Transcript
I'm Nader Pouratian. I am from UCLA in the Department of Neurosurgery. I'm pleased to introduce this symposium sponsored by Brainlab, the Novalis Circle on technological advances in the treatment of brain metastases. I'm joined up here by Cory Knill from Beaumont Health as well as Yel Mardoor [SP] from Chaim Sheba Medical Center in Israel.
I'll get started with...each of us will give 20-minute presentations. And hopefully, we'll have some time for questions at the end. So I'll be speaking about improving radio surgical outcomes, going through the spectrum of the timing of radiosurgery planning, tracking our outcomes, and analyzing outcomes through various initiatives that we have in collaboration with Brainlab.
So the UCLA approach to treating brain metastases is very similar to many other centers in that radiosurgery does come first. But there're certain situations where surgery comes first and we do surgery plus radiation. In cases where we need a tissue diagnosis, for example, in this case with a patient with remote history of renal cell carcinoma greater than 10 years prior, we didn't know what this was, it was also a large tumor, so not a good candidate for upfront radiation, and then large and symptomatic tumors like this patient who has a history of salivary cancer who now presents with seizures as the mental status changes.
And we know quite well from our history of using adjuvant stereotactic radiosurgery what we expect outcomes to be, which is somewhere around 70% to 80% control rate at 1 to 2 years with a certain rate of radiation necrosis on the order of 10% or so. And there's a certain rate of leptomeningeal disease as well because when these tumors are not taken out en bloc, there's a chance of tumor spillage. And that's the same that we've seen in our series at UCLA about a 15% rate of leptomeningeal disease. And if we take that information along with what we know from other series, which is that the difference between postoperative radiosurgery versus postoperative whole-brain radiation from metastases, the major differences in terms of distinct control rates with out of field failures as well as if we look at preoperative SRS.
So if we do radiosurgery upfront versus post-op of whole-brain radiation, we can control leptomeningeal disease where there's a higher chance of leptomeningeal disease when we do postoperative stereotactic radiosurgery. So this leads us to the question of the challenge of the timing of radiosurgery the way we do it when there's surgery required. Postoperative radiosurgery has a risk of leptomeningeal disease. There's challenges with target delineation as we see in this case where preoperatively the tumor is very well defined, but postoperatively you have a large resection cavity, you have enhancement that can be postoperative enhancement versus residual tumor, and then you have a vascular compromise in the region also that can affect the efficacy of radiosurgery.
And so one potential solution, which you've heard about earlier in this meeting, and I'm sure many of you heard about is the use of neoadjuvant stereotactic radiosurgery with better target delineation. A decreased treatment volume as well because generally the resection volume or the resection cavity tends to be larger than the actual tumor. We can potentially avoid fractionation because we're lowering the dose, decrease the risk of tumor spillage in leptomeningeal disease. There's also a cost-efficiency factor because you only need to get one MRI. It's very efficient because we go from treatment to surgery, in our case, in one day. And it's convenient for the patient as well. There's also the risk, and it's happened when we have surgeons who don't participate in the radiosurgery program, where someone will resect the tumor, and then they're lost for two or three weeks, sometimes six to eight weeks before they get radiation. And so there's a decreased chance of that occurring as well if you get preoperative radiation.
So in the series that have been published, and we've heard a little bit about it at this meeting as well. Local control is not very different with preoperative SRS versus postoperative SRS, but what we do see is a decrease or the early studies show there's a decreased rate or probability of radiation necrosis in patients who get preoperative SRS. Probably because we're taking out a large volume of the tissue that's radiated, as well as a smaller treatment volume. And then, the risk of leptomeningeal disease seems to be smaller as well. So just to explain that, you're radiating these tumor cells before they spil so you don't have to go chasing them at the time of radiation.
So this is the approach that's been described and we use at UCLA, which is a 20% dose reduction from RTOG dosing standards, which allows us to minimize fractionation, we use no margins, minimizing dose for normal brain. And we'll do surgery, it says one to five days post-SRS, but generally, we actually do it the very next day after SRS. And there's various schema for doing trials. And we have our own trial that we've just activated at UCLA for randomizing patients who do require surgery to pre versus...or neoadjuvant versus adjuvant radiosurgery with primary endpoints of leptomeningeal disease as well as secondary endpoints as described there.
