Video
library

  • 453
Comments (0)
    Transcript
    Bogdan: Hello, everyone. My name is Bogdan Valcu, I'm the director of Novalis Circle. Welcome to a new webinar with us today. We have the pleasure of welcoming Dr. Dheerendra Prasad from Roswell Park in Buffalo, New York. Today Dr. Prasad will provide a review of our Elements contrast clearance analysis, in particular looking at response assessment maps to differentiate tumor from treatment effects.

    Dr. Prasad is the Medical Director of the Department in Radiation Medicine at Roswell Park in Buffalo, New York. He's also the director of the CNS and Pediatric Radiation Medicine, as well as the Gamma Knife Center. And he is both a radiation oncologist, as well as a neurosurgeon, and he will provide a review today of various Brainlab enrichment elements that he utilizes in conjunction with his Gamma Knife program.

    As always, we are providing CE credits, if you'd like to receive CAMPEP, MDCB, or ASRT credits, please follow up with us at info@novaliscircle.org upon successful completion of this webinar. And don't forget to sign up for our upcoming webinar with Dr. Michael Girvigian from Kaiser Permanente in Los Angeles, who will review their experience utilizing brain lab software in their Brain Mets program, as well as also cover their experience utilizing some of our response assessment techniques.

    And lastly, don't forget to log in via either Google Chrome or Safari. And should you have any technical difficulties throughout the webinar, you can simply refresh your window. Please utilize the chat interface to send us questions. We will answer those questions upon completion of Dr. Prasad's lecture. We will also ask you some questions via our polling interface, and as always, should you like to follow us on social media, please utilize the hashtag provided. With that said, I'd like to turn it over to Dr. Prasad.

    Dr. Prasad: Good morning, everyone. Thank you for that introduction, Bogdan. And it's a great pleasure to share our experience with the Brainlab Elements suite in conjunction with our Gamma Knife program with all of you. Of course, these are strange times and I'm coming to you from my own study at home, sitting in my son's gaming chair. So if you happen to notice, that gamer's share in the background, that's not me, that's other member in my family. I'm strictly all business and all work. Our talk today will focus on our integration of the Brainlab Elements suite into our radiosurgery practice, with a particular focus on the contrast clearance analysis as a tool for evaluating response as well as complications of radiosurgery, in particular, for brain metastatic tumors.

    The title indicates that we are a high volume radiosurgery practice and therefore, we are a team of multiple people, which includes physicians, physician assistants, both neurosurgeon and radiation oncologists and, of course, medical physicists to run the program. I have been doing radiosurgery for a very long time now, almost from 1992 to today, and nearly 12,000 patients that I've treated personally over the years. And in the last two-plus years, we've had the opportunity to use Brainlab Elements in our program. And you'll discover pretty quickly why we integrated it into our workflow. We are based in a cancer center but we treat almost all indications for brain radiosurgery that there are out there. And Elements has been helpful across the board for multiple indications for us. We have the Gamma Knife icon, which is the current version of the Leksell Gamma Knife.

    Over the years, we have played a significant role in modernizing the paradigms of managing brain metastatic disease, particularly when it comes to patients with multiple tumors who are just beginning therapy. Gone are the days when we counted the number of lesions and restricted the use of radiosurgery only to patients who had four or fewer lesions, which clearly was a old time count-what-you-can approach to this technology. The fact that radiosurgery can now be performed with great speed and is available to more and more patients and especially with software developments like Elements, the whole process of delivery has become much more efficient as as the process of follow up these patients. There is no reason to restrict the application of this technology to only a certain selected group of patients.

    In fact, in my opinion, all brain metastatic disease should be managed with radiosurgery, except when you're dealing with very limited life expectancy or leptomeningeal disease. And which is kind of what this study tells you, that we focus more and more on negative neurocognitive effects of whole brain radiation therapy and try to limit its use as much as possible, because we want to maintain the highest quality of life. And we also see these patients much earlier in the diagnosis of their cancer, since brain MRIs are no longer a difficult test to obtain. And many of these patients present with very small lesions, which fit nowhere in the historical understanding of brain mets, and certainly are not symptomatic.

    But you will realize that despite all of this understanding and the availability of technology from multiple vendors, there has been a very slow increase in the penetration of radiosurgery as a modality for treating patients with brain metastatic tumors. And this table you see here stops at 2014, and I think we're doing a little bit better now, nationally and internationally. By comparison, at Roswell Park Cancer Institute, the utilization of radiosurgery for brain metastatic disease will fall in the high 80s rather than these low numbers that you see here. And that is because, like many other leading centers in the country, we have taken it upon ourselves to change this dialogue.

    There are many reasons for this disparity, but many of them read the patient education, reimbursement strategies, and we see that there are disparities even in proven diagnosis, like renal cancer, where conventional whole brain radiation therapy in small low-dose fractionation is not very effective, that the utilization can vary considerably, depending on where the patient's healthcare plan and socioeconomic status. So there are certainly a lot of work to be done in this field. But we treated this point at last count roughly 700 patients a year with a single Gamma Knife, which makes us one of the busiest Gamma Knife practices anywhere in the world.

