Transcript
Thank you very much. Let me first apologize by saying that what I will show you is not a real integration of RapidArc on the Novalis Tx. What I will do is give the results on the Trilogy machine. And this is due to the fact that we've put all our efforts on implementing RapidArc on the Trilogy machine and had to delay the installation of the Novalis Tx to later this year. But still, I think that the things I'm going to show you are quite interesting, also related to radiosurgery. As you know, there's been a good collaboration now between Varian and Brainlab combination for the RapidArc and the high-definition MLC in combination with the ExacTrac system and the 6D couch.
You may have seen in the newspapers that earlier in May, three centers in U.S., Denmark, and Holland started treating patients using RapidArc technique. And RapidArc is effect IMRT in a single gantry rotation. RapidArc is a rapid treatment planning with a new optimization algorithm. Cohan Sanger [SP] from Varian already mentioned it briefly when he talked about the new developments at Varian. And in this system, simultaneously, the gantry angle, the rotation speed of the gantry, the shape of the MLC setting, and the dose rate can be modified. And the treatment is delivered in a very rapid way, typically within 90 seconds.
This is where a movie should come in. We've tried it several times. Sometimes it worked. Sometimes it didn't. No returns. Because of time, let me move on. In this movie, you would see a typical treatment of a patient RapidArc starting with the gantry on one side, and within 80 to 90 seconds moving to the other side. And as I said before, there's a continuous modulation of the dose rate, a continuous movement of the MLC shape. And that leads to a dose distribution which is at least as good as what we see with IMRT.
I know there are a few people in the audience who use tomotherapy. Maybe I should also say a few words about that and what we think are benefits of this system over tomotherapy. And it's, first of all, the fact that you can use non-coplanar beams if you want to. Another thing is that you are able to use kilovolt imaging. And the third issue is, of course, that effect that you are much more versatile with a linear accelerator where you also have excess electrons and can treat other patients, than with the tomotherapy unit. At least those are our arguments to buy the Novalis Tx and the Trilogy machine both with RapidArc System.
Okay, we can skip this one. You've probably seen. This is the same movie as has been shown on Wednesday morning. It's already stopped. It shows continuous movement of the MLC. And we move on. But maybe I should also mention here that the fast delivery also results in possibly less intrafraction motion of the patient. This morning, Dr. Wong [SP] presented. In his presentation you saw data from Princess Margaret Hospital showing that if the time during which the patient is on the couch increases, the chance that the patient moves increases considerably. And it can go up to 10 millimeters if you wait long enough. When the treatment can be delivered in a faster way, there's less chance of patient movement. Besides, it's more comfortable for the patient. And from an economical point of view, you can treat more patients pulling your accelerator. Another issue is that the number of monitor units used by standard IMRT is much higher than with RapidArc. So with RapidArc, the dose to the rest of the body is much lower.
The optimization methods is different than what's being used in standard IMRT and Eclipse. It uses initially relatively coarse points on an arc which would go even to here. It starts with a number of points and then refines it. And, initially, there are large variation in MLC setting and in the dose rate, but that decreases when points in between are being calculated.
What we did at my center is, we started with this product information about more than a year ago. We were a member of a council with four European centers and a few centers in the U.S. And every three to four weeks we received new software releases and tested it. Over that time we measured improvements and earlier this year, a clinical release has become available. And we have started early May on the Varian Trilogy machine to start patients with it. And we will do the same at the beginning of next year on the Novalis Tx.
What I will show you are some planning studies done still with a preclinical version, one of the latest preclinical versions and also dose metric validation. This is an example of a patient with a head and neck tumor. You see that with RapidArc, it's possible to get a good dose distribution. It appears to be a little better here. In our experience, the differences are not that large. You cannot say RapidArc is much better than IMRT with certain fields because, of course, it also depends on how hard do you work to get the seven-field IMRT plan optimal. What is important here is the number of monitor units. But only 30% of the number of monitor units were used with standard seven-field IMRT. An example of patients with glioma, it's possible to get a very good sparing of normal tissues in here and a high conformality.
