So I'm going to talk to you about radiosurgery for brain metastases, try to make up some time. Here you can see the time. The stated time is 9.35 a.m. for this talk. This is my disclosure here. Learning objectives to review prognostic factors for patients with brain metastases using treatment selection and to review the indications of toxicity of radiosurgery and whole brain radiation for brain metastases. I'm going to focus on whole brain radiation therapy and radiosurgery because I think it's essential to understand how these two modalities are used in conjunction for the optimal management of brain metastases. This is my personal forecast of the evolving relative roles of whole brain radiation and radiosurgery. So this is an opinion slide. But as we know, whole brain radiation therapy has a long history, about 50 years, in being used for treatment of brain metastases. But with the introduction of radiosurgery, we've seen that there's been a gradual and maybe more of an acceleration of the adoption of this modality. And with more sophisticated technologies and more targeted therapies, I believe there's going to be an exponential increase in the utilization of radiosurgery. I don't think there's going to be an elimination of whole brain radiation anytime soon, but there's a declining utilization. Secondly, there's going to be a role for patients with leptomeningeal disease, uncontrolled disease, and low performance status. So it's important to judiciously use these two modalities when appropriate. This is a demographic slide demonstrating the aging baby boomer population that I'm sure many of you are familiar with. And what this portends for is that there's going to be increasing incidence of cancers and therefore increased incidence of secondary tumors from their cancers. And therefore, we're going to be facing this increasing problem of metastatic brain disease. This slide is courtesy of Kathy Gilmore, who's in the audience right now. I show this slide just to show the increasing utilization of radiosurgery for brain metastases worldwide using GammaKnife as a surrogate platform for all types of radiosurgery platforms. But you can see the increasing utilization all the way up to 2013. I imagine this is going to continue, this trend, you know, in the following years. Another trend to be aware of is that our colleagues in medical oncology are developing more effective systemic therapies that are translating into increased CNS metastases. One example is Receptin, which is unmasking a population of CNS progression leading to morbidity and mortality. Whereas previously this wasn't noted as much, now it's really a noted problem amongst breast cancer medical oncologists, and so we see this trend. Now we're seeing a similar trend in melanoma patients, where previously there really weren't very many systemic agents. Now we're seeing very exciting agents like anti-PD-1. We're seeing BRAF inhibitors. We're seeing ipilimumab. So all these systemic targeted therapies are going to need to be integrated with the management of the CNS, and the radiosurgery certainly plays an integral part. Important to remember that when you're talking to a patient, whole-brain radiation therapy doesn't completely eliminate all the brain metastases, even though it comprehensively treats the entire brain. So even after delivering a course of whole-brain radiation, there is the situation where you are dealing with progressive or recurrent metastatic brain disease, and these patients are living longer because of the systemic therapies we just talked about, and so there can be a reshowering of the brain. You can show this diagrammatically. You can give whole-brain radiation therapy, say, for five brain metastases, and then maybe six months to a year later, you now have four new brain metastases. So what are your options? You could do radiosurgery. You could consider repeat whole-brain radiation or, you know, palliative care, comfort care, or supportive care. There are multiple platforms that will be you'll get hands-on experience during this meeting, and these are very sophisticated technologies that are increasing access to radiosurgery and also the efficiency of radiosurgical treatments. Most notably, we now have the capability to treat high numbers of multiple brain metastases. Whether that's appropriate or not is a subject of intense debate. So there's this issue of technological feasibility versus evidence-based practice. Current sophisticated radiosurgery units, whether it's Gamma Knife, Linac-based, they allow for treatment of a larger number of brain metastases in a single session than was previously imaginable. So when Linac-based radiosurgery was in its infancy, you know, treating one was, you know, kind of a very new thing. Treating multiple brain mets was really on the fringe, and then, you know, really the cutoff was about three or four. But now we can treat five, 10, 20, 25. So we have this capability, but it begs the question, just because the technology allows us to do this doesn't mean we should. And so the challenge is to develop the next generation of head-to-head randomized trials. One such trial is currently going on at MD Anderson, which is looking