    And so when we had looked at what would make a quantum difference in our ability to perform these treatments, for more and more patients with many, many more brain mets, we identified several areas of weakness. One of the biggest weaknesses, of course, was segmenting of normal anatomy, and quickly identifying and segmenting brain tumors rather than doing it painstakingly slice by slice. And Brainlab Elements has provided us that functionality very well. Not only is it a good tool to auto segment normal anatomy, which we use for every single brain radiation patient in our department, and certainly for a large number of our radiosurgery population, it also provides very rapid three-dimensional brush painting and contouring off our targets, which allows drawing of multiple maps very quickly.

    It also happens to be the one software package where translation of contours from it to our gamma planning software is with high fidelity. And we can use the contours as such with minimal change. But not only that, we have the ability now, with the use of Elements, to perform contrast clearance analysis, which I will go to in a little bit more detail, allowing us to now identify with a greater degree of precision, those of our patients who are undergoing radiosurgical side effects versus those who are undergoing tumor response or tumor recurrence. Because when you treat this many patients, and that's just number of patients I mentioned to you, many of them will harbor three to five to many more brain mets.

    There's a lot of lesions that are being treated, and they need to be followed on a very accurate basis. We have new tools to manage radiation effects, we have new tools to manage patients with recurrence. And the sooner we find out which category a given MRI falls into, the faster we can take action and plan things accordingly for the patient. Just to give you an idea, radiation effect from radiosurgery is not an unknown, it occurs. As you can see in this slide, about 80% of the patients treated with radiosurgery using any radiosurgical technique out there, and certainly with the Gamma Knife, won't respond without any evidence of radiosurgical adverse effect, and also no failure. In other words, you control the tumor and you don't produce any side effects.

    Twenty percent of the patients then remain in categories, which include patients who fail to respond to the treatment, and that's about 9% to 10% in most practices, those who develop adverse radiation effect, or ARE, seen about 5% to 6% of patients, and then there are those who have failing tumor as well as adverse radiation effect, and that becomes a really difficult thing to tease out with conventional existing imaging. And there will be a small number of patients who will not have any determinate outcome, but don't fit into any of these various categories. This is a UCSF data sec and it varies pretty closely what we have seen in our practice. Adverse radiation effective, we've looked at quite carefully.

    It is a significant issue for a patient who is undergoing radiosurgery for brain mets, because if they develop it, they often require the use of steroids, which is going to block them from receiving other treatments like immunotherapy and sometimes enrollment in trials. So we make every effort that our treatment if that's kept to a minimum. If they can't be avoided, they're identified early and they're handled accordingly. When we look at what happens in a tumor that has been treated with radiosurgery, you will always find a spectrum of changes which includes necrosis and, of course, persistent tumor islands. And in patients with active radio necrosis, you will see proliferation, macrophage infiltration, new vessel formation, which is all going to produce almost a mass effect-like phenomenon.

    That being said, with nearly 700 patients treated a year, typically in a year we operate upon 3 to 4 patients for mass effect coming from adverse radiation effect that cannot be controlled by existing therapies. So it's a fairly rare event that you need to intervene. But when you do, you get histology like this, which helps you identify what's going on in a treated tumor field. And, of course, correlating that with just an MRI finding, whether or not accompanied with flare, perfusion studies, and existing MRI only modalities can be very difficult. And therefore, a new tool clearly is needed. And I think that in our experience, contrast clearing analysis has proved to be a very valuable tool. Described initially for glioblastoma multiforme, it depends upon the fundamental rate at which contrast flows through the brain and clears out of the brain.

    It makes common sense that it will flow quickly in and out of a blood vessel, which is the way that the normal is set in this technique. It will also go through highly vascular structures, which includes tumors, at a fairly rapid rate and clear out pretty quickly. Where it will penetrate slowly and linger will be areas of brain where the blood-brain barrier is deficient, or the brain itself has been injured as a result of the treatment. And so, that inherently produces a difference in rate of clearance of contrast. As you can see, in this curve, you see slow accumulation of contrast agent in areas where injury has occurred and the BBB is deficient. Normal brain tissue will pretty much not show much activity, the blood vessels will spike with contrast and clear pretty quickly.

    And the tumor tissue will spike just a little bit slower than normal blood vessel within a few minutes, and then clear out reasonably fast. So this differential ability to watch the contrast get in and wash out of the brain, results in a contrast clearance map, which can then be cryptographically depicted on the screen. Essentially, the only additional study that a patient requires in order for a contrast clearance analysis to be performed is a repeat of the T1-weighted MRI that was done post contrast, repeating it for the same setting, same slice setups, as well as the same exact position, will make this analysis more accurate. And it's usually not hard to reproduce that.

    So essentially, the patient receives a standard brain MRI with contrast, hangs out in the MRI department for a period of about 60 to 90 minutes, gets back on the table, on the same MRI, with the same settings, and has a redo of just the one sequence, which is the...in our case, it is the 3D Brain Volume [inaudible 00:13:57] or the 3D BRAVO sequence made by GE, it could be called the 3D spoiled GRASS or SPGR, or the 3D MP-RAGE, depending on the software you're using. But essentially, it'll be a 3D brain optimization matching the T1 post contrast. If the T1 post contrast was a spinnacle non 3D set, you can use a non 3D...you, of course, use the same sequence in the delayed acquisition. But in general, since we're trying to reconstruct this data three-dimensionally, it's always ideal if a 3D sequence is used.