This is another case, a patient with a glioblastoma where we used a simultaneous integrated boost technique, the gross tumor volume received 70 Gy, in this area an equivalent of 60 Gy and using a single rotation of the gantry is possible to get a very conformal dose distribution.
Here in color wash. And in this case, it's planned with a dose of 20x3 Gy to the macroscopic tumor and equivalent of 60 Gy to the whole PTV. And all of this showing again, a very good conformality.
It's also possible to treat multiple brain metastases in a single arc using RapidArc. This is a patient who has four brain metastases. We contoured them and then started planning with RapidArc. And as you can see, the blue one here is the 60% isodose and this is the 40%. In other case, dose maximum below 104%.
In some slices, the dose to the normal tissue appears to be large, but as you can see, most of it receives a fairly low dose.
Where I get really interested is when we have to combine the treatment of brain metastases during stereotactic radiotherapy with whole-brain radiotherapy. And with RapidArc, it's possible to combine these two. So in a single rotation, you give whole-brain radiotherapy and an extra dose to the metastases where you want to give a higher dose. And in our current protocol, we give 5x8 Gy to the metastases, and 5x4 to the whole brain.
Here you see the same in color wash.
We played around a little also with the 120HD collimator. On this side, you see RapidArc plan. And here five dynamic conformal arcs and one dynamic conformal arc.
I think it's too early to draw any conclusions on this issue. And I think that in the next Novalis Circle meeting, there's probably a whole session defaulted to the use of RapidArc. In our experience, we also compared the use of the Millennium MLC and the 120HD MLC and we didn't see so much difference. But we only did it in a very few cases. So this is something to be studied in the future. There's been a lot of physics work involved and I will mention Wilko Verbakel who was the physicist in my department who did most of the work. And this is work done using film measurements as well as ionization chambers. I'm not a physicist, but if you have to compare two lines or two profiles, we have to look in two directions, the dose and the distance, and you end up with a gamma value. And this gamma value should be ideally below one. Am I correct? More or less. It gets even simpler in one of the next slides. So we used two millimeters and six cGy as the tolerance values.
So here you see, excuse me, using ionization chamber measurement, what happens when the gantry rotates and delivers the dose. And when that's ready, we get this dose distribution.
On this slide, you see here, the measured dose distribution in a certain plane and here the plan. On this slide, you see, this one should start, the two lines measured and planned. And here you see the gamma value. And on this side, on this image, what is blue is okay, and it shouldn't be red or be... You shouldn't see too much red. Red indicates the gamma value higher than one. So we did this for a lot of plans. This case, the nasal pharyngeal tumor, and in general there was a very good dose distribution with gamma values below one.
Here's an example for a patient with multiple brain metastases. Again, only in one of these we saw a gamma value above one. I'll go through this a little quicker. What we noticed was the fact that Eclipse gives the impression that the dose distribution is very small, but in reality, one millimeter apart, there can be large differences. This is one millimeter before and one millimeter after the plane, before and after the plane. And here you see, in the middle, it's okay. But one before and one after, there are relatively large areas, which are red indicating a gamma value above one. Of course, it can be due to uncertainty and phantom positioning and also the highly variable spatial dose distribution which has been used.
So in fact, what we, in general, found in six plans in five coronal planes per plan an average gamma value of 0.29, and on the film measurements, the area with a gamma over one indicating poor correlation was about 1% with a maximum of 3%.
So in conclusion, RapidArc offers a very rapid treatment planning, typically less than 60 minutes for a complicated plan using this new optimization algorithm. The delivery is also very fast. For a dose of about 200 cGy, it takes about 80 to 90 seconds. There's less intrafraction motion of the patient. It's more comfortable for the patient. You could treat more patients per LINAC. And what I also think, IMRT is, at this time, at least in my department, limited to only part of the patients who might benefit from it because it takes much more time, and using RapidArc, it's probably possible to offer a technique as IMRT to a much larger percentage of patients. Last point, the number of monitor units being used with RapidArc is much lower.
