Mark Shiroishi, MD, a radiologist at Keck Medicine of USC, provides information about the state-of-the-art neuroimaging services and customized patient care offered at the USC Brain Tumor Center.
Yeah. Hi, I'm Mark Shiraishi, and I am the director of brain tumor imaging here at USC and I'll giving, uh, overview of the brain tumor imaging we have at the USC Brain Tumor Center. So this talk will outline some concepts regarding brain tumor diagnosis and how we use nor radiology to help the doctors figure out what kind of lesion you have. And then we'll be talking about some details about post therapy monitoring and some of the unique aspects of that and pre surgical um, planning imaging that we have in order to personalize your brain tumor imaging here at the U. S. A brain tumor center. So first we'll be talking about brain tumor diagnosis using neuro radiology. There are certain basic concepts that have stood the test of time, including information about a patient's age, the location of the lesion and the image ing appearance. Um, here we have two slices from a CT scan in a 68 year old female who presented with headache and seizures. Now, C T scans are very often the first imaging modality that patients will undergo, especially in the acute setting. C T scans are easily attainable. They are available in emergency rooms. 24 7. Uh, this is opposed to Marie scanners, which may not always be available around the clock. And so CT scans are very important, Uh, in initial imaging step that ah patient would need to undergo in order to make sure that they don't have anything emergent going on. So in this case, you can see that the patient has a lesion, um, the right side of the patient spray. Now, if you can see in these images that the right side is actually on the left side of the image, that's just an image ing convention. And then you could see that the right side of the image actually is the left side of the patient spring. So when I say right and left, that's sort of the imaging convention that will use. But on the right side of the brain, you can see this hyper dense, pretty well circumscribed mass that has some sharp borders, and it is hyper dense or hyper attenuating compared with the density or the attenuation of the rest of the brain. So it appears a little bit brighter. One slice up on the image to the right. You can see that there are some very bright areas or hyper dense or hyper attenuating areas that correspond to calcifications. Another important aspect of this is that it looks like this This tumor is arising from outside of the brain, actually pushing into the brain rather than arising in the brain. So in that case, we call it extra axial if a lesion is arising from inside the brain, referred to as interaction. In this case, we have an extra axial mass that looks well circumscribed and is partially calcified. And given that this is an older patient, this seems most ah, consistent with them and Ngoma meningioma or benign tumors, generally speaking, that are not usually very aggressive and can be dealt with in a pretty straightforward matter in patients they're usually cured of of the tumor after surgical resection, although sometimes they can appear in more troublesome areas and those air more complicated cases. But by and large, um, tumors like this can be, uh, could have a relatively good outcome. So after the initial head, C t um, if they there's concern, uh, for something going on. In that case, we definitely saw what looked like a meningioma. Patients will go to Emory, and with Emery that is the test of choice in neuroimaging, particularly with regard to the imaging of brain tumors. And, as you can see in the top left image, that is a T one weighted image, Um, where you can see that the white matter, which is thestreet chur of the brain that is mawr Central, is brighter or more hyper intense than the gray matter, which is overlying, uh, the white matter in the cortex, which is a bit dark or hypo intense. And you can see the men NGO or suspected meningioma that we see overlying the right frontal cerebral hemisphere convexity. And it looks very well circumscribed sharp borders, and it looks definitely extracts. He'll thean Midge below that on the bottom left. That is, after the administration of the contrast agent. We often will, uh, give a contrast agent to make sure that the lesion, um doesn't enhance or may may enhance, and the characteristics of the enhancement pattern can also help differentiate different types of lesions. In this case, meningioma is characteristically have uniformed contrast enhancement are generally uniforms on t one weighted imaging as well as on T two weighted imaging, which is the bottom right image on T two. Weighted imaging is also very helpful because it gives an idea of the composition of a lesion. And the you can see also that the imaging characteristics a T two weighted image are quite different from that of a T one weighted image and that the white matter is relatively darker than the cortical gray matter that we see. And it's actually the opposite of T two weighted imaging in the upper right, uh, corner. We