    The software essentially will then co-register these two sets of images and automatically code areas of rapid contrast wash-in, wash-out as blue, which would be representative of tumor, and accumulating areas as red, which is more commonly seen in both radiation change. So, essentially, without much enrollment on the part of the user, in fact, the user has no control on this process, it will generate a map for you. As I said, this was earlier approved for glioblastoma multiforme by doing biopsies in the regions that were quoted as clearing and accumulating to show that there was a preponderance of GBM cells in that area. We integrated this into our Gamma Knife practice, including that for, say, a GBM. So what you see here is a small area of GBM that has been seen to appear after conventional surgery and conventional chemo-radiation therapy, and the patient undergoes a radiosurgical treatment.

    At follow up, we see until at that point in time, as you can see, this is the first followup of this patient, and there is a significantly reduced amount of enhancement on the left side. And a contrast clearance analysis performed shows a small area of stippling blue in the middle, but for the most part, you see red treatment effect. However, as you follow this patient out serially, you start to see, on the MRI, an area of nodular enhancement that starts to appear right here. And this is now superior and medial to our prior radiosurgical field. There is also a small area of enhancement more anteriorly in the frontal lobe. When you perform a contrast clearance analysis, you can see a clear evidence of recurrent tumor here, which is coded in blue, and the early appearance of some contrast accumulation in the area where you were seeing faint contrast enhancement.

    Given enough time, we've seen these small patch areas to then evolve into frank tumor. So I feel that not only is the blue and red encoding a part of it, there are going to be focal changes around it, which will match your brain T1 and flare findings, and together provide you one additional tool to identify areas of recurrence. Clearly, this could have been thought of as radiation effect, although it was just outside the field. But even in a GBM where the index of suspicion is very high, we would often consider this as tumor. So if it did not have contrast clearance analysis, to be honest, I would have considered this as a failure and treated it. But this just makes that much easier and firmer for us to add radiation on top of radiation. Remember that this patient already received conventional radiation therapy to 5940 centigray. So additional treatment is not without its potential side effects.

    Here's another example of a recurrent area of a patient with GBM, where you can see that we followed this patient out, we knew there is active tumor present, patient underwent Avastin therapy, and on Avastin, you can often see progression, pseudo progression, and pseudo response. What is interesting is the contrast clearance can actually help you pick out what part of this abnormal appearing area is actually potentially active tumor cells. And so on the whole, you start to see a much better way of determining what aspects of a brain MRI represent actively growing tumor. The translation of this technology to brain mets was not commonly done when we acquired the software. And we were very clear that that was the main reason we really wanted to explore contrast clearance in our practice. And the reason is this, that when you have a long standing radiosurgery practice, you begin to see your failures and you begin to retreat them.

    So if you look at this plan, you're seeing, essentially, three dose lines, the turquoise dose line is one and two are previously delivered. And here is the isodose line for this third treatment. So this patient is now heavily treated in this area for a lesion that keeps recurring in a cavity that was previously a post-surgical cavity. At this point in time, with the recurrence even after the third treatment, the question is, are we looking at brain radiation effects given all the treatment we've done? That's entirely a possibility, and the likelihood is much higher than 5% of the setting. Or am I looking at recurrent tumor? So what we essentially did was we performed a TRAM analysis and we ended up operating on this patient in order to remove this area of abnormality.

    And essentially, before we went on and did this, we decided to run a contrast clearance analysis. And our read based on our understanding of CCA is that there is a nodule of tumor located more posteriorly here towards the sinus and then there is an area of field necrosis radiation effect, etc., some of which on the MRI is very obviously necrotic, salvage not so much. It's hard to tell the difference between this area and this much more enhancing area further in. So we did pointed biopsies, and it's very interesting that we were able to correlate the areas that appeared coated red on contrast clearance were largely necrotic. There were small areas of active tumor in there, but very small. Seventy to eighty percent of this microscopic field represented necrosis. By comparison, it targeted resected specimen sent to the pathologist from the blue coded area on contrast clearance was nearly 80% tumor.

    So we know that even in the case of metastatic disease, there is a correlation between contrast clearance and the appearances on pathologic examination. Most of us have always been taught to say necrosis is a pathology diagnosis not to be made on an MRI or any kind of imaging study. Well, it is possible with contrast clearance to bridge that gap and to say it's certain. There's certainly a greater degree of confidence, what parts of enhancement on an MRI better present necrosis, and what parts represent residual or recurrent tumor? We did the same in a glioblastoma just for kicks, although we realized this is published data by other centers. But we were able to confirm quite clearly that blue appearing areas on contrast clearance analysis show predominantly glioblastoma cells, whereas areas that are coded red are predominantly necrotic.

    And so we essentially ended up pathologically sort of ratifying what we were seeing on the imaging findings on contrast clearance. In the next several slides, I will try to walk you through the various types of outcomes that you see in a radiosurgically treated patient, and how those correlate with the appearances on contrast clearance. So here's a simple example, one of the desirable outcomes that you want in every patient, tumor response and you have no adverse effect. And you can see progressively decreasing amounts of enhancement with the radiation field overlaid, and you can see that the contrast clearance map essentially is silent. There's a tiny speck of blue in the center, probably corresponding to this enhancement, probably still some residual tumor there, but there was no adverse effect in this picture.