The other interesting thing from a scientific standpoint is to gain insight into the radiobiology of treatment in that we now have tumors that will be treated with radiation the day before as well as tumors that have not been treated. And we can start to look at different patterns of gene activation that can be attributed to the radiation treatment itself. So while we do have a trial, there are certain lesions that I would say are particularly high risk and deserve special consideration for neoadjuvant therapy, which would be very large tumors that cannot be resected en bloc. Cystic tumors, which there's a high rate of spillage of tumor. And in our experience, posterior fossa tumors seem to have a higher rate of leptomeningeal disease as well.
I've just talked about the timing of radiation. I think there is a big move to work towards this neoadjuvant radiosurgery, and that could potentially improve our outcomes from radiosurgery. There are other things we can do to improve radiosurgical outcomes as well. And a lot of it has to do with on the software planning and the treatment planning and treatment administration side. So I want to talk a little bit about the Elements Multi-Brain METs program that Brainlab has provided. And I think, you know, as I look at this checklist that the Elements Multiple Brain METs interface provides, we at UCLA have, as many of us do, a very high level of checks and quality controls before a patient ever gets treated. It's all very manual, but we are religious about it.
But we are all human and we're all susceptible to errors. And in the 10 years that I've been a part of the program at UCLA, errors have still been made despite the fact that we have checklists. So to work towards an automated program, a system that actually does the checks for you and will give you indicators of acceptable thresholds or not is absolutely imperative in terms of improving outcomes and improving the safety of the treatments that we deliver. It also ensures more consistent outcomes. So we've all heard about the checklist manifestos and making sure that everything is within protocol.
So the problem really is, when everything is manual, there's a chance that you fall out of protocol, and this is not particularly specific to radiosurgery but any kind of treatment protocol and there is a chance that you can have major deviations from a certain treatment protocol and that it can affect the outcomes of treatment. And it seems that centers that are low volume or centers that are newer tend to have a higher probability of making these types of errors. And there's a lot of literature that supports the fact that high-volume centers have better outcomes is probably related to deviations from treatment protocols.
So that's the problem that we face and we're looking for a solution. And the Multi-METs Elements that Brainlab provides does help us in terms of being more consistent in automating this process, and also increasing the efficiency of treatments. So we have improved patient experience because we're treating multiple METs in the brain as a single ISO center with significant reduction treatment times. You know, when we used to treat 8 to 10 METs, not only 2 hours, but sometimes we'd have people come back 3 or 4 days in a row to have all the different treatments because it took so long. Now we treat up to 10 METs in about 20 minutes time.
There's provider benefits to us because it's more efficient in terms of getting the patient positioned and fewer adjustments between fractions and increased throughput and eligibility also. So there's decreased time on the treatment table means you can treat more patients. There's decreased time in treatment planning means that you can treat more patients. Everything increases efficiency. But that increase in efficiency and throughput is not at the cost of compromising on patient care or quality of the plans. In fact, it's probably the opposite, it's better quality of plans.
So just gonna fly through some numbers here. In terms of treatment planning time, you almost can't see it, but you can see this is the number of METs on the X-axis. You go from 1 through 16 METs. And, you know, to plan a case with 10 METs, this says about an hour's time in the old software system. And now it's about 10 minutes which is true, based on our experience. And it does not increase that much based on the number of METs that you treat. But that's, you know, a multi-fold decrease in treatment planning time. Treatment delivery also, whereas it used to be pretty linear, the more number of METs you treat, the more time it takes, now it's much flatter over time in terms of the number of METs...the amount of time needed for the number of METs treated.
Treatment attendance time. So both physician and physicists, you can see a significant reduction in amount of time of personnel. And personnel are some of our most expensive costs. And increasing our availability increases our ability to deliver care to more patients. And, you know, all this is important, but then you need to make sure that when it comes together, when you improve, when you implement all these changes, that it actually turns into savings and time as well as sustained improvements. And I can say from personal experience that this is true. But you can see in this type of paradigm, which is illustrated here, when you implement various changes and you see sustained changes over time with implementing the software and various changes along the way, which we're gonna look at here.
