supertemporal, I mean, supercellular cystic mass with some calcifications. So craniopharyngioma, sure, but what you really need to pay attention to is the dilation of the temporal horns indicating the obstructive hydrocephalus. And that's what can be, you know, acutely life-threatening potentially. Also, what tumors are doing is something that you need to pay attention to. Like I tell my trainees, yeah, meningiomas are benign and they can be small, but they, I mean, meningiomas can be really problematic. Sometimes they're hard to go get out, they go away. Even small meningiomas, if they're in proximity to like a sagittal sinus, probably should be taken out because over time they're going to grow. And if they grow into the sinus, that's going to really complicate the surgery. Also, you know, for doing a surgery, obviously you want to pay attention to something like that. No going in there. You don't want to be surprised during the operation as you're removing this thing that it's, you know, invading the sinus. So if it's like a, you know, an 80-year-old lady, 90-year-old lady, and she's got a little meningioma here, you're probably not going to do anything about it. But even if it's like a 20-year-old and it's small, if it's close to a sagittal sinus, or, you know, if it's a sphenoid wing and it can get into the cavernous sinus, or just go somewhere where it's going to be difficult to get it out later, that's important to note. And then, of course, posterior fossa masses. Any posterior mass effect can be life-threatening. Like this is a cyst with a mural nodule, classic comangioblastoma, not a, you know, not an acute, I mean, not a malignancy. But remember, these patients are presenting with symptoms generally, and you don't know where they are in terms of like crossing their threshold, right? Like it might have been growing for a long time, but now even a slight increase in size can potentially cause a life-threatening herniation. And this patient was in the OR later that day, even though this is a benign tumor because of the symptoms they were having. Again, big posterior circulation infarcts with edema can also have this problem. Surgeries with, you know, hematomas, things like that. I mean, we're talking about this in the context of neuro-oncological imaging, but just have a very low threshold for mass effect in the posterior fossa for close, you know, very close clinical monitoring and potential for intervention. So just to recap that section, remember, if it's unsuspected but looks like cancer, it's still probably cancer. Young people obviously do get cancer. Vasogenic edema is highly suspicious for an underlying lesion. Be able to differentiate between vasogenic and cytotoxic edema. Does your impression make comprehensive sense and account for all findings? You know, if like something seems wrong, listen to that voice in your head. That's called experience. Be extra cautious in patient with nonmalignancy. You know, check all the corners. DWI is important for more than just strokes and be aware of pertinent positives. Again, this is more for radiologists, but know when to communicate directly. Like not only to say in that case, yes, this is a, you know, a benign tumor in the posterior fossa, but this is a benign tumor in the posterior fossa. And by the way, this patient's at risk of herniation, so they need to be evaluated, you know, urgently. Okay, so let's speak very briefly now to issues related to treatment and follow-up. The first one I think I would have to say is to understand pseudoprogression. Pseudoprogression, and this is something that we generally talk about in the treatment of GBM and high-grade myeloma, it simulates tumor progression. It's an increase in the solid enhancing portion of the lesion following treatment. Classically in high-grade gliomas receiving both chemotherapy with hemizolemide and radiation therapy, you see it in up to 30% of patients receiving both. Most cases by three months, but it can occur in up to six months. There's some information about, you know, whether it's more likely in wild-type IDF mutations, but I think that's kind of in the weeds and beyond the scope of this lecture. And then the flip side of that, you have a similar issue of pseudoresponse in patients being treated with Avast and Bevacizumab. It's a VEGF inhibitor that can lead to decreased flare, which looks like improvement, but it's really just because you're not seeing that edema, so you have to be careful in those interpretations. This is an example of a pretreatment GBM. This is obviously the same patient, post-chemo radiation. It looks worse. For me, the only thing I can say about pseudoprogression, like I said, is you should know about it. And it always has this like less solid, more feathery appearance is how I describe it. To me, this is like very clearly pseudoprogression. Now, again, does that mean that there's not residual disease there? No, but predominantly what you're seeing is pseudoprogression. I mean, I always tell people just remember, it's GBM. If they don't get hit by a bus or have a heart attack, they're going to die of this disease. Like there's not really a cure for GBM, so the tumor is there. It's just like, what are you predominantly seeing here? You also want to take into account clinical findings. You know, advanced imaging