    You could see also the converse, in which case there is no residual tumor left. However, given the size and somewhat necrotic nature of the tumors to begin with, as you see in this field, it's no surprise that the contrast clearance analysis at response time, you would have read this MRI by itself without contrast clearance analysis as a response with some residual tumor. And you would be right, except in this case, we can tell you that based on contrast clearance analysis, more than 90% of this tumor has essentially been destroyed. And there is a field of necrosis, partly tumor necrosis and partly radiation effect that is left in its place. So here's an example of ARE with minimal or no residual tumor left. Most patients in the early stages after treatment will show a mixed response.

    So here's the treated lesion, about three months after radiosurgery, and you can still see residual tumor. It is smaller, I don't have the pretreatment slide for you, the tumor is smaller, but you're starting to see a halo of adverse effect or certainly contrast leakage up here, and in the same patient, another lesion and the same pattern. So that's early to two month follow-up showing you that you will see shrinkage in tumor, which will correlate to the contrast clearance analysis, and will also show some early radiation effects and changes in the CCA. Here's an example of a mixed response that is quite common in larger tumors. You will see areas of red, but significant fields of blue. So here's the tumor at the time of treatment, and here is a sequential progression of the contrast clearance in this patient.

    So you see the tumor responded pretty quickly. And over the course of the 6-month to 10-month scans, there still remain islands of blue, which were visible but largely necrotic tumor, and correlate that with the MRI on top. And you can see as the MRI starts to clear up and the lesion is involuting at about a year out, you're still left with a small area of radiation effect on the brain with very limited areas of blue in the region where the tumor used to be. So that's useful for you to see. Also important to recognize is both enlarging and shrinking areas and contrast enhancement in a treated field can be not representative of anything but radiation effect. So on the slide set above, you'll see a tumor. That actually appeared to be enlarging, but some sort of a necrotic area appearing in it, but all of it appears red on contrast clearance with virtually no evidence of active tumor remaining.

    Similarly, the one below is shrinking. And again, you can see there was no evidence of active tumor. So the idea that you can purely based on shrinking enhancement predicted the patient is responding. And for those of us who follow our patients volumetrically, as we do, we've hung our hat a lot on volumetric reduction as a marker for tumor response, it may not be the case. It may very well be that the geographic area that you created continues to leak contrast, but it's merely contrast leakage as a result of either the radiation effect that we have produced, or a combination of other therapies that the patient might be on. We have not yet fully explored but are keenly looking at the changes in contrast clearance when patients receive immunotherapy.

    Many patients who have been treated for lesions that have disappeared, when placed on certain agents, particularly immunotherapy agents and B-RAF agents, will start the show enhancement again. And that kind of recall effect that we see is very fascinating for us, especially if you want to analyze it on a contrast clearance to understand how that will read out on a contrast clearance analysis. And, again, another example of adverse effect with partial response, that's quite a common scenario early in the follow-up. Now, this slide is more of a question for the audience to think off, here is a patient treated, this is the tumor close to the surface of the brain, the radiation field on it. And you see the MRI of this patient, and this would have been labeled as ARE, plain and simple.

    We have said the tumor has responded, and we're seeing some radiation effects. Contrast clearance, however, shows there's still a significant amount of tumor cells still visible, there's some tumor necrosis, there's some radiation effect, but a part of this area is still present. Now, this becomes significant when we're retreating a patient for other lesions. And perhaps we have one area like this where the response appears mixed from the prior Gamma Knife treatment. Typically, we would have let these people be followed, sometimes meaning that you have to bring them back and do another procedure for this area that you called as effect, but indeed, was residual tumor and starts to grow. But nowadays, we're much more careful.

    So in our standard formula now, if we see anyone with a lesion that has mixed response, or only a partial response prior a to radiosurgery treatment, and they're presenting for a second procedure, we will perform a contrast clearance scan on the morning of the procedure, which means the patient will go to MRI, get their MRI scan done with or without a stereotactic frame attached to their head, and they return 60 to 90 minutes later and complete their contrast clearance scan, and we will simultaneously analyze on the contrast clearance analysis tool within Elements, as well as, of course, proceed to plan this patient. Now you can always in Elements draw out this area We cannot export these clearance maps to other software because they're not DICOM ready.

    But you can certainly use this map and draw out the area of interest that you want to retarget with radiosurgery and you can export that contour to the radiosurgery tool and proceed with the treatment plan. You can also see evidence, so to illustrate that point, here is a case in point. A patient has come for a new treatment, and I'm looking at an area that I've already targeted twice. Clearly my motivation is not to treat this area. And so when you perform a TRAM analysis, which is shown here, actually, when we did the TRAM on this, this was all...contrast clearance analysis, it was all essentially red. And so at that point, we call this as a adverse radiation effect, continued to follow the patient, you can see that there is very little enhancement left. And this is when the patient had come back for yet another procedure, and then sometime later, there was an occurrence again.

    So you catch that this would have been perhaps confused with late effects, who knows what this is? Fairly fuzzy margins, and then there you see on the contrast clearance analysis, clear evidence of tumor. If you do serial follow-up on these patients, you will see an evolution of change that starts with the appearance of this red area, significant areas of blue. And then as you go along, you will find here, for example, an enlarging area of radiation effect, minimal tumor, but on this other side, we see an enlarging area of radiation effect, but we also see recurrence of tumor. So I'm only clearly showing you a fraction of our practice. If you listen to me talk for the last 30 minutes, you're assuming every one of my patients is failing, or that conference clearance is calling these as failures much earlier than others. That is not the case.