see T two weighted flare image in a flare image is a T two weighted image where the cerebral spinal fluid, that bright stuff that you see in the south side of the brain, the folds of the brain in between the folds. That's the cerebral spinal fluid that bathes the brain and spinal cord On t. One weighted imaging. It's dark, but on T two weighted imaging it's very bright. Flare imaging is t two weighted imaging, but the fluid is made dark, and so in that way it really accentuates the contrast between the brain and the CSF and the ventricles and the soul side. Uh, flare imaging is very helpful because most pathologies that affect the brain are hyper intense, are are bright. And if you just look at T two weighted imaging, it's possible that if he had a relatively small lesion that if it's right next to the street, will spot the very bright serial spinal fluid. You may not see it very well, so doing flare imaging is is quite useful. And so that's one of the first things we go to. We look at the flare imaging and then we look at the pre and the Post contrast enhanced T one weighted imaging to characterizes. And this all the imaging characteristics here are consistent with with him in Ngoma. Also on the T one weighted post contrast Enhanced image on the bottom left, you can see at the very anterior and posterior aspects of the lesion, you could see a little area What we refer to as a dural tail dural tail is a, uh is, ah, relatively unique hallmark of an NGO HMAS, and it is highly suggestive of that. And in this case, things turned out to be a man NGO and the patient did quite well after surgical resection Now, the next case that I'm going to show you is a little bit different and a little bit more difficulty in its in its treatment in its treatment on and the outcome. In this case, we have a 60 year old male who has who had presented with a headache. And we see that we have a memory, uh, image on the on the very left here, that is a T two weighted flare image. And in the right temporal lobe, situated more centrally in the white matter is ah, hyper intense lesion. It looks mass like, um, there was concern that this could be, ah, low grade glioma. There are other things that potentially could look like this some infection and inflammatory conditions. But a glioma was was a big concern based on the location and the age of the patient. And you look in the center image and in the image to the right. The pre and the Post contrast enhanced t one weighted image that the lesion does not really enhanced. Unlike the meningioma that we saw in the previous scan, the image, uh, shows that it doesn't really, uh, light up on the post contrast and scan, so that really doesn't enhance and as opposed to an extra axial lesion with them in Ngoma. This is an interactive lesion because we could see it's clearly arising from the substance of the brain itself and because it didn't really enhance the consideration that was that this could be a low grade glioma, as opposed thio high grade gliomas. High grade gliomas usually demonstrate some contrast enhancement, but there are some high grade gliomas that can demonstrate very little contrast enhancement. So this was followed up with imaging. Uh, and then three months later, you can see that lesion started to get a little bit bigger, and it was even more concerning for a low grade glioma. And so then it was decided to undergo surgical resection, too, really, the symptoms and also established a histological diagnosis. And you can see there's a big cavity. Now where the where the, uh, surgery surgical resection was accomplishing along the anterior and posterior margins, You could see some blood products there, that which you often see after surgery. Unfortunately, the diagnosis for this was not a low grade glioma and was an Anna plastic Astro cycle, which is a great three. Uh, Astra Saitama. And so this is a high grade glioma. Eso Then we're going to segue into post therapy monitoring Now. Now, with, uh, patients that are in the post therapy setting the treatment for high grade gliomas maximum safe surgical resection with T Mazzola might chemo radiation. And then after that, uh, close interval, uh, memory follow up is what is done. And in this case, you can see three months after the surgery, the patient did quite well. You can see that the surgical margins looked very clean and on the left you have pre contrast t one weighted images and on the right, you have a post contrast t one weighted image and you see along the margins of the resection cavity, you see a little bit of linear contrast enhancement on the post contrast enhanced scan. There's nothing really nodule er about it in. This is just post surgical appearance that you see that usually will resolve over time and is not concerning for any tumor progression at all. The patient did quite well for several years with follow up with conventional memory imaging. However, almost three years later, you see ah small kidney being shaped area of nodule er contrast enhancement, actually anterior to the resection cavity in the right temporal lobe. This is concerning because progressive tumor can