So again, I wanna emphasize improved efficiency, which is extremely important, but it's not at the cost of quality. So we have...and, in fact, it's the opposite, it's improved plan quality. So this is an analysis that was done by my colleagues, just comparing 21 patients with 114 lesions and comparing what it was with actually the earlier version of Multi-METs versus the newer version, which is now in use. And we replanned everything. It took us less than three minutes per patient for the replanning. So if we look at a couple of plan metrics, we look at Normal Brain Dose finds, we have a lower V12 with the newer software, we have a lower Conformity Index, and a lower Gradient Index. So we're improving our plans.
And then what we also see is, you know, we started...we're a very cautious group at UCLA and we were skeptical of how accurate this might be. So when we started, we added a 2-millimeter PTV to all of our targets, all over the tumors that we're treating. And now with improved validation, improved confidence in the system, and also looking at our clinical outcomes, we now have changed to basically 1 milliliter PTV or, if we're clustered, very close to another 0.5-millimeter PTVs. And that has significant implications going down from 2 millimeters to 0.5 or 1 millimeter in terms of treatment planning, in terms of the dose to the brain, which we can appreciate here.
You see over on the yellow, a 2-millimeter PTV significantly will increase...Let me actually go to the next slide because it'll show it better. If you look on the left side of the screen, let's look at the yellow bar, the V12 volume. When you have a 1-millimeter margin is 14.9 CCs across 51 targets. But if we go to 2 millimeters, it's nearly twice as much, it's 70% higher at 24 CCs. And then you can look at the dose impact throughout the rest of the brain. And then we can also, now if we are a little bit more confident what we do, and we use a different margin...so the tumors that are less than 6 centimeters from the isocenter or from the center of the treatment area, we can even get more conservative or safer treatment volumes to the whole brain.
So it's great that we are improving both, you know, the timings that we decrease our leptomeningeal disease, we're improving...hopefully, gonna decrease our radiation necrosis rate by doing neoadjuvant therapy, to be supported by future studies. We're improving the efficiency of our care, the quality of our plans. But really what we need to do is track our outcomes. And although there's always a great desire to do a randomized control trial to support everything we do, there's power in numbers. And there's been a significant investment, both on the corporate side as well as from Neuropoint Alliance, which is a neurosurgery quality outcomes group, which is primarily driven by the AANS. And then, also investment from corporate with Brainlab making the biggest investment to develop a registry that in its first iteration was very good, and its next iteration is much easier for people to participate in, contribute their data to.
Now, should highlight that this software or the registry is actually available that people could use this and Brainlab can make it available to use it internally for measuring outcomes locally. Or you can actually subscribe and participate in the larger network and contribute to our national quality database so we can look at outcomes. So these are the participating centers right now. There are several throughout the country and the Neuropoint Alliance, as well as Brainlab, are still actively looking for additional sites to expand upon this because we want more data.
Currently, there are 23 sites. There's over 4000 patients that have had treatments entered and then follow-ups as well. So it takes a little bit of a strategy to get a center up and going. So you can see the number of events uploaded. It's still over 4000 and what we are focusing on is still new events, but also follow-up events. Because just knowing that someone is treated is no good if you don't know what happens to them afterwards. And the number one event or the number one type of treatment that's being deposited into the registry are metastatic tumors, which is why we're here talking about this. And again, it's because it's the number one thing that we treat with stereotactic radiosurgery.
And there are pathways, and I'm not going to go through all of it because I can't, but there are pathways that integrate, especially if you have a Brainlab system that integrate with the uploading of data into the Quentry, which is a cloud-based system for depositing the data in a HIPAA compliant manner. And then there is another interface for not just new patients, but for uploading new data, so follow-up data. And the follow-up data, the real significant advance here is that there's auto contouring, and it starts matching up previously treated tumors with newly found tumors, or basically, you can track the same tumor a long time, which is really important to know the response to treatment.
And then there's a very friendly graphical interface where you select subjects, there's a data upload through images, there's easy interface for updating information. The way we deal with this at UCLA, and I think it's going to happen more broadly, is that we use a template so that we know that we collect the same information in every single subject so that we can...it's not only good for registry but it's good for our own tracking of patient outcomes. And you can enter this data very easily. So it has both patient information and treatment information. And a lot of this can be uploaded automatically from the planning software.