And from the physics measurements, we were able to demonstrate that RapidArc accurately delivers what has been planned. And let me end then by saying that RapidArc has now successfully been introduced into clinical practice. And as you can see, we've already modified the signs for the patients which are being treated. Thank you very much
You may have seen in the newspapers that earlier in May, three centers in U.S., Denmark, and Holland started treating patients using RapidArc technique. And RapidArc is effect IMRT in a single gantry rotation. RapidArc is a rapid treatment planning with a new optimization algorithm. Cohan Sanger [SP] from Varian already mentioned it briefly when he talked about the new developments at Varian. And in this system, simultaneously, the gantry angle, the rotation speed of the gantry, the shape of the MLC setting, and the dose rate can be modified. And the treatment is delivered in a very rapid way, typically within 90 seconds.
This is where a movie should come in. We've tried it several times. Sometimes it worked. Sometimes it didn't. No returns. Because of time, let me move on. In this movie, you would see a typical treatment of a patient RapidArc starting with the gantry on one side, and within 80 to 90 seconds moving to the other side. And as I said before, there's a continuous modulation of the dose rate, a continuous movement of the MLC shape. And that leads to a dose distribution which is at least as good as what we see with IMRT.
I know there are a few people in the audience who use tomotherapy. Maybe I should also say a few words about that and what we think are benefits of this system over tomotherapy. And it's, first of all, the fact that you can use non-coplanar beams if you want to. Another thing is that you are able to use kilovolt imaging. And the third issue is, of course, that effect that you are much more versatile with a linear accelerator where you also have excess electrons and can treat other patients, than with the tomotherapy unit. At least those are our arguments to buy the Novalis Tx and the Trilogy machine both with RapidArc System.
Okay, we can skip this one. You've probably seen. This is the same movie as has been shown on Wednesday morning. It's already stopped. It shows continuous movement of the MLC. And we move on. But maybe I should also mention here that the fast delivery also results in possibly less intrafraction motion of the patient. This morning, Dr. Wong [SP] presented. In his presentation you saw data from Princess Margaret Hospital showing that if the time during which the patient is on the couch increases, the chance that the patient moves increases considerably. And it can go up to 10 millimeters if you wait long enough. When the treatment can be delivered in a faster way, there's less chance of patient movement. Besides, it's more comfortable for the patient. And from an economical point of view, you can treat more patients pulling your accelerator. Another issue is that the number of monitor units used by standard IMRT is much higher than with RapidArc. So with RapidArc, the dose to the rest of the body is much lower.
The optimization methods is different than what's being used in standard IMRT and Eclipse. It uses initially relatively coarse points on an arc which would go even to here. It starts with a number of points and then refines it. And, initially, there are large variation in MLC setting and in the dose rate, but that decreases when points in between are being calculated.
What we did at my center is, we started with this product information about more than a year ago. We were a member of a council with four European centers and a few centers in the U.S. And every three to four weeks we received new software releases and tested it. Over that time we measured improvements and earlier this year, a clinical release has become available. And we have started early May on the Varian Trilogy machine to start patients with it. And we will do the same at the beginning of next year on the Novalis Tx.
What I will show you are some planning studies done still with a preclinical version, one of the latest preclinical versions and also dose metric validation. This is an example of a patient with a head and neck tumor. You see that with RapidArc, it's possible to get a good dose distribution. It appears to be a little better here. In our experience, the differences are not that large. You cannot say RapidArc is much better than IMRT with certain fields because, of course, it also depends on how hard do you work to get the seven-field IMRT plan optimal. What is important here is the number of monitor units. But only 30% of the number of monitor units were used with standard seven-field IMRT. An example of patients with glioma, it's possible to get a very good sparing of normal tissues in here and a high conformality.
This is another case, a patient with a glioblastoma where we used a simultaneous integrated boost technique, the gross tumor volume received 70 Gy, in this area an equivalent of 60 Gy and using a single rotation of the gantry is possible to get a very conformal dose distribution.