at whole-brain irradiation versus radiosurgery in a randomized fashion. The primary endpoint here is not survival. Wouldn't expect there to be a survival difference. It's really neurocognition. And so the hypothesis is that by doing radiosurgery, you would have improved neurocognition as compared to whole-brain radiation therapy. Until we have this type of, you know, level one evidence, we're still going to continue to have these active debates. Another point to know is that we really have no level one data at all for addressing recurrent brain metastases. So really, it's left up to the clinicians and teams' best judgment. This evidence-based guideline done by the AANS really, you know, showed there's really no level one data. So it's all retrospective data. So for the trainees in the room, this is a very fruitful area for investigation. Recurrent brain metastases, so patients who are living longer, we have more effective systemic therapies. If you want to develop an academic career, this is a very good, you know, area to focus on because we really don't have good evidence on how to manage these patients who are living longer and longer. Now, the issue of volume of brain metastases is something to consider. One thing is number, but what about volume? And so, do tiny metastases, say five of them, is that equivalent to one big one? Is, you know, what is the better scenario? We really don't know. This paper from the University of Pittsburgh seems to suggest that there is a survival benefit using radiosurgery for four or more intracranial metastases, but it's the total intracranial volume that was most significant predictor of survival. And total volume was more important than the number of brain metastases and should be considered in identifying appropriate radiosurgical candidates. This is a very recent paper from Japan by Yamamoto and colleagues looking at radiosurgery for patients with multiple brain metastases. It's a multi-institutional, prospective, non-randomized observational study done at multiple Japanese centers. And what this trial or study seek to show, that treatment of five to 10 brain metastases initially with radiosurgery was not inferior to patients with two to four brain metastases. And so, this is the trial looking at survival here, and you can see the two to four brain metastases in blue and the five to 10 is in red here. Another important point to look at in this trial is what is the incidence of new lesions developing after radiosurgery, six months, 12 months, and 24 months. And so, if you look at the incidence for the patients who had two to four brain metastases, it was 40% to 65%, and it was slightly higher in the five to 10 group, 45% to 72%. So not terribly worse. It's very, I guess, this is a provocative study, and I think it's starting to open the door to treat higher number of lesions. It's my opinion that it's probably not the number of lesions that's so important as a cutoff, but the total intracranial volume. And the number of lesions is three, one to three, or one to four versus more, was kind of set up historically because of our technological limitations. That's really where that comes from, but we really need to look at this from a statistical standpoint and look at, see if there's statistic cutoffs that make sense. And so, this is kind of where the current state of the field is right now. Getting back to basics now, factors used to direct therapy, of course, we just talked about number, size, volume, eloquent areas. This has been talked about in earlier talks today. Total intracranial target volume, neurological deficits, if patients are symptomatic from their brain metastases, they have mass effect, those patients are going to be favored with surgical resection. Younger age, performance status, these are important variables that compose the recursive partitioning analysis, RPA, and more recently, the GP or graded prognostic assessment system based on the RTG databases. And the GPA also takes into account primary tumor now, so not all brain metastases are the same, even though most of them are based on non-small cell lung cancer, and we kind of treat them very similarly. The GPA, diagnostic-specific GPA, suggests that really they are different entities, and so when we design clinical trials, it should be stratified according to histology, or we should have histology-specific trials for brain metastases. And of course, extracranial disease is an important factor, and then finally, patient's input, what they want, surgery versus radiated surgery versus whole brain, these all need to be discussed. These are the ASTRO guidelines for brain metastases, which were endorsed by the Congress of Neurosurgeons, and these are published in the Practical Radiation Oncology Journal. There are key questions that are asked, like what prognostic factors are important for assessing and managing patients with brain metastases, and the RPAs is important. This is a classic system for assessing the prognosis for brain metastases patients. This is the DSGPA score of newly diagnosed brain metastases, so you can see there's quite a wide variation between a patient with, say, breast cancer, who has an overall median survival