And why do we do this? We do this because we want to actually analyze our outcomes. So this is the initial data that we have looking at survival rates and people treated and patients with metastases are treated with stereotactic radiosurgery. Ideally, where we want to get this to is to then compare it with other outcomes as shown there. And if we can show that we are improving what we're doing, or we have specific impact in a certain subgroup of patients, then it can drive reimbursement. And eventually, with Medicare and payers going towards quality-based reimbursements, it might be something that actually has to be required. So tracking outcomes is gonna be extremely important.
Also, internally as well as nationally, it's good for quality monitoring to make sure that looking at compliance with treatment protocols. Making sure that all the treatments steps are met, and then comparing outcomes for people for different treatment patterns to assess whether different treatment protocols results in different types of outcomes. And then we get into this field of big data analytics where registries are actually useful. Again, we all like our randomized control trials, but there's a lot of data to be gathered from real human, real-life experience.
We can...and this talk is beyond big data discussion or explanation of what big data is. But when you start getting into multiple thousands of patients and multiple time points, you can start looking for drivers of therapy and drivers of outcome, whether it be based on the location of the tumor, the type of the tumor, the timing of the tumor, the disease status, other factors that we can't even start to think about the interactions of which could drive outcomes. And then looking at, for example, what doses might actually optimize outcomes, you know, I think every center generally uses the same doses but they might be a little bit different, and looking at how that might affect outcomes as well.
And this is something I was just speaking with a colleague from USC, looking at the probability maps of where tumors might be in the brain. But this is something that could be done as well where if everyone is depositing their data into a large registry, you can then start to map them out because we have spatial information with imaging information that's all annotated and ready to look at. And last but not least is gonna be toxicity evaluation. So we don't just want to know where tumors are, how we're treating, and what the best treatment is, but how safe we are and what we can do to improve the safety of our treatments.
So I am going to stop there just to summarize, you know, I think there's a lot we can do and we need to think about the continuum of care of these patients through the radiosurgery program. Thinking about before we ever treat them from the timing of treatment, the experience of the treatment, and then tracking our outcomes. I think there's a lot of tools here that we can work with to get us closer to that ideal of improving patient outcomes.
I'll get started with...each of us will give 20-minute presentations. And hopefully, we'll have some time for questions at the end. So I'll be speaking about improving radio surgical outcomes, going through the spectrum of the timing of radiosurgery planning, tracking our outcomes, and analyzing outcomes through various initiatives that we have in collaboration with Brainlab.
So the UCLA approach to treating brain metastases is very similar to many other centers in that radiosurgery does come first. But there're certain situations where surgery comes first and we do surgery plus radiation. In cases where we need a tissue diagnosis, for example, in this case with a patient with remote history of renal cell carcinoma greater than 10 years prior, we didn't know what this was, it was also a large tumor, so not a good candidate for upfront radiation, and then large and symptomatic tumors like this patient who has a history of salivary cancer who now presents with seizures as the mental status changes.
And we know quite well from our history of using adjuvant stereotactic radiosurgery what we expect outcomes to be, which is somewhere around 70% to 80% control rate at 1 to 2 years with a certain rate of radiation necrosis on the order of 10% or so. And there's a certain rate of leptomeningeal disease as well because when these tumors are not taken out en bloc, there's a chance of tumor spillage. And that's the same that we've seen in our series at UCLA about a 15% rate of leptomeningeal disease. And if we take that information along with what we know from other series, which is that the difference between postoperative radiosurgery versus postoperative whole-brain radiation from metastases, the major differences in terms of distinct control rates with out of field failures as well as if we look at preoperative SRS.
So if we do radiosurgery upfront versus post-op of whole-brain radiation, we can control leptomeningeal disease where there's a higher chance of leptomeningeal disease when we do postoperative stereotactic radiosurgery. So this leads us to the question of the challenge of the timing of radiosurgery the way we do it when there's surgery required. Postoperative radiosurgery has a risk of leptomeningeal disease. There's challenges with target delineation as we see in this case where preoperatively the tumor is very well defined, but postoperatively you have a large resection cavity, you have enhancement that can be postoperative enhancement versus residual tumor, and then you have a vascular compromise in the region also that can affect the efficacy of radiosurgery.