look like this. On the other hand, though, what confuses all this is that post treatment radiation effect can also look exactly the same as contrast enhancing tumor. Now there are different ways of dealing with this. The conventional method is to you follow with close interval follow up. Now, if it's post treatment radiation effect than the contrast enhancing lesion will, either stabilizer will disappear. If it's a tumor, it's generally going to keep getting bigger and bigger and bigger as we can see on the very right. The, uh, contrast enhancing lesion was starting to get very big, and the concern for tumor was very high. But at the same time, some of this could also be post radiation effect, and we wanted to really be sure. Now we at USC have been doing profusion imaging for quite a while. Uh, it's an advanced imaging modality that that has been around for many years. However, the way it's being, uh, the way the image is required in the way that we post process. Them has really changed a lot, and we're doing some of the most advanced techniques now at USC. And these are just some cereal images just to show you what these images look like when we are acquiring the profusion images now profusion. Emery is a dynamic form of imaging as opposed to the conventional imaging that we saw, which is more like snapshots in time. This is actually kind of like a movie, and so we acquire very fast images before, during and after the administration of a contrast agent. And this has many applications and brain tumors from differentiating tumor from non tumor, trying to differentiate different tumor types, trying to differentiate the grade of the tumor. And also biopsy guidance and what we're talking about now. Evaluation in the post therapy setting the this dynamic susceptibility contrast method that uses T two star waiting uh, bits. Um, physics details that won't really go into is the most common method. This is the same method, by the way, that is used in stroke images. So the two major applications of this type of profusion Emery is both in the evaluation of tumors and also in the evaluation of patients with acute stroke symptoms. Now this there is a metric, uh, that you'll hear it's called our CBB, which stands for relatives through blood volume. And so this is the main, um, number or value that we get from profusion. Emery uh, that is most relevant to brain tumors and higher numbers indicate a higher degree. Aggressiveness and malignancy and lower numbers indicate a lower grade or potentially post treatment radiation effect. Basically, what you do is you take these rapid, dynamic images. You look at changes in the signal intensity over time, and you do mathematical calculations with different algorithms that different Softwares have, and you get your relatives through volume measurements. However, one important thing to remember about it is it's all predicated on the idea that the lesion that you're evaluating doesn't have contrast enhancement. In other words, the blood brain barriers intact and the dye that we injected is not leaking out into the lesion, but a zoo we saw with this case of this post therapy monitor, we saw that there was contrast enhancement and that the guy was leaking out from the capital areas out into the extra vascular extra cellular space, And this then can introduce some, uh, complications when we do relative super blood volume measurements and can affect the accuracy of these things. So the two situations that you can have, depending on the what's going on the biophysical level of the tumor or you can either get underestimation. Or you could get overestimation of your relatives through the blood volume, depending on what's going on, UH, locally at the at the cellular level. And to make things even more complicated, you can have both things going on when you get under and over estimation. And so this is a real troubling aspect of, uh, doing profusion, imaging and brain tumors that people have been working on for many years. This isn't really an issue in stroke imaging, because in acute strokes they don't. The lesions basically don't enhance. But in brain tumors, many brain tumors do enhance. And in the last few years has been mawr recognition of this lack of standardization with how people deal with this. Over the years, profusion imaging has been around for quite a while, but the way that people acquire the images and the way that people post process of the images has been quite variable. And so this has made it very difficult to use profusion imaging routinely in clinical practice and in clinical trials because of the varied, uh, methodology by which this, um, is being performed. And I was part of the American Society of Functional Radiology White paper. What we talked about this dilemma. Uh, about 2015, we wrote this white paper talking about some of the major issues that are still remaining unanswered about how to do profusion imaging, uh, for for brain tumors. Now, most recently in the last year or two. My colleagues Kathleen sh mind uh Leland who? Jerry Boxer and Chad quarrels have been working on some, uh, both clinical and mawr