can be useful, but my experience really just short-term follow-up and seeing how it progresses is the best thing. My advanced imaging, I'm talking about MR perfusions, spectroscopy. I find those things to be useful only in hindsight and not particularly useful progressively, prospectively. I mean, pseudoprogression should be self-limited and then generally regressed. So other things, recognize the sequela of remote whole brain radiation and investigate, take that into consideration. This is again for interpretation. This woman, this patient, 27-year-old female, note she's young. She presented with what's kind of an aggressive looking lesion here. Well, she has some, you know, a couple of foci of susceptibility signal loss in the white matter. She also has, it's hard to appreciate this, but she has a craniotomy here. So to me, this says this lady had a prior resection for something. She has findings that are suggestive of post-radiation treatment, like long-term post-radiation treatment, multiple cavernous malformations, and that turned out to be the case. She ended up having a medulloblastoma resection when she was younger. She got a whole brain radiation. Now she's got a radiation-induced osteosarcoma and multiple cavernous. This one's important. Always look at more than the most recent prior, because tumors can grow slowly and, you know, you can end up seeing something that says no significant change from prior over and over and over. But if you go back and look at the original tumor, you know, maybe there's actually been some pretty significant change. So again, if you look at the numbers here, this was like 28.9 millimeters. This was 29 millimeters. That's nothing, right? So they just said no significant change. But if you go back to several years prior, it's actually increased by about a third, and you can see that. So if you don't jump from the most recent prior, you can miss that a tumor has indolent growth. And again, I mean, you know, generally we're talking about meningiomas here, but a meningioma is the most common intracranial tumor. So that's what you're going to deal with the most. And again, like I said earlier, this one is in a place where, you know, at a certain point, it's going to be difficult to resect this, right? If it gets into the cavernous sinus, if it goes into the orbital apex, you know, you lose your opportunity to have an easy resection. And it's important to know, I think, that these tumors are growing. This is our unfortunate lady with the anaplastic oligodendroglioma that got missed. Originally, we failed her again. Unfortunately, I didn't read the study for the record. But this is her, she had a resection, okay? This was the original resection in 2018 for her anaplastic oligo. You know, you see that this is immediate post-op, right? She's got fluid here and a resection cavity, etc. So she shows up a couple years later, she's pregnant, and she has a seizure. And they, because she's pregnant, they didn't give contrast. And the radiologist who read it was a little bit dismissive, and kind of, you know, pointed at this and said, well, this signal's the same. There's volume loss here. Yeah, there's edema here. I mean, there's, there's signal here, but it's probably, it's probably just gliosis. You know, we can't really say because there's no, because we didn't give enhancement, so it's limited. Well, there's a couple problems with that. One, it was an oligo, so it wasn't enhancing to begin with, right? So the lack of enhancement really shouldn't affect your diagnostic acumen in this case. Secondly, and this is the real point of this, be aware of transcortical hyper intense T2 slash flare signal. And this is the part up here that I don't really like. And I've showed it on the sagittals, because I think that makes the case a lot more. Here was the resection cavity before. Here was her cortex. Here's the resection cavity now. And you have this infiltrative, it's expansile. Yes, it's hypo intense on T1 and hyper intense on flare. To me, I would have not even suggested, I would have come down hard on this is disease recurrence. And also don't forget, it was a, it was a high grade at the time it was resected, glioma. You know, it was a grade three oligo. So, you know, it makes sense that it's still there, it came back. Anyways, unfortunately, she, you know, didn't have that treated, although, and so this was her several months later with disease progression. Now it is enhancing, although maybe it was enhancing at the time. Again, we don't know. It's gotten larger, etc. So don't rush to dismiss cases because of things like contrast or some patient motion. Often you can still make a definitive diagnosis. And most importantly, be aware of expansile T2 flare or T1 signal, particularly if it's transcortical. Flare is your friend, okay? And when you really gain a lot of experience, the best radiologists and people interpreting studies pay attention to subtle things, okay? This is pretreatment, this is post-treatment. The distribution of flare here is, it's basically the same. However, the degree of the flare in that distribution, the mass effect are subtly improved from pretreatment to post-treatment, right? Like if you look at the flare here, it's going all the way into the subcortical white matters. Now it's backed off from that a little bit. Here