    In the vast majority, I'd say 80-plus percent of our patients, our clinical judgment based on an MRI correlates very well with contrast clearance analysis. But it's these challenging cases that are brought to show you that really highlight the value of an additional tool. For us, determining adverse radiation effect is not enough. It also means that we have therapeutic interventions we can offer these patients. And we use contrast clearance as a tool to measure the efficacy of the intervention that is planned. Now the very simplest level of steroids are an intervention commonly used. But in some patients, as an example of this person with invasive ductal carcinoma of the breast, who presented multiple brain mets seven years after her initial diagnosis. She underwent Gamma Knife for 11 lesions.

    She was then seen to follow up, and came back and I treated 22 lesions, including one area that was progressing in her postero-lateral right cerebellum. So that area got treated a second time. After that, the patient complained of some side effects, particularly balance difficulties pertaining to cerebellar influence. So, when we performed the contrast clearance analysis, we saw an area of very tiny residual tumor, but we did see a significant area of adverse radiation effect. And we based this patient on Avastin, and now you can see how on Avastin, over the course of the next six months, the adverse effect responded. And this is something we've seen repeatedly. We have seen patients undergo treatment with steroids and Avastin, and you can see the adverse radiation effect response.

    We also had a couple of patients in whom when we discontinued the Avastin, we started to see a recurrence of the adverse effect, and those patients either went back on Avastin, or after a suitable interval, if there was mass effect or if there were clinical symptoms that persisted, even undergo a resection and removal of the necrotic mass. But in general, we are able to now monitor, not only the effect, the side effect, but also the influence of therapeutic interventions on that side effect. And this did correlate with improvement in patient's clinical condition. So as the ARE got better, as seen on the contrast clearance analysis, so did the patient's symptoms, and the patient continued to be free of brain mets and the ARE results of the Avastin and doing well.

    So we went back and looked at...early on, I have many, many more cases right now that we'll analyze and we're still in the process of crunching that data. But as you can see, we've done contrast clearance analysis in almost all different indications. The most important thing in this slide is that the interpretation of radiologists whether or not they made a comment on any kind of radiation effects on these patients. And when you look at this slide, you see that about 24% of the time in this patient group, the patient was recorded on the MRI report as having abnormal effects or treatment effects. Most of the time, no comment was made in that factor. Whereas when you look at the same patients analyzed with contrast clearance analysis, you begin to see that you can catch the treatment effect, it comes to your attention and you catch it much earlier 62% of the time. So that's a greater number of patients that we found.

    And in a number of patients that we may have labeled as only treatment effect, we did see it as tumor evidence, which means that if those things, those areas don't resolve, you have the opportunity to go back and retreat them. But this time around, you will not be retreating the entire volume. So if we go back just for a second, you will realize that if I go back and treat this patient, based purely on MRI, there's a good chance that my targeting volume will be much larger than the small contained blue area. So our standard practice is, if we're retreating an area, and we have contrast clearance analysis on it, then we will go back and target only the region shown in blue here. And that I have found reduces the targeted volume down to almost 40% to 50% of what it would have been without the evidence that the contrast clearing analysis provides us.

    And although it is early to say this, but we have not seen any significant change in failure rates in those patients. So it's not like we're missing disease by containing our field within an area of enhancement, which traditionally would have been in its entirety targeted with the retreatment, sometimes targeted with a hypofractionated retreatment, so that we split those down just a little bit and reduce the likelihood of adding more adverse effects from the second round of radiation. And then as [inaudible 00:34:42], the Elements, so it has actually a suite that has many features in it, and it's certainly not a one-trick pony. So, moving away from contrast clearance for just a few minutes before I conclude this presentation, here's an example integration with a patient with essential tremor undergoing a thalamotomy.

    The Elements software was able to map the nuclei, the dentate, the red nucleus, the ventralis intermedius with great speed and accuracy. And we were able to analyze the fiber tracts using the DTI component of Elements without requiring a ton of post processing and certainly not involving a whole bevy of radiologists. So this was done just right in the Gamma Knife suite, performed by fairly uneducated people like myself, who are not trained radiologists. So here is the track. Here is the targeting, we targeted with fairly standard atlas-based coordinates. That's the standard practice for Gamma thalamotomy, and 130 gray dose. And here you can see the targeting on the left panel, you can see the lesion as it appeared at six months and the lesion at one year, a patient has completely resolved his essential tremor on that side, and there is no weakness.

    Although the size of this lesion is starting to worry me just a little, I'm hoping we reach the maximum lesion point and it's going to turn right around. But at this point, I'm happy to say the patient does not have any adverse side effects. But what happened to the tractography? So it's interesting that you see the fiber tract that I showed you earlier, I can go back and remind you, that's the left-sided DRT on this patient when the patient had tremor. And here is the DRT after thalamotomy. So even though we can't use the tractography as a tool for targeting, although you're capable of converting the tracts that we mapped into a structure and exporting them to your radiosurgery software, it's certainly not how we planned this. But it's interesting also that the VIM nucleus and the targeted lesion, you can map it and you can see in three-dimension how beautifully the two correlate. And so we treated the medial VIM, which is exactly what our intent was. And here we are with the reduction in the tract.