radiologic physics level types of studies trying to look at what are the optimal, optimal acquisition parameters for this type of profusion, imaging and rain tumors. And they've had some really good results where they standardized protocol that we have now adopted at USC. Just recently, we've I've completely overhauled all the brain tumor protocols here to adhere to these new acquisition parameters, so the images are acquired in a standardized way that's backed up by the literature and then we're also using the latest post processing software that can also compensate for that leakage effect they talked about in order to make to make the relative suitable blood on the RCB values as accurate as possible. And another really interesting outgrowth of this are what we call these F t B or fractional tumor burden maps. And so this is that same image of that patient I showed you where we were concerned about post treatment radiation effect versus progressive tumor. We were suspecting progressive tumor, but we weren't quite sure how much of it was tumor and how much of it was post treatment radiation effect before the advent of these F T B maps. We were very often given these RCB V color maps where way had an idea of where some hot areas are or tumor. But generally we kind of characterized the entire lesion as either post treatment, radiation effect or tumor. Now, in this case, you can see that a good deal of the area along the anterior margin is red and red is an indication of its tumor. You see some purple mixed in, which is an admixture, saying that it's both post treatment radiation effect and tumor, but post eerily along that rim along the rim of the resection cavity. You see that all that stuff around there is post treatment radiation effect, and you could see to the left that hissed a gram gives your distribution of the number of, uh, boxers or pixels in that and shows you how much of that lesion is. Tumor versus mixture versus, uh, post treatment radiation effect. And so then we're able to get a MAWR nuanced view of this lesion. And so when the when we talk to our radiation oncology and neurosurgical colleagues who are better able to tell them, Look, this is not all tumor. There's some of it around the posterior margin, especially around the resection cavity that looks like it's post treatment radiation effect. However, if you're contemplating further surgery or radiation therapy, you really need to concentrate on that anterior aspect there, and this could give you higher confidence and really personalized the amount of therapy, uh, personalized the the therapeutic approach to each patient. Instead of just treating the whole thing as one entity, you can see that it's quite more complex than that. A couple of things I also want to mention are the advanced imaging techniques we utilize in pre surgical planning. So very often you will have tumors that are located in areas that could affect, uh there post operative outcome. For example, the motor, uh, pathways in the brain are very often either infiltrated or displaced by tumors here on the very bottom, on the left of the image, you see this T two weighted image where you see a very large, hyper intense mass. And the big question for the neurosurgeons are if they're going to respect this, where are the motor pathways and in the middle and to the right? You see these diffusion tensor imaging images which show that the the cortical spinal track, the motor pathways actually pushed over immediately and that the tumor is actually along the lateral aspect of the cortical spinal tract. And so this gives the surgeon ah lot of information and that they should be very cautious as they proceed along the more medial aspect of this tumor. And so this, in addition to the conventional imaging, is something that we routinely utilize here to make sure that we have optimal outcomes. After surgery. In addition to that, we also routinely performed functional Emery Thio Act as a complement to the structural diffusion tensor imaging tracked. Ah, graffiti images toe. Also, look for where the motor, uh, motor areas are the brain and also language. And so, in this case, to the very left, you see a large contrast enhancing lesion that could be near the motor, uh, stripped for the patient, and then the slice in the middle shows a cut just above that area where there's still some low signal areas. That is the Idema part of the of the tumor. And on the right is an activation map showing the patient tapping their fingers, showing where the motor strip is, and you could see it's right along the anterior aspect that that a demon. So this tells the patient, uh, this tells a surgeon that when they go in to operate, that they need to be very conservative in that area in order to decrease the chances of any motor deficits after surgery. So I just want to give it just a quick overview of some of the research areas that we're talking about with regards the brain tumors. Uh, we have a seven Tesla Emery scanner here at USC and that picture in the upper right. You see them lowering in this this massive scanner here, uh, most conventional clinical imaging