And so one potential solution, which you've heard about earlier in this meeting, and I'm sure many of you heard about is the use of neoadjuvant stereotactic radiosurgery with better target delineation. A decreased treatment volume as well because generally the resection volume or the resection cavity tends to be larger than the actual tumor. We can potentially avoid fractionation because we're lowering the dose, decrease the risk of tumor spillage in leptomeningeal disease. There's also a cost-efficiency factor because you only need to get one MRI. It's very efficient because we go from treatment to surgery, in our case, in one day. And it's convenient for the patient as well. There's also the risk, and it's happened when we have surgeons who don't participate in the radiosurgery program, where someone will resect the tumor, and then they're lost for two or three weeks, sometimes six to eight weeks before they get radiation. And so there's a decreased chance of that occurring as well if you get preoperative radiation.
So in the series that have been published, and we've heard a little bit about it at this meeting as well. Local control is not very different with preoperative SRS versus postoperative SRS, but what we do see is a decrease or the early studies show there's a decreased rate or probability of radiation necrosis in patients who get preoperative SRS. Probably because we're taking out a large volume of the tissue that's radiated, as well as a smaller treatment volume. And then, the risk of leptomeningeal disease seems to be smaller as well. So just to explain that, you're radiating these tumor cells before they spil so you don't have to go chasing them at the time of radiation.
So this is the approach that's been described and we use at UCLA, which is a 20% dose reduction from RTOG dosing standards, which allows us to minimize fractionation, we use no margins, minimizing dose for normal brain. And we'll do surgery, it says one to five days post-SRS, but generally, we actually do it the very next day after SRS. And there's various schema for doing trials. And we have our own trial that we've just activated at UCLA for randomizing patients who do require surgery to pre versus...or neoadjuvant versus adjuvant radiosurgery with primary endpoints of leptomeningeal disease as well as secondary endpoints as described there.
The other interesting thing from a scientific standpoint is to gain insight into the radiobiology of treatment in that we now have tumors that will be treated with radiation the day before as well as tumors that have not been treated. And we can start to look at different patterns of gene activation that can be attributed to the radiation treatment itself. So while we do have a trial, there are certain lesions that I would say are particularly high risk and deserve special consideration for neoadjuvant therapy, which would be very large tumors that cannot be resected en bloc. Cystic tumors, which there's a high rate of spillage of tumor. And in our experience, posterior fossa tumors seem to have a higher rate of leptomeningeal disease as well.
I've just talked about the timing of radiation. I think there is a big move to work towards this neoadjuvant radiosurgery, and that could potentially improve our outcomes from radiosurgery. There are other things we can do to improve radiosurgical outcomes as well. And a lot of it has to do with on the software planning and the treatment planning and treatment administration side. So I want to talk a little bit about the Elements Multi-Brain METs program that Brainlab has provided. And I think, you know, as I look at this checklist that the Elements Multiple Brain METs interface provides, we at UCLA have, as many of us do, a very high level of checks and quality controls before a patient ever gets treated. It's all very manual, but we are religious about it.
But we are all human and we're all susceptible to errors. And in the 10 years that I've been a part of the program at UCLA, errors have still been made despite the fact that we have checklists. So to work towards an automated program, a system that actually does the checks for you and will give you indicators of acceptable thresholds or not is absolutely imperative in terms of improving outcomes and improving the safety of the treatments that we deliver. It also ensures more consistent outcomes. So we've all heard about the checklist manifestos and making sure that everything is within protocol.
So the problem really is, when everything is manual, there's a chance that you fall out of protocol, and this is not particularly specific to radiosurgery but any kind of treatment protocol and there is a chance that you can have major deviations from a certain treatment protocol and that it can affect the outcomes of treatment. And it seems that centers that are low volume or centers that are newer tend to have a higher probability of making these types of errors. And there's a lot of literature that supports the fact that high-volume centers have better outcomes is probably related to deviations from treatment protocols.
