Implementation of End-Expiration Breath-Hold for Liver SABR Patients Using SGRT Alone as a Motion Management Tool

Orla McKivitt (00:04):

So my name is Orla and I’m the SABR and SRS Clinical Specialist Radiation Therapist from Cork University Hospital. And I’m going to talk about our liver SRS program and how we’ve managed to implement an SGRT workflow for these patients. So I’m going to quickly go through our vision, our journey with Vision RT and how we implemented a liver SABR free breathing and an end expiration breath-hold pathway for these patients. I’ll then go through the motion management and the challenges that we have faced. So just quickly, a bit of background, I won’t go too much into this, but we went live with a line RT back in 2019. We went tattoo-free, we used it for our DIBH breasts. We went live with liver, lung SABR, then liver SABR, and then we moved on to SRS. And recently, we went live with prostate SABR.

Orla McKivitt (00:50):

In the new year, we will be doing bone nodes and adrenals as well. So when we began to discuss implementing liver SABRE in our centre, we began our meetings in 2023. And this is the team here in the photograph. So includes our radiation oncologists, our RT’s, physicists and dosimetrist. And our main concern when we began meeting was how were we were going to do our motion management. So from the literature, we know that these abdominal SRE patients should be having end expiration breath-hold, where possible, as that’s the most reproducible phase of breath-hold. So we were considering how we would implement this. We have been using Vision RT for so long. I really felt strongly that we could use this within our motion management, but there were concerns from the wider MDT, especially as we had new radiation oncologists join our department who had no experience, we were using it, and that kind of trust maybe wasn’t there yet with the system.

Orla McKivitt (01:49):

So the two systems you might be familiar with are ABC and RPM. And I’m sorry if you are a fan of ABC I am not. We had used ABC in our old center for our breast DIBH patients. And look, it works. But we found that there was a lot of patients who couldn’t tolerate the ABC system. And when we considered the abdominal SR cohort of patients, these people are in a lot of pain. They’re uncomfortable, they’ve often been refused surgery, they have had multiple taste procedures and microwave ablation. So, if we had healthy breast patients not being able to tolerate ABC, how were our abdominal SABR patients going to feel using ABC? Also with the ABC we would have to buy new systems. We all know money is tight.

Orla McKivitt (02:40):

We really wanted to get liver SABR up and running, as it was not a treatment option for patients in Ireland in a public center at the time. We were also concerned about the extra staff training, but there is a lot of publications supporting its use out there and its ability to reproduce the breath-hold amplitude. Same with RPM. We’re an elected department. But again, we didn’t look at RPM too much because of that. But there is data out there supporting the use of RPM and its ability to reproduce breath-hold amplitude. So my question was why not use AlignRT? There’s no published data. And kind of echoing what Mike said earlier, though, sometimes you just have to take a jump. We were confident in it. We knew that it worked for our breast DIBH, but we had done some lymphoma thorax patients in DIBH. So we knew that this could work. There were a lot of technical aspects we would need to consider. There would be some staff training, but then, how accurate is it going to be with these patients? We want to implement a margin of five millimeters where possible and we going to reproduce the breath-hold accurately.

Orla McKivitt (03:52):

But as I say, we were successful in using it. The training we would need for RT’s were limited and Vision RT put us in contact with the Alfred in Melbourne Australia, who had been using it for quite some time and they were really confident in using it. And they said, look guys, this works. We may not have published our data, but it does work. And they gave us a lot of tips and support when we looked into to implementing this. So the key considerations that we had to make when we were meeting to implement this were: What would our monitoring tolerances might be, what would be realistic for an end expiration breath-hold patient is one millimeter is a two millimeters. What, what’s going to be realistic on set but also what’s going to give us that accuracy, what our I is going to look like?

Orla McKivitt (04:39):

So we’ve discussed ROIs a lot here today, and I’m going to go through later what we decided on our ROIs. So the camera occlusion. So, the camera block, what’s going to look like? So you’re used to doing a DIBH patient on a breast board, which is at an angle, so you’re not getting that shadow that you would get when the patient is flat. So we were concerned as to how our treatment captures would look. Onset CT simulations, we have the CRT package as well. So how was our capture going to look in ct? Where were we going to put our tracking patch? And also, how are we going to implement a multi-phase IV contrast CT protocol in end expiration breath-hold using vision rt? It’s complicated. So that was something that we really had to do a lot of work on. Then what kind of BCTs were we going to acquire, what our capture quality was going to be?