it's expanding this sulcus again. You know, it's really all the way up to the cortex and here it's not. So it is improving. And the reverse can be true. Like let's say this was free and this was post, I would say like, hey, this is worsening. This disease is worsening. So that's subtle, but it's there. And you need to pay attention to that level of nuance if you really want to make accurate reports. All right. Where's the met in this study? And again, this gets into my argument about patient motion. You know, it's hard to say. I would say, you know, you could say like, well, there's no obvious metastatic disease within the limitations of patient motion. There's obviously some flare over here. Ignore that. This is what I'm looking at. I say, you know, there's some signal here. What does that correspond to? And when you look, there is a peripherally enhancing met that once you see it, really hard to unsee. Like, oh yeah, that's there for sure. So flare is your friend and don't forget to cross check with the post to look. Here's another example. Again, some motion. T1 post. You know, is that a, is that a met? And again, obviously there's a lesion over here. Ignore that. Is that a met? Well, there's some flare there. So it probably is real. And the more you look, oh, you know what, there's a lesion there in DWI. Oh, you know what? And this gets into my argument about things moving. This is the same spot as this, although on the CT, it really, it looks like it's more anterior, but that's the same, that's the same lesion. So it's, you've got hemorrhagic, additional hemorrhagic metastasis there. Okay. And this is just, don't over rely on advanced imaging. I, I honestly don't put a lot of stock in it myself. It's been my overwhelming experience that it's not helped. FMRI, useful. Spectroscopy and MR perfusion, they're a toss up. You know, they can give some corroborating evidence, but I would never make a primary diagnosis based on what those said and override what conventional sequences are saying. Findings are often nonspecific. The interpretation is incredibly subjective. Like on spectroscopy, you just get to pick what you want to look at. Like, are you going to look at the single voxel? Are you going to look at the multi-voxel? Which voxel are you going to look at? There's obviously, there's, there's usually conflicting information and people will pick one that supports, you know, like what they think is going on anyways. And then there's a lot of underlying limitations. And if you don't understand, you know, how those studies are acquired, you know, you can, you can make some, some blunders in their interpretation. This is again, just like a classic case. Side with common sense over advanced imaging findings. This was a patient with known tumor effective MS presenting with worsening right-sided weakness. This, I mean, yeah, it could be a tumor, but he's got tumor effective MS. He's had lesions like this before. Like this is an, this is just a prior. So neurology asked me if they should do spectroscopy. I said explicitly, don't get spectroscopy. It's going to be confusing and it's not going to help. Just treat him and follow it up. So let me show you the spectroscopy. Here it is. Reversal of normal Hunter's angle with a metabolite ratio suggesting a high-grade tumor. And yet here's the conventional imaging showing that all these lesions improved after, you know, one month treatment with a biological agent. And of course, you know, like we, they didn't think it was a tumor. They went with this, but this is my point. Like, why did you even get the spectroscopy? He just told you the exact wrong thing. And you never know, like this case, he happened to have a known diagnosis. So it was easy, but if you're going to put a lot of weight on this, you know, it can lead to a lot of patient anxiety, the way you convey that, what you do for follow-up and stuff like that. So I, I don't think advanced imaging is super useful. So let's just recap real quick. Be aware of pseudo-progression, pseudo-response. Look at the most recent prior, you know, at more than the most recent prior to rule out indolent tumor growth. Give the flare it's due. It's so important. Expansile transcortical T2 flare for non-enhancing tumor. Low-grade gliomas, right? Pay close attention to distribution degree of vasogenic edema, white matter flare signal. Always cross-check different sequences to corroborate findings and understand, but don't overly rely on advanced imaging. That's my presentation. And I think we have a little bit of time left for questions. So I will open the floor now if anybody has any questions for me. Thank you so much, Dr. Hyman for such an excellent radiology review. I know as APPs, we are trained to look at imaging ourselves, but our attendings where I work always encourage us to review the neuroradiology reports for misses, such as what you suggested with that trauma patient, that young trauma patient, you know, had they, you know, probably just compared the CT, said it didn't change and not, you know, gotten MRI, but it had, they followed maybe the directives of, you know, consider MRI study, you know, that could have been hopefully, you know, not a miss. The other thing is too, sometimes on imaging, you know, neuroradiologists are so great about mentioning adjacent structures