So that's the problem that we face and we're looking for a solution. And the Multi-METs Elements that Brainlab provides does help us in terms of being more consistent in automating this process, and also increasing the efficiency of treatments. So we have improved patient experience because we're treating multiple METs in the brain as a single ISO center with significant reduction treatment times. You know, when we used to treat 8 to 10 METs, not only 2 hours, but sometimes we'd have people come back 3 or 4 days in a row to have all the different treatments because it took so long. Now we treat up to 10 METs in about 20 minutes time.
There's provider benefits to us because it's more efficient in terms of getting the patient positioned and fewer adjustments between fractions and increased throughput and eligibility also. So there's decreased time on the treatment table means you can treat more patients. There's decreased time in treatment planning means that you can treat more patients. Everything increases efficiency. But that increase in efficiency and throughput is not at the cost of compromising on patient care or quality of the plans. In fact, it's probably the opposite, it's better quality of plans.
So just gonna fly through some numbers here. In terms of treatment planning time, you almost can't see it, but you can see this is the number of METs on the X-axis. You go from 1 through 16 METs. And, you know, to plan a case with 10 METs, this says about an hour's time in the old software system. And now it's about 10 minutes which is true, based on our experience. And it does not increase that much based on the number of METs that you treat. But that's, you know, a multi-fold decrease in treatment planning time. Treatment delivery also, whereas it used to be pretty linear, the more number of METs you treat, the more time it takes, now it's much flatter over time in terms of the number of METs...the amount of time needed for the number of METs treated.
Treatment attendance time. So both physician and physicists, you can see a significant reduction in amount of time of personnel. And personnel are some of our most expensive costs. And increasing our availability increases our ability to deliver care to more patients. And, you know, all this is important, but then you need to make sure that when it comes together, when you improve, when you implement all these changes, that it actually turns into savings and time as well as sustained improvements. And I can say from personal experience that this is true. But you can see in this type of paradigm, which is illustrated here, when you implement various changes and you see sustained changes over time with implementing the software and various changes along the way, which we're gonna look at here.
So again, I wanna emphasize improved efficiency, which is extremely important, but it's not at the cost of quality. So we have...and, in fact, it's the opposite, it's improved plan quality. So this is an analysis that was done by my colleagues, just comparing 21 patients with 114 lesions and comparing what it was with actually the earlier version of Multi-METs versus the newer version, which is now in use. And we replanned everything. It took us less than three minutes per patient for the replanning. So if we look at a couple of plan metrics, we look at Normal Brain Dose finds, we have a lower V12 with the newer software, we have a lower Conformity Index, and a lower Gradient Index. So we're improving our plans.
And then what we also see is, you know, we started...we're a very cautious group at UCLA and we were skeptical of how accurate this might be. So when we started, we added a 2-millimeter PTV to all of our targets, all over the tumors that we're treating. And now with improved validation, improved confidence in the system, and also looking at our clinical outcomes, we now have changed to basically 1 milliliter PTV or, if we're clustered, very close to another 0.5-millimeter PTVs. And that has significant implications going down from 2 millimeters to 0.5 or 1 millimeter in terms of treatment planning, in terms of the dose to the brain, which we can appreciate here.
You see over on the yellow, a 2-millimeter PTV significantly will increase...Let me actually go to the next slide because it'll show it better. If you look on the left side of the screen, let's look at the yellow bar, the V12 volume. When you have a 1-millimeter margin is 14.9 CCs across 51 targets. But if we go to 2 millimeters, it's nearly twice as much, it's 70% higher at 24 CCs. And then you can look at the dose impact throughout the rest of the brain. And then we can also, now if we are a little bit more confident what we do, and we use a different margin...so the tumors that are less than 6 centimeters from the isocenter or from the center of the treatment area, we can even get more conservative or safer treatment volumes to the whole brain.
So it's great that we are improving both, you know, the timings that we decrease our leptomeningeal disease, we're improving...hopefully, gonna decrease our radiation necrosis rate by doing neoadjuvant therapy, to be supported by future studies. We're improving the efficiency of our care, the quality of our plans. But really what we need to do is track our outcomes. And although there's always a great desire to do a randomized control trial to support everything we do, there's power in numbers. And there's been a significant investment, both on the corporate side as well as from Neuropoint Alliance, which is a neurosurgery quality outcomes group, which is primarily driven by the AANS. And then, also investment from corporate with Brainlab making the biggest investment to develop a registry that in its first iteration was very good, and its next iteration is much easier for people to participate in, contribute their data to.