Orla McKivitt (05:31):

And also, how are we going to implement our moves? So we know with Vision RT, when we’re implementing our moves after we’ve done our image registration, we lose that tracking of the patient’s surface. And when you’re using the hex pod system with Elekta, when you’re correcting in six degrees of freedom, it can take a considerable amount of time if you’ve got rotations near three degrees and sometimes you have to apply a vector move. So there are two different sets of motion. So we were concerned about how we would verify that the patient has maintained the breath during implementation of the moves? So this is the very simple pathway that we came up with for our patients. So just to say that we do have an index probation breath-hold pathway, but we also have a free breathing pathway. And I’ll briefly touch on our free breathing pathway as well.

Orla McKivitt (06:19):

So first of all, our motion assessment. So this is the biggest change in our pathway, right? So we’ve never done a motion assessment appointment before. And I know we’ve touched on it earlier. Some questions about how you know your patient is for DIBH? So first of all, we bring them in. This is done in the linac. So this is about a 45-minute appointment inside in the linac. We make the vac bag there and then, and then we use a generic motion assessment patient file on the SGRT terminal in the room. We take a capture of the patient in free breathing and then we coach the patient into end-expiration breath-hold. So they have to be able to hold a consistent end expiration breath-hold for 15 seconds. That is our minimum allowed time. And then once they’re coached into it and they can, we’re happy that they are maintaining it for that minimum of 15 seconds.

Orla McKivitt (07:08):

We take five KV cine fluoroscopy style images while the patient maintains end-expiration breath-hold while indicated by the SGRT system. On those fluoroscopy images, we are assessing the intra breath-hold motion and also the in breath-hold motion. So during those 15 seconds, you’re having a look to see is your diaphragm position is holding for that whole 15 seconds, but also you’re looking between each of the five breath holes is the diaphragm is reproducible to the same level. So each breath-hold, are they going to the same amplitude? So we have different tolerances for that. So the ideal is not to two millimeters but if they are three to five millimeters, we will still continue an end expiration breath-hold, but we will consider greater margins if they fail that they’ll move on to the free breathing pathway. The CT simulation workflow that we then came up with.

Orla McKivitt (08:05):

So the patient is set up as per the motion assessment. This is done the following day. We don’t do this on the same day due to patient fatigue. We use the SimRT module to take a surface capture and we place the tracking patch on the sternum. The most important thing with putting your tracking patch and placing it is that you know your patient, how they breathing? And you have to sit and watch your patient and let them relax. A patient quite often will start breathing through their chest and then change to breathing through their abdomen if they get quite anxious as they’re lying on the bed or vice versa. So understanding your patient and understanding their breathing pattern is paramount. We do a free-breathing CT first for the free-breathing surface. So that is done after contrast injection.

Orla McKivitt (08:54):

So this is the arterial phase M IV contrast that we’re doing for the free breathing scan. This is where it gets tricky. So you now want to catch the patient in the venous phase of the IV contrast, but you also need to move your couch to the start position. You need to take a new surface capture, and you need to place your tracking patch where it needs to be. You need to coach the patient into end expiration breath, hold to the correct amplitude, and you have about 25 seconds. So this, this is challenging, but the breath-hold takes precedent over everything. So sometimes you may sacrifice your scan delays in order to make sure that your end expiration breath-hold is accurate and correct. We do tend to have diagnostic CTs and sometimes MRI that we will fuse if there is problem with target delineation.

Orla McKivitt (09:42):

If the contrast scan isn’t the best quality. We do a delayed phase scan as well for our primary liver cancer patients. So that’s done four to six minutes after contrast injection. It’s then planned. I’m not going to go into that. This goes into the magical world of dosimetry and comes back out. And then we’re ready to go. So this is just an example of what you will see in your SimRT module. So just an inverted DIBH trace there. So you coach the patient to take a deep breath in, breathe out and hold. And this is an example of the surface capture that we would get in. The MRT generally, you can see that we’re missing some of the surface on the top, and that’s generally just because they’re lying flat rather than on a breast board.