such as, you know, kidney cysts or, you know, possible kidney disease, you know, tumors. So we all need to pay attention to your reports for things like that. One question I had was about radiation necrosis, you know, that's such a beast in treating some of the tumor patients and, you know, recurrent, you know, can't rule out, you know, tumor progression versus radiation necrosis, you know, and you get all this edema. And sometimes you go back in and just resect the whole thing. But can you talk some about the radiation necrosis issues? It's challenging. And it's one of those things that I feel like is like the golden grail, the holy grail, sorry, for like AI, right? Like people talk about AI and computers can do this better than we can. And I think that that's something because there's just that hopefully they come up with some way that the AI can interpret things that we're not seeing, patterns that we're not seeing, and lend some weight on that because where we are now, it's very difficult. And it's one of the things that conventional imaging is conventional imaging. I mean, you just look at it. Like I said, like for me, it's just like this, the enhancement look feathery. Did adjacent edema also worsen or improve? Like when was the treatment done? Like how far out are you? IDF wild type, you know, like these issues clinically, how's the patient doing? Those are the kind of things that you end up saying, you know, this is likely sort of progression, but recommend short-term follow-up, or this really looks like, you know, if you get multiple kind of short-term follow-ups and it continues to progress, this is more likely to be, you know, recurrent disease or residual disease. You can, some places, you know, I say high-end, like big research institutions, like I trained at Northwestern Memorial, we did the whole shebang, you know, we did spectroscopy and MR perfusion for everything. And, you know, there will be, they will tell you like, well, if you have, you know, increased cerebral blood flow, that should be more akin to residual disease as opposed to radiation necrosis, or, you know, they'll give you different metabolite ratios. But it's, it's, I have one case, you know, I'm limited in what I can show here, but it's just like, a tumor was treated, it was followed for like, six months, okay. And it showed up in a different place, there's mass effect, it kept progressing. And, you know, at some point, neurosurgery's hands are tied, they're like, okay, we really have to go in and take this out. And boom, that was also a pseudoprogression. Like, even on the conventional imaging, everybody was kind of like, well, I mean, it keeps getting worse. So, I think it's just, it's one of those deals where we don't, we don't have, there's no concrete way to say it's this or that. I think just like, if there's a suspicion, you know, continued short term follow up is really where we're at right now in determining whether it's, you know, radiation necrosis or pseudoprogression. Sure. Great. Thank you. There's actually a question from our audience too, wondering, you know, are PET scans useful in, you know, looking at radiation necrosis to determine that? PET scans? So, PET scans rely on, and we don't do a ton of PET here. So, let me just offer as a caveat, there may be, I may be derelict in my knowledge base about nuclear medicine, imaging of the brain for oncology, but generally speaking, when you think about PET for the interpretation of, or for finding malignancies, PET uses radioactive or radio tracer labeled FDG glucose. And the idea is that tumors are hypermetabolic. And so, they're taking up a bunch of glucose. They take up all this, you know, radio tracer, and then you image and you're like, oh, look at all that radio tracer there. That's a tumor. Problem with that in the brain is the brain predominantly uses glucose. So, if you look at a PET on a brain, the brain just, the whole thing's glowing, right? So, the brain is not a place where we typically get a lot of use from PET, at least FDG PET, which is what my experience is in oncology because there's too much background. Thank you for that. Another quick question too, do you have any resources for us as an APP team for imaging for, you know, our own education and as a resource? Well, you know, I would say an APP is kind of like in the vein of a radiology trainee, right? And there are, in today's world, infinite resources. So, the AUR, the Association of University Radiologists actually has a free, an entire lecture series. I think you'll have to play around on their website, but it's under like educational resources, trainee resources, and it's broken down by section. So, you can find what you, you know, what you are interested in, like, you know, they'll have neuro, you know, if you do spine or if you do brain, you know, there'll be specific trauma, you know, it's all in there and you can watch some lectures and these are free lectures. I think they're like 20 to 30 minutes, so they're very digestible and they're done by like, you know, top tier lectures. They created that during COVID to kind of like help trainees. Websites, you know, depending on what, where you are and what you have access to, obviously, StatDx, but that's expensive. The Radiology Assistant, I think, is pretty good. And I mean, that's just a