of 13.8 months, versus a small cell lung cancer patient has an overall median survival of 4.9 months. And then, so if you score according to the GPA, zero to one, all the way up to four, you can make an estimate of what that patient's survival might be. In terms of the guidelines, these two tables, this is the first one, is really the key table. If you read nothing else, just look at these tables. Essentially, you look at the patient's prognosis. You look at whether resection is possible or not, and then you look at the size of the lesion, and then these are your treatment options, and then these are the level of evidence that support these various treatment options here. So these are all the things that you could potentially be discussing with your patient. For multiple brain mets and the initial management scenario, again, you're looking at good prognosis, poor prognosis, size, presence of mass effect, and then these are the treatment options that are potentially available for patients in this situation. So RPA, GPA, some other stereotactic index for radiosurgery, BSBM, these are systems to be aware of for prognosticating brain metastasis patients. These are treatment strategies. Corticosteroids like to focus on, you know, pass on some teaching. When I was a trainee, I was taught, get the patients off the steroids as quickly as possible. I think that's still an important message to pass on to the residents because too often patients get put on steroids, no one's paying attention. They stay on them, nobody takes them off, and then they develop all these sequelae, they get cushing noise features, they get infections, they get, you know, cataracts, they get all these horrible things, they get mood disorders, and they have insomnia problems. So very simple thing is just if they're on steroids and they don't need to be on steroids, take them off the steroids, taper them off, or just take them off. Steroids very good in the short term, very bad in the long term. So I think that's a real important message because oftentimes they get neglected. Whole brain radiation therapy is the standard of care still for brain metastasis. Surgery plus whole brain radiation therapy for the single brain metastasis is standard of care based on the PATCH-L data published in the New England Journal in 1990. Surgery plus or minus localized radiation therapy. This is still cutting edge, although there are multiple single institutional reports talking about this approach. There is a randomized trial, which I'll talk about more, addressing this particular question. Whole-brain radiation therapy plus radiation sensitizers. John Su has led the front on this in investigating this area with multiple radiation sensitizers, including motexamine gadolinium and efraproxyrel. Unfortunately, these Herculean efforts have not led to a situation where we can say that we have a radiation sensitizer that improves overall outcomes. Whole-brain radiation therapy plus chemotherapy. No real high-quality evidence to support this. Whole-brain radiation surgery plus radiation surgery standard of care based on the RTOG 9508 study. Radio surgery alone, I would say, still is cutting edge. Believe it or not, when I go to various places and talk, there's still a lot of people who believe in giving whole-brain radiation therapy when you poll the audience. So this is still a very controversial area. And then chemotherapy in certain situations that can be offered, although there's really not as much data to support that. This is the Cleveland Clinic data looking at dose response data. This study is by Mike Vogelbaum looking at 202 patients treated from 1997 to 2003. And what you can see clearly is that when you drop the dose, your local control rates go down. And so here's the hazard ratios for local failure, 24 versus 15 or 18 gray. Patients do much better with higher dose, but we can only prescribe that for smaller tumors. And so hence, this is the actuarial curve showing the outcomes. And so this is why we say when we are facing tumors that are larger, 2.5, 3 centimeters or higher, we tend to want to favor surgery because we know our ability to control these tumors is worse with radiosurgery for the larger lesions, so we prefer to give them surgery. However, a very, very tiny lesion, less than a centimeter, I think most neurosurgeons prefer to go to radiosurgery. Possibly it's very hard to see those lesions described as trying to find a needle in a haystack. You can actually do a craniotomy and miss. See, the tumor's still there. I've seen this happen. So that's kind of the selection criteria. One to 2.5 cm, I think there's a lot of room for active debate between surgery or radiosurgery. Radiosurgery toxicities are quite infrequent and rare, but they do occur. So if you are a high-volume center, you will see these. Acutely, 24 to 48 hours, patients can develop severe headaches, nausea, vomiting, aphasia, motor problems, seizures, bleeding, strokes, swelling, necessitate them coming back to the emergency room. They can develop late radiation necrosis. They can also develop radiation-associated secondary malignancies, and even death can occur. So I think all of these things need to be talked about if you're gonna be complete about your informed consent. This is the classic RTOG 9508 trial, looking at whole-brain radiation therapy plus or minus radiosurgery. So whole-brain radiation therapy was given to the RTOG standard dose of 37.5 gray. With the radiosurgery, or just whole-brain radiation therapy alone. And what this trial showed was that there was a survival benefit for the application of radiosurgery in addition to whole-brain radiation therapy of 6.5 months versus 4.9 months. This is a Japanese study published in JAMA looking at the question of radiosurgery plus whole-brain radiation therapy versus radiosurgery alone. 