    So as you can see, it's not just one part of Elements, we actually have used multiple aspects of the Elements software. It's all integrated into not just our radiosurgery practice, it is also integrated into my practice as a radiation oncologist. And as I said, 100% of our brains get processed through Elements, we're auto contouring and identifying the normal structures, as well as for delineating target anatomy, which is then shipped over to our external planning system and checked for accuracy and then handed over to those planners for planning. So needless to say, my residents are now experiencing some atrophy of their skill set when it comes to delineating model structures. There are certainly some shaping and correcting that needs to happen.

    But the idea that you can quickly segment the entire brain in matter of just a couple of minutes and have a lot of functional anatomy also auto segmented is a tremendous value add to us. It saves us a lot of time, we were spending a lot of our time in exactly that aspect of radiosurgery. And to me, the more quickly I can achieve a plan, the more quickly I can then treat the patient, which means I can make room for more patients to be treated with our existing technology. And for us, that is a certainly a very critical part of our practice. With the newer optimization and planning software that we are in the process of acquiring for our Gamma Knife, the GammaPlan Lightning software, even the planning portion of this will become considerably faster.

    As I'm sure other webinars will cover, Elements has an expedited path for planning multiple lesions together in a single radiosurgical screen, and I'm sure other users who are more familiar with that aspect of Elements. And the planning software, we'll be discussing that with you. From our point of view, this is a Elements integration into GammaPlan, and it really works very well for us. We were able to, as I showed you, follow the response of these patients, detect failure early, use that failure to trigger either retreatment or, in the case of adverse effects, therapeutic interventions to correct the adverse effects. And we have also fortunately cleared the hurdle, which I get a lot of questions on with our radiology department. We are, as I said, a high volume practice. Clearly, it's symbiotic for them to have our business as much as it is necessary for us to have time on their machines as well as their expertise.

    And so we now have a standardized ordered set within our EMR. And any of my team members can order a contrast clearance analysis scan with a click of a button, and the technologists know to hold the patients in the waiting area, put them back on the table, usually after completing one or more scans to follow and quickly scan. We do not have any reimbursement or billing codes for this that we could share with you. We do it as a part of our idea that our goal is to provide the best possible care for our patients. And certainly, I think Elements really takes that bar and raises it. It allows the user to develop a fine tune sense. If you're a beginner and starting to do radiosurgery, having one more way of knowing where there's recurrent tumor versus response and adverse effect is a great help. We've all scratched our heads, we all think there's a lot of subjectivity in this. And certainly, this provides a modicum of objectivity.

    Not to say that this is perfect ,not to say that this will be something that will replace all of our other techniques, but we often correlate contrast clearance with a brain perfusion imaging. And I can tell you that when we see adverse effects and most of the radiation affecting our GBM patients, those are profusion silent, and ARPT and perfusion changes tend to be more correlated with the blue areas. So clearly, multiple things point to the same place. And in the world of glioblastoma multiforme where targets are bigger, you possibly can use multiple modalities to achieve these conclusions. But in brain mets, in particular, where the lesions are smaller and multiple, and followed over an extended period of time, having contrast clearance analysis provides a significant value add to our practice.

    Thank you for your attention, please stay safe, these are challenging times. And hopefully, every one of you is able to take care of the patients that we look after who need our help, no matter what comes and goes. And Brainlab and Bogdan will provide you with my contact details, please feel free to call me, contact me, email me, ask me about your challenging cases, ask me about input as to how a certain case is handled. And I have, on past occasions, even helped folks who are trying to interpret the contrast clearance analysis and make clinical decision, I'm more than happy to be a partner in that process, if that's of help to you.

    Bogdan: Thank you very much.

    Dr. Prasad: Thank you very much.

    Bogdan: Thank you very much for your lecture, Dr. Prasad. And we do have a few questions that hopefully you could answer. So maybe I'll start with a general one looking at management of adverse radiation effects. At least the people that answered the poll today, it seems to be a high utilization of Avastin, what is your formula on selecting what kind of treatment to apply in the management of AREs? Given that you have access to everything, and also, maybe you can discuss financial implications of different solutions.

    Dr. Prasad: So I still think that the utilization of steroids as a first pass is where everyone should go. They're readily available, they're inexpensive, they're time tested, and, in the short course, have minimal side effects. We do it under difficulty sometimes when there needed to be a prolonged course of steroids, or if we're not getting control on the adverse effect. That's not someone that we consider retreatment to be a solution either, especially if the CCA response is mixed. You could argue that persistence of edema is an evidence of persistence of disease, as much as adverse effect from the radiation that has been given. So that's always a difficult needle to thread.

    But generally, we start with steroids, and I'd say the vast majority of our patients are managed with steroids. Roughly twice a month in a mature practice, you will find a patient who has difficult amounts of edema, and for various reasons, is someone who will be benefited from one to three cycles of Avastin. The most we've used is six cycles of Avastin, typically two does the tract, and they are then able to settle down.

    You have to be careful, however, on the amount of disease you're seeing because if it's mass that is sizable and surgery wouldn't be a consideration downstream, then the Avastin decision is a little bit more complex, because once you're given Avastin, you're going to have to forego any surgical intervention, at the very least, from the most aggressive practices, two to three weeks, but most people wait four to six weeks before operating.

    So we have to weigh all of those factors when we decide, cost is certainly a factor. There's no question that we have to definitely weigh that for patients with either insurance to authorize it, or for patients with other more complicated situations. So I can't answer that and it's generally been successful getting authorization to use Avastin for this treatment.