website that you can look at, but there's a lot of stuff out there. You just have to search around for it. Great. Thank you. That's some great tips. I'm sorry, I'm kind of dual looking at two different areas for questions. A quick question, a follow up on the radiation necrosis question. When you mentioned short-term follow-up for that, as far as with repeat imaging, are you talking six weeks, three months? Just a quick question on that for follow-up. Yeah, I think like around three months, you know, assuming that, and again, it's a multifactorial issue, right? Like if the patient is deteriorating, you know, continuing to deteriorate, maybe more quickly, but I think three months, three to six months is reasonable to follow to see where things are going. And by the way, I didn't kind of specifically talk about it, it doesn't come up too much, but I mentioned the deal about the subacute infarct and tumors. Follow-up is also great for that. Like I had an argument a number of years ago with a surgeon here about whether what we were looking at was an infarct or a tumor, and he wanted to go in and biopsy it. And I said, just wait. I said, I think this is an infarct. I turned out to be right, for the record. And I said, if it is a tumor, it's not going to kill him in the next, you know, it's not going to change that much, right? These, the pathways, the imaging pathways of these will diverge significantly in a relatively short period of time. So just hold on and re-image and see what it looks like. And, you know, it turned out to be an evolving infarct. But particularly for like posterior fossa infarcts, sometimes the clinical picturum is weird, so they don't present acutely, you know, and they don't present necessarily with like classic stroke signs. And they end up getting imaged and there's just this enhancing mass with edema around it, enhancing, you know, something with edema around it. And it can be confusing, certainly can. But a lot of times, just like waiting and re-imaging is the, you know, it's the best way to kind of see what something is. Absolutely. Thank you for that. Well, this has been an excellent talk and a great discussion. Are there any additional questions from our attendees? I'm just looking at a couple of different places to make sure I haven't missed anything. I think the next session is we're going to have a Q&A. So thanks again, Dr. Hyman. This has been wonderful, very educational, and we really appreciate you helping us since, what, 2015? I think that's what you said. Yeah. Well, my son was born. He was born a week after I attended this and when y'all were in Los Angeles. The last thing I'll say, I know you got to keep to a schedule, is thank you for having me. I really enjoy speaking to this group of people. And just remember when you're interpreting imaging, I'm a, you know, CAQ neuroradiologist with over 10 years experience, and I miss things. So, you know, a lot of you, depending on where you're working, are reading studies that are being interpreted by general radiologists who don't even particularly have, you know, experience in neuroimaging. And we're, it's, you know, it's all the same team. So like I, if you have a question, call your radiologist, you know, like ask, be like, hey, I'm worried about this. You know, like I would much rather have somebody be like, what about this? And be like, oh, don't worry about that. Or say like, oh yeah, you know what, now that you mention it, I probably should have said something about that. I think you're right, that is, that is worrisome. So it is important that everybody treating a patient, you know, is looking at these images and provides that, that background. I mean that, that second layer of protection. Absolutely. And to your point, as far as going back as many years as possible to the original imaging, we had that issue as well, where just from year to year, something can appear stable. But when you go back to, you know, 2012, and it definitely, you know, changed. Thank you for that. Real quick, you have some super fans in the audience. And so, as am I, there's some questions. Do you have a YouTube channel? Or is there a platform where we can find some of your videos? Tell us more about this. The only thing I can say is I created some content for AANS a number of years back. I forget it was a lecture on the cervical spine. I stay away from social media because, you know, I'm very professional in this regard. But I work for an academic hospital, you know, UTMB, that has a certain level of decorum. And I'll be honest with you, I'm always worried that if I turn myself loose on social media, that I'm going to, you know, say something or do something that's going to end up getting me in trouble. So I just don't do it. But if AANS keeps inviting me back, you know, I'll keep coming back. So I do enjoy lecturing for y'all. That's wonderful. And we're hoping our group will meet in person next year. So would love to have you. Thank you so much. Yeah, I look forward to it. I enjoy the in-person even more. Great. Thanks, Dr. Hyman. We really appreciate you today. Thanks so much. All right. Enjoy the rest of y'all's meeting. All right. You too, sir.

neuro-oncological imaging

gliomas

brain malignancies

pseudoprogression

spectroscopy

flare imaging

radiation necrosis

PET scans

tumor progression