Here in color wash. And in this case, it's planned with a dose of 20x3 Gy to the macroscopic tumor and equivalent of 60 Gy to the whole PTV. And all of this showing again, a very good conformality.
It's also possible to treat multiple brain metastases in a single arc using RapidArc. This is a patient who has four brain metastases. We contoured them and then started planning with RapidArc. And as you can see, the blue one here is the 60% isodose and this is the 40%. In other case, dose maximum below 104%.
In some slices, the dose to the normal tissue appears to be large, but as you can see, most of it receives a fairly low dose.
Where I get really interested is when we have to combine the treatment of brain metastases during stereotactic radiotherapy with whole-brain radiotherapy. And with RapidArc, it's possible to combine these two. So in a single rotation, you give whole-brain radiotherapy and an extra dose to the metastases where you want to give a higher dose. And in our current protocol, we give 5x8 Gy to the metastases, and 5x4 to the whole brain.
Here you see the same in color wash.
We played around a little also with the 120HD collimator. On this side, you see RapidArc plan. And here five dynamic conformal arcs and one dynamic conformal arc.
I think it's too early to draw any conclusions on this issue. And I think that in the next Novalis Circle meeting, there's probably a whole session defaulted to the use of RapidArc. In our experience, we also compared the use of the Millennium MLC and the 120HD MLC and we didn't see so much difference. But we only did it in a very few cases. So this is something to be studied in the future. There's been a lot of physics work involved and I will mention Wilko Verbakel who was the physicist in my department who did most of the work. And this is work done using film measurements as well as ionization chambers. I'm not a physicist, but if you have to compare two lines or two profiles, we have to look in two directions, the dose and the distance, and you end up with a gamma value. And this gamma value should be ideally below one. Am I correct? More or less. It gets even simpler in one of the next slides. So we used two millimeters and six cGy as the tolerance values.
So here you see, excuse me, using ionization chamber measurement, what happens when the gantry rotates and delivers the dose. And when that's ready, we get this dose distribution.
On this slide, you see here, the measured dose distribution in a certain plane and here the plan. On this slide, you see, this one should start, the two lines measured and planned. And here you see the gamma value. And on this side, on this image, what is blue is okay, and it shouldn't be red or be... You shouldn't see too much red. Red indicates the gamma value higher than one. So we did this for a lot of plans. This case, the nasal pharyngeal tumor, and in general there was a very good dose distribution with gamma values below one.
Here's an example for a patient with multiple brain metastases. Again, only in one of these we saw a gamma value above one. I'll go through this a little quicker. What we noticed was the fact that Eclipse gives the impression that the dose distribution is very small, but in reality, one millimeter apart, there can be large differences. This is one millimeter before and one millimeter after the plane, before and after the plane. And here you see, in the middle, it's okay. But one before and one after, there are relatively large areas, which are red indicating a gamma value above one. Of course, it can be due to uncertainty and phantom positioning and also the highly variable spatial dose distribution which has been used.
So in fact, what we, in general, found in six plans in five coronal planes per plan an average gamma value of 0.29, and on the film measurements, the area with a gamma over one indicating poor correlation was about 1% with a maximum of 3%.
So in conclusion, RapidArc offers a very rapid treatment planning, typically less than 60 minutes for a complicated plan using this new optimization algorithm. The delivery is also very fast. For a dose of about 200 cGy, it takes about 80 to 90 seconds. There's less intrafraction motion of the patient. It's more comfortable for the patient. You could treat more patients per LINAC. And what I also think, IMRT is, at this time, at least in my department, limited to only part of the patients who might benefit from it because it takes much more time, and using RapidArc, it's probably possible to offer a technique as IMRT to a much larger percentage of patients. Last point, the number of monitor units being used with RapidArc is much lower.
And from the physics measurements, we were able to demonstrate that RapidArc accurately delivers what has been planned. And let me end then by saying that RapidArc has now successfully been introduced into clinical practice. And as you can see, we've already modified the signs for the patients which are being treated. Thank you very much