Orla McKivitt (10:30):

When you have a male patient, you get less than that, but you can see here it’s perfectly fine. You can see the patches where it needs to be, and there’s enough information there. So just to touch on the free breathing pathway, so again, a similar T surface capture is taken. We put the tracking patch on the sternum, and we make sure the patient is coached into a regular breath. Where possible we’ll actually coach the patient into a shallow, regular breath. And this allows us to set tighter tolerances on the SGRT tracking. And that just makes sure that we’re minimizing that motion, and we are doing an ITV method on this planning. But it does just keep that motion down. Some patients are not eligible for coaching into a shallow regular breath and it’s safer just to let them do their own thing.

Orla McKivitt (11:17):

We do a 4DCT scan, and we do our multi-phase contrast on our 3D scan. So no offence to the physicist in the room, but on the left here, we see that this is our trace that you get from your QA. So our physics sometimes live in an Instagram world where everything is perfect and beautiful but on the left is reality. So the RT’s, so this is a real trace from a real patient, but just to highlight that, although it doesn’t look pretty, this is a perfectly usable trace, and this produces a lovely four d ct. So just to, to highlight that, sometimes it might not look, look exactly how you expect it to look, but this is, this is perfectly fine. So once everything is planned, we do a day zero for all our end-of-expiration breath-hold patients.

Orla McKivitt (12:03):

So on day zero, it’s a dummy run of the treatment. And the reason that we kept this appointment, although, you know, we capacity tighten departments and, and you’re looking for time anywhere, everywhere you can, what’s so important about keeping the day zero and the end expiration breath-hold patients is that you need to test your ROIs. So the patients generally have different ROIs depending on their body shape and depending on their breathing type. So it’s really important that you can spend that time in testing the ROI with the pre without the pressure of delivering treatment on that day. We also check where our camera block is going to be so that we can predict during our cone beam CT where we’re going to have that block and where we’re going to lose that tracking of the patient so that we can, if it does drop out, we know that it is camera block and we know to expect it, we check clearance.

Orla McKivitt (12:54):

Generally not a problem with our liver patients and it’s a good opportunity for the patients to practice end expiration breath-hold again because it’s been about two weeks since they’ve had their CT simulation and you find that sometimes they kind of have forgotten or they’ve, or they’ve got a bit too anxious about it. So it just gives them that, that opportunity. We do acquire a CBCT during the end expiration breath, hold on day zero. And that’s just to make sure that there haven’t been any significant internal changes between CT, so the customizing of our ROIs. So here are just a few examples of the different ROIs that we have. So we do go with arms up for our patients standardly but we are very quick to put arms down. As I say, the ability for the patient to reproduce their end-expiration breath-hold is paramount.

Orla McKivitt (13:43):

So we’re quite quick to put arms down when needed, and sorry to the Dosimetrists the room, I know it’s a bit harder to plan. You can see here that we’ve got one ROI that’s actually quite large and that would generally be if the patient is breathing through their belly a bit more. So you’d include a bit more of the abdomen. Some of them are just a normal thorax, ROI, but what you probably notice is that it does come down over the rib in all patients to catch that motion. Just under the ribs of where the, when the diaphragm is moving. One thing that we have found is that we sometimes have to take a little circle out of the center. If you’ve got quite a slim patient, the heartbeat and the motion there can actually be quite significant. So sometimes you have to take the ROI out there just to, to stop that from being counted into your motion.

Orla McKivitt (14:30):

So for treatment, they’re set up as per your CT simulation and we do a 3D CBCT in end expiration breath-hold. We do an image registration and six degrees of freedom, and we coach the patient’s intent and expiration breath-hold and apply all the moves in six degrees. So I mentioned earlier the challenges with applying the moves in six degrees, especially with the hex pod system. So what we found is if you have a large move to make, if your rotations are close to three degrees on any of your rotations, you’ll have to apply a vector move. So your bed will move, and you’re a lot longer and your vert first, and then you’ll do your hex pod move. So what we decided to do was to do the moves in two stages. So we first the patient into end expiration, do the first shift in the lot long in the vert, take a surface capture and give the patient a break.

Orla McKivitt (15:16):

Once the patient is ready, we then coach them back into that surface, apply the HEXA pod move and then take another surface capture. And that works quite well. Some patients are well able to hold the breath but you kind of get to know your patient and you know if they will or not. We then delivered the treatment and our first, probably more than our five patients, actually, we did post-treatment BCTs, which validated the pathway. So just to briefly touch on our free breathing pathway. So if you can see here the orange and the blue color on the CBCT here. So the orange is the liver max and the blue is the liver min. So these are contoured on our four DCT. So you have a maximum and minimum position of your liver based on your maximum inhale and exhalation.