Now, should highlight that this software or the registry is actually available that people could use this and Brainlab can make it available to use it internally for measuring outcomes locally. Or you can actually subscribe and participate in the larger network and contribute to our national quality database so we can look at outcomes. So these are the participating centers right now. There are several throughout the country and the Neuropoint Alliance, as well as Brainlab, are still actively looking for additional sites to expand upon this because we want more data.
Currently, there are 23 sites. There's over 4000 patients that have had treatments entered and then follow-ups as well. So it takes a little bit of a strategy to get a center up and going. So you can see the number of events uploaded. It's still over 4000 and what we are focusing on is still new events, but also follow-up events. Because just knowing that someone is treated is no good if you don't know what happens to them afterwards. And the number one event or the number one type of treatment that's being deposited into the registry are metastatic tumors, which is why we're here talking about this. And again, it's because it's the number one thing that we treat with stereotactic radiosurgery.
And there are pathways, and I'm not going to go through all of it because I can't, but there are pathways that integrate, especially if you have a Brainlab system that integrate with the uploading of data into the Quentry, which is a cloud-based system for depositing the data in a HIPAA compliant manner. And then there is another interface for not just new patients, but for uploading new data, so follow-up data. And the follow-up data, the real significant advance here is that there's auto contouring, and it starts matching up previously treated tumors with newly found tumors, or basically, you can track the same tumor a long time, which is really important to know the response to treatment.
And then there's a very friendly graphical interface where you select subjects, there's a data upload through images, there's easy interface for updating information. The way we deal with this at UCLA, and I think it's going to happen more broadly, is that we use a template so that we know that we collect the same information in every single subject so that we can...it's not only good for registry but it's good for our own tracking of patient outcomes. And you can enter this data very easily. So it has both patient information and treatment information. And a lot of this can be uploaded automatically from the planning software.
And why do we do this? We do this because we want to actually analyze our outcomes. So this is the initial data that we have looking at survival rates and people treated and patients with metastases are treated with stereotactic radiosurgery. Ideally, where we want to get this to is to then compare it with other outcomes as shown there. And if we can show that we are improving what we're doing, or we have specific impact in a certain subgroup of patients, then it can drive reimbursement. And eventually, with Medicare and payers going towards quality-based reimbursements, it might be something that actually has to be required. So tracking outcomes is gonna be extremely important.
Also, internally as well as nationally, it's good for quality monitoring to make sure that looking at compliance with treatment protocols. Making sure that all the treatments steps are met, and then comparing outcomes for people for different treatment patterns to assess whether different treatment protocols results in different types of outcomes. And then we get into this field of big data analytics where registries are actually useful. Again, we all like our randomized control trials, but there's a lot of data to be gathered from real human, real-life experience.
We can...and this talk is beyond big data discussion or explanation of what big data is. But when you start getting into multiple thousands of patients and multiple time points, you can start looking for drivers of therapy and drivers of outcome, whether it be based on the location of the tumor, the type of the tumor, the timing of the tumor, the disease status, other factors that we can't even start to think about the interactions of which could drive outcomes. And then looking at, for example, what doses might actually optimize outcomes, you know, I think every center generally uses the same doses but they might be a little bit different, and looking at how that might affect outcomes as well.
And this is something I was just speaking with a colleague from USC, looking at the probability maps of where tumors might be in the brain. But this is something that could be done as well where if everyone is depositing their data into a large registry, you can then start to map them out because we have spatial information with imaging information that's all annotated and ready to look at. And last but not least is gonna be toxicity evaluation. So we don't just want to know where tumors are, how we're treating, and what the best treatment is, but how safe we are and what we can do to improve the safety of our treatments.
So I am going to stop there just to summarize, you know, I think there's a lot we can do and we need to think about the continuum of care of these patients through the radiosurgery program. Thinking about before we ever treat them from the timing of treatment, the experience of the treatment, and then tracking our outcomes. I think there's a lot of tools here that we can work with to get us closer to that ideal of improving patient outcomes.