has done at 1.5 or three Tesla, but seven. Tesla's scanners are very powerful scanner, the most powerful clinical scanners in the world, and not many places have them. They're still primarily used in research applications. There's very little known about the use of seven tests, memory and brain tumors, particularly with advanced, uh, with respect, advanced imaging like brain tumor profusion imaging. And so we're undertaking a program to see what sort of advantage is that we can get with the higher spatial resolution you could get with this very, very powerful seven test memory scanner were also embarking on a program using a technique called magnetic residency last, ah, graffiti to determine how hard tumors are. Uh, for example, meningioma, uh, can be very software. They could be very hard, and it's very important for the neurosurgeon to know how what the consistency is before they go in and operate. This could help the patient understand what their expectations are with respect to how much of that tumor we can cut out, because if the tumor is rock hard, it may be very difficult to get it all out. On the other hand, if it's just it's soft is Jello. It's relatively straightforward to cut out the tumor in an entirety and would have also have minimal blood laws. So that is a new area that we are working on is well as trying to develop these F T B maps for metastases. The F T B. The fractional tumor burden map that I showed you for that primary brain tumor is, uh, relying on tissue samples taken from patients with glioblastoma and and high grade gliomas. On the other hand, metastases, which are tumors that spread to the brain from other parts like breast cancer, lung cancer and so on. The the the blood vessel physiology of that is very different from that of a primary brain tumor. And so we really need to understand what are the profusion, the R C B B values of the blue post treatment radiation effect of the purple, that it's the admixture of post treatment radiation effect, and the red, which is the which is Frank tumor. And we don't know that because the different primary tumors all have different physiology. And so we need to understand. Um, that And so I'm were participating in a multi center study where the developers of the TV maps, uh, the research colleagues that I talked about Onda were collecting cases, and we're trying to trying to figure that out so we can create these maps for these different metastases. And finally, um, I am part of the Enigma Consortium, which does the world's largest brain tumor, or brain brain imaging studies, UH, that have ever been done on. I'm focusing on cancer related cognitive impairment. And we're so we're trying to collect cases from around the world of patients who have cancer, UH, that are not yet in the brain but are outside of the brain to see what what portions of the brain are affected structurally and also functionally determine the basis of cancer related cognitive impairment. One, uh, major aspect that people may have heard about his chemo brain. So people have this fog, uh, either temporarily or long lasting after they survived cancer, and they undergo chemotherapy and other kinds of cancer therapy but have problems, Um, coming back to normal with regard to their cognitive functioning. And this is all, uh, in partnership with the other institutes at USC, like the Stevens Neuroimaging Institute, the image ING Genetic center and the UK Bio Bank. And also, some of the funding has been obtained from the N i H. So, in conclusion, I think I've given you, uh, an overview of the brain tumor imaging program we have here at USC. I've gone over three use of new radiology and brain tumor diagnosis, and I spent quite a bit of time dealing with the critical role that neuroimaging has in post therapy moderating. And I've highlighted some of the unique aspect of the advanced imaging programs here at USC, with particular respect to the fractional tumor burden maps that we now have, which we just added here at U S. C. As well as some of the pre surgical planning with respect to the diffusion tensor imaging and the functional memory that we could do to optimize uh, the safety of the brain tumor surgeries, thes air all in effort in order to personalize the brain tumor imaging for each patient and not treat all brain tumor patients as if they're one sort of monolith. And at the at the very end here, I just wanna thank my colleagues here in the U S. C. Brain tumor center on my colleagues of the U. S. C. Division of neuro radiology. My research collaborators. And last but not least, I want to thank the patients, their families. Um, I personally have had some close, uh, family relatives who are cancer survivors, and I realized the devastating impact that having a cancer diagnosis can have on a patient. And I think we all here at USC are doing our very best to do whatever we can to ensure that you have the best outcomes. And I am leaving my email here. If you have any questions about our program or in particular, the neuroimaging brain tumor program we have here, please feel free to reach out for me. Thanks for your attention