Orla McKivitt (16:01):

So they’re contoured and exported to the CBCT system. We then take a 3D, CBCT and we apply all moves in six degrees. Our tolerance is on SGRT two millimeters and two degrees for our free-breathing patients. So once the moves are applied, we take a confirmation 4D CBCT and we deliver the treatment. But if this plays, so this is our 4D CBCT with our two millimeter tolerances. So you can see here that the diaphragm is staying quite nicely within the two lines. So you’re just ensuring that your patient breath is staying within what you’ve expected from your 4D CBCT, from your planning.

Orla McKivitt (16:41):

So one of the most important tools and one of the most important developments in the AlignRT is the deformation tool. So this allows us to really understand where the patient is breathing and you can see where the motion is actually happening. So, where this is really important is the motion assessment. So you can, we have now changed our protocol to use the deformation tool to actually identify where the patient is moving. So instead of doing it by eye, you can use the deformation tool to help you see where that’s coming from. And also then, when you’re on set, it gives you a really nice indication to if the patient’s breath is incorrect or if they have moved somewhere else, and it’s actually not their breath, that’s the problem. So if I play this here, you’ll see the patient is getting coached into taking a deep breath in and they slowly start breathing out.

Orla McKivitt (17:35):

There you go. Down comes, the bark goes green, everything goes green and you beam on. And it’s just really lovely to demonstrate here how well the system can work for end expiration breath-hold without having an invasive option for making sure they achieve that breath-hold. But it’s not been easy, it’s not that easy. It sounds easy, it’s not. We’ve had a lot of challenges and we’ve had to learn as we went. We didn’t have the benefit really of having many centres to reach out to for support. So we had to learn as we went along. So the breathing types that we, I touched on earlier, you can have a patient that starts off breathing through their chest and halfway through treatment, they decide to start breathing through their belly, be they become uncomfortable or, you know, become even more relaxed and start breathing through their chest.

Orla McKivitt (18:26):

So understanding where the motion is happening and where the breath is coming from is really important. The ROIs, we were really strict with our ROIs at the start. We thought, okay, we’re going to have a free breathing ROI, and then we’re going to have a breath-hold ROI and then we’re going to do this, this, and this, and you’re not to touch it. And that was all great until it wasn’t. So our RT’s now are really experienced in editing your ROIs to maximize your breath-hold reproducibility. Our camera block. So again, we’ve just got really used to where this generally occurs and knowing when it’s camera block and knowing when it’s the patient that’s moving. And somebody touched on earlier the different tools available in SDRT that when you can see the patient, you can see that they haven’t breathed out on your, on your cameras.

Orla McKivitt (19:16):

And then finally, the coaching training for the RT’s. And it sounds like something small, but if you have somebody coaching you through your end expiration breath-hold and suddenly there’s a different voice on the mic the next day and they’re telling you to breathe com, you know, with using different languages and different words, it can really impact the patient. Some patients really benefit from breathe in, breathe out and hold, but some patients need and hold and hold and hold and hold. You have 10 more seconds. You can do it, you can do it, you can do it. Some patients benefit from that. And being able to learn who will benefit from that and who won’t is a real skill that the RT’s need to develop. And it is very different to DIBH coaching somebody into end expiration breath-hold. It really is not the same. We also liaised with our respiratory team in the hospital and on how best to achieve end expiration breath-hold. So it’s really important to include your physios and your respiratory team and how best to get the patient to reproduce the breath-hold.

Orla McKivitt (20:17):

So we’ve successfully implemented this, it says here in 19 liver SRE patients, but we’re up to 22 since I wrote this slide. The vast majority of our liver patients SABR patients, are in end expiration breath-hold. We have had three free breathing patients who are unable to do end expiration, but it just shows with the correct coaching and time and patients that you really can get these patients into end expiration. breath-hold. We are currently running a reproducibility study for end expiration breath-hold so that we can directly compare it to

Orla McKivitt (20:46):

ABC and RPM. So we’re aiming to publish this next year, but our preliminary results show that the int factional difference in breath-hold amplitude is less than two millimeters between CBCT, which is actually slightly better than the data on ABC and RPM. So thanks so much for listening and thanks so much to all the SABR team at Cork that went with me on this. It was a hard sell at the beginning but with us working together on good collaboration, we managed to implement this pathway. Thank you.