132 patients. Is whole-brain radiation therapy worse than the disease or is uncontrolled occult disease in the brain the greatest risk? These authors concluded that radiosurgery alone could be a treatment option, provided that frequent monitoring of the brain status is conducted. These are their rates of failure. As you might expect, control rates are always better when you add whole-brain radiation therapy versus radiosurgery alone, but at what cost? This is the MD Anderson trial looking at radiosurgery versus radiosurgery plus whole-brain radiation therapy for one to three brain metastases. And with the primary endpoint of this trial was cognitive decline in learning and memory as assessed by the Hopkins Verbal Learning Test. There need to be a greater than or equal to five-point drop at four months from baseline, which is enrollment into the trial, utilizing this reliable change index here. And it's been determined to be both statistically and clinically meaningful. So this trial was halted early, just like the ARUBA trial we heard earlier today by the Data Monitoring Committee because there was a big difference between the two arms where there was a almost doubling or doubling of the risk of development of neurocognitive decline as assessed by the Hopkins Verbal Learning Test, a 52% versus 24% at the four-month time point. And these are the control rates of local control, again, better with whole-brain radiation therapy and distant brain met control, better with whole-brain radiation therapy, but worse neurocognitive outcome. This trial has been looked at in terms of the societal economic impact. So generally, when health economists look at interventions, things that are between the 50 to 100%, excuse me, $100,000 quality of life year, cost-effectiveness range is what they look for. This study showed that the application of RAID surgery and observation with subsequent neurosurgical management of recurrence was a reasonable treatment modality for brain metastases. So in conclusion, patients who were randomized to RAID surgery plus whole-brain radiation therapy had twice the risk of cognitive decline in learning and memory compared to RAID surgery patients alone. Now I just wanna review the spectrum of whole-brain radiation therapy for brain metastases and its indications based on metastasis volume. So we can look at one extreme where there's no visible brain metastases at all in a cancer patient. This is referred to as the prophylactic cranial irradiation. When would we do this? The only time we would ever do this is in a small-cell lung cancer patient. It's proven to be survival benefit by giving prophylactic whole-brain radiation therapy in a limited setting as well as extensive setting of their small-cell lung cancer. Otherwise, we really wouldn't do prophylactic cranial irradiation. There's been studies by the RTG looking at this for non-small-cell lung cancer, which I'll show you those studies, but really wouldn't, I don't think really anybody would do this. For small-volume disease, one to three brain metastases, radiosurgery plus whole-brain radiation therapy, maybe people are doing this still. It's still controversial. For large-volume disease, 10 plus brain metastases, should we be doing whole-brain radiation therapy? I think the answer is still yes. So this is the PCI study for non-small-cell lung cancer done by the RTG 0214. This is no visible brain metastases in a non-small-cell lung cancer patient. These patients were randomly allocated. There were 356 of these patients. These are the results. There was no difference in overall survival. There was a benefit in terms of disease-free survival with the PCI application. And there was also a benefit in terms of CNS metastases failure with PCI. But again, there was no overall survival difference, and I think that's what would need to happen in order for people to accept giving PCI. This is showing the deterioration of the neurocognition measured by the Hopkins verbal learning test with PCI for non-small-cell lung cancer. And so you can see, this is PCI group. This is observation. At three months, there was a 45% decrement or deterioration versus 13% in the observation, statistically significant. At 12 months, a longer follow-up, 26% versus 7%, and with a p-value of .03. This is showing you diagrammatically what's happening here. You're seeing this drop here in the PCI raw score. You're seeing a drop here in the delayed recall, and then you're seeing a rebound, and then you're seeing a drop again at 12 months, and you're seeing a drop here