    Bogdan: Right. Thank you. Raj Jenna [SP] has a very timely question. So maybe I'll start with that before we go into all the other ones. Has COVID-19 affected your ability to bringing contrast clearance analysis scans, he says that they rely on radiology to hold patients for 60 minutes, as you know, before scans and during this time, they have refused to do so. And how have you nailed that?

    Dr. Prasad: Well. We are fortunate, firstly, that our radiology department has had a long relationship with our practice, in particular. We started with one MRI scanner 15 years ago, we have 3 now. And I know that the financial model for that has depended a lot on our growing practice, and I think they're following [inaudible 00:45:58]. So we have a little bit of an advantage with our radiology department to begin with. So I've never had difficulty getting the 60, 90-minute delay scan, as long as we are able to request it in time. We are also able to give them as a sort of last minute clock, if a patient comes down for follow up and has a scan within the last 60 minutes.

    And I think a TRAM analysis or a CCA would be useful, so they go right back up. That's a little harder, but we've achieved it. As far as COVID is concerned, we really had no trouble because COVID tended to, in a cancer center, drop the overall volume for us. We are not a city with a big leap of COVID. So I don't know where Raj practices. I saw his question earlier. But I can assure you, if this was New York City, not only contrast clearance be a difficult thing to do, I think honestly having enough physicians to take care of these patients would be hard, because many physicians were really taken on to take care of COVID patients. So it's a difficult answer to give in general. We were lucky with what happened.

    Bogdan: We have a good number of questions related to trying to identify tumor from treatment effects. So we start with some of them. Rogels [SP] was asking to make subtle assessments of tumor recurrence in close proximity to blood vessels, you are requiring a baseline contrast clearance analysis scan prior to re-irradiation or radiosurgey. Is it your protocol to perform contrast clearance analysis scans for all radiation planning on your patients?

    Dr. Prasad: So, yeah, we would, in the ideal world, like a baseline, we don't. And as far as proximity of lesions is concerned, as I mentioned earlier, the calibration of this process, and there are those who actually designed the software who can tell you, the calibration is based on vessel clearance. So the vessel should not light up as a spurious result here. Except if they are tumor vessels in which the flow is slower. Neovascular directories tend to have slower flow and lingering contrast. And those are essentially the same process as ARE. And in terms of actually picking out a mixed response, I saw a bunch of questions about thin rooms of blue and small areas of blue.

    So the answer to that is that if you go back and ask your surgical colleagues who have removed both necrotic looking masses, as well as recurrent tumors, they will tell you in both those situations, the pathologist actually says that there is a mixture of response. The tumors that look frankly recurrent showed a lot of necrosis and [inaudible 00:48:34], and the areas of frank so-called radiation necrosis, when resected surgically, also showed viable tumor. So that's exactly what the CCA is indicating to you, any field, as a mixture of both tumor cells and areas of the brain and tissue that have been adversely affected by either the treatment or the tumor as an involution necrosis. So that's really what CCA shows.

    So essentially, it becomes a judgment call. If you see areas of thin mixed blue, I generally regard that as a mixed response, and I follow those patients further along. I don't act on them, unless there is only one area and it is significant, or the pathology I'm dealing with is such that the other systemic options are limited, then I may get a little bit more aggressive, especially if it's a small nodule of blue. But in general, mixed response is typical, and that's why I showed you so many cases of evolving response, and over time, you will see things get better.

    Bogdan: There's another question. How important is it to perform a contrast clearance analysis scan before radiotherapy or radiosurgey treatment?

    Dr. Prasad: Like I said, a baseline on every patient would be ideal. I think if we did that, we would then run into difficulty without [inaudible 00:49:49] because we're treating so many patients. And even if we weren't, honestly, we can't get every single test on everything. I can't get DTI on everybody, but I'm at least now able to get DTI sequences on everybody with trigeminal, everybody with an acoustic in the hope of accumulating data in addition to our patients with thalamotomy. So I think it's always a volume question. I think getting a baseline on everyone is very difficult to achieve.

    Ideally, yes, the fact that at least in my satellites, we have satellites two and three hours away, these protocols are now available. And people ask a lot of questions about whether MRI manufacturer is supporting. They don't need to. It's a similar scan to what they're always doing. There is just simply a delay of one hour and repeat of the same sequence. So it is innominate to the manufacturer of your magnet. We have in the system, GE, Siemens, everything that could always work. So my colleagues in the satellites are even able to obtain the studies and just ship them to me, and we can perform the analysis right here on our station.

    Bogdan: Right, and thank you for answering that question, too. Let's move on to...and so you showed a lot of brain mets and glioma tumors, how effective is contrast clearance analysis for other brain tumors, and do you have radiology readings on other types of tumors?

    Dr. Prasad: So we have not yet done it, for instance, on benign tumors. So partly because the incidence of adverse effect is much rare in those patients, and we have a fairly good sense, those are not tumors in whom the tumor change and the radiation change can be mixed as easily. So we've not done it in those. Other malignant tumors tend to be a smattering of cases. So I'd say that the very fact that we have a large volume of metastatic disease, analyzed with contrast clearance, I think that's our value add to this database.