at 12 months. So sometimes people will argue, well, there's not enough time. You know, we're just looking at a four-month time point. There's rebound effect, but this study's showing, well, they rebound, but then they drop again. So, you know, it is having an effect. There's a European study looking at radiosurgery or surgery, randomized to observation or whole-brain radiation therapy. This is looking at the case-two scenario of one to three brain metastases, and this study was published in JCO and showed, essentially, that you cannot do surgery plus observation. So this yellow line here, which represents surgery plus observation, shows that probably an unacceptably high local failure rate here. So you have to give some form of adjuvant treatment, and when you give surgery plus whole-brain radiation therapy, it drops you down to here in terms of your failure rates. Excuse me, I'm on switch. It should be here. This is just showing your local failure rate with surgery alone and with local, the local control improves considerably with the application of whole-brain radiation therapy. So I think that's one important take-home point from this trial. The other thing is that there was no survival difference between the two arms in terms of overall survival or survival with a performance status of less than or equal to two. So that's kind of a surrogate for quality of life. So I think these investigators kind of concluded that, well, maybe whole-brain radiation therapy didn't need to be given. There are neurocognitive sparing strategies which are being practiced at various centers as well as the RTOG. One is the RTOG 0614, which is a trial that looked at randomizing patients to a drug called memantine versus placebo. There's a placebo-controlled randomized trial in brain metastases. Class one and class two stratification, whole-brain radiation therapy that is standard 37.5 gray and 15 fractions. Very rapidly accruing study. 554 patients over two years. They were well-balanced. And this resulted in this publication in neuro-oncology which shows that there is a decreased incidence of cognitive function failure and those patients were randomly assigned to get the memantine versus their placebo counterparts. And this was statistically significant at 0.01 with a hazard ratio of 0.784. So at least for me, I feel this is practice changing and so when I give whole-brain radiation therapy, I give memantine now. I give these patients an opportunity to have improved neurocognition based on the results of these trial. And the conclusions were that this is a well-tolerated drug and there is better cognitive function over time. It delays the time the cognitive decline, reduces the decline in memory, executive function and processing speed. Now this is the N107C trial which is very important because it's looking at the issue of radius surgery to the post-operative cavity. It's randomizing patients to either get radius surgery after a resection of a brain metastasis or whole-brain radiation therapy. So this trial is designed such that you're stratified according to age, number of brain metastases, extracranial disease, histology, cavity size. You either get radius surgery to the surgical bed and then radius surgery to any remaining unresected lesions or you get whole-brain radiation therapy and then radius surgery to any remaining unresected lesions. And so this trial is currently ongoing and its assessments are neurocognition, looking at the Hopkins verbal learning test, the control oral word association test and the trail making test A and B for attention, processing speed, executive function. It also has quality of life endpoints here, MRI evaluation and also some serum and urine evaluation as well. So the primary goal of this study is to look at neurocognitive progression at six months post-radiation with radius surgery and see that if there's 85% power to detect a 20% difference in the neurocognitive progression free at six month rate. And also to ascertain if there's an improved survival with radius surgery. So very interesting study. These are the guidelines for prescription dose for radius surgery based on the cavity volume here. And so it's nice to have some sort of standard that we can go by in the meantime. And finally, I will just talk very briefly about adding margin to cavities or post-operative cavities. So this is a study from Stanford which looked at doing resection cavity radius surgery and what the outcomes were by adding a margin to that cavity. And what they found, I think, in a previous study was that with radius surgery delivery where there was less conformality, they seemed to achieve better results. And so that was probably because there was more dose being given to the periphery of the cavity. And so they formalized this and looked at, well, if we add a margin now to the cavity, what does that do to our local failure rate versus no margin? And you can see there's a much lower cumulative incidence of local failure with the application of a margin. So that's all I have for you and thanks for your attention. Thank you.

radiosurgery

brain metastases

treatment

whole brain radiation therapy

cognitive decline

individualizing treatment