    GBMs were the initial subject of this study and we continue to see that in our GBMs. But that's not adding new information. I think the metastatic disease is new info. Should we be using it for other conditions, we would like to see it used in our vascular malformations and in other cases, where we'll basically be looking at fast flow through the vascular areas versus slow flow in an area possibly affected by radiation. But we have not yet reached the mature data to share with you. We're looking at it, but the big category we applied it to is brain mets.

    Bogdan: In regards to comparison of the contrast clearance analysis technique to other modalities, how do you see the CCA technique against things like standard T1s or perfusion, and Dr. [inaudible 00:52:40] was saying that methionine PETs is better distinguishing tumor recurrence versus necrosis. So how do you rate all these techniques in place?

    Dr. Parsad: Hello to Masaki, [SP] thank you. This is how we're gonna meet looks like for the next several years, because we've shared many a platform and talked about these cases. I do agree he was early in pointing out defining that as a useful agent. He talked about it, we've all tried to implement it. Believe it or not, getting new software is easier than getting a new isotope. And I have not put together slides to share with you, but we're doing a lot of gallium 68. That's for malignant meningiomas, anaplastic meningiomas. And this ties into the previous question. The next big discussion that we will probably have is the application of contrast clearance along with [inaudible 00:53:29] PET in atypical meningioma size. So that's where I see some value add there.

    And Masaki, we don't do methionine PET. You asked about Bev therapy, Bevacizumab is essentially a raster. We talked about that. And uncertainly, we've done correlation with blood flow, CBF studies, APT analysis, and perfusion. And hyperperfusing areas do correlate APT areas, do correlate the blue areas. I think that because CCA is derived from a high resolution 3D dataset, T1 series, its resolution, especially, is actually much higher than many of these other [inaudible 00:54:07.814]. So in a way I like the CCA, if I can trust it, which is the trust we've been trying to establish and share with you, then honestly, it gives me a much finer resolution three-dimensionally, it actually gives me a target direction, I can draw it and I can radiosurgically treat in with steep gradients, sparing areas of adverse effect.

    That's something that we just can't do with a PET and even with our [inaudible 00:54:33.057], but yes, there is correlation. I also heard a question which I think must be from an American consumer because it had the word medicolegal in it. And the question was, do I have a radiologist read it? Well, radiologists are slow to learn new things, but they're likely, they have been talking...presented this at multiple tumor boards. And both my neuroradiology colleagues understand and like the idea of this contrast clearance analysis, will occasionally send me pictures that they have analyzed using their own software, try to tell me that I don't need Bogdan's story story there, but in general, I like the fact that I can do it myself. So, yes, they are on board with this.

    Since it's not a DICOM friendly set, I can't push it back to PETs, and I can't have them over read it. But if we could, and that's, I think, for Bogdan to solve, that if this can be re-compressed into a DICOM friendly set, it will help a lot of clients. It can certainly help me get radiology to incorporate it into their reading. And they would obviously caveat it with the limited data we have. But that could be some protection for those of us who are asking. In the meanwhile, if you have a case that you're concerned about, one good way is to do a peer review, send it over, let us look at it, have Brainlab coordinate some way of getting these pictures over from one user to another, and then we can provide each other some support in terms of operating one.

    Bogdan: Yes. And at Brainlab, of course, we'll continue to facilitate these kind of technologies. And there are some questions pertaining to how you can use some of this information in the active treatment planning. The first one was, whether you're considering some sort of differential dose painting based on some of these maps. And then the second question for retreatment with hyperfractionated techniques, how do you apply a margin to your TTDs?

    Dr. Prasad: So anytime the volume of retreatment exceeds the initial treatment or crosses...and this is a rule of thumb for us, I mean, if I'm going to cross about two cubic centimeters of retreatment volume, this comes from a lot of ABM data, years of work with Professor Steiner back in Virginia, that showed us that retreating those volumes can have difficult consequences.

    And so we adjusted those in the past, now with the icon, we actually had to fractionate sometimes. So that is certainly done. In terms of the other question, which is whether or not you can use this to actually create a smart dose plan, well, it's interesting because as the icon [inaudible 00:57:10] become available, one of the major motivators of the next generation planning software is dose painting.

    And one of the requirements for dose painting with radiosurgery will be a biological activity map. That is going to be difficult in most conditions. But I see contrast clearance as perhaps a useful tool in retreatment to create cold and hot regions in a treatment plan, not only for writing steep gradients outside the blue into the red, but perhaps cooling the red down more effectively, so that we can provide more quick without providing more injury.

    And so I think your question is spot on, you can already, with the contrast clearance tool that you have, grow into doing this, send them over to your planning software, and plan just the residual or recurrent tumor, and be as careful as you can with the adverse effect region, that we already do. But in terms of more creative planning, and actually having a dose plan three-dimensionally accounts for all those major issues, that will require actually software and hardware activities, which I think are pretty much the next holy grail for any radiosurgery provider.

    Bogdan: Great. Dr. Prasad, thank you very much. Always a pleasure to have you present. And I'd like to thank all of you for participating today. Stay healthy, and we'll see you on a future webinar. Thank you and goodbye.

    Dr. Prasad: Thank you, everybody.
    Info
    Title:
    Contrast Clearance Analysis for Differentiating Tumor From Treatment Effects
    Topic:
    Cranial neoplasia (benign)
    Year:
    2020
    Speaker:
    Prasad Dheerendra
    Language:
    English
    Category:
